Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Spark plasma sintering of cerium dioxide and its composites Prasad R, Anil 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2017_september_Prasad R_Anil.pdf [ 3.36MB ]
Metadata
JSON: 24-1.0355271.json
JSON-LD: 24-1.0355271-ld.json
RDF/XML (Pretty): 24-1.0355271-rdf.xml
RDF/JSON: 24-1.0355271-rdf.json
Turtle: 24-1.0355271-turtle.txt
N-Triples: 24-1.0355271-rdf-ntriples.txt
Original Record: 24-1.0355271-source.json
Full Text
24-1.0355271-fulltext.txt
Citation
24-1.0355271.ris

Full Text

Spark Plasma Sintering of CeriumDioxide and its CompositesbyAnil Prasad RB.Tech., Indian Institute of Technology Madras, India, 2015A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF APPLIED SCIENCEinThe College of Graduate Studies(Mechanical Engineering)THE UNIVERSITY OF BRITISH COLUMBIA(Okanagan)August 2017© Anil Prasad R 2017The undersigned certify that they have read, and recommend to the College of GraduateStudies for acceptance, a thesis entitled: Spark Plasma Sintering of Cerium Dioxide and itsComposites.submitted by Anil Prasad R in partial fullment of the requirements of the degree of Masterof Applied Science.Dr. Lukas Bichler, Applied Science/School of EngineeringSupervisor, ProfessorDr. Dimitri Sediako, Applied Science/School of EngineeringSupervisory Committee Member, ProfessorDr. Jian Liu, Applied Science/School of EngineeringUniversity Examiner, ProfessorDr. Kasun HewageExternal Examiner, Professor29 August 2017(Date Submitted to College of Graduate Studies)iiAbstractCerium dioxide (CeO2) is an important electroceramic material with a wide array of applica-tions in the fuel cell industry. Recently, cerium dioxide was also found to have similar ther-mophysical properties as common nuclear fuel materials such as uranium dioxide, thoriumdioxide and plutonium dioxide. Thus, it can be used as a surrogate material for simulatingthe processing of nuclear fuel materials, without the risks associated with radioactivity. How-ever, similar to other ceramic materials, processing of cerium dioxide is challenging, due toits high melting point and low ductility. Consequently, powder metallurgy techniques arewidely employed to overcome the limitations of ceramic processing. In this thesis, studies onthe processing of cerium dioxide and its composites using a novel sintering technique calledSpark Plasma Sintering (SPS) were carried out. The eect of SPS process parameters on thephysical and chemical changes of cerium dioxide were investigated.The results indicate that an optimal combination of sintering temperature, pressure andtime are required in order to fabricate high integrity CeO2 coupons. Challenges associatedwith chemical expansion and stoichiometric instabilities were observed and related to SPSprocessing conditions, as well as the electric current inherent to the SPS fabrication process.iiiPrefaceThis research thesis titled "Spark Plasma Sintering of Cerium Dioxide and its Composites"was written for the partial fulllment of the Master of Applied Science (MASc) degree at theUniversity of British Columbia.The contents of the Results and Discussion chapter presented in this thesis have beenpublished in parts in the following papers:• A. Prasad, L. Malakkal, L. Bichler and J. A. Szpunar; "Eect of Spark Plasma Sinter-ing Process Parameters on Density and Microstructure of Cerium (IV) oxide" publishedin the proceedings of the Canadian Society for Mechanical Engineering InternationalCongress 2016.• A. Prasad, L. Malakkal, L. Bichler and J. A. Szpunar; "Challenges in Spark Plasma Sin-tering of Cerium (IV) oxide", under review for publication in Ceramic Material Trans-actions 2017.Some parts of the chapter were also presented at the following conferences:• A. Prasad, L. Malakkal, L. Bichler and J. A. Szpunar; "Eect of Spark Plasma SinteringProcess Parameters on Density and Microstructure Of Cerium (IV) oxide" was presentedat Canadian Society for Mechanical Engineers (CSME) International Congress 2016,June 26-29, 2016, Kelowna, BC, Canada.• A. Prasad, L. Malakkal, L. Bichler and J. A. Szpunar; "Challenges in Spark Plasma Sin-tering of Cerium (IV) oxide" was presented at Material Science & Technology (MS&T)2016, October 23-27, 2016, Salt Lake city, Utah, U.S.A.• A. Prasad, L. Malakkal, L. Bichler and J. A. Szpunar; "Eect of particle size distributionon the sinterability of Cerium (IV) oxide using Spark Plasma Sintering" was presented ativPrefaceMaterial Science and Technology (MSAT) 2016, December 13-15, 2016, Bangkok, Thai-land.The contributions from the authors and other collaborators are listed below: All X-RayDiraction and Dierential Scanning Calorimetry experiments were carried out at the Uni-versity of Saskatchewan, Saskatoon, Canada by L. Malakkal under the supervision of Dr. J. A.Szpunar. The particle size analysis experiments were carried out at the Nanotechnology Labin Indian Institute of Technology Madras, Chennai, India by A. Prasad under the supervisionof Dr. B. S. Murty. The Spark Plasma Sintering experiments were carried out at the SPS Facil-ity at the University of British Columbia Okanagan, Kelowna, BC, Canada by A. Prasad underthe supervision of Dr. L. Bichler. All the manuscripts were written and edited by A. Prasadunder the supervision of Dr. L. Bichler.vTable of ContentsAbstract iiiPreface ivTable of Contents vList of Tables xList of Figures xiList of Equations xviList of Abbreviations xviiList of Acronyms xixAcknowledgements xxDedication xxiiChapter 1: Introduction 11.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 Research objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Chapter 2: Literature Review 62.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.2 Ceramics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.2.1 Cerium dioxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.3 Powder metallurgy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92.3.1 Spark Plasma Sintering . . . . . . . . . . . . . . . . . . . . . . . . . . . 102.4 Sintering of cerium dioxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122.5 Materials characterisation techniques . . . . . . . . . . . . . . . . . . . . . . . 14viTABLE OF CONTENTS2.5.1 Powder owablity test . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.5.2 Nelson-Riley method for precision lattice parametermeasurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.5.3 Thermal analysis techniques . . . . . . . . . . . . . . . . . . . . . . . . 15Chapter 3: Experimental Procedure 183.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183.2 Precursor powders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193.2.1 Sigma Aldrich® micro cerium dioxide . . . . . . . . . . . . . . . . . . . 193.2.2 Inframat® nano cerium dioxide . . . . . . . . . . . . . . . . . . . . . . 193.2.3 Inframat® micro cerium dioxide . . . . . . . . . . . . . . . . . . . . . . 203.2.4 Silicon carbide powder . . . . . . . . . . . . . . . . . . . . . . . . . . . 203.3 Material processing techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . 213.3.1 Powder blending . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213.3.2 Spark Plasma Sintering . . . . . . . . . . . . . . . . . . . . . . . . . . . 233.3.3 SPS program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283.3.4 Electrochemical studies . . . . . . . . . . . . . . . . . . . . . . . . . . . 293.4 Material characterisation techniques . . . . . . . . . . . . . . . . . . . . . . . . 303.4.1 Density measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303.4.2 Particle size analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313.4.3 Powder owablity test . . . . . . . . . . . . . . . . . . . . . . . . . . . 313.4.4 X-Ray Diraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333.4.5 Scanning Electron Microscopy . . . . . . . . . . . . . . . . . . . . . . . 343.4.6 Light stereo microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . 353.4.7 Dierential Scanning Calorimetry andThermogravimetric Analysis . . . . . . . . . . . . . . . . . . . . . . . . 363.5 Sample preparation techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . 36viiTABLE OF CONTENTSChapter 4: Results and Discussions 384.1 Characterisation of precursor powders . . . . . . . . . . . . . . . . . . . . . . 384.1.1 X-Ray Diraction of precursor powders . . . . . . . . . . . . . . . . . 394.1.2 Particle size analysis of precursor powders . . . . . . . . . . . . . . . . 414.1.3 Micrography and chemical analysis of precursorpowders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464.2 Densication behaviour of CeO2 during SPS . . . . . . . . . . . . . . . . . . . 534.2.1 Repeatability of displacement data . . . . . . . . . . . . . . . . . . . . 534.2.2 Eect of sintering temperature . . . . . . . . . . . . . . . . . . . . . . . 544.2.3 Eect of sintering pressure . . . . . . . . . . . . . . . . . . . . . . . . . 554.2.4 Eect of pressure ramp rate . . . . . . . . . . . . . . . . . . . . . . . . 554.2.5 Eect of sintering atmosphere . . . . . . . . . . . . . . . . . . . . . . . 564.2.6 Eect of particle size distribution . . . . . . . . . . . . . . . . . . . . . 574.3 Thermochemical stability of CeO2 . . . . . . . . . . . . . . . . . . . . . . . . . 604.3.1 Visual examination of as-sintered pellets . . . . . . . . . . . . . . . . . 604.3.2 X-Ray diraction of as-sintered CeO2 pellets . . . . . . . . . . . . . . . 654.3.3 Micrography and chemical analysis of as-sintered CeO2pellets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664.3.4 Thermal stability of pure CeO2 . . . . . . . . . . . . . . . . . . . . . . . 704.3.5 Mechanism of SPS processing of CeO2 . . . . . . . . . . . . . . . . . . 724.4 Electrochemical stability of CeO2 . . . . . . . . . . . . . . . . . . . . . . . . . . 774.5 SPS processing of SiC-CeO2 composites . . . . . . . . . . . . . . . . . . . . . . 844.5.1 Densication of SiC-CeO2 blends . . . . . . . . . . . . . . . . . . . . . 844.5.2 Characterisation of as-sintered SiC-CeO2 composites . . . . . . . . . . 854.5.3 Micrography and chemical analysis of as-sinteredSiC-CeO2 pellets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 864.5.4 Thermal analysis of SiC-CeO2 blends . . . . . . . . . . . . . . . . . . . 90viiiTABLE OF CONTENTSChapter 5: Conclusions and Future Work 925.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 925.2 Research Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 945.3 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95Bibliography 96AppendicesAppendix A 113Appendix B 117Appendix C 119ixList of Tables1.1 Energy density of fuels [1][2]. . . . . . . . . . . . . . . . . . . . . . . . . . . . 22.1 Conventional sintering of CeO2 . . . . . . . . . . . . . . . . . . . . . . . . . . . 133.1 Summary of all blends prepared in this thesis. . . . . . . . . . . . . . . . . . . 234.1 Precision lattice parameter calculation. . . . . . . . . . . . . . . . . . . . . . . 394.2 Average value of peaks in particle size distribution of precursor powders. . . . 414.3 Chemical analysis of precursor powders used for sintering. . . . . . . . . . . . 474.4 Eect of processing parameters on the densication during SPS. . . . . . . . . 544.5 Average value of angle of repose for dierent powders from the powder ow-ablity test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 574.6 Chemical analysis of SPS processed CeO2 pellets. . . . . . . . . . . . . . . . . . 81xList of Figures1.1 Flowchart illustrating scope of research. . . . . . . . . . . . . . . . . . . . . . . 52.1 SPS process schematic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112.2 An example of a DSC plot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163.1 Precursor powders used in this research: a) Sigma Aldrich® CeO2, b)Inframat®nano CeO2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203.2 Precursor powders used in this research: a) Inframat® CeO2, b) Silicon carbide. 213.3 Lab equipment used for blending powders: a) Fritsch® Pulversette 7 ball mill,b) Zirconia balls and vial. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223.4 Hot plate used for drying wet powders. . . . . . . . . . . . . . . . . . . . . . . 223.5 Thermal Technlogies® 10-3 SPS system. . . . . . . . . . . . . . . . . . . . . . . 243.6 Modules within SPS equipment: a) SPS sintering chamber with loaded die, b)Process control module. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253.7 Tooling components used for SPS process: a) Graphite die, b) Grafoil lining ofthe die, c) Grafoil disks, d) Graphite punches (20 mm and 12.5 mm). . . . . . . 263.8 Stages of SPS die preparation: a) Die ready for lling powder inside the weigh-ing balance, b) Die ready for sintering. . . . . . . . . . . . . . . . . . . . . . . . 273.9 Focusing of pyrometer on the die surface: a) Not focused, b) Focused. . . . . . 273.10 A screenshot of the iTools® software. The red box indicates the rst step ofthe program. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283.11 Spark plasma sintering die set-up for electric current insulation. . . . . . . . . 303.12 Microtrac® particle size analyser. . . . . . . . . . . . . . . . . . . . . . . . . . . 31xiLIST OF FIGURES3.13 Angle of repose test: a) Test set-up, b) Schematic for calculation of angle ofrepose. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333.14 X-Ray diraction equipment: a) Bruker® D8, b) Bragg-Brentano type dirac-togram arrangement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343.15 Tescan® Mira3 XMU eld emission Scanning Electron Microscope. . . . . . . . 353.16 Ancansco® stereo microscope. . . . . . . . . . . . . . . . . . . . . . . . . . . . 353.17 TA Instruments® DSC/TGA equipment. . . . . . . . . . . . . . . . . . . . . . . 363.18 SEM sample preparation equipment: a) Metallographic polishing machine, b)Sputter coating machine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.1 XRD patterns of precursor powders: a) CeO2, b) SiC-CeO2 blend. . . . . . . . . 394.2 Magnied XRD plot of SiC-CeO2 blends showing the variation with SiC content. 404.3 Particle size distribution for: a) Sigma Aldrich®, b) Inframat® micro CeO2 pow-der. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424.4 Particle size distribution for: a) Pure Nano powder, b) 1:3 Nano-Micro CeO2blend. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434.5 Particle size distribution for: a) 1:1, b) 3:1 Nano-Micro CeO2 blend. . . . . . . . 444.6 Particle size distribution for: a) Pure SiC powder, b) 5wt% SiC-CeO2 powder. . 454.7 Particle size distribution for: a) 10 wt%, b) 15 wt% SiC-CeO2 powder blend. . . 464.8 Scanning Electron Micrograph of: a) Sigma Aldrich®, b) Inframat® CeO2 powder. 484.9 Scanning Electron Micrograph of Inframat® nano CeO2 powder. . . . . . . . . 494.10 Scanning Electron Micrograph of 1:3 Nano-Micro CeO2 blend. . . . . . . . . . 504.11 Scanning Electron Micrograph of pure SiC powder . . . . . . . . . . . . . . . . 514.12 Elemental Map of 5 wt% SiC-CeO2 blend. . . . . . . . . . . . . . . . . . . . . . 524.13 Repeatability test on SPS data. . . . . . . . . . . . . . . . . . . . . . . . . . . . 544.14 Eect of sintering pressure on the densication of CeO2: a) Minor variationin pressure, b) Major variation in pressure. . . . . . . . . . . . . . . . . . . . . 554.15 Eect of pressure ramp rate on the densication of CeO2. . . . . . . . . . . . . 56xiiLIST OF FIGURES4.16 Eect of sintering atmosphere on the sintering of CeO2. . . . . . . . . . . . . . 564.17 Eect of particle size distribution on SPS processing of CeO2: a) Angle of re-pose for various Nano-Micro blends, b) Green density for various Nano-Microblends. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 584.18 Eect of particle size distribution on SPS processing of CeO2: a) Sinteringcurves, b) Final densities of as-sintered CeO2. . . . . . . . . . . . . . . . . . . . 594.19 As-sintered CeO2 pellets: a) 900 ◦C, 60 MPa, 5 minutes, b) 1000◦C, 60MPa, 5minutes, c) 1050 ◦C, 60 MPa, 5 minutes, d) 1150 ◦C, 65 MPa, 5 minutes, e) 1200◦C, 50 MPa, 5 minutes, f) 1500 ◦C, 60 MPa, 25 minutes. . . . . . . . . . . . . . 614.20 As-sintered CeO2 pellets: a) 1050 ◦C, 65 MPa, 5 minutes, b) 1300◦C, 50MPa, 5minutes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 624.21 As-sintered CeO2 pellets: a) 1150 ◦C, 65 MPa, 5 minutes, b) 1200◦C, 50MPa, 5minutes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 634.22 As-sintered CeO2 pellets: a) 1500 ◦C, 50 MPa, 25 minutes, b) 600◦C, 50MPa, 5minutes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 634.23 As-sintered CeO2 pellets: a) 1200 ◦C, 50 MPa, 5 minutes, b) 1200◦C, 60MPa, 10minutes, c) 1150 ◦C, 65 MPa, 2 minutes, d) 1100◦C, 50 MPa, 5 minutes. . . . . . 654.24 X-Ray Diractograms of precursor CeO2 powder and pellets sintered at vari-ous sintering temperatures of 900◦C, 1100 ◦C, 1300 ◦C, 1500◦C. . . . . . . . . . 664.25 Microstructure of as-sintered CeO2 pellets: a) 950 ◦C, b) 1050◦C, c) 1150◦C. . . 674.26 Microstructure of as-sintered CeO2 pellet at 1062 ◦C, 70 MPa, 5 minutes. . . . 684.27 Microstructure of as-sintered CeO2 pellet at 1200 ◦C,50 MPa, 5 minutes. Thecircled regions in high magnication image show defects. . . . . . . . . . . . . 694.28 Thermogravimetric analysis for pure CeO2 in Argon: a) DSC plot and b) TGAplot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 724.29 Summary of results from as-sintered pellets. . . . . . . . . . . . . . . . . . . . 73xiiiLIST OF FIGURES4.30 Fracture surface of CeO2 pellet cross section sintered at 1400 ◦C, 50MPa and5 minutes without alumina encapsulation. . . . . . . . . . . . . . . . . . . . . 774.31 Corroded edge of positive punch during SPS of CeO2 at 1400 ◦C, 50MPa and5 minutes without alumina encapsulation. . . . . . . . . . . . . . . . . . . . . 784.32 Stereo micrograph of CeO2 pellet cross section sintered at 1400 ◦C, 50MPa and5 minutes without alumina encapsulation. . . . . . . . . . . . . . . . . . . . . 784.33 CeO2 pellet sintered at 1400 ◦C, 50MPa and 5 minutes with alumina encapsu-lation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 794.34 X-ray Diractograms of CeO2 powders and SPS processed pellets for studyingeect of electric current on the chemical stablity of CeO2. . . . . . . . . . . . . 804.35 The cross section of CeO2 pellet after sintering. The red arrows represent thepossible path of oxygen ion transport. . . . . . . . . . . . . . . . . . . . . . . . 824.36 Microstructural variation across the pellet sintered without alumina encapsu-lation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 834.37 Eect of SiC content on densication: a) Densication curves of SiC-CeO2composites, b) Final displacement during SPS versus wt% SiC. . . . . . . . . . 854.38 Cross-section of: a) Pure CeO2, b) 10 wt% SiC-CeO2 composite, c) 15wt% SiC-CeO2 composite. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 854.39 X-Ray diractograms from as sintered SiC-CeO2 blends. . . . . . . . . . . . . . 864.40 Low magnication image of the 10 wt% SiC-CeO2. . . . . . . . . . . . . . . . . 874.41 Microstructure of 10 wt% SiC showing the SiC particles in: a) SE mode, b) BSEmode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 874.42 Microstructure of 10 wt% SiC showing: a) White matrix, b) Black spots at highmagnication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 884.43 X-EDS maps of 10 wt% SiC-CeO2 pellet: a) Low magnication, b) High mag-nication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 894.44 Combined a)DSC and b)TGA plot of SiC-CeO2 blends in Argon atmosphere . 90xivLIST OF FIGURESA.1 As-sintered CeO2 pellets: a) 1200◦C, 50MPa, 5minutes, b) 1500◦C, 50MPa,25minutes, c) 1200◦C, 50MPa, 5minutes, d) 1100◦C, 65MPa, 5minutes, e) 1150◦C,65MPa, 5minutes, f) 1150◦C, 65MPa, 2minutes. . . . . . . . . . . . . . . . . . . 113A.2 As-sintered CeO2 pellets: a) 1150◦C, 65MPa, 5minutes, b) 1:3 Nano-Microblend, 1200◦C, 50MPa, 5minutes, c) 1:1 Nano-Micro blend, 1200◦C, 50MPa,5minutes, d) 3:1 Nano-Micro blend, 1200◦C, 50MPa, 5minutes, e) Nano-Microblend, 1062◦C, 70MPa, 5minutes, f) Inframat® nano CeO2, 1062◦C, 70MPa,5minutes, g) 1050◦C, 70MPa, 5minutes, h) 1050◦C, 70MPa, 5minutes. . . . . . . 114A.3 As-sintered CeO2 pellets: a) 1000◦C, 50MPa, 5minutes, b) 1062◦C, 70MPa,5minutes, c) 1250◦C, 50MPa, 5minutes, d) 1062◦C, 70MPa, 5minutes, e) 1:3Nano-Micro blend, 1062◦C, 70MPa, 5minutes, f) Inframat® nano CeO2, 1062◦C,70MPa, 5minutes, g) 1500◦C, 50MPa, 25minutes, h) 1200◦C, 50MPa, 5minutes. . 115A.4 As-sintered CeO2 pellets: a) 1062◦C, 70MPa, 5minutes, b) 1200◦C, 60MPa,5minutes, c) 1200◦C, 60MPa, 10minutes, d) 1200◦C, 60MPa, 5minutes. . . . . . 116B.1 Elemental map of 10 wt% SiC-CeO2 blend. . . . . . . . . . . . . . . . . . . . . . 117B.2 Elemental map of 15 wt% SiC-CeO2 blend. . . . . . . . . . . . . . . . . . . . . . 117B.3 Microstructure of of 15 wt% SiC-CeO2 as-sintered pellet. . . . . . . . . . . . . 118B.4 Elemental map of 15 wt% SiC-CeO2 as-sintered pellet. . . . . . . . . . . . . . . 118C.1 Scanning Electron Micrograph of 1:1 Nano-Micro CeO2 blend powder. . . . . . 119xvList of Equations2.1 Nelson-Riley function for precision lattice parameter calculation. . . . . . . . . 153.1 Density measurement using Archimedes’ principle. . . . . . . . . . . . . . . . . . 303.2 Calculation of radius of a circumcircle. . . . . . . . . . . . . . . . . . . . . . . . . 323.3 Calculation of angle of repose. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33xviList of AbbreviationsCAS Chemical Abstracts ServiceCeO2 Cerium dioxide/Cerium (IV) oxideCeO2- Non stoichiometric cerium dioxide/cerium (IV) oxideCe2O3 Cerium (III) oxideDAQ Data AcquisitionDMF N,N Dimethyl FormamideDSC Dierential Scanning CalorimetryECAS Electric Current Assisted/Activated SinteringFCC Face Centered CubicIPA Isopropyl AlcoholMSDS Material Safety Data SheetPECS Pulse Electric Current SinteringPM Powder MetallurgyPuO2 Plutonium dioxiderpm rotations per minuteRS Resisitive SinteringxviiList of AbbreviationsSEM Scanning Electron MicroscopeSiC Silicon CarbideSPS Spark Plasma SinteringSOFC Solid Oxide Fuel CellTGA Thermogravimetric AnalysisThO2 Thorium dioxideUO2 Uranium dioxideXRD X-Ray DiractionX-EDS X-Ray Energy Dispersive SpectroscopyxviiiList of Acronyms Angle of repose2 Angle between incident and refracted ray in Bragg’s law Wavelength Densitya; b; c Lengths of triangle sides in angle of repose calculationW Base of cone in angle of repose calculationH Height of cone in angle of repose calculationn Order of refractioncRfunution Nelson Riley functionWair Weight in airWwatwr Weight in distilled waterxixAcknowledgementsI would like to oer my deepest gratitude to my supervisor Dr. Lukas Bichler for his supportand guidance throughout the course of this program. I acknowledge him for oering me thiswonderful opportunity which has helped me grow as a student and researcher, and I owe itto him. Thank you, for supporting every trivial idea that I had no matter what the cost was.Thank you, for letting me learn from my own mistakes. I also thank you, for the all the helpwith writing this thesis and the research papers. I am also deeply grateful for giving me theopportunities to present my work at many venues. I am sure no other supervisor would havesupported their students to this extent. I look forward to continue working with you on myPhD project.I would like to thank Dr. B.S. Murty of IIT Madras, India for inspiring me to take up thisresearch project. Also, I thank him for teaching me the fundamentals of materials engineering.In addition to this, thank you for permitting me to carry out some of my experiments at yourlab in Chennai.Any research thesis is incomplete without the support of colleagues. Thank you Ms. SomiDoja, for all the help with the experiments, writing and proof reading. I truly lost track ofall your support, throughout the duration of this project. Also, thank you for the fruitfuldiscussions that inspired many new ideas for this research work. Thank you Mr. Tyler Davis,Mr. Sebastian Lemus Fonseca, Mr. Justin Mok, Mr. Levi Lafortune and Mr. Mathew Smith forall your support. I am also thankful to the research sta at the Nanotechnology Lab in IITMadras for their support with some experiments.I would also like to thank Dr. Jerzy Szpunar and Mr. Linu Malakkal for their crucialsupport in this project. Thank you, for all the help with characterisation and analysis ofsintered samples. Also, thank you for all the inputs which helped in shaping this project toxxAcknowledgementsits current form. I look forward to your continued support for the PhD work as well.Also, I would like to thank Mr. Durwin Bossy for allowing us to use the machine shop forthe manufacture of the tooling needed for this research work, and Mr. David Arkinstall forhis expert help on the Scanning Electron Microscope.xxiDedicationTi Famcfs&Flcehdm& nhe Five fil Sccehce.xxiiChapter 1: IntroductionThis chapter introduces the topic of Spark Plasma Sintering of cerium dioxide. The current energyscenario and the research and development in the area of nuclear power are discussed to providea framework for the objectives and motivation behind this research.1.1. BackgroundWith the rising awareness about global warming and fossil fuel shortage, there has been anunprecedented need to shift towards alternative energy sources [1][2][3]. One of the optionsfor this crisis is the development of renewable energy sources, such as wind, hydro, solar andtidal energy, etc. [4]. However, these energy technologies are still in their infancy stage, andhave a low eciency and per-capita economic return [5]. Also, each of these energy sourceshas technical challenges that are yet to be fully resolved [5]. For example, the unpredictabilityof solar and wind energy due to their dependence on climatic conditions [6][7] or the environ-mental cost incurred for hydroelectric energy pose signicant challenges [8]. Further, there isa need to modify or build several new infrastructures to generate enough renewable energy toreplace fossil fuel entirely [9]. Hence, the shift from fossil-fuel energy sources to alternativeenergy sources remains challenging and not necessarily a short term solution [10]. However,the transition from fossil fuel energy sources to renewable energy sources can be gradual andthe use of nuclear energy can help in facilitating the transition [11][12][13].Nuclear energy is also a non renewable energy source [14]. However, nuclear power isa zero carbon emission energy source [13][11]. A nuclear power plant and a fossil poweruse steam to run turbines and generate electricity [15]. However, a nuclear plant uses heatgenerated from a nuclear reaction, rather than combusting fuel to generate heat [16]. Hence,the technology used for these two types of power plants are similar up to the stage of heat11.1. Backgroundexchange [15]. Thus, transitioning from fossil fuel power to nuclear power would not be aschallenging as the transition from fossil fuel to renewable energy. Though nuclear fuel is alsoexhaustible, it still has a higher energy density in comparison to any other fuel [17] as seen inTable 1.1. The energy density of uranium, thorium and plutonium is higher than the commonfuel materials including gasoline, natural gas and diesel [1][2]. Hence, the exhaustion of thesefuels would take longer than for fossil fuels.Table 1.1: Energy density of fuels [1][2].Material Energy density (MJ/kg)Wood 16Coal 24Ethanol 27Biodiesel 38Crude oil 44Diesel 45Gasoline 46Natural gas 55Uranium 80,620,000Thorium 79,420,000Plutonium 2,239,000Despite numerous advantages, nuclear energy has challenges as well. Nuclear disasterssuch as Chernobyl and Fukushima are reminders that the nuclear technology is not entirelysafe [18]. Detailed investigations revealed that the root cause for these two disasters wasoverheating of the nuclear fuel leading to a meltdown of the reactor [19][20]. In Chernobyl,the overheating was caused by the control rod failure [19]. However, in Fukushima, the melt-down was caused by the loss of coolant due to failure of water pumps [20]. There are variousways to improve the safety of nuclear power plants. From a material scientist’s perspective,21.2. Research objectivesone of the solutions is to improve the thermal conductivity of the nuclear fuel pellets to facili-tate easier and faster conduction of heat away from the fuel pellets to prevent their meltdown[21].Oxides of uranium, thorium and plutonium have very poor thermal conductivity [22][23].Recognizing this limitation, nuclear materials scientists actively research methods to improvethe thermal conductivity of nuclear fuel pellets. Given that the current technology for pro-cessing nuclear fuel powder into fuel pellets is several decades old [21], there are opportuni-ties for innovation in terms of introducing additives and changing the processing techniqueto improve thermal conductivity of the nal products [21].Research and development in this eld is controlled and relatively slow due to the safe-guards associated with handling radioactive materials. Cerium dioxide is very similar to nu-clear fuel in terms of its crystal structure and thermal conductivity, which makes ceriumdioxide a good surrogate material for studying nuclear oxides and advance the developmentof novel safer nuclear fuel [24][25]. Thus, the key motivation of this research was to usecerium dioxide as a surrogate material for experimentation on fabrication of nuclear fuels.1.2. Research objectivesThe research objectives for this thesis were governed by the need to fabricate samples for ther-mal conductivity studies (future research). Since SPS of cerium dioxide is rarely researched,the objectives for this thesis were as follows :1. Study the eect of SPS processing parameters on densication of the cerium dioxide.2. Study the eect of cerium dioxide powder morphology on its densication during SPS.3. Examine the physical and chemical state of cerium dioxide after the SPS processing.4. Process cerium dioxide-silicon carbide composite pellets using SPS and study the impactof silicon carbide content on the sintering kinetics.31.2. Research objectivesTo achieve these objectives, the research was divided into three stages as shown in Figure 1.1:1. Spark Plasma Sintering Experiments: This involved the experiments carried outwith the SPS equipment. There were two sub-parts in this stage. First, the eect ofprocessing parameters was investigated where the eect of temperature, pressure andholding time was studied. Second, the eect of powder characteristics on the sinteringbehaviour of cerium dioxide was explored.2. Characterisation: This stage of the research consisted of characterisation of the rawpowders and the sintered pellets. Common characterisation techniques such as EDS,SEM, XRD, DSC and powder testing techniques were employed.3. Mechanism and Sintering Kinetics: The analysis of all the results from the rst twoparts enabled development of a model for evaluating the sintering kinetics for ceriumdioxide during SPS.This thesis has been divided into the following chapters:1. Chapter 1: Introduction2. Chapter 2: Literature Review: This chapter provides selected literature relevant tothis thesis.3. Chapter 3: Experimental Procedure: This chapter outlines the details of all experi-ments, materials and methods used in this thesis.4. Chapter 4: Results and Discussion: This chapter contains all the results from theexperiments carried out in this research. The chapter also provides a discussion ofthese results.5. Chapter 5: Conclusions and Future Work: This chapter oers a concise summaryof generated the results. This chapter also provides suggestion for future work.41.2. Research objectivesFigure 1.1: Flowchart illustrating scope of research.5Chapter 2: Literature ReviewThis chapter provides an overview of available literature pertinent to this study. The "introduc-tion" and "ceramics" sections summarize literature on cerium dioxide and its processing. Thesection on powder metallurgy discusses sintering process and various sintering routes that arecommercially available for ceramic materials.2.1. IntroductionMaterial processing remains the milestone for evolution of humanity ever since the inceptionof tools [26]. For example, each of the prehistoric eras in the human history was named afterthe popular material used during that era (e.g., Stone Age, Iron Age and Bronze Age) [27].With the use of re, humans learned how to melt materials and mould them in a desir-able shape [28]. During the Iron Age and Bronze Age, humans invented forging and castingtechnologies [29]. The scientic highlight of these two periods was the understanding thatapplication of heat during material processing eases material shaping. This understandingforms the basis for many modern day processing techniques, including sintering of ceramics[30].Materials and their processing play a critical role in technological development. Thisthesis focuses on cerium dioxide, which is a ceramic material with versatile applications.Therefore, the following section discusses some literature relevant to the eld of processingof cerium dioxide ceramic.62.2. Ceramics2.2. CeramicsUnlike most metals, ceramics are usually found as compounds rather than being elementalmaterials. For example, common ceramics are oxides, carbides, borides and nitrides of metallicand non metallic elements. Ceramics usually have melting points as high as 4000 ◦C, whichmakes them suitable as refractory materials [31].In general, depending on their applications, ceramics can be categorised as: structuralceramics and functional ceramics. Structural ceramics are used exclusively due to their highmechanical properties and thermal stability at high temperatures [32]. These ceramics areemployed in various applications, such as cutting and grinding tools and thermal barriers.Examples include silicon carbide cutting blades [33], alumina and silica grinding wheels [34],yttria stabilized zirconia thermal barrier coatings, etc [35].Functional ceramics have gained popularity with the advent of electronic age. Since mostceramics have ionic or covalent bonding, they are also good electrical insulators [36], whichmakes them suitable for insulators in electrical and electronic applications [37]. Also, ceram-ics have good dielectric property, hence they are used as a dielectric medium in capacitors[37].Some ceramics also have unique magnetic properties. For example, ferrites exhibit ferro-magnetic property [37]. Barium titanate is used in piezoelectric applications, and can convertmechanical pressure to electrical energy, which makes it useful as a pressure sensor and ac-tuator [37].2.2.1. Cerium dioxideSolid oxide fuel cells (SOFCs) have shown a good promise as an alternative energy source[38]. High temperature stable electroceramics play a critical role in solid oxide fuel cells,since electroceramic solid state electrolytes such as zirconia [39] and cerium dioxide functionto break down fuel and generate energy [40]. This is one of the most critical applications of72.2. Ceramicscerium dioxide.Cerium dioxide is also used as a catalyst for oxidation of unburned carbon in combustionprocesses [41]. In some cases, cerium dioxide is added as a powder additive to diesel to oxi-dise harmful pollutants in automobile exhausts [42]. Also, cerium dioxide is used in oxygensensors of internal combustion engines [43].Most of the applications of cerium dioxide are due to its wide stoichiometry range [44].Cerium dioxide is classied as a non-stoichiometric oxide [45]. Non-stoichiometric oxidesexist as a single phase over a wide range of compositions [46]. For example, consider anoxide BOx; if BO is a nonstoichiometric oxide, it could exist as BOx- [46], which suggests thatBOx has the capability to remain stable with a high concentration of oxygen vacancies withinits lattice.Recent studies indicate that cerium dioxide has similar thermophysical properties as tho-rium dioxide (ThO2), uranium dioxide (UO2) and plutonium dioxide (PuO2) [24][25][47], be-cause the atomic radii of cerium atom (2.42 Å) is very close to that of thorium (2.45 Å), uranium(2.41 Å) and plutonium (2.43 Å) atoms. The thermal conductivity of a material depends onphonon transport across the material [48], which is related to the atomic vibrations and thesize of the atoms in the crystal lattice [48]. Thus, cerium dioxide is often studied as a model/-surrogate material in experiments related to ThO2, UO2 and PuO2 [24][25][47], which avoidsthe risk associated with radioactivity associated with experiments concerning these materials[49].Cerium dioxide at room temperature and normal atmospheric conditions exists as CeO2and is pale yellow in color [50]. However, at high temperatures and when exposed to highlyreductive environments, CeO2 tends to get reduced to CeO2- and turns to bluish grey color[51]. At high temperatures, oxygen ions tend to leave the vacancies in the lattice points [52].Cerium dioxide has the ability to sustain vacancies within itself, without transforming into alower oxide [53]. Some researchers reported CeO2- remained stable for  values as high as0.3 [53].82.3. Powder metallurgyThe ability of cerium dioxide to remain stable as a non-stoichiometric oxide is the reasonbehind many of its electrochemical properties [54]. At high temperatures, cerium dioxide be-haves as a solid state electrolyte with oxygen ions free to move anywhere within the material[55]. However, unlike liquid state electrolytes, atoms are not free to move randomly in solidstate electrolytes [56], where they move from a lattice position to a nearby vacancy site [57].Since this happens in a sea of identical oxygen ions, it is easier to perceive this movementas a vacancy diusion rather than the diusion of the ions itself [58]. The vacancy diusionhelps in the transport of ions from one end of the material to the other [59]. This is essentiallythe role of a solid state electrolyte in a SOFC, i.e. transportation of oxygen ions [60]. Hence,non-stoichiometric oxides are ideal for such applications due to their ability to retain theirstability even at high concentration of vacancies. Though this study does not focus on theelectrochemical properties of cerium dioxide, the non-stoichiometric nature of cerium dioxideis very important for the discussion in Chapter 4 on the chemical stablity of cerium dioxideduring the SPS processing.2.3. Powder metallurgyPowder metallurgy (PM) is a material manufacturing process where the starting material isa powder [61][62][63][64]. PM is applicable to diverse types of materials including metals,ceramics and polymers [65][66][67]. However, powder metallurgy is particularly relevant forceramics due to their high melting points and low ductility, which makes forging and castingof ceramics very challenging [68].Sintering is commonly used for the manufacture of ceramic bulk parts [69][70][71] [72].Sintering involves the application of pressure and heat during powder processing [73][74].There are several types of sintering processes, based on the technology involved [69][75].Conventional Sintering (CS) process involves a two stage process, where the powder to besintered is rst pre-pressed in a die [76][77] and then the pre-pressed green pellet is placedinside a oven and heat treated until the powder particles in the green pellet join together92.3. Powder metallurgyto form a solid pellet [76][77]. The more advanced version of this process, called Hot Iso-static Pressing (HIP) consists of lling the die and simultaneously pressing it while heating it[74][78][74][78].CS and HIP techniques have several technological limitations, such as a long processingtime and high power consumption rate [79]. Hence, there is a need for developing innovativeprocesses to overcome the limitations of conventional sintering techniques [80], which ledto research and development of several novel advanced sintering technologies [74], such asElectric Current Assisted Sintering (ECAS) techniques [81] and microwave sintering, whereinmicrowaves are used to heat the powder while simultaneously pressing it [82].2.3.1. Spark Plasma SinteringThere are several sintering processes that fall under the category of Electric Current Assist-ed/Activated Sintering (ECAS) [83], Pulsed Electric Current Assisted Sintering (PECS) [84],Resistive Sintering (RS) [85] and Flash Sintering [86].Spark Plasma Sintering (SPS) falls under the general category of Pulsed Electric CurrentSintering [87]. In SPS, the powder to be sintered is lled inside a graphite die-punch system[88]. This die-punch system is placed between hydraulic rams and electric current is passedthrough the die-punch system, as shown in Figure 2.1 [89]. The electric current heats up thedie and powder and the simultaneous pressure joins the powder particles together [90].SPS is preferred over conventional sintering processes, since it is a single step process,uses less power, and has shorter processing times [91]. It was also reported that materialsprocessed under optimal SPS conditions have densities very close to the theoretical density[88]. Hence, like other PECS processes, SPS also has very high densication rates [83]. Also,SPS is benecial for sintering nanocrystalline materials [92][93][94], because under normalsintering conditions, nanocrystalline materials experience excessive grain growth; however,the high heating rate and short processing time in SPS prevent such grain growth [95][96].There are several theories for the reason behind the high densication rate in SPS [97]. In102.3. Powder metallurgyFigure 2.1: SPS process schematic.general, the following mechanisms are the most often discussed:1. Spark [91][89][98]: According to the spark theory, there is a spark generated betweenthe powder particles during SPS. The region where this spark strikes the powder par-ticles is subjected to spark impact pressure, which also heats up the powder particlesurface. This facilitates the adhesion between powder particles.2. Plasma [99][90]: Some researchers believe that under the inuence of the electricpotential during the SPS process, the sintering atmosphere forms plasma. This plasmacleans the powder particle surfaces, and assists in the sintering process [97].3. Joule heating [100]: According to this theory, the powder particles are heated due toJoule heating. Since the powder particle surface resistance is higher than the powderbulk resistance, the interface between powder particles is subjected to very high tem-peratures. It is claimed that the powder surfaces experience temperatures as high asthe melting point of the material itself. However, a major shortcoming of this theory isthat it is valid only for electrically conductive materials. This theory still cannot explain112.4. Sintering of cerium dioxidewhy SPS works eectively for non-conductive materials.4. Electric eld enhanced diusion [89][98]: According to basic diusion theories,diusion is enhanced in the presence of an electric eld [101]. The electric eld presentin SPS is believed to enhance powder densication.SPS has been proved benecial for processing of a wide variety of materials includingceramics, metals and polymers [92][102][103][104]. There are several reports on SPS of alu-mina [105], silicon carbide [106] and zirconia [107] where SPS was demonstrated as superiorto other processing techniques.2.4. Sintering of cerium dioxideThere are only a few studies reported in published literature on the sintering of cerium diox-ide using conventional sintering (CS) techniques [108][109][50]. Table 2.1 summarises thesestudies. It is evident that in some of the experiments, the density of the as-sintered materialwas as low as 56%. For experiments where the relative density reached above 90%, the process-ing time was 3-6 hours, with the highest density that was achieved being 96% (experimentinvolved sintering temperature of 1600 ◦C and sintering pressure of 300 MPa).122.4. Sintering of cerium dioxideTable 2.1: Conventional sintering of CeO2 .Reference Technique Pressure(MPa)Temperature(◦C)Holdingtime (h)Relativedensity(%)Panhanset al.[108]Cold pressingfollowed bybakingN/A N/A N/A 80N/A N/A N/A 75Podor etal. [109]Cold pressingfollowed bybaking200 1300/1400 6 90Hammoudet al. [50]Cold pressingfollowed bybaking30 1050 4 5650 1250 0 6050 1400 0 6650 1500 0 7450 1500 1 8750 1500 3 90300 1600 3 96Mori et al. reported results from SPS of Dysprosium (Dy) doped cerium dioxide. Theauthors reported that SPS processing of cerium dioxide at 1100-1200 ◦C resulted in a pelletdensity of 85% [110]. The authors claimed that carbon diusion from graphite die hinderedthe densication of cerium dioxide. It was also suggested that a combination of SPS and CScould yield pellets of density higher than 95% [110][111].Wang et al. developed a novel SPS die design to carry out SPS processing of Gadolin-ium doped cerium dioxide (Ce0.9Gd0.1O2) at a high pressure and low temperature [112]. Thismethod was adopted from a dierent SPS study on tungsten carbide powder [113]. With thissetup, the pellet sintered at 700 ◦C and 600 MPa had a relative density of 95% [112].Choi et al. carried out a similar high pressure approach for sintering pure cerium dioxide[25] and reported that sintering at 700 ◦C and 500 MPa yielded a pellet of 98.5% density [25].The pellet sintered under these conditions had a diameter of 5 mm and thickness of 1mm [25].132.5. Materials characterisation techniquesPlapcianu et al. synthesised "fully dense" pellets of Lanthanide (Sm, Gd, Yb) doped Ceriumdioxide (Ce0.8Ln0.2O2) [95]. The processing conditions used were 1100 ◦C and 70 MPa pres-sure [95]. However, there was no discussion on the quantitative value of the density or thechemical stability of cerium dioxide during spark plasma sintering.These studies reveal that most of the research was done on doped cerium dioxide ratherthan pure cerium dioxide [95][112][110]. This was because the electrochemical applicationsof cerium dioxide involve adding dopants to the material [114]. Also, the research on purecerium dioxide was performed under high pressure conditions using specialised die set-up[25]. From a practical perspective; however there may be problems in up-scaling such a pro-cess. As a result, there is a lot of interest in carrying out sintering of pure cerium dioxide atlower pressures and higher temperatures. At this point, to the best of author’s knowledgethere is no additional literature on the SPS of cerium dioxide.2.5. Materials characterisation techniquesComprehensive and in-depth characterization of materials is equally important as materialprocessing. In general, material characterisation is categorised based on the property of ma-terial that is being characterised [115].In this research, characterisation techniques relevant to powders and solid materials wereof relevance. There are universal techniques, such as Scanning Electron Microscopy (SEM)and X-Ray diraction (XRD), which are applicable to both powder and solid samples. As perthe following non-traditional material characterisation techniques used in this research wereemployed to provide additional information on material structure and properties.2.5.1. Powder owablity testPowder owablity is a powder characteristic of critical interest to powder metallurgists [116].When considering powder synthesis and powder processing [116]. Powder particles with142.5. Materials characterisation techniquesspherical or rounded edges tend to ow easily, whereas powder particles with rough edgesare retarded by friction [117]. The powder owablity can be modied based on the powderfabrication technique [118]. For example, powder atomisation results in spherical particles,whereas powder milling results in irregular particles [119][120]. Sintering of powders relieson powder owablity. Powders that ow easily generally yield denser parts [121].2.5.2. Nelson-Riley method for precision lattice parametermeasurementNelson Riley technique is a statistical model, which is used to measure crystallographic lat-tice parameter from XRD peak positions [122]. This method needs at least three known peakpositions (). In this method, the lattice parameter calculated from Bragg’s law and is plot-ted against the Nelson Riley function shown in Equation 2.1 [122]. The intersection in theresulting plot, at cRfunution is equal to zero, is the precise lattice parameter [122]. A latticeparameter calculated using this method is precise upto 0.0001 Å [122].cRfunution 5cos2 san +cos2 (2.1)2.5.3. Thermal analysis techniquesThermal analysis techniques are employed to study the variation of some properties withtemperature [123]. In this study, since the sintering was carried out at high temperature, itis relevant to study the impact of temperature on various properties [124]. There were twotechniques that were used for this study, and are discussed in detail below. Typically, ther-mal analysis is carried out under inert atmosphere due to experimental limitations; however,there are several new techniques, which allow experiments to be carried out under reactiveatmosphere [125][126].152.5. Materials characterisation techniquesDierential Scanning CalorimetryDierential Scanning Calorimetry (DSC) involves two crucibles with individual heating ele-ments and thermocouples attached to each crucible [126]. One of the crucibles contains thesample that has to be studied and the second crucible acts as a reference [127]. The referencecrucible is either empty or lled with a known material [128]. Both the crucibles are heatedsuch a way that they both are maintained at the same temperature [128]. The dierence inpower needed to maintain this temperature is measured over the desired temperature range[126]. This power dierence is directly proportional to the dierence in specic heat betweenthe sample and the reference [129]. In most cases, the reference is empty hence any suddenvariation (peak or trough) could be due to the material undergoing a chemical or physicalchange that either absorbs or releases heat energy [127].Figure 2.2 shows a typical DSC plot, where Peak A represents an exothermic change andTrough B represents an endothermic change.Figure 2.2: An example of a DSC plot.162.5. Materials characterisation techniquesThermogravimetric analysisThermogravimetric analysis (TGA) is complimentary in a DSC instrument [130]. In contrastto DSC, TGA measures the variation of mass of the sample as a function of temperature[131]. Similar to DSC, the mass change data could be utilised to study the change the sampleundergoes with temperature change [132]. For example, a sudden drop in mass could be dueto a process similar to evaporation or sublimation [133]. A gain in mass could be related to achemical reaction with the atmosphere, such as oxidation [134].TGA and DSC have several limitations, such as just using these two techniques alonewould be insucient to study the thermal stability of the sample [126]. DSC/TGA needs to besupported with other characterisation techniques such as XRD or SEM to provide conclusiveevidence for mechanisms responsible for the chemical change [135]. Also, since DSC/TGA iscarried out under a controlled atmosphere they are not always identical to experimental orservice conditions [123].17Chapter 3: Experimental ProcedureThis chapter provides a detailed overview of all experiments conducted in this thesis. The sectionon precursor powders includes the details pertinent to all powders used in this study, followedby a section on powder processing techniques, where powder blending and sintering is described.The section on materials characterisation reports on the instruments used for characterising theraw powders and sintered materials.3.1. IntroductionAs discussed in Chapter 1, the major objective of this thesis was to process cerium dioxideand its composites using Spark Plasma Sintering. As a result, the experiments can be dividedinto three major stages:1. Sample Preparation: The sample preparation techniques include all post processingtechniques employed to prepare the powders and pellets into specimens, which wereready for characterisation2. Material Processing Techniques: This included Spark Plasma Sintering and ballmilling operations.3. Characterisation Techniques: This included techniques such as X-Ray Diraction(XRD), Scanning Electron Microscopy (SEM) and particle size analysis, which were usedto characterise the powders and sintered pellets.Before discussing the specics of each experiment, the details of the raw precursor pow-ders are described. These details are important, because the characteristics of the raw pow-ders varied based on the powder supplier. These variations were seen to aect the nal pelletcharacteristics.183.2. Precursor powders3.2. Precursor powdersLaboratory grade cerium dioxide powder was used in all experiments in order to avoid any im-pact of impurities or other additions often found in industrial-grade powders. However, thereare various laboratory grade cerium dioxide powders available commercially. Three dierentpowders of varying powder characteristics were obtained from two dierent manufacturersfor this study.3.2.1. Sigma Aldrich® micro cerium dioxideThis is one of the most commonly available (and published) laboratory grade cerium dioxidepowders. The CAS (Chemical Abstracts Service) number for this powder was 1306-38-3, andthe product identier was 22390-500G-F. The powder was claimed to be 99.5% pure on tracemetal basis.Figure 3.1a) shows Sigma Aldrich® cerium dioxide powder. The raw powder had a paleyellow coloration. This was in agreement with existing literature, which conrms that fullyoxidized ceria has a pale yellow coloration. The coloration of raw cerium dioxide is dependenton the oxygen stoichiometry, as mentioned in Section 2.2.1. Please note that this powder wasused in all sintering experiments unless specied otherwise.3.2.2. Inframat® nano cerium dioxideNano-powder (particle size <100nm [136]) of cerium dioxide powder was obtained from Inframat®.The product identier for this powder was 58N-0802. As per the Material Safety Data Sheet(MSDS), the powder was 99.97% pure with a particle size of 50-80nm. Due to the hazards re-lated to handling nanometric powders, all experiments involving handling of these powderswere carried out inside an airtight glovebox. As seen in Figure 3.1b), the nano ceria was alsoa ne pale yellow powder.193.2. Precursor powdersa bFigure 3.1: Precursor powders used in this research: a) Sigma Aldrich® CeO2, b)Inframat®nano CeO2.3.2.3. Inframat® micro cerium dioxideIn order to compare the Sigma Aldrich® powder to other commercially available powders,micron sized powder was obtained from Inframat®. This powder was claimed to be 99.95%pure with 0.1% non rare earth trace impurities. The product ID for this powder was 58R-0803.As shown in Figure 3.2a), the Inframat micro cerium dioxide was also pale yellow powder.However, the Inframat® micro cerium dioxide powder had appeared to have agglomerated incomparison to Sigma Aldrich® powder.3.2.4. Silicon carbide powderThe cerium dioxide-silicon carbide composites were made by blending the Sigma Aldrich®micro cerium dioxide and silicon carbide powders. The CAS number of the silicon carbidepowder was 409-21-2 and was obtained from Fisher Scientic®. Figure 3.2b) shows the siliconcarbide powder.203.3. Material processing techniquesa bFigure 3.2: Precursor powders used in this research: a) Inframat® CeO2, b) Silicon carbide.3.3. Material processing techniquesThe processing techniques used in this study are documented in this section. There were twoprocessing techniques that were used in this study. First, the powder blending techniques aredescribed followed by description of methods for SPS processing.3.3.1. Powder blendingThe composite powder blends were prepared in a planetary ball mill Fritsch® Pulverisette 7(Product ID: 07.5000.00, shown in Figure 3.3). Also, blends of nano and micro powders wereprepared using the ball mill to study the impact of particle size distribution on the sinterablityof cerium dioxide.213.3. Material processing techniquesa bFigure 3.3: Lab equipment used for blending powders: a) Fritsch® Pulversette 7 ball mill, b)Zirconia balls and vial.The Process Control Agent (PCA) for blending micro cerium dioxide and silicon carbideused was 1,1,1,2,3,4,4,5,5,5-Decauoropentane (Manufacturer: Sigma Aldrich®, CAS Number:138495-42-8). The major role of the PCA during any blending process was to enhance theeciency of dispersion and blending. The PCA also prevented any localised heating of thepowder particles during the blending process by distributing the heat throughout the medium.10 mm diameter zirconia balls (Manufacturer: Fritsch®, Product ID: 55.0200.27) of 10 mmdiameter were used as milling media in a Zirconia coated jar of 80 ml capacity (Manufacturer:Fritsch®, Product ID: 50.9660.00).Figure 3.4: Hot plate used for drying wet powders.223.3. Material processing techniquesThe weight of total powder being blended was limited to 10 grams. Since the objective ofthe process was only to mix the cerium dioxide and silicon carbide particles homogeneously,the processing parameters were limited to a rotational speed of 150 rpm (rotations per minute)for 2 minutes [137]. This cycle was repeated twice with a cooling time of 1.5 minutes betweenthe cycles. The direction of rotation between cycles was altered.After blending, the powder-PCA slurry was separated from the balls. This mix was driedin a fume-hood in a petri dish. The surface temperature of hotplate (refer Figure 3.4) wasmaintained at 150 ◦C for a time duration of 20 minutes.The same procedure was used for the mixing of nano and micro cerium dioxide powderblends, with the exception that the PCA used was N,N Dimethyl formamide (DMF). DMF(CAS No: 68-12-2; Product ID: BP1160-4) was found to be a good solvent for dispersing nanoparticles.Table 3.1: Summary of all blends prepared in this thesis.BlendSigmaAldrich®micro CeO2(wt%)Inframat®nano CeO2(wt%)Fisher® SiC(wt%)1:3 Nano-Micro CeO2 75 25 N/A1:1 Nano-Micro CeO2 50 50 N/A3:1 Nano-Micro CeO2 25 75 N/A5 wt% SiC-CeO2 blend N/A 95 510 wt% SiC-CeO2 blend N/A 90 1015 wt% SiC-CeO2 blend N/A 85 153.3.2. Spark Plasma SinteringThroughout this research, over 100 sintering experiments were carried out. However, due tothe limited scope of this thesis, only results from selected experiments are provided in the233.3. Material processing techniquesdocument. The SPS system used in this study was the 10-3 SPS machine manufactured byThermal Technologies®. Figure 3.5 shows the SPS machine at the SPS facility in UBCO, withits four major modules:Figure 3.5: Thermal Technlogies® 10-3 SPS system.1. Sintering chamber: The sintering was carried out inside this chamber. Figure 3.6a)shows the sintering chamber with a fully set-up die ready for sintering. The chamberwas evacuated and back-lled with Argon during sintering.2. Hydraulic system: This module includes the hydraulic pump and piston. This modulecontrols the pressure applied on the die-punch assembly.3. Vacuum system: Since the SPS process is carried out at very high temperatures, mate-rials, which are sensitive to oxidative atmospheres might be prone to oxidation. There-fore, prior to each run, the chamber was evacuated to remove any atmospheric air. This243.3. Material processing techniquesa bFigure 3.6: Modules within SPS equipment: a) SPS sintering chamber with loaded die, b)Process control module.also protected the steel chamber and rams from undesirable oxidation. The vacuum sys-tem is not shown in the gure, since it is behind the sintering chamber.4. Process control and data acquisition system: As shown in Figure 3.6b), the tem-perature and chamber pressure were controlled individually by loop control method bya Eurotherm® controller. The sintering program was compiled using iTools® softwarein a computer connected to the SPS machine and then synchonised with the processcontrol module. The Data Acquisition (DAQ) software SpecView® recorded the data(temperature, pressure, vacuum, displacement, current and voltage) during the sinter-ing process.Sintering procedureThe procedure listed below was followed sequentially for all the sintering experiments carriedout during the course of this research:1. Step 1: A graphite die, as shown in Figure 3.7, was prepared by placing a 0.005" thickgrafoil in the inner diameter of the die. The grafoil was added to reduce the frictionbetween the die and powder during the sintering process.2. Step 2: Once the grafoil was placed on the inner surface of die, the bottom punch wasinserted. This was followed by the addition of a grafoil disk in order to protect the253.3. Material processing techniquesbottom punch surface. The disc acted as a barrier between the powder and the surfaceof the puch that was in contact with the powder. Figure 3.7c,d) show the grafoil disksand graphite punches that were used for the sintering process.a bc dFigure 3.7: Tooling components used for SPS process: a) Graphite die, b) Grafoil lining ofthe die, c) Grafoil disks, d) Graphite punches (20 mm and 12.5 mm).3. Step 3: The die was placed inside an analytical balance for lling with the powder, asshown in Figure 3.8a). This enabled accurate die ll. Figure 3.8b) shows a completelylled die set-up.263.3. Material processing techniquesa bFigure 3.8: Stages of SPS die preparation: a) Die ready for lling powder inside the weighingbalance, b) Die ready for sintering.4. Step 4: Once the die was lled with the powder, the top punch was inserted into thedie with a grafoil disk in between the punch and the powder. The completely lled diewas then placed inside the SPS chamber with a carbon sleeve around it. The carbonsleeve was used to reduce the radiation heat loss at high temperatures.5. Step 5: The completely lled die was placed inside the sintering chamber. Once the diewas placed inside the chamber the pyrometer was focused at a hole on the die surface,as seen in Figure 3.9.a bFigure 3.9: Focusing of pyrometer on the die surface: a) Not focused, b) Focused.6. Step 6: Once the die was set up inside the chamber, the chamber was evacuated andlled with inert gas, and the sintering process was started.273.3. Material processing techniques3.3.3. SPS programThe SPS process was operated in automatic mode by using a software within the computerconnected to the SPS control system. This software was used to create a sintering recipe/pro-gram based on the desired sintering parameters. Once the program was started, the sinteringprocess ran uninterruptedly till completion.Figure 3.10: A screenshot of the iTools® software. The red box indicates the rst step of theprogram.Figure 3.10 shows one of the programs from the iTools® software. The program was com-pleted in six steps:1. Step 1: Ramp the temperature from room temperature to 600 ◦C at 100 ◦C/min heatingrate. Dwell the pressure at 5 MPa.2. Step 2: Dwell at 600 ◦C for 15 seconds. Dwell at 5 MPa for the same duration. This stagewas intended to give the pyrometer enough time to catch up with the actual temperatureincrease.3. Step 3: Ramp the temperature to 1200 ◦C at 100 ◦C/min. Ramp the pressure to 50 MPa283.3. Material processing techniquesat 8 MPa/min.4. Step 4: Hold the temperature and pressure under these peak conditions for 5 minutes.5. Step 5: Ramp down temperature to 20 ◦C at 200 ◦C/min and ramp down pressure to5MPa at 20 MPa/min.6. Step 6: End the program and reset the controller.3.3.4. Electrochemical studiesThis section summarises the experimental setup for a special experiment designed to study theeect of electric current on the chemical stability of cerium dioxide during SPS. The purposeof this experimentation was to explore if the electric current imposed on the CeO2 powder hadany eect on the properties of as-sintered cerium dioxide. This section compares a speciallydesigned experimental set-up for sintering cerium dioxide without electric current to a controlexperiment, which was carried out under normal SPS conditions.As shown in Figure 3.11, unlike in all the experiments performed in this thesis, the ceriumdioxide pellet in this particular experiment was insulated from coming in contact with thegraphite die. A cerium dioxide pellet of 10 mm diameter was made using cold pressing tech-nique. Cerium dioxide powder was lled in a steel die-punch setup and pressed using a hy-draulic press under a load of 2 metric tonnes. The cold pressed cerium dioxide pellet wasplaced inside a 20 mm graphite die and was encapsulated with an alumina powder. Alu-mina being an electrical insulator blocked any electrical contact between cerium dioxide andthe graphite die. An experiment was carried out at 1400 ◦C, 50 MPa pressure and 5 min-utes of holding time, in order to enable direct comparison to other experiments with non-encapsulated cerium dioxide.293.4. Material characterisation techniquesFigure 3.11: Spark plasma sintering die set-up for electric current insulation.3.4. Material characterisation techniques3.4.1. Density measurementArchimedes’ principle was used to measure the density of as-sintered pellets. The weight ofpellet in air, Wair, and the weight in water, Wwatwr, was used to calculate the density () ofthe pellet using Equation 3.1: 5WairWair −Wwatwr (3.1)Green density measurementAll green density values were calculated by taking the ratio of weight and volume of thepowder before sintering. The weight of the powder was measured before lling the die usingthe set-up shown in Figure 3.8a). The volume of the pellet was calculated using the heightof the pellet and the radius of the pellet. The height and radius of the pellet was measuredphysically by using digital callipers (Manufacturer: Mitutoyo®).303.4. Material characterisation techniques3.4.2. Particle size analysisLaser diraction technique was used for particle size analysis. The tests were carried out usingMicrotrac® particle size analyser, as shown in Figure 3.12, at the Indian Institute of TechnologyMadras, Chennai, India. These measurements were carried out in Isopropyl Alcohol (IPA)medium. The test was carried out after 120 seconds of ultrasonication of the powder in theIPA medium. Each measurement was repeated thrice.Figure 3.12: Microtrac® particle size analyser.3.4.3. Powder owablity testThe owablity of a powder was quantied by measuring the angle of repose of powder heapformed when the powder was allowed to ow freely under its own weight. The angle ofrepose was measured by following the xed funnel method. This procedure has been adaptedfrom the ASTM standard for basic angle of repose measurement [138]. A starting powderweight was xed at 100g for all the powders. The detailed procedure for angle of reposemeasurement involved the following steps:1. Step 1: A funnel was setup as shown in Figure 3.13a) above a white paper. It was set-upsuch that the tip of the funnel was 30 mm above the surface of the white sheet.313.4. Material characterisation techniques2. Step 2: The powder was let to ow freely through the funnel till the tip of the heap justtouched the bottom of the funnel neck.3. Step 3: At least three points falling in the circle at the bottom of the heap were markedon the white sheet.4. Step 4: These three points were joined to form a triangle.5. Step 5: The radius r of the circumcircle (circle formed at the bottom of heap) wascalculated using Equation 3.2:r 5abcp(a+ b+ c)(a+ b− c)(a− b+ c)(b+ c− a) (3.2)where a, b and c were the lengths of the three sides of the triangle.6. Step 6: The angle of repose was calculated using the base and height of the cone shownin Figure 3.13b). Here, the base of the cone was equal to the radius of the circumcirclecalculated from Step 6.323.4. Material characterisation techniquesa bFigure 3.13: Angle of repose test: a) Test set-up, b) Schematic for calculation of angle ofrepose.7. Step 8: The following Equation 3.3 was used to calculate the angle of repose. 5 tan−1hb(3.3)where b was the base length of the cone, h was the height of the cone and  was theangle of repose.3.4.4. X-Ray DiractionX-Ray Diraction (XRD) experiments were carried out at a facility in Univeristy of Sasketch-wan, Saskatoon. The XRD experiments were carried out using Bruker® D8 X-Ray Dirac-tometer. The radiation used for the XRD experiments was Chromium K-alpha radiation ofwavelength 2.2898 Å. A step size of 0.02° with a time per step of 100 °/s was used for all the333.4. Material characterisation techniquesscans. Figure 3.14 shows the XRD machine and the Bragg-Brentano geometric arrangementused for the XRD experiments.a bFigure 3.14: X-Ray diraction equipment: a) Bruker® D8, b) Bragg-Brentano type dirac-togram arrangement.3.4.5. Scanning Electron MicroscopyA Tescan® Mira3 XMU eld emission Scanning Electron Microscope was used for all imag-ing and chemical analysis (Figure 3.15). This microscope had an Oxford® AZtec X-MAX X-Ray Energy Dispersive Spectroscopy (X-EDS) attachment for chemical analysis. The Oxford®AZtec software was used for all post characterisation data processing.343.4. Material characterisation techniquesFigure 3.15: Tescan® Mira3 XMU eld emission Scanning Electron Microscope.3.4.6. Light stereo microscopyThe light stereo microscopy data reported in this study were obtained from a Ancansco®stereo microscope seen in Figure 3.16.Figure 3.16: Ancansco® stereo microscope.353.5. Sample preparation techniques3.4.7. Dierential Scanning Calorimetry andThermogravimetric AnalysisDierential Scanning Calorimetry was carried out at the University of Saskatchewan, Saska-toon using a TA Instruments® SD Q600 equipment, as seen in Figure 3.17. A sample of approx-imately 30 mg was used for all DSC/TGA experiments. The samples were heated to 1400°Cat a steady heating rate of 20 °C/min in an alumina crucible. The samples were allowed tofurnace cool after reaching the peak temperature without any holding time. The referenceused for the DSC/TGA measurements was an empty alumina crucibleFigure 3.17: TA Instruments® DSC/TGA equipment.3.5. Sample preparation techniquesMetallography samples were prepared by using the standard metallography practice. Roughpolishing was carried out sequentially with 120, 240, 400 and 600 grit silicon carbide polishingpapers. Following this, the nal polishing was carried out with the polishing cloths with 9, 6,3 and 1 µm diamond paste. The nal step was polishing in colloidal silica suspension. In caseswhere the sample size was too small, the samples were mounted by cold mounting technique.363.5. Sample preparation techniquesA resin-adhesive mixture was poured onto a plastic mold with the sample placed inside themold. The mounted sample was popped out of the die after 30 min of curing time.Due to the fact that the samples were ceramics, performing SEM on non-conductive sam-ples was challenging. The low electrical conductivity of the ceramic samples caused problemsin imaging (e.g. charging and streaking). Hence, before performing SEM imaging, the sam-ples were coated with a highly conductive and inert platinum, palladium, gold or rhodiumlayer. This coating was approximately 5-10 nm thick and did not aect the micro structuralfeatures of the material being observed. Figure 3.18b) shows the sputter coating machineused for preparing the samples. In case of very small and powder samples, the samples wereattached to aluminum stubs using an adhesive carbon tape.a bFigure 3.18: SEM sample preparation equipment: a) Metallographic polishing machine, b)Sputter coating machine.37Chapter 4: Results and DiscussionsThis chapter presents the results obtained from the Spark Plasma Sintering experiments and char-acterisation of cerium dioxide and its composites. The results are organised into ve sections asfollows:• 4.1 Characterisation of precursor powders: The section summarizes the results of pre-cursor powder characterisation.• 4.2 Densication behaviour of CeO2: The section focuses on the eect of various pro-cessing conditions on the density of as-sintered cerium dioxide.• 4.3 Thermochemical stability of CeO2: The results from thermal and chemical stabilityevaluation of cerium dioxide during SPS processing are discussed in this section.• 4.4 Electrochemical stability of CeO2: This section reports on the eect of electricalcurrent on the stability of cerium dioxide during SPS processing.• 4.5 Spark Plasma Sintering of SiC-CeO2 composite: The section consists of resultsfrom sintering of SiC-CeO2 composites.4.1. Characterisation of precursor powdersThe rst section of this chapter focuses on the the characterisation of precursor powdersthat were used for this study. Characterisation of these powders was carried out using vari-ous techniques, such as particle size analysis, powder owablity test, Scanning Electron Mi-croscopy and X-Ray Diraction.One of the objectives of this research was to study the eect of precursor powder charac-teristics on the physical and chemical conditions of the as-sintered pellet. Current literature384.1. Characterisation of precursor powdersclaims that powder characteristics aect the sintered pellet properties in all powder metal-lurgy techniques [139]. As was outlined in Section 3.2, three dierent commercially availablelaboratory grade cerium dioxide powders were therfore studied in detail.4.1.1. X-Ray Diraction of precursor powdersAs seen in Figure 4.1a), all three cerium dioxide powders had the same FCC (Face CenteredCubic) crystal structure. This is consistent with literature on the raw cerium dioxide powder[25][110][112]. Lattice parameters were measured for all these powders from the XRD pat-terns using the Nelson-Riley method for precision lattice parameter measurement. The latticeparameter values obtained from this analysis are summarised in Table 4.1.a bFigure 4.1: XRD patterns of precursor powders: a) CeO2, b) SiC-CeO2 blend.Table 4.1: Precision lattice parameter calculation.Powder Lattice parameter (Å)Sigma Aldrich® micro CeO2 5.4080Inframat® micro CeO2 5.4072Inframat® nano CeO2 5.4116394.1. Characterisation of precursor powdersThe silicon carbide powder has a hexagonal crystal structure as seen in Figure 4.1b), whichalso shows the XRD patterns of SiC-CeO2 blends. The XRD analysis revealed that the inten-sity of silicon carbide peaks increased with increasing weight percentage of silicon carbideadditive, as seen in the enlarged plot in Figure 4.2.Figure 4.2: Magnied XRD plot of SiC-CeO2 blends showing the variation with SiC content.According to the results in Table 4.1, it is clear that the lattice parameter values werewithin the range of± 0.005 Å of each other. The micron sized powders had a lattice parameterof 5.4080 Å and 5.4072 Å. However, the nano powder had a relatively higher lattice parametervalue of 5.4116 Å. This was probably due to the particle size eect on the lattice parametervalues. It has been shown both experimentally and theoretically that at very low crystallitesizes, the lattice parameter increases with decreasing crystallite size [140]. At lower particlesizes, the lattice opens up due to increasing surface area and surface energy, which leads to acorresponding increase in the lattice parameter.404.1. Characterisation of precursor powders4.1.2. Particle size analysis of precursor powdersThe results of particle size analysis of the three raw CeO2 powders are summarised in Table4.2. Some powders were observed to have more than one peak in the particle size distri-bution plots. The tabulated values have been obtained by averaging at least three dierentmeasurements.Table 4.2: Average value of peaks in particle size distribution of precursor powders.Powder Peak 1 (µm) Peak 2 (µm) Peak 3 (µm)Sigma Aldrich® CeO2 18.69±(:)) - -Inframat® CeO2 2.28±(:() 0.62±(:(( -Nano Inframat® CeO2 12.58±):26 5.17 2.16±(:++Nano-Micro 1:3 CeO2 blend 13.96±(:05 - -Nano-Micro 1:1 CeO2 blend 13.48±(:(5 2.22±(:() -Nano-Micro 3:1 CeO2 blend 13.95±(:60 - -Silicon carbide 12.60±(:(0 1.95±(:() 0.45 ±(:(25 wt% SiC-CeO2 blend 10.95±(:(0 - -10 wt% SiC-CeO2 blend 10.52±(:(( - -15 wt% SiC-CeO2 blend 11.25±(:(1 - -Sigma Aldrich® micro CeO2The particle size distribution of micro CeO2 powder obtained from Sigma Aldrich® is shownin Figure 4.3a). The powder particles had a normal bell shaped particle size distribution curve.The peak for this distribution was at 18.7 µm, as mentioned in Table 4.2. The blue line rep-resents the cummulative distribution, i.e. percentage of particles below a certain particle sizevalue. In the case of this powder, the curve was a simple sigmoid curve and was consistentwith the particle size distribution curve.414.1. Characterisation of precursor powdersInframat® micro CeO2The particle size distribution of the Inframat® micro CeO2 powder is shown in Figure 4.3b).Unlike the Sigma Aldrich® micro CeO2 powder, the Inframat® micro CeO2 powder had twopeaks in its particle size distribution curve. This suggests that the Inframat® micro CeO2powder consisted of particles of two dierent values. The two peaks in Inframat® micro CeO2powder particle size distribution curve were around 0.6 µm and 2.3 µm. Hence, in generalthe CeO2 particles of the Inframat® micro CeO2 powder were smaller than the particles of theSigma Aldrich® micro CeO2 powder.a bFigure 4.3: Particle size distribution for: a) Sigma Aldrich®, b) Inframat® micro CeO2 pow-der.Inframat® nano CeO2According to the MSDS (Material Safety Data Sheet) obtained from the manufacturer, the nanoCeO2 powder was expected to have particles of 80nm. However, the particle size distributioncurve, shown in Figure 4.4a), had only peaks within the micron sizes. This could be dueto agglomeration of the nano particles to form larger clusters. Apart from the two largepeaks at 2.2 µm and 12.6 µm, there was an additional peak observed at 5.17 µm in one of themeasurements. The peak at 12.6 µm had a larger standard deviation (1.3 µm) compared toany other measurement carried out in any other powder. Thus, the non-uniform particle size424.1. Characterisation of precursor powdersdistribution suggests that the particles in this nano powder were agglomerated and formedclusters of varying sizes.1:3 Nano-Micro CeO2 blendBlends of nanometric and micron particles were prepared to study the eect of particle sizedistribution on the densication of CeO2 during SPS processing. The experimental procedureused for this was outlined in Section 3.4.2. The particle size distribution of 1:3 Nano-MicroCeO2 blend is shown in Figure 4.4a). This particle size distribution had two peaks, 2 µm and13.9 µm. The larger peak might be due to the micron powder and smaller peak due to thenano powder. A better understanding of this was obtained through microscopy (discussedfurther in Section 4.1.3).a bFigure 4.4: Particle size distribution for: a) Pure Nano powder, b) 1:3 Nano-Micro CeO2blend.1:1 Nano-Micro CeO2 blendThe particle size distribution of 1:1 Nano-Micro blend powder is shown in Figure 4.4b). Thepowder blend had two peaks at 2.2 µm and 13.5 µm. The second peak was clearly visible incomparison to the larger peak of the 1:3 blend.434.1. Characterisation of precursor powders3:1 Nano-Micro CeO2 blendThe particle size distribution of 3:1 Nano-Micro powder blend is shown in Figure 4.5. Similarto 1:3 and 1:1 blends, the 3:1 blend also had two peaks: one at 2 µm and 13 µm. The rst peakwas probably due to the nano powder particles and the second peak due to the micro powderparticles.a bFigure 4.5: Particle size distribution for: a) 1:1, b) 3:1 Nano-Micro CeO2 blend.To summarise, the Nano-Micro blends had a mean particle size higher than that of nanopowder and lower than that of micron powder.Silicon carbideThe particle size distribution of pure silicon carbide is shown in Figure 4.6. There were threepeaks at 0.45 µm, 1.95 µm and 12.6 µm with standard deviations as low as 0.08 µm. However,the peak at 12.6 µm was small, suggesting that there were only a small percentage of particlesat this particle size scale.5 wt% SiC-CeO2 blendThe experimental procedure mentioned in Section 3.3.1 was used for blending cerium diox-ide and silicon carbide. All experiments were carried out using Sigma Aldrich® micro CeO2444.1. Characterisation of precursor powderspowder.The particle size curve for the 5 wt% SiC-CeO2 blend is shown in Figure 4.6. Althoughautomatic peak detection of the particle size analyzer software detected only one peak at 10.9µm, it is evident there were also two small peaks at 2.0 µm and 0.5 µm. The Sigma Aldrich®CeO2 powder had only one peak at 18.0 µm whereas pure silicon carbide had peaks at 12.6µm, 1.9 µm and 0.4 µm. Thus, the peaks at 2.0 µm and 0.5 µm likely corresponded to the SiCparticles. Whereas the peak at 10.9 µm would be a result of Sigma Aldrich® CeO2 particles.a bFigure 4.6: Particle size distribution for: a) Pure SiC powder, b) 5wt% SiC-CeO2 powder.10 wt% SiC-CeO2 blendThe 10 wt% SiC-CeO2 blend had similar particle size distribution curve as the 5 wt% SiC-CeO2.However, unlike 5 wt% SiC-CeO2, the 10 wt% SiC-CeO2 had a smaller peak at 11 µm, as seenin Figure 4.7. This suggests that the fraction of particles at 11 µm decreased, i.e. in this casethe fraction of cerium dioxide particles.15 wt% SiC-CeO2 blendThe particle size distribution curve is shown in Figure 4.7. The curve was similar to that of5wt% and 10 wt% SiC-CeO2 blends with minor variation in the peak intensities.Therefore, the results from particle size analysis revealed that the particle size distribution454.1. Characterisation of precursor powdersof blends was a mixture of particle sizes corresponding to cerium dioxide and silicon carbideratios. Also, the results reveal a modulation of peak intensity with changing wt% of siliconcarbide.a bFigure 4.7: Particle size distribution for: a) 10 wt%, b) 15 wt% SiC-CeO2 powder blend.4.1.3. Micrography and chemical analysis of precursorpowdersThe micrography and chemical analysis for all precursor powders was carried out using aScanning Electron Microscope. The results obtained from X-ray Energy Dispersive Spec-troscopy (X-EDS) chemical analysis are summarised in Table 4.3. It was evident that the threedierent CeO2 powders had a very similar oxygen composition (within the range of 3 at%).The results reveal that the stoichiometry of CeO2 was approximately 2. However, the stoi-chiometry of SiC varied from the expected number, which was possibly due to the unreliabilityof detecting carbon using the X-EDS technique.464.1. Characterisation of precursor powdersTable 4.3: Chemical analysis of precursor powders used for sintering.PowderIdenticationCerium(at%)Oxygen(at%)Silicon(at%)Carbon(at%)OxygenstoichiometrySigma Aldrich®CeO230.7 ±0.6 69.3 ±0.6 N/A N/A 2.25Inframat® CeO2 28.6 ±2.8 71.4 ±2.8 N/A N/A 2.49Inframat® nanoCeO231.4 ±5.4 68.6 ±5.4 N/A N/A 2.181:3 Nano-Micro CeO2 30.7 ±3.2 69.3 ±3.2 N/A N/A 2.251:1 Nano-Micro CeO2 33.4 ±3.8 66.6 ±3.8 N/A N/A 2.003:1 Nano-Micro CeO2 30.7 ±0.6 69.3 ±0.6 N/A N/A 2.25Silicon Carbide N/A N/A 12.0 ±3.4 88.0 ±3.4 N/A5 wt% SiC-CeO2blend 14.9 ±5.3 5.3 ±6.9 5.0 ±0.9 44.9 ±12.1 N/A10 wt% SiC-CeO2blend 8.3 ±8.4 26.4 ±8.6 3.6 ±1.6 61.6 ±17.3 N/A15 wt% SiC-CeO2blend 12.4 ±0.9 27.0 ±0.8 16.5 ±1.1 44.0 ±2.3 N/ASigma Aldrich® micro CeO2Figure 4.8a) shows the electron micrograph of the Sigma Aldrich® micro CeO2 powder. Itwas evident from the micrographs that the particles were acicular or needle shaped, and theparticles were approximately 20 µm long and 5 µm wide. This was consistent with particlesize measurements discussed in Section 4.1.2. Some studies suggest that needle shaped par-ticles have poor sinterablity [141]. Hence, the relation between the needle morphology anddensication of the pellet was explored later in this study.474.1. Characterisation of precursor powdersInframat® micro CeO2SEM image of the Inframat® micro CeO2 powder is shown in Figure 4.8b). The morphology ofthe particles was dierent from the Sigma Aldrich® powder particles. The Inframat® powderparticles had a polyhedral shape, which possibly facilitated easier sintering than the acicularshaped powder particles. This is due to the fact that such a powder has better packing fractionthan the powder with acicular shaped particles. The needle shaped particles often interlockeach other leaving closed pores; thus, reducing densication. The particles in this powderwere less than 1 µm in size. Apart from a few particles as large as 5 µm, all other particlesappeared to be less than 2 µm. These results correlate with the particle size measurementa bFigure 4.8: Scanning Electron Micrograph of: a) Sigma Aldrich®, b) Inframat® CeO2 powder.Inframat® nano CeO2The Inframat® nano CeO2 particles (refer to Figure 4.9) were approximately 100 nm in size.However, the particles appeared to be rod shaped, but unlike the Sigma Aldrich® micro CeO2particles, the Inframat® nano CeO2 particles had rounded edges and not sharp corners. Also,as discussed in Section 4.1.2, there was a clear evidence of powder agglomeration, whichwould have resulted in erratic particle size distribution values, leading to particle size values484.1. Characterisation of precursor powdersin micron range.a bFigure 4.9: Scanning Electron Micrograph of Inframat® nano CeO2 powder.Nano-Micro CeO2 BlendsScanning electron microscopy was carried out for the Nano-Micro CeO2 blends. As shownin Figure 4.10, at low magnication only the larger needle shaped CeO2 particles were vis-ible. However, at high magnication it was evident that these particles were coated withnano particles. A similar observation was made for all Nano-Micro CeO2 blends too (refer toAppendixC).494.1. Characterisation of precursor powdersa bFigure 4.10: Scanning Electron Micrograph of 1:3 Nano-Micro CeO2 blend.Silicon carbideThe composition of pure silicon carbide appeared to vary from the ideal composition (Si- 50at%, C- 50 at%). As discussed in Section 3.5, the powder samples were subjected to SEM aftersticking them on a carbon tape. Hence, when X-EDS was performed under these conditions,the carbon from the underlying carbon tape likely articially increased the carbon concen-tration recorded by the X-EDS system. This explains why the carbon composition was higherthan the expected carbon composition of the silicon carbide powder.SEM micrographs of pure silicon carbide particles are shown in Figure 4.11. The particlesappeared to have straight edges and plain facets, suggesting that each particle was probablya single crystal. There was a slight inconsistency between these micrographs and the particlesize analysis results (refer to Section 4.1.2). The laser particle analysis suggested the presenceof particles as big as 12 µm. However, such large particles were not detected in the micro-graphs even at low magnications. This suggests that there could have been agglomerationduring particle size measurements. However, the other two peaks at 2 µm and 0.5 µm inparticle size distribution could be the actual SiC particles itself.504.1. Characterisation of precursor powdersa bFigure 4.11: Scanning Electron Micrograph of pure SiC powderSiC-CeO2 blendsSiC-CeO2 blends were observed in the SEM. However, only the 5 wt% SiC-CeO2 is discussedhere. The results observed in the 5 wt% SiC-CeO2 was representative of the results observedin 10 and 15 wt% SiC-CeO2. Elemental mapping using X-Rays enabled the identication ofthe silicon carbide particles. As shown in Figure 4.12, the lower corner of the elemental maphad green particles that were rich in Silicon. Hence, it is possible that these particles wereSiC particles. Additional results can be found in Appendix B.514.1. Characterisation of precursor powdersFigure 4.12: Elemental Map of 5 wt% SiC-CeO2 blend.524.2. Densication behaviour of CeO2 during SPS4.2. Densication behaviour of CeO2 during SPSOne of the objectives of this phase of study was to understand the eect of the processingparameters on the Spark Plasma Sintering behaviour of CeO2 and its composites. The resultsof these experiments are discussed in this section. All the results and discussions from thissection are based on the data that was collected during the SPS process by the Data AcquisitionSystem. The SPS processing parameters were varied independently to study their impact onthe nal as-sintered material.4.2.1. Repeatability of displacement dataAll the discussions in this section are based on the displacement data obtained from the SPSmachine. To verify the statistical signicance of the obtained readings, three independentrepeat trials of the same experiment were carried out.The displacement measurements are plotted in Figure 4.13. It was evident that the vari-ation of displacement data was within ±0.5 mm, suggesting that any variation in the dis-placement within 0.5 mm could be considered as instrumental error. However, this analysisconrmed that the observed signicant variation in displacement (as discussed in subsequentsections) was caused by the powders being sintered, or by the SPS process parameters, ratherthan by experimental error.534.2. Densication behaviour of CeO2 during SPSFigure 4.13: Repeatability test on SPS data.4.2.2. Eect of sintering temperatureTable 4.4 summarises some of the experiments carried out to study the eect of the peaksintering temperature on the relative density of the as-sintered pellet. It was evident thatthe pellets sintered at 1000 ◦C had a lower density than the pellet sintered at 1400 ◦C and1500◦C. This is in agreement with several studies, which reported that an increase in sinteringtemperature increases the density of the nal pellet [105][106][107].Table 4.4: Eect of processing parameters on the densication during SPS.Sinteringtemperature (◦C)Sinteringpressure(MPa)HoldingTime(min)PressureRamp rate(MPa/min)Finalrelativedensity(%)Increasein densityduringdwelltime (%)1000 50 25 12 79 91400 50 25 7 92 101500 10 5 2 76 51500 50 5 7 78 41500 50 25 7 90 11544.2. Densication behaviour of CeO2 during SPS4.2.3. Eect of sintering pressureThe peak pressure during sintering also aected the relative density of the pellet, as seenin Figure 4.14a). The higher the peak pressure, the higher the relative density of the sinteredpellet. This was likely due to an increase of pressure which escalated the deformation of pow-der particles; thus, aiding in the coalescence of pores between the powder particles leadingto higher densities. However, it was also observed that minor variations in pressure such as±5MPa did not seem to aect the density signicantly, (as seen in Figure 4.14b).a bFigure 4.14: Eect of sintering pressure on the densication of CeO2: a) Minor variation inpressure, b) Major variation in pressure.4.2.4. Eect of pressure ramp rateThe rate at which the pressure was ramped was seen to aect the sintering kinetics. Figure4.15 shows three sintering curves for three dierent pressure ramp rates of 11.25 MPa/min,5.6 MPa/min and 5 MPa/min, respectively. It was evident that the higher the pressure ramprate, the higher the instantaneous relative density of the pellet.The results indicates that in the temperature range investigated, the powder particle mo-bility was relatively high, and no resistance due to external applied pressure was observed inthe temperature range investigated.554.2. Densication behaviour of CeO2 during SPSFigure 4.15: Eect of pressure ramp rate on the densication of CeO2.4.2.5. Eect of sintering atmosphereIn this research, all the experiments were carried out in Argon atmosphere unless speciedotherwise. The two other sintering atmospheres that were tried were vacuum and Helium.The sintering curves for three environments are shown in Figure 4.16. There did not appearto be any considerable dierence between the sintering curves obtained from three dierentsintering atmospheres.Figure 4.16: Eect of sintering atmosphere on the sintering of CeO2.564.2. Densication behaviour of CeO2 during SPS4.2.6. Eect of particle size distributionThe Nano-Micro CeO2 blends were created to study the eect of particle size distribution onthe sinterablity of CeO2 using SPS. Before the start of the SPS experiments, the owablity andgreen density of the pellets were measured.Angle of repose testAs discussed in Section 2.5.1, the angle of repose test was performed to quantify and comparethe owablity of the powders. The smaller the angle of repose, the more uid the powderwas. A summary of the results is provided in Table 4.5.Table 4.5: Average value of angle of repose for dierent powders from the powder ow-ablity testPowder Angle of repose (degrees)Sigma Aldrich® micro CeO2 53.12±):02Inframat® micro CeO2 35.75±):67Inframat® nano CeO2 41.97±(:67Nano-Micro 1:3 CeO2 blend 44.99±(:44Nano-Micro 1:1 CeO2 blend 44.24±2:))Nano-Micro 3:1 CeO2 blend 40.34±(:+)As seen in Figure 4.17a), the owablity of the Sigma Aldrich®micro CeO2 powder wasfound to be lower than that of the blends and the Inframat® nano CeO2 powder. This wasevident from the high angle of repose for Sigma Aldrich® micro CeO2 powder. Earlier inSection 4.1.3, it was found that the Sigma Aldrich® micro CeO2 powder particles had a needleshaped morphology. Needle shaped particles tend to interlock and cause the poor owablity.In contrast, the Inframat® nano CeO2 particles had well rounded edges, which were good for574.2. Densication behaviour of CeO2 during SPSowablity. In the case of the blends, except for 3:1 Nano-Micro blend, the addition of SigmaAldrich® microAs seen in Figure CeO2 lead to a reduction in owablity.As seen in Figure 4.17b), the green density of the Inframat® nano CeO2 powder was lowercompared to the blends and the Sigma Aldrich® micro CeO2 powder. This could be the resultof the Inframat® nano CeO2 powder having a higher surface area, which reduced the actualforce on the particle surfaces (note: the green density was measured upon applying a 5 MPapressure to the sintering die). Thus, the nanopowder was able to sustain this pressure withoutthe need to rearrange the powder particles.a bFigure 4.17: Eect of particle size distribution on SPS processing of CeO2: a) Angle of reposefor various Nano-Micro blends, b) Green density for various Nano-Micro blends.The densication curves obtained during the sintering of these powders are shown in Fig-ure 4.18. The densication curves for all powders except the nanometric CeO2 powder tracesmooth curves which were very close to each other. It was also observed that the blends hadsteeper densication rates than the pure Sigma Aldrich® micro CeO2 powder. The nanomet-ric CeO2 had steeper densication curve than the blends. This was probably because in thecase of nanometric powders, the sintering activation happened earlier than in case of regularpowders. Nano powders have a higher specic surface area (surface area per unit volume)compared to micro powders. Consequentially, they have a higher surface energy too, whichleads to lower activation energy for sintering [142].584.2. Densication behaviour of CeO2 during SPSHowever, the nal density results are not in agreement with this argument. It was evidentfrom Figure 4.18 that the Nano-Micro blends had a higher nal density than the pure nano-metric and micron powders. This could be due to the fact that for the blends the nanometricpowder lled the pores in between micron sized CeO2 particles; thus increasing packing e-ciency.a bFigure 4.18: Eect of particle size distribution on SPS processing of CeO2: a) Sinteringcurves, b) Final densities of as-sintered CeO2.594.3. Thermochemical stability of CeO24.3. Thermochemical stability of CeO2This section discusses all the results from the characterisation of the as-sintered CeO2 pellets.The rst subsection is on the characterisation of as-sintered pellets based on visual examina-tion, X-Ray Diraction and Scanning Electron Microscopy. The following subsection presentsthe results of DSC and TGA experiments. The nal subsection critically presents a proposedmechanism for SPS of CeO2.4.3.1. Visual examination of as-sintered pelletsThe following discussion focuses on representative results which were consistently observedduring repeat trials. Additional results can be found in the Appendix A.The visual examination of as-sintered CeO2 pellets played a key role in identifying severalfeatures of processing CeO2 via SPS. These visual observations had been categorised undercolor change, integrity and homogeneity of the as-sintered pellet.Color changeCeO2 pellets sintered under dierent sintering conditions had dierent coloration after sin-tering. As shown in Figure 4.19, the pellet color varied from yellowish white to dark brown-ish green. The precursor CeO2 powder, as discussed in Section 4.1.3, was yellowish whitein color. The pellet turned bluish, as was observed for pellets sintered at 900 ◦C, 1000 ◦C,1050◦C, 1150◦C and 1200 ◦C. Once the pellet has changed to dark blue, a further increase ofthe sintering temperature changed the color to brown green.604.3. Thermochemical stability of CeO2a b cd e fFigure 4.19: As-sintered CeO2 pellets: a) 900 ◦C, 60 MPa, 5 minutes, b) 1000◦C, 60MPa, 5minutes, c) 1050 ◦C, 60 MPa, 5 minutes, d) 1150 ◦C, 65 MPa, 5 minutes, e) 1200 ◦C, 50 MPa, 5minutes, f) 1500 ◦C, 60 MPa, 25 minutes.Integrity of sintered pelletMany of the sintered pellets had a tendency to fragment once ejected from the graphite die. Asshown in Figure 4.20, some pellets fractured into very ne pieces, while other pellets fracturedinto just two pieces.614.3. Thermochemical stability of CeO2a bFigure 4.20: As-sintered CeO2 pellets: a) 1050 ◦C, 65 MPa, 5 minutes, b) 1300◦C, 50MPa, 5minutes.These observations are in contrast to observations made where an increase in the sinteringtemperature resulted in pellets with higher density (Section 4.2.2). As shown in Figure 4.20,the pellet sintered at 1300 ◦C fractured into several pieces, while the pellet sintered at 1050◦C fractured to just two pieces. Hence, from this view point, the increase in the processingtemperature had a negative impact on the integrity of the as-sintered CeO2. Such an obser-vation is uncommon in SPS, since an increase in the sintering temperature typically yieldspellets with a higher density [92][105][102][103][104]. Hence, it was evident that additionalmechanisms were operative during SPS processing.Figure 4.21 shows pellets sintered at 1150 ◦C and 1200 ◦C. It was observed that thesepellets (though haven’t disintegrated completely) were at the verge of disintegration. Veryne cracks were visible on the surface of these pellets. These pellets probably represent aintermediate state between the pellet sintered at 1050 ◦C and at 1300 ◦C.624.3. Thermochemical stability of CeO2a bFigure 4.21: As-sintered CeO2 pellets: a) 1150 ◦C, 65 MPa, 5 minutes, b) 1200◦C, 50MPa, 5minutes.Figure 4.22 shows two other extreme cases, wherein the pellet sintered at 600 ◦C disinte-grated easily as powder. In contrast, the fragment of a pellet sintered at 1500 ◦C seemed toremain as a single piece. This piece also survived the cutting process which was carried outto create a specimen for measuring its thermal conductivity (seen in Figure 4.22a).a bFigure 4.22: As-sintered CeO2 pellets: a) 1500 ◦C, 50 MPa, 25 minutes, b) 600◦C, 50MPa, 5minutes.To summarize these observations, the CeO2 sintered at very low temperatures (such as 600◦C) fragmented into ne powdery pieces. However, the integrity of the pellets improved withincreasing temperature upto around 1050 ◦C. Further, increase of the temperature to around1200 ◦C enabled ne cracks to develop on the pellet surface. With further increase of thetemperature, the cracks started spreading and the pellet fractured into several small pieces.634.3. Thermochemical stability of CeO2At around 1500 ◦C, a consolidated pellet was formed again. This suggests that the integrityof CeO2 is a strong function of the sintering temperature, and an optimal range exists to yieldwell sintered high density pellets.Inhomogeneities in sintered pelletThe as-sintered pellets had visible inhomogeneities on their surface (based on color and tex-ture). Figure 4.23 shows some examples of inhomogeneities observed in samples all through-out this research. The pellets often had a dark coloration at the edges in comparison to thecenter of the pellet. Similarly, Figure 4.23 b) shows a layered inhomogeneity, where there weretwo layers of yellowish white and bluish white bands along the fracture surface of the pellet.Figure 4.23c) shows the cross-section of the pellet sintered at 1150 ◦C, 65 MPa, 2 minutes,where the top part of the pellet was whitish, while the bottom part was dark bluish. Similarobservations were made in several sintered pellets. It was also observed simultaneously thatthe darker side of the pellet was usually the side that was in contact with the negative termi-nal of SPS heating circuit. This observation was very signicant for the mechanism proposedfor the reduction of CeO2 during SPS and will be discussed further in Section 4.3.5.To understand the inhomogeneous colour of the pellets, XRD analysis was carried to ex-plore possible phase transformations during processing.644.3. Thermochemical stability of CeO2a bc dFigure 4.23: As-sintered CeO2 pellets: a) 1200 ◦C, 50 MPa, 5 minutes, b) 1200◦C, 60MPa, 10minutes, c) 1150 ◦C, 65 MPa, 2 minutes, d) 1100◦C, 50 MPa, 5 minutes.In addition to the above, it was also observed that at the end of the sintering processes thechamber pressure increased above atmospheric level, indicating that there was gas evolvingduring the SPS processing of cerium dioxide.However, analysis of outgassing was beyond thescope of this thesis.4.3.2. X-Ray diraction of as-sintered CeO2 pelletsX-Ray diraction was carried on the as-sintered pellets. Representative results are reportedin the following text.Patterns in Figure 4.24 are from SPS pellets sintered at varying sintering temperature(900 ◦C to 1500 ◦C) in 200 ◦C increments. The XRD patterns reveal that the precursor CeO2powder and pellets sintered at 900 ◦C, 1100 ◦C and 1300 ◦C retained a FCC crystal structure.In contrast, the pellet sintered at 1500 ◦C had a mixture of hexagonal phase and a FCC phase.654.3. Thermochemical stability of CeO2It was evident that this FCC phase had a lattice parameter higher than the FCC phase found inthe precursor CeO2 powder and the other pellets sintered at lower sintering temperature. Thiswas inferred from the observation that the peaks were shifted to the left [143]. The hexagonalphase was found to be Ce2O3, which is a lower oxide of Cerium. Thus, these results suggestthat CeO2 has undergone chemical reduction during SPS processing.Figure 4.24: X-Ray Diractograms of precursor CeO2 powder and pellets sintered at varioussintering temperatures of 900◦C, 1100 ◦C, 1300 ◦C, 1500◦C.4.3.3. Micrography and chemical analysis of as-sintered CeO2pelletsThe micrography of as sintered pellets was carried out using a scanning electron microscope.All the images were taken on the fracture cross-section of the pellets unless specied other-wise.Figure 4.25 shows the microstructure of the SPS pellets processed at dierent sinteringtemperatures. It was found that in the pellet sintered at 950 ◦C, the powder particles weredistinctly visible, hence proving the pellet was fully sintered.664.3. Thermochemical stability of CeO2As discussed in Section 3.2.1, the precursor CeO2 powder had a needle shaped morphology.The needles were retained in the sintered microstructure as well, as seen in Figure 4.25b).However, in contrast to the 950 ◦C pellet, the 1050 ◦C pellet had small undulations growingon its surface. This might be related to the chemical stability of CeO2 during sintering. Itwas evident from the XRD results that CeO2 undergoes reduction during SPS process. Thereduction process results in oxygen vacancies which possibly agglomerate to form porosity.a b cFigure 4.25: Microstructure of as-sintered CeO2 pellets: a) 950 ◦C, b) 1050◦C, c) 1150◦C.The microstructure of the pellet sintered at 1150 ◦C was signicantly dierent from thetwo pellets prepared at lower temperatures. This pellet had polyhedral shaped particles onits fracture surface. This is an indication that the powder particles were starting to deformand join together. There are several reports that suggest this particle geometry as the onset ofsintering [144][145][146]. At this particular stage, the needle shaped powder particles starteddeforming under the application of load from all directions and formed polyhedral particles,as seen in Figure 4.25c). Hence, the conclusion from these three micrographs was that theonset of sintering in SPS of CeO2 was around 1150 ◦C.Figure 4.26a) shows the microstructure of CeO2 pellets sintered at 1062 ◦C. This mi-crostructure also had several features that were found in the 1050 ◦C pellet. Hence, highmagnication microscopy was carried out to examine the features in detail. Figure 4.26b)shows the high magnication image of these features. It was found that these undulation-s/growth on the surface of the particles were actually pores that have formed, suggesting a674.3. Thermochemical stability of CeO2possibility of a recrystallisation mechanism which caused the formation of the pores [147].The recrystallisation mechanism may have contributed to the change in powder morphologyfrom needle shaped to the polyhedral, which continued on further increase of the tempera-ture.a bFigure 4.26: Microstructure of as-sintered CeO2 pellet at 1062 ◦C, 70 MPa, 5 minutes.Figure 4.27 shows the microstructure of the pellet sintered at 1200 ◦C. In this pellet, therewas a clear evidence that the CeO2 particles have changed their morphology from needleshaped to polyhedral. Also, in this pellet, the polyhedral particles were found to join togetheras a result of sintering. However, there were also spots where there were clear gaps in betweenthe powder particles. These "unfused grain boundary" regions are marked by the white circlesin the Figure 4.27b). These "unfused grain boundaries" results in weakening of the pellets andloss of integrity.684.3. Thermochemical stability of CeO2a bFigure 4.27: Microstructure of as-sintered CeO2 pellet at 1200 ◦C,50 MPa, 5 minutes. Thecircled regions in high magnication image show defects.Summary of observationsThe observation of the as-sintered pellets, yielded the following results:• The coloration on the surface of the pellet changed based on the sintering temperature,from yellowish white to bluish to dark blue black to greenish yellow color.• The integrity of the pellet was aected by the sintering temperature. The pellets sin-tered at low temperatures (600 ◦C) disintegrated as powdery fragments. The pellet sin-tered at 1050 ◦C had considerable integrity, whereas beyond 1150 ◦C, the pellet startedcracking again. The pellet sintered at around 1300 ◦C started to break into ne pieces.However, the pellet sintered at 1500 ◦C remained as a solid pellet.• There were some inhomogeneities found in the pellets. The most common observationwas that the pellets had a dark coloration towards the periphery. In addition to thisdiscoloration, the pellets also had darker coloration near the positive terminal of thegraphite punch than compared to the negative terminal punch.694.3. Thermochemical stability of CeO2• X-Ray Diractograms revealed that cerium dioxide remained as an FCC phase until asintering temperature of 1300 ◦C. However, at 1500 ◦C, the hexagonal phase of Ce2O3and FCC phase of CeO2- were present.• The microstructure of pellets sintered at 950 ◦C revealed needle shaped powder parti-cles, which were found in the microstructure of precursor powder.• The microstructure of pellets sintered at around 1050 ◦C also had needle shaped powderparticles; however, these particles had porosities on their surface. These pores were notfound in the precursor powder, hence they were formed during the sintering process.• The onset of sintering in SPS of cerium dioxide was around 1150 ◦C. In the pellet sinteredat 1200 ◦C, polyhedral particles seemed to join together.4.3.4. Thermal stability of pure CeO2In order to study the thermal stability of cerium dioxide powder, Dierential Scanning Calorime-try (DSC) and Thermogravimetric Analysis (TGA) were carried out for the pure cerium diox-ide powder. The results obtained from these experiments are shown in Figure 4.28. As dis-cussed in Section 3.4.7, this experiment was carried out in oxygen decient atmosphere, be-cause the main aim of this experiment was to understand the reversible stoichiometric trans-formations of cerium dioxide. The experiment was carried out for the three dierent ceriumdioxide powders available: Sigma Aldrich® micro, Inframat® micro and Inframat® nano CeO2powders.The DSC curves from the three dierent powders had the same peaks and troughs exceptfor slight dierence in the height of peaks and the depth of troughs. Since the plot is drawnas DSC (power) vs. temperature, the depth of the troughs and the height of the peaks repre-sent the energy required to complete a particular phase transformation. Since the location ofthe peaks and the troughs was the same for all three powders, the processes (e.g., evapora-tion/boiling of solvent) responsible for the peaks and troughs was likely similar for all three704.3. Thermochemical stability of CeO2powders.The DSC curve had two endothermic peaks: rst at 200 ◦C and second at 680 ◦C. The rstendothermic peak consisted of series of sub-peaks between 100 ◦C and 200 ◦C. These peakswere very shallow, possibly suggesting the evaporation of moisture or any trace organic im-purities present in the powder. Usually cerium dioxide is synthesised by the decompositionof cerium carbonate or cerium oxalate [148][149], traces of which could be retained in thepowders. Also, in some cases the cerium dioxide is made by solution state precipitation re-actions. Hence, the peaks at 200 ◦C could correspond to the degradation and evaporationof trace solvents [95]. However, since they are present in traces, these chemicals might notbe perceptible in X-ray diractograms. Hadi et al. reported the same in the thermal analy-sis experiments on cerium dioxide nanoparticles [150]. This is also consistent with the TGAexperiments carried out on cerium oxalate [151].The peak at 680 ◦C maybe due to the reduction of cerium dioxide to a lower stoichiometricoxide. However, there is no reference from literature to suggest that the peak at 680 ◦C is dueto the reduction process. Also, beyond 1300 ◦C the DSC curve dropped suddenly. This couldbe due to the phase transformation of CeO2 to Ce2O3. The drop in DSC curve suggests that thisphase transformation is possibly exothermic in nature. Hence, the energy absorbed duringthe process is the enthalpy change for the phase transformation of CeO2 to Ce2O3.The TGA curve clearly shows weight reduction all throughout the temperature range,hence it is well in agreement with the mechanisms discussed above. The initial weight reduc-tion could be due to evaporation of the precursors and the further weight reduction could bedue to loss of oxygen atoms due to chemical reduction in cerium dioxide [152].714.3. Thermochemical stability of CeO2abFigure 4.28: Thermogravimetric analysis for pure CeO2 in Argon: a) DSC plot and b) TGAplot.4.3.5. Mechanism of SPS processing of CeO2Figure 4.29 visually summarises the observations of as-sintered pellets and the DSC/TGAresults of cerium dioxide. It was established in Section 2.2.1 that cerium dioxide is a non-stoichiometric oxide and can exhibit a range of stoichiometries subject to varying temperatureand reductive atmosphere. Also, cerium dioxide is susceptible to chemical expansion, whichis caused by lattice expansion due to changes in composition [153].724.3. Thermochemical stability of CeO2Figure 4.29: Summary of results from as-sintered pellets.As mentioned in Section 3.3.2, the SPS sintering process is carried out in graphite tool-ing. It is well known that graphite being an allotrope of carbon is a very good reducing734.3. Thermochemical stability of CeO2agent [154][155][156]. In contrast, cerium dioxide at high temperatures is an oxidising agent.Hence, there is a possibility of cerium dioxide being reduced at high temperature (i.e, duringSPS processing).The change in coloration of the pellets supports the hypothesis that CeO2 undergoes re-duction. When cerium dioxide undergoes reduction, it changes its color from yellowish white(fully oxidised state) to a white blue and nally dark blue black colour [157]. A further reduc-tion of cerium dioxide enables transformation from CeO2 to Ce2O3, which results in a goldenyellow colour. These observations are consistent with literature [157]. The only contradictingobservation was that the pellet sintered at 1500 ◦C, which was greenish brown in color. Thiscan be explained by the results of XRD analysis, which suggests that the pellet sintered at1500 ◦C was a mixture of non stoichiometric CeO2- and Ce2O3. This means that this pelletwould be mixture of dark blue black and golden yellow colors, which is probably the reasonwhy the pellet appeared greenish brown in color.The observations on the integrity of the pellet can be explained by splitting the sinteringtemperature range into three stages. During the rst stage, at low temperatures (<1000 ◦C)the cerium dioxide remains as CeO2, the integrity of the pellet is governed by the density ofthe pellet. With increasing sintering temperature, the density and integrity increase at around1050 ◦C, when the pellet developed sucient strength to arrest cracking.During the second stage, cerium dioxide undergoes a chemical reduction and a simulta-neous chemical expansion. Chemical expansion may cause atomic level defects, which couldagglomerate and cause macroscopic defects [158]. These macroscopic defects in turn causethe cracking of the pellets, which would explain why the pellet started cracking only whensintered above 1100 ◦C. This mechanism also explains why the cracks started as ne hairlinecracks, which did not necessarily break the pellet apart. However, with increasing tempera-ture, the chemical expansion continued and the pellet fragmentation was accelerated.The third stage of sintering involved CeO2 transfroming to Ce2O3. During this stage, thecerium dioxide would have been reduced signicantly. Thus, converting most of the CeO2 to744.3. Thermochemical stability of CeO2Ce2O3. This is probably the reason why the pellet sintered at 1500 ◦C didn’t break part.The microstructural observations support the above mechanisms. At low temperature,the microstructure of the pellet was lled with needle shaped particles, which were also ob-served in the raw cerium dioxide powder. After increasing the sintering temperature, the rawparticles started to deform and grow/recrystallise to form polyhedral grains, which enhancedthe density. Also, during this stage, the particles were found to have pores on their surface.These pores were partly caused by oxygen vacancies, which clustered together to form pores[158]. With further increasing of the sintering temperature, the polyhedral particles startedjoining together; however, the microstructure still had cracks, probably due to chemical ex-pansion. The microstructure of the pellet sintered at 1500 ◦C appeared to be plain, with noparticles distinctly visible.The hypothesis of chemical reduction was supported by the DSC/TGA experiments aswell. The DSC plot had two endothermic peaks at 680 ◦C and 200 ◦C. It was concluded, withsupport of literature, that the peak at 200 ◦C was due to disintegration of precursor materialsused for synthesis of cerium dioxide powder. The peak at 680 ◦C was a very broad peak (400◦C to 1000 ◦C), hence it could be the peak for the endothermic reduction of CeO2 to CeO2-.There was also an exothermic peak at 1400◦C, which could be due to transformation of CeO2to Ce2O3. The TGA plot indicated an ever decreasing mass, which was due to the loss ofoxygen atoms during the heating process. The oxygen atoms that were released from ceriumdioxide might be forming oxygen gas, or could be reacting with the graphite die to formcarbon monoxide or carbon dioxide gas.Hence, it could be concluded that the chemical reduction and simultaneous chemical ex-pansion makes the Spark Plasma Sintering of pure cerium dioxide very challenging. However,one of the unique features of SPS is the application of electric current to aid in the sinteringprocess. There are several reports which study the eect of electric current in SPS on sinteredmaterials [100][159][87]. Since cerium dioxide is a high temperature ionic conductor [160],the role of the electric current in the chemical reduction of cerium dioxide was investigated.754.3. Thermochemical stability of CeO2However, for a detailed evaluation of the SPS mechanism, the reduction process shouldbe examined in atomic level. When CeO2 undergoes reduction, the unit cell of cerium dioxideundergoes change. The reduction process causes oxygen vacancies within the lattice [161].The absence of O-2 ions in CeO2- lattice causes the Ce+4 ions around the oxygen vacancy toget reduced to Ce+3. This reduction process causes the modication of various properties ofCeO2 [161]. The optical properties of Ce+4 and Ce+3 ions are dierent due to the dierence invalence shell electron conguration. In addition, Ce+4 and Ce+3 has ionic radius of 0.11 nm and0.097 nm respectively [161]. Hence, the increase in ionic radius of Cerium atom on reductioncauses an expansion of the lattice itself. This expansion process is called chemical expansionand it is very common in electroceramic materials such as cerium dioxide. This process couldcause the loss of integrity of the pellet with increasing sintering temperature, which results inincreased chemical reduction. This was evident from colour of the pellet, which had changedwith the reduction of the pellet. However, on further increase of sintering temperature theCeO2 transforms to Ce2O3 which is not susceptible to chemical expansion. Hence, the pelletsstop breaking beyond a certain temperature [161].764.4. Electrochemical stability of CeO24.4. Electrochemical stability of CeO2This section summarises a special experiment designed to study the eect of electric currenton the chemical stability of cerium dioxide during SPS. The purpose of this experimentationwas to explore if the electric current imposed on the CeO2 powder had any eect on the prop-erties of as-sintered cerium dioxide. This section compares a specially designed experimentalset-up for sintering cerium dioxide without electric current to a control experiment whichwas carried out under normal SPS conditions. The experiment was carried out at 1400 ◦C, 50MPa pressure and 5 minutes of holding time, in order to enable direct comparison to otherexperiments with non-encapsulated cerium dioxide. The temperature was xed at 1400 ◦Cbecause in case of the experiment without alumina encapsulation carried out at 1400 ◦C, somepart of the CeO2 transformed to Ce2O3.The as-sintered pellet, without alumina encapsulation, had a density of 6.56 g/cm3. Whilethe CeO2 powder had a pale yellow colour (refer to Figure 3.1) , the as-sintered pellet wasdark green-brown in colour, as seen in Figure 4.30.Figure 4.30: Fracture surface of CeO2 pellet cross section sintered at 1400 ◦C, 50MPa and 5minutes without alumina encapsulation.The CeO2 pellet without encapsulation was manually fractured in order to reveal its crosssection. Though the as-sintered pellet was solid without any visible cracks, it fractured easily.Figure 4.30 shows the fracture surface of the pellet and reveals two distinct regions. Nearthe core of the pellet, the fracture surface was porous and granular with a dark blue colour,while the remaining regions of the pellet were yellow-green with a clearly visible columnar774.4. Electrochemical stability of CeO2morphology. It is important to note that the granular region was in contact the positiveterminal of the graphite SPS tooling. The positive graphite punch appeared to be corroded atits interface, which was near CeO2, as seen in Figure 4.31.Figure 4.31: Corroded edge of positive punch during SPS of CeO2 at 1400 ◦C, 50MPa and 5minutes without alumina encapsulation.A stereo micrograph of the interface is shown in Figure 4.32. The columnar portion ofthe cross section was seen to either originate or terminate at the granular region. Moreover,the columnar part consisted of two dierent colored columnar crystals with the upper layerbeing greenish in color, while the lower layer appeared yellowish.Figure 4.32: Stereo micrograph of CeO2 pellet cross section sintered at 1400 ◦C, 50MPa and5 minutes without alumina encapsulation.The as-sintered pellet from the experiment with alumina encapsulation was dierent fromthe normal SPS pellet. It had a uniform bluish black coloration (Figure 4.33) and had a density784.4. Electrochemical stability of CeO2of 6.83 g/cm3 and did not exhibit any particular coloration or grain texture. Further, there wasno corrosion on the surface of the graphite punches used during SPS processing.Figure 4.33: CeO2 pellet sintered at 1400 ◦C, 50MPa and 5 minutes with alumina encapsula-tion.To understand the reason for the dierence in color, density and morphology of the pelletswhen the alumina encapsulation was used, XRD analysis of the pure cerium dioxide powderand the as-sintered pellets was performed. As shown in Figure 4.34, the precursor powder hadpeaks corresponding to only face-centered cubic (FCC) CeO2. However, the cross-section ofthe pellet sintered without alumina encapsulation was a mixture of FCC CeO2 and hexagonalCe2O3. The top surface of this pellet where the granular morphology was observed had onlypeaks associated with FCC CeO2, while the bottom surface contained peaks associated withFCC CeO2 and also hexagonal Ce2O3.794.4. Electrochemical stability of CeO2Figure 4.34: X-ray Diractograms of CeO2 powders and SPS processed pellets for studyingeect of electric current on the chemical stablity of CeO2.In the case of the cerium dioxide sintered with alumina encapsulation, only peaks corre-sponding to FCC CeO2 were observed. These peaks often had a shoulder or a small secondarypeak, suggesting that two phases of FCC CeO2 evolved with a slight variation of the lattice pa-rameter. Using the Nelson-Riley approach, the precision lattice parameter for the new phasewas calculated to be 0.5522 nm, while for the majority phase it was 0.5413 nm (comparableto the CeO2 powder; i.e. 0.5411 nm). As discussed in Section 2.2.1 in Chapter 2, the reductionof CeO2 may enable the formation of CeO2- with a higher lattice parameter. This is a clearproof of chemical reduction happening within the material during SPS processing. Therefore,the results conrmed that the encapsulated pellet was a mixture of CeO2 and CeO2-, whilethe normal pellet was a mixture of Ce2O3 and CeO2-.One possible reason behind the compositional dissimilarity between pellets sintered withand without alumina encapsulation can be attributed to the oxygen vacancy mobility andthe concentration gradients in these samples. It has been reported that cerium dioxide has amoderate stoichiometry range within which its FCC crystal structure is stable, but when thestoichiometry drops beyond a certain value, cerium dioxide becomes unstable and transforms804.4. Electrochemical stability of CeO2to hexagonal Ce2O3 [162]. The limiting value was reported to be 1.5 and relates to the oxygenstoichiometry of Ce2O3, i.e. 1.5. Thus, the sample without alumina encapsulation had moreoxygen vacancies than the encapsulated specimen. Cerium dioxide was shown to act as areservoir of mobile oxygen ions at high-temperatures [54]. In this state, the oxygen ions tendto migrate towards the positive terminal (in this case the graphite punch connected to thepositive terminal of the SPS machine). In the sample with encapsulation, the passage of ionswas interrupted, hence reduction of cerium dioxide was hindered. However, in this case, theow of oxygen ions might not be a direct result of the ow of electric current. Instead, theelectric eld across the pellet enabled the movement of oxygen ions. As a result, cutting oelectrical contact didn’t necessary cease the ion movement, but hindered it to a certain extent.Consequently, the cerium dioxide was reduced despite encapsulating in alumina powder.This conjecture is validated by the chemical analysis performed on the specimens. Asseen in Table 4.6, the precursor cerium dioxide powder had oxygen stoichiometry of 2.8. Theoverall stoichiometry of the pellet without alumina encapsulation was 1.5; thus, confrmingthat a reduction occurred during the sintering process. However, the alumina encapsulatedpellet had a stoichiometry of 1.8, suggesting that the chemical reduction occurred, but to alesser extent.Table 4.6: Chemical analysis of SPS processed CeO2 pellets.PureceriumdioxidepowderCross-Sectionof pelletwithout en-capsulationTop surfaceof pelletwithout en-capsulationBottomsurfaceof pelletwithout en-capsulationCross-Section ofpellet withencapsula-tionCerium (at%) 69.5 ±0.7 60.6 ±1.1 65.0 ±1.2 55.5 ±0.5 64.3 ±0.3Oxygen (at%) 30.5 ±0.7 39.4 ±1.1 35.0 ±1.2 44.5 ±0.5 35.7 ±0.3Oxygen Stoi-chiometry 2.8 1.53 1.85 1.25 1.80814.4. Electrochemical stability of CeO2This hypothesis is also supported by the observed corrosion of the positive terminal of thegraphite punch, which likely resulted from oxygen ions moving towards it and oxidising it toform carbon dioxide/carbon monoxide. Formation of CO or CO2 would explain the reductionof oxygen composition in the nal pellets, as measured by SEM X-EDS and XRD.Table 4.6 summarises the chemical analysis on the samples with and without aluminaencapsulation. It is evident that the top region (region "A" in Figure 4.35) of the pellet with-out alumina encapsulation had higher oxygen concentration than the bottom region (region"C" in Figure 4.35). This suggests that the reduction happened from the negative terminaltowards the positive terminal. During SPS processing, the negatively charged oxygen ionswere attracted towards the positive punch and were repelled by the negative punch. Hence,this caused mass migration of oxygen ions towards the positive punch. This probably causedthe higher concentration of oxygen ions near the positive terminal. Thus, with a higher oxy-gen concentration, this region was still CeO2 and retained the blue-black color. However,the bottom region was exhausted of oxygen ions and transformed into Ce2O3, with a colourchange towards yellowish. The middle region (region "B" in Figure 4.35) was a mixture ofboth stoichiometries leading to its greenish appearance.Figure 4.35: The cross section of CeO2 pellet after sintering. The red arrows represent thepossible path of oxygen ion transport.824.4. Electrochemical stability of CeO2It was also observed that the top corners appeared to be more yellowish than the centerregion. This was possibly due to the oxygen ion preferentially owing through corners of thepellet than through the centre region. Electric current (and ions) always ow through the leastresistive path. Literature investigating the path of electrical current through powder materialsduring SPS processing revealed that in the case of electrically non-conductive powders, thecurrent passing through the powder is negligible in comparison to the current ow throughthe die [163]. In contrast, for conductive materials, the current density through the material isconsiderable. However, irrespective of the conductivity of the material, at the powder-punchinterface the current starts owing through the material into the die. Hence, as shown inFigure 4.35 (the red arrows the indicate the possible path of oxygen ions) the oxygen ion pathat the centre of the pellet was expected to vary across the pellet.In the case of the alumina encapsulated sample, the alumina disrupted the path of oxygenions; thus, reducing the volume fraction of oxygen ion vacancies. This didn’t prevent thereduction entirely, but has signicantly inhibited the reduction process.Figure 4.36 shows the microstructural variation in the top surface and bottom surface ofthe pellets. The microstructure on the top surface of the pellet had particles with a clear gap inbetween them. However, the particles in the bottom surface of the pellet were clearly uniformwhich no distinct visible boundaries.Figure 4.36: Microstructural variation across the pellet sintered without alumina encapsula-tion.834.5. SPS processing of SiC-CeO2 composites4.5. SPS processing of SiC-CeO2 compositesThe focus of this section was to understand the physical and chemical behavior of ceriumdioxide and silicon carbide blends processed via SPS. Three silicon carbide blends of 5 wt%,10wt% and 15 wt% were prepared and processed using SPS. The subsection on characterisationof sintered pellets covers results from XRD and SEM analysis, while the thermal analysis ofSiC-CeO2 blends subsection provides the results of DSC/TGA results.SiC-CeO2 composites may be used as surrogate materials for thermal conductivity evalu-ation of nuclear material-silicon carbide composites. However, as discussed in Chapter 4, theSPS of cerium dioxide and its composites is challenging. In this work, even minor variations(±10 ◦C, ±5 MPa, ±2 minutes) in peak temperature, peak pressure and holding time were seento cause the densication behaviour to change.4.5.1. Densication of SiC-CeO2 blendsSpark Plasma Sintering of SiC-CeO2 composites was carried out at a sintering temperature of1200 ◦C and a pressure of 50 MPa. The ramping rate of pressure was coupled with the heatingrate. The pellet was cooled down at 200 ◦C/min after holding for 5 minutes.The densication curves from all three blends traced each other, as seen in Figure 4.37a),with exception of the 5 wt% SiC curve. In contrast to the blends, the pure cerium dioxide sin-tered under same conditions followed an entirely dierent densication curve. This suggeststhat that the sintering kinetics of the blends were dierent from the that of pure cerium diox-ide. There was also a dierence in the displacement during the holding time; however, thiswasn’t noticeable in the densication curves, but created a dierence in the nal displace-ment, as shown in Figure 4.37b). The nal displacement increased linearly with increasingwt% SiC in the composite powder. This suggests that the densication of cerium dioxideincreased with increasing silicon carbide content.844.5. SPS processing of SiC-CeO2 compositesa bFigure 4.37: Eect of SiC content on densication: a) Densication curves of SiC-CeO2composites, b) Final displacement during SPS versus wt% SiC.Figure 4.38 shows the as-sintered pellets of pure CeO2 and 10 wt% and 15 wt% SiC-CeO2composites. The composites had a dark black coloration, suggesting that the blends wereundergoing a physical or chemical change during the sintering process. Thus, XRD and SEManalysis were performed to study the material composition in further detail.a b cFigure 4.38: Cross-section of: a) Pure CeO2, b) 10 wt% SiC-CeO2 composite, c) 15wt% SiC-CeO2 composite.4.5.2. Characterisation of as-sintered SiC-CeO2 compositesXRD analysisFigure 4.39 shows the X-Ray diractograms of the as-sintered SiC-CeO2 composites. Therewas clear evidence from the diractograms that there was a change in crystal structure of854.5. SPS processing of SiC-CeO2 compositesthe pellets due to the addition of SiC. All the three patterns had peaks corresponding toCeO2, Ce2O3, SiO2 and Ce2Si2O5. However, there were no peaks corresponding to siliconcarbide. This suggests that the CeO2 has reacted with the SiC powder to form Ce2O3, SiO2and Ce2Si2O5. Since CeO2 (as discussed in Section 2.2.1) is a powerful oxidising agent at hightemperatures, it likely reduced the SiC to form SiO2 and Ce2Si2O5.Figure 4.39: X-Ray diractograms from as sintered SiC-CeO2 blends.4.5.3. Micrography and chemical analysis of as-sinteredSiC-CeO2 pelletsFigure 4.40 shows a low magnication back scattered electron mode image of as sintered 10wt% SiC-CeO2 composite. There were a few black spots visible in a white matrix. The whitematrix was cerium dioxide, while the black spots were the silicon rich regions.864.5. SPS processing of SiC-CeO2 compositesFigure 4.40: Low magnication image of the 10 wt% SiC-CeO2.Figure 4.41 shows these black spots at high magnication. These particles were approx-imately 100 µm in size. Since, the black spots were not visible in the SE mode the imagingcontrast was due to elemental contrast and not morphological contrast.a bFigure 4.41: Microstructure of 10 wt% SiC showing the SiC particles in: a) SE mode, b) BSEmode.Figure 4.42 shows the SEM images of the white matrix and black spots at high magni-cation. There was a clear dierence in the morphology of the particles in these two regions.874.5. SPS processing of SiC-CeO2 compositesThis suggests that there is some inhomogeneity in the microstructure of the SiC-CeO2 blendpellet.a bFigure 4.42: Microstructure of 10 wt% SiC showing: a) White matrix, b) Black spots at highmagnication.X-EDS mapping was done at dierent sites in the 10 wt% SiC-CeO2 sample. Figure 4.43shows two X-EDS maps taken from dierent sites and dierent magnication levels. Themaps show evidence that the black part was rich in Silicon, while the white region was rich inCerium and Oxygen. Interestingly, the carbon shows a uniform distribution across the sample.However, detection of carbon in SEM is debatable according to literature [164]. Hence, thedata obtained from SEM on Carbon could be unreliable.884.5. SPS processing of SiC-CeO2 compositesabFigure 4.43: X-EDS maps of 10 wt% SiC-CeO2 pellet: a) Low magnication, b) High magni-cation.The interface between the cerium dioxide and the silicon rich region looks distinct andthere was no evidence of diusion of silicon into the cerium dioxide matrix or cerium andoxygen into silicon carbide particles. Hence, it is possible that the silicon carbide reactedwith cerium oxide only at the interface.894.5. SPS processing of SiC-CeO2 compositesThe SEM results suggest that there was a separation between the silicon carbide regionand the cerium dioxide particles, which is contradictory to the XRD results that proved thatthe cerium dioxide and silicon carbide reacted with each other. Therefore, additional workneeds to be carried out to fully characterise the reaction between cerium dioxide and siliconcarbide. Additional results can be found in Appendix B.4.5.4. Thermal analysis of SiC-CeO2 blendsThermal analysis was performed using DSC and TGA methods to determine if there was areaction between the cerium dioxide and silicon carbide during the sintering process. Theexperiment was performed under Argon atmosphere.As discussed in Section 4.3.4, the DSC curve of pure cerium dioxide has two exothermicpeaks: rst at 200 ◦C and a second at 680 ◦C. The same observation was made in the purecerium dioxide, as seen in Figure 4.44. However, in SiC-CeO2 blends, there was one additionalpeak at 1300 ◦C. This observation was evident in 5 wt%, 10 wt%C and 15 wt% SiC-CeO2 curves.abFigure 4.44: Combined a)DSC and b)TGA plot of SiC-CeO2 blends in Argon atmosphereThe only dierence between the DSC curves for the pure cerium dioxide and the com-posites was an additional endothermic peak. The endothermic peak at nearly 1300 ◦C was astrong indicator that a reaction between cerium dioxide and silicon carbide occurred. This904.5. SPS processing of SiC-CeO2 compositespeak was endothermic in nature, hence it absorbed heat, and the cerium silicate, cerium (III)oxide and silicon dioxide observed in XRD were the likely by-products of reaction.The TGA curves of 5 wt%, 10 wt% and 15 wt% SiC-CeO2 also had peaks at 1300◦C. Thesepeaks represent a large loss of mass of the powder during the exothermic reaction discussedabove. This observation supports the fact that there was some loss of mass in the form ofgaseous products, possibly due to the formation of carbon dioxide or carbon monoxide. Thiscarbon dioxide or carbon monoxide evolution during the reaction would possibly be the rea-son for the observed positive chamber pressure observed during SPS.These observations suggest that the SPS of SiC-CeO2 composites leads to a reaction be-tween these two powders. It could be concluded that these conditions are not advisable if theobjective was to add silicon carbide as an inert additive rather than a reactive additive.91Chapter 5: Conclusions and FutureWork5.1. ConclusionsThis study focused on the processing of CeO2 and SiC-CeO2 composites using Spark PlasmaSintering. The major conclusions from this study are listed below:Densication Behavior of Pure Cerium Dioxide1. The sintering displacement curves appeared to be highly repeatable with a maximumvariation of ±0.5 mm.2. The higher the sintering pressure and temperature, the higher the relative density ofthe pellet.3. The higher the pressure ramp rate, the higher the instantaneous relative density.4. The sintering atmosphere did not seem to create any noticeable dierence in sinteringcurves.5. Prior to sintering, the nano CeO2 powder pellet had a lower green density than otherblends and micro CeO2 powder.6. The blending of nano and micro CeO2 aided in increasing the density of the nal sin-tered pellet.Stability of Cerium Dioxide During SPS1. There was a colour and textural gradient observed in the SPS pellet which stronglysuggested that the reduction happened in a direction opposite to the ow of electric925.1. Conclusionscurrent.2. The center region of the top surface of pellet seemed to be less reduced than the areasurrounding it. This was probably due to the higher current densities at the corners ofthe sample.3. In the case of CeO2 processed with an alumina encapsulation, the reduction decreasedsignicantly. Apart from trace amounts of CeO2- , the remaining material was unaf-fected; thus proving that cutting o the electric eld helped in improving sinterabilityof the cerium dioxide.4. The precision lattice parameter measured from the XRD patterns conrmed that thenonstoichiometric ceria had a higher lattice parameter than CeO2. This suggests thatthe material was undergoing a chemical expansion.Sintering of Crium Dioxide-Silicon Carbide Composites1. The precursor SiC and CeO2 powders remained inert and non reactive during the blend-ing process.2. The silicon carbide powder had a trimodal particle size distribution with peaks at 0.5µm,2 µm and 13 µm. The SiC-CeO2 blends also had a trimodal distribution. However, theintensity of the peaks was reduced.3. The sintering curves of all three blends traced each other until the sintering temperaturewas reached.4. The displacement of the rams during the sintering stage changed with silicon carbidecontent, increasing linearly with increasing silicon carbide content5. The SiC-CeO2 composites changed color from grey to black color, suggesting that theblends, have undergone a chemical change.935.2. Research Limitations6. XRD analysis preformed on the as-sintered pellets supports the conjecture that ceriumdioxide reacted with silicon carbide during the sintering process. The XRD patternsshow peaks corresponding to cerium silicate, cerium (III) oxide and silicon dioxide aspossible byproducts.7. It is possible that cerium dioxide, which is a strong oxidising agent at room temperature,self reduced and also oxidised silicon carbide to silicon dioxide during sintering.8. The microstructure of the as-sintered materials contained black spots in a white matrix.The black spots were rich in silicon, whereas the white region was rich in cerium andoxygen.9. Thermal analysis concluded that the cerium dioxide and silicon carbide reacted at 1300◦Cvia an endothermic reaction. The reaction of cerium dioxide and silicon carbide resultedin a mass loss.5.2. Research LimitationsThe following limitations pertaining to the generated results:1. The Spark Plasma Sintering could only be carried out in vacuum or inert conditions toprotect the parts of equipment from high temperature oxidation. If the sintering processcould be carried out in oxygen atmosphere, then the cerium dioxide might not haveundergone chemical reduction and dierent densication kinetics would be observed.2. The chemical reduction of ceria was enhanced with at higher sintering temperatures.However, low sintering temperature sintering required high pressure, which is detri-mental to the integrity of the graphite tooling used during SPS.945.3. Future Work5.3. Future WorkThe following are suggestions would help provide additional data to further expand on theresults of this thesis:1. The pellets sintered using SPS could be subjected to heat treatment in a furnace to re-oxidise them. Mori et.al. has suggested that such a process could help in synthesis ofdense cerium dioxide pellets [110].2. The pellets could be sintered in other atmosphere, such as carbon dioxide. Carbondioxide was found to oxidise cerium dioxide at high temperature. This reaction couldavoid the self reduction of cerium dioxide.3. The SPS of cerium dioxide-silicon carbide blends could be carried out at temperatureslower than 1200 ◦C to prevent their reaction.95Bibliography[1] I. Hore-Lacy, Nuclear energy in the 21st century : the World Nuclear University primer.World Nuclear University Press, 2006.[2] “Computing the energy density of nuclear fuel,” Whatisnuclear.com, 2006.[3] E. D. Coyle and R. A. Simmons, Understanding the Global Energy Crisis. Purdue Uni-versity Press, 2014.[4] N. Haluzan, “Why is renewable energy important for the future?,” Renewables-info.com,2010.[5] D. Fridley, “Nine challenges of alternative energy,” Resilience, 2010.[6] K. Mathiesen, “What is holding back the growth of solar power?,” The Guardian, 2016.[7] L. V. Bremen, “Large-Scale Variability of Weather Dependent Renewable EnergySources,” pp. 189–206, 2010.[8] G. Walker, “Renewable energy and the public,” Land Use Policy, vol. 12, pp. 49–59, Jan-uary 1995.[9] D. Timmons, J. M. Harris, and B. Roach, “The Economics of Renewable Energy A GDAETeaching Module on Social and Environmental Issues in Economics,” tech. rep., 2014.[10] A. Nogee, S. Clemmer, B. Paulos, and B. Haddad, “Barriers to Renewable Energy Tech-nologies,” Union of Concerned Scientists, pp. 16–18, 1999.[11] D. Biello, “How Nuclear Power Can Stop Global Warming,” Scientic American, 2013.[12] C. Lo, “Flexible fossil fuels: bridging the energy gap,” Power Technology, 2012.[13] M. Krancer, “Nukes And Coal: The Surprising Clean Energy Bridge to Obama’s LowCarbon Future,” Forbes Energy, 2014.96Bibliography[14] R. Price and J. Blaise, “Nuclear fuel resources: Enough to last?,” NEA News, vol. 20,pp. 10–13, 2002.[15] S. J. Odell, Comparative Assessment of Coal-Fired and Nuclear Power Plants. PhD thesis,Rensselaer Polytechnic Institute, 2011.[16] D. Mills, “Comparison of solar, nuclear and wind options for large scale implementa-tions,” Clean Energy! - Can do, 2006.[17] “Fuel comparison,” https://www.euronuclear.org/info/encyclopedia/f/fuelcomparison.htm.[18] G. Steinhauser, A. Brandl, and T. E. Johnson, “Comparison of the Chernobyl andFukushima nuclear accidents: A review of the environmental impacts,” Science of TheTotal Environment, vol. 470-471, pp. 800–817, February 2014.[19] N. Beresford, S. Fesenko, A. Konoplev, L. Skuterud, J. Smith, and G. Voigt, “Thirty yearsafter the Chernobyl accident: What lessons have we learnt?,” Journal of EnvironmentalRadioactivity, vol. 157, pp. 77–89, June 2016.[20] Z. Su’ud and R. Anshari, “Preliminary analysis of loss-of-coolant accident in Fukushimanuclear accident,” pp. 315–327, 2012.[21] S. Yeo, E. Mckenna, R. Baney, G. Subhash, and J. Tulenko, “Enhanced thermal con-ductivity of uranium dioxide–silicon carbide composite fuel pellets prepared by SparkPlasma Sintering (SPS),” Journal of Nuclear Materials, vol. 433, pp. 66–73, February 2013.[22] T. R. G. Kutty, J. Banerjee, and A. Kumar, Thoria-based Nuclear Fuels. 2013.[23] Q. Yin, A. Kutepov, K. Haule, G. Kotliar, S. Y. Savrasov, and W. E. Pickett, “Electronic cor-relation and transport properties of nuclear fuel materials,” Physical Review B, vol. 84,p. 195111, November 2011.97Bibliography[24] H. S. Kim, C. Y. Joung, B. H. Lee, J. Y. Oh, Y. H. Koo, and P. Heimgartner, “Applicability ofCeO2 as a surrogate for PuO2 in a MOX fuel development,” Journal of Nuclear Materials,vol. 378, pp. 98–104, August 2008.[25] K. Choi, W. Tong, R. D. Maiani, D. E. Burkes, and Z. A. Munir, “Densication of nano-CeO2 ceramics as nuclear oxide surrogate by spark plasma sintering,” Journal of NuclearMaterials, vol. 404, pp. 210–216, September 2010.[26] M. Miodownik, “Why the story of materials is really the story of civilisation,” TheGuardian, 2014.[27] M. A. Morse, “Craniology and the Adoption of the Three-Age System in Britain.,” Pro-ceedings of the Prehistoric Society, vol. 65, pp. 1–16, February 1999.[28] “Stone Age man used re to make tools - 50,000 years earlier than we scientiststhought,” The Telegraph, 2009.[29] B. S. Ottaway, “Innovation, Production and Specialization in Early Prehistoric CopperMetallurgy,” European Journal of Archaeology, vol. 4, pp. 87–112, 2001.[30] A. Reardon, Metallurgy for the Non-Metallurgist, Discovering Metals— A HistoricalOverview. 2 ed., 2011.[31] D. W. Budworth, An introduction to ceramic science. Pergamon Press, 1970.[32] T. Mason, “Advanced structural ceramics,” Britannica Encylopedia, 1998.[33] K. Miyoshi and D. H. Buckley, “Tribological properties of silicon carbide in the metalremoval process,” Wear, vol. 82, no. 2, pp. 197–211, 1982.[34] D. Jiao, S. Zheng, Y. Wang, R. Guan, and B. Cao, “The tribology properties of alu-mina/silica composite nanoparticles as lubricant additives,” Applied Surface Science,vol. 257, no. 13, pp. 5720–5725, 2011.98Bibliography[35] D. R. Clarke , M. Oechsner and N. P. Padture, “Thermal-barrier coatings for more e-cient gas-turbine engines,” MRS Bulletin, vol. 37, no. 10, pp. 891–897, 2012.[36] “Structure and Properties of Ceramics,” American Ceramic Society, 2014.[37] Q. Yin, B. Zhu, and H. Zeng, Microstructure, property and processing of functional ce-ramics. Metallurgical Industry Press, 2009.[38] T. I. John and C. Paul, eds., Solid Oxide Fuels Cells: Facts and Figures: Past Present andFuture Perspectives for SOFC Technologies, vol. 2. Springer Science & Business Media,2012.[39] S. Badwal, “Zirconia-based solid electrolytes: microstructure, stability and ionic con-ductivity,” Solid State Ionics, vol. 52, pp. 23–32, May 1992.[40] H. Inaba and H. Tagawa, “Ceria-based solid electrolytes,” Solid State Ionics, vol. 83, no. 1,pp. 1–16, 1996.[41] H. C. Yao and Y. F. Yu Yao, “Ceria in Automotive Exhaust Catalysts I. Oxygen Storage,”Journal of December, vol. 86, pp. 254–265, 1984.[42] J. Zhang, Y. Nazarenko, L. Zhang, L. Calderon, K.-B. Lee, E. Garfunkel, S. Schwander,T. D. Tetley, K. F. Chung, A. E. Porter, M. Ryan, H. Kipen, P. J. Lioy, and G. Mainelis, “Im-pacts of a nanosized ceria additive on diesel engine emissions of particulate and gaseouspollutants.,” Environmental Science & Technology, vol. 47, pp. 13077–85, November 2013.[43] N. Izu, S. Nishizaki, W. Shin, T. Itoh, M. Nishibori, and I. Matsubara, “Resistive oxygensensor using ceria-zirconia sensor material and ceria-yttria temperature compensatingmaterial for lean-burn engine.,” Sensors (Basel, Switzerland), vol. 9, no. 11, pp. 8884–95,2009.99Bibliography[44] D. Marrocchelli, S. R. Bishop, H. L. Tuller, and B. Yildiz, “Understanding Chemical Ex-pansion in Non-Stoichiometric Oxides: Ceria and Zirconia Case Studies,” AdvancedFunctional Materials, vol. 22, pp. 1958–1965, May 2012.[45] R. Panlener, R. Blumenthal, and J. Garnier, “A thermodynamic study of nonstoichiomet-ric cerium dioxide,” Journal of Physics and Chemistry of Solids, vol. 36, no. 11, pp. 1213–1222, 1975.[46] A. D. Wadsley, Nonstoichiometric Metal Oxides. January 1963.[47] Y. Zhong, R. C. O’Brien, S. D. Howe, N. D. Jerred, K. Schwinn, L. Sudderth, and J. Hund-ley, “Spark Plasma Sintering of Fuel Cermets for Nuclear Reactor Applications,” MRSProceedings, vol. 1383, pp. 01–08, February 2012.[48] N. Pala, A. N. Abbas, C. Rockstuhl, C. Menzel, S. Mühlig, F. Lederer, J. J. Brown, V. M.Bright, S. Curley, Y. K. Koh, S. Pourkamali, J. P. Heremans, A. M. Chamoire, K. Shankar,A. Risbud, A. Lakhtakia, M. H. Bartl, M. Carrière, C. Larue, H. Kohl, K. Yang, D. Hitch-cock, J. He, and A. M. Rao, “Thermal Conductivity and Phonon Transport,” in Encyclo-pedia of Nanotechnology, pp. 2704–2711, 2012.[49] J. A. Webb and I. Charit, “Fabrication of Cermets via Spark-Plasma Sintering for NuclearApplications,” Journal of Materials, vol. 66, pp. 943–952, April 2014.[50] H. Hammoud, S. Vaucher, and F. Valdivieso, “Dielectric and thermal properties ofcerium dioxide up to 1000°C and the eect of the porosity for microwave processingstudies,” Thermochimica Acta, vol. 617, pp. 83–89, October 2015.[51] S. Bishop, K. Duncan, and E. Wachsman, “Defect equilibria and chemical expansion innon-stoichiometric undoped and gadolinium-doped cerium oxide,” Electrochimica Acta,vol. 54, pp. 1436–1443, February 2009.100Bibliography[52] S. Wang, H. Inaba, H. Tagawa, and T. Hashimoto, “Nonstoichiometry of Ce0.8Gd0.2O1.9x,”Journal of The Electrochemical Society, vol. 144, no. 11, p. 4076, 1997.[53] M. Hillert and B. Janson, “Thermodynamic Model for Nonstoichiometric Ionic Phases- Application to CeO2-x,” Journal of the American Ceramic Society, vol. 69, pp. 732–734,October 1986.[54] M. Panhans and R. Blumenthal, “A thermodynamic and electrical conductivity study ofnonstoichiometric cerium dioxide,” Solid State Ionics, vol. 60, pp. 279–298, April 1993.[55] G. B. Peter, Solid State Electrochemistry. 1997.[56] K. Meier, T. Laino, and A. Curioni, “Solid-State Electrolytes: Revealing the Mechanismsof Li-Ion Conduction in Tetragonal and Cubic LLZO by First-Principles Calculations,”The Journal of Physical Chemistry C, vol. 118, pp. 6668–6679, April 2014.[57] S. Bao, C. Ma, G. Chen, X. Xu, E. Enriquez, C. Chen, Y. Zhang, J. L. Bettis, M.-H.Whangbo, C. Dong, and Q. Zhang, “Ultrafast Atomic Layer-by-Layer Oxygen Vacancy-Exchange Diusion in Double-Perovskite LnBaCo2O5.5+ Thin Films,” Scientic Reports,vol. 4, pp. 1212–1217, April 2014.[58] G. E. Murch and A. S. Nowick, Diusion in crystalline solids. Academic Press, 1984.[59] H. Inaba, “Ceria-based solid electrolytes,” Solid State Ionics, vol. 83, pp. 1–16, January1996.[60] A. Chroneos, B. Yildiz, A. Taranc, D. Partt, and J. A. Kilner, “Oxygen diusion in solidoxide fuel cell cathode and electrolyte materials: mechanistic insights from atomisticsimulations,” Energy & Environmental Science, vol. 4, no. 8, pp. 2774–2789, 2011.[61] G. S. Upadhyaya, Powder Metallurgy Technology. Cambridge International Publishingcompany, 2002.101Bibliography[62] P.C.Angelo and R. Subramanian, Powder Meetallurgy: Science, Technology and Applica-tions. PHI Learning Pvt. Ltd, 2008.[63] I. Chang and Y. Zhao, Advances in powder metallurgy : properties, processing and appli-cations. Elsevier, 2013.[64] H. H. Hausner, ed., Modern Developments in Powder Metallurgy. Boston, MA: SpringerUS, 1995.[65] R. M. German, A-Z of powder metallurgy. Elsevier, 2005.[66] H. H. Hausner, Modern Developments in Powder Metallurgy : Volume 1 Fundamentalsand Methods. Springer US, 1995.[67] F. Thummler, R. Oberacker, and E. Institute of Materials (London, An introduction topowder metallurgy. Institute of Materials, 1993.[68] A. R. Cooper and L. E. Eaton, “Compaction Behavior of Several Ceramic Powders,”Journal of the American Ceramic Society, vol. 45, pp. 97–101, March 1962.[69] R. M. German, Sintering theory and practice. Wiley, 1996.[70] Z. Z. Fang, Sintering of advanced materials : fundamentals and processes. WoodheadPub, 2010.[71] G. S. Upadhyaya, Sintered metallic and ceramic materials : preparation, properties, andapplications. Wiley, 2000.[72] C. Burt, Sintering of ceramics : prominent processes and methods. Ny Research Pr, 2015.[73] S.-J. L. Kang, Sintering : densication, grain growth, and microstructure. ElsevierButterworth-Heinemann, 2005.[74] S.-J. L. Kang, R. Bordia, E. Olevsky, and D. Bouvard, eds., Advances in Sintering Scienceand Technology II. Hoboken, NJ, USA: John Wiley & Sons, Inc., September 2012.102Bibliography[75] S. Smiya and Y. Moriyoshi, eds., Sintering Key Papers. Dordrecht: Springer Netherlands,1990.[76] M. N. Rahaman, Ceramic processing and sintering. M. Dekker, 2003.[77] M. N. Rahaman, Sintering of ceramics. CRC Press, 2008.[78] P. Guyot, G. Antou, N. Pradeilles, A. Weibel, M. Vandenhende, G. Chevallier, A. Peigney,C. Estournès, and A. Maître, “Hot pressing and spark plasma sintering of alumina: Dis-cussion about an analytical modelling used for sintering mechanism determination,”Scripta Materialia, vol. 84-85, pp. 35–38, August 2014.[79] C. A. Handwerker, J. E. Blendell, and R. L. Coble, “Sintering of Ceramics,” in Science ofSintering, pp. 3–37, Boston, MA: Springer US, 1989.[80] R. K. Bordia and E. A. Olevsky, eds., Advances in Sintering Science and Technology. Hobo-ken, NJ, USA: John Wiley & Sons, Inc., December 2009.[81] E. Zapata-Solvas, D. Gómez-García, A. Domínguez-Rodríguez, and R. I. Todd, “Ultra-fast and energy-ecient sintering of ceramics by electric current concentration.,” Sci-entic reports, vol. 5, p. 8513, January 2015.[82] D. E. Clark and W. H. Sutton, “Microwave Processing of Materials,” Annual Review ofMaterials Science, vol. 26, pp. 299–331, August 1996.[83] S. Grasso, Y. Sakka, and G. Maizza, “Electric current activated/assisted sintering ( ECAS): a review of patents 1906–2008,” Science and Technology of Advanced Materials, vol. 10,p. 053001, October 2009.[84] Z. A. Munir, D. V. Quach, and M. Ohyanagi, “Electric Current Activation of Sintering:A Review of the Pulsed Electric Current Sintering Process,” Journal of the AmericanCeramic Society, vol. 94, pp. 1–19, January 2011.103Bibliography[85] A. G. Bloxam, “1906 GB Patent No. 27,002 [5],” 1906.[86] R. Todd, E. Zapata-Solvas, R. Bonilla, T. Sneddon, and P. Wilshaw, “Electrical charac-teristics of ash sintering: thermal runaway of Joule heating,” Journal of the EuropeanCeramic Society, vol. 35, no. 6, pp. 1865–1877, 2015.[87] A. M. Locci, A. Cincotti, S. Todde, R. Orrù, and G. Cao, “A methodology to investigatethe intrinsic eect of the pulsed electric current during the spark plasma sintering ofelectrically conductive powders,” Science and Technology of Advanced Materials, vol. 11,p. 045005, February 2010.[88] R. S. Dobedoe, G. D. West, and M. H. Lewis, “Spark plasma sintering of ceramics: under-standing temperature distribution enables more realistic comparison with conventionalprocessing,” Advances in Applied Ceramics, vol. 104, pp. 110–116, June 2005.[89] M. Tokita, “Trends in Advanced SPS Spark Plasma Sintering Systems and Technology.Functionally Gradient Materials and Unique Synthetic Processing Methods from NextGeneration of Powder Technology.,” Journal of the Society of Powder Technology, Japan,vol. 30, pp. 790–804, April 1993.[90] W. Chen, U. Anselmi-Tamburini, J. Garay, J. Groza, and Z. Munir, “Fundamental in-vestigations on the spark plasma sintering/synthesis process,” Materials Science andEngineering: A, vol. 394, pp. 132–138, March 2005.[91] M. Omori, “Sintering, consolidation, reaction and crystal growth by the spark plasmasystem (SPS),” Materials Science and Engineering: A, vol. 287, pp. 183–188, August 2000.[92] M. Nygren and Z. Shen, “On the preparation of bio-, nano- and structural ceramics andcomposites by spark plasma sintering,” Solid State Sciences, vol. 5, pp. 125–131, January2003.104Bibliography[93] W. Liu and M. Naka, “In situ joining of dissimilar nanocrystalline materials by sparkplasma sintering,” Scripta Materialia, vol. 48, pp. 1225–1230, May 2003.[94] D. H. Shim, S. S. Jung, H. S. Kim, H. Cho, J. K. Kim, T. G. Kim, and S. J. Yoon, “Eect OfCarbon Nanotubes On The Properties Of Spark Plasma Sintered ZrO2/CNT Compos-ites,” Archives of Metallurgy and Materials, vol. 60, pp. 1315–1318.[95] Plapcianu, Valsangiacom, Schaer, J. E, Wieg, Garay, and Stanciu, “Spark plasma sin-tering studies of nanosize lanthanide-doped ceria obtained by sot-gel method,” Journalof Optoelectronics and Advanced Materials, vol. 13, no. 9-10, pp. 1101–1108, 2011.[96] S. C. Tjong, Nanocrystalline materials : their synthesis-structure-property relationshipsand applications. Elsevier, 2006.[97] D. M. Hulbert, A. Anders, D. V. Dudina, J. Andersson, D. Jiang, C. Unuvar, U. Anselmi-Tamburini, E. J. Lavernia, and A. K. Mukherjee, “The absence of plasma in “spark plasmasintering,” Journal of Applied Physics, vol. 104, August 2008.[98] G.-D. Zhan, J. Kuntz, J. Wan, J. Garay, and A. K. Mukherjee, “A Novel Processing Routeto Develop a Dense Nanocrystalline Alumina Matrix (<100 nm) Nanocomposite Mate-rial,” Journal of the American Ceramic Society, vol. 86, pp. 200–2002, January 2003.[99] S. Muñoz and U. Anselmi-Tamburini, “Temperature and stress elds evolution duringspark plasma sintering processes,” Journal of Materials Science, vol. 45, pp. 6528–6539,December 2010.[100] N. Chawake, L. D. Pinto, A. K. Srivastav, K. Akkiraju, B. Murty, and R. S. Kottada,“On Joule heating during spark plasma sintering of metal powders,” Scripta Materialia,vol. 93, pp. 52–55, 2014.[101] P. G. Shewmon, Diusion in solids. Minerals, Metals & Materials Society, 1989.105Bibliography[102] J. R. Groza and A. Zavaliangos, “Nanostructured bulk solids by eld activated sintering,”vol. 5, pp. 24–33, October 2003.[103] C. Manière, L. Durand, A. Weibel, and C. Estournès, “Spark-plasma-sintering and niteelement method: From the identication of the sintering parameters of a submicronic-alumina powder to the development of complex shapes,” Acta Materialia, vol. 102,pp. 169–175, 2016.[104] T. Hungría, J. Galy, and A. Castro, “Spark Plasma Sintering as a Useful Technique to theNanostructuration of Piezo-Ferroelectric Materials,” Advanced Engineering Materials,vol. 11, pp. 615–631, August 2009.[105] D. Jiang, D. M. Hulbert, U. Anselmi-Tamburini, T. Ng, D. Land, and A. K. Mukherjee,“Optically Transparent Polycrystalline Al2O3 Produced by Spark Plasma Sintering,”Journal of the American Ceramic Society, vol. 91, pp. 151–154, December 2007.[106] J. Wan, R. Duan, and A. Mukherjee, “Spark plasma sintering of silicon nitride/siliconcarbide nanocomposites with reduced additive amounts,” Scripta Materialia, vol. 53,pp. 663–667, September 2005.[107] H. Zhang, B.-N. Kim, K. Morita, H. Y. Keijiro Hiraga, and Y. Sakka, “Eect of sinteringtemperature on optical properties and microstructure of translucent zirconia preparedby high-pressure spark plasma sintering,” Science and Technology of AdvancedMaterials,vol. 12, October 2011.[108] M. Panhans and R. Blumenthal, “A thermodynamic and electrical conductivity study ofnonstoichiometric cerium dioxide,” Solid State Ionics, vol. 60, no. 4, pp. 279–298, 1993.[109] R. Podor, N. Clavier, J. Ravaux, L. Claparede, N. Dacheux, and D. Bernache-Assollant,“Dynamic aspects of cerium dioxide sintering: HT-ESEM study of grain growth andpore elimination,” Journal of the European Ceramic Society, vol. 32, pp. 353–362, Febru-ary 2012.106Bibliography[110] T. Mori, T. Kobayashi, Y. Wang, J. Drennan, T. Nishimura, J.-G. Li, and H. Kobayashi,“Synthesis and Characterization of Nano-Hetero-Structured Dy Doped CeO2 Solid Elec-trolytes Using a Combination of Spark Plasma Sintering and Conventional Sintering,”Journal of the American Ceramic Society, vol. 88, pp. 1981–1984, July 2005.[111] T. Mori, J. Drennan, T. Mori, and J. Drennan, “Inuence of microstructure on oxide ionicconductivity in doped CeO2 electrolytes,” J Electroceram, vol. 17, pp. 749–757, 2006.[112] S. Wang, S. Chen, A. Navrotsky, M. Martin, S. Kim, and Z. A. Munir, “Modied polyol-mediated synthesis and consolidation of Gd-doped ceria nanoparticles,” Solid State Ion-ics, vol. 181, no. 8, pp. 372–378, 2010.[113] U. Anselmi-Tamburini, J. Garay, and Z. Munir, “Fast low-temperature consolidationof bulk nanometric ceramic materials,” Scripta Materialia, vol. 54, pp. 823–828, March2006.[114] M. Mogensen, “Physical, chemical and electrochemical properties of pure and dopedceria,” Solid State Ionics, vol. 129, pp. 63–94, April 2000.[115] E. N. Kaufmann, ed., Characterization of Materials. Hoboken, NJ, USA: John Wiley &Sons, October 2002.[116] L. Svarovskly, Powder Testing Guide. Springer, 1987.[117] J. Schwedes and D. Schulze, “Measurement of ow properties of bulk solids,” PowderTechnology, vol. 61, pp. 59–68, April 1990.[118] A. Bodhmage, Correlation between physical properties and owability indicators for nepowders. PhD thesis, University of Saskatchewan, 2006.[119] S. Özbilen, “Inuence of atomising gas pressure on particle shape of Al and Mg pow-ders,” Powder Technology, vol. 102, pp. 109–119, March 1999.107Bibliography[120] R. Ho, M. Naderi, J. Y. Y. Heng, D. R. Williams, F. Thielmann, P. Bouza, A. R. Keith,G. Thiele, and D. J. Burnett, “Eect of Milling on Particle Shape and Surface EnergyHeterogeneity of Needle-Shaped Crystals,” Pharmaceutical Research, vol. 29, pp. 2806–2816, October 2012.[121] C. Perego, R. Millini, C. Perego, G. Girotti, G. Terzoni, R. Ryoo, H. Wang, X. Zhang,M. Tsapatsis, F.-S. Xiao, F.-S. Xiao, D. Zhao, and Y. Wei, “Porous materials in catalysis:challenges for mesoporous materials,” Chem. Soc. Rev., vol. 42, no. 9, pp. 3956–3976,2013.[122] J. B. Nelson and D. P. Riley, “An experimental investigation of extrapolation methods inthe derivation of accurate unit-cell dimensions of crystals,” Proceedings of the PhysicalSociety, vol. 57, pp. 160–177, May 1945.[123] M. E. Brown, Introduction to Thermal Analysis. Dordrecht: Springer Netherlands, 1988.[124] P. Gabbott, Principles and applications of thermal analysis. Blackwell Pub, 2008.[125] R. F. Speyer, Thermal analysis of materials. Marcel Dekker, 1994.[126] M. Brown and P. Gallagher, Handbook of Thermal Analysis and Calorimetry. 2007.[127] M. Reading and D. J. Hourston, Modulated temperature dierential scanning calorimetry: theoretical and practical applications in polymer characterisation. Springer, 2006.[128] M. Reading, A. Luget, and R. Wilson, “Modulated dierential scanning calorimetry,”Thermochimica Acta, vol. 238, pp. 295–307, June 1994.[129] G. W. H. Hohne, W. F. Hemminger, and H.-J. Flammersheim, Dierential ScanningCalorimetry. Springer Berlin Heidelberg, 2003.[130] C. Schick, D. Lexa, L. Leibowitz, C. Schick, D. Lexa, and L. Leibowitz, “Dierential Scan-ning Calorimetry and Dierential Thermal Analysis,” in Characterization of Materials,Hoboken, NJ, USA: John Wiley & Sons, Inc., October 2012.108Bibliography[131] S. Vyazovkin, “Thermogravimetric Analysis,” in Characterization of Materials, Hobo-ken, NJ, USA: John Wiley & Sons, Inc., October 2012.[132] R. B. Prime, H. E. Bair, S. Vyazovkin, P. K. Gallagher, and A. Riga, “ThermogravimetricAnalysis (TGA),” in Thermal Analysis of Polymers, pp. 241–317, Hoboken, NJ, USA: JohnWiley & Sons, Inc.[133] “A Beginner’s Guide to Thermogravimetric Analysis,” Perkin Elmer.[134] A. W. Coats and J. P. Redfern, “Thermogravimetric analysis. A review,” The Analyst,vol. 88, no. 1053, p. 906, 1963.[135] B. Wunderlich, Thermal analysis. Academic Press, 1990.[136] AZO Nano, “Nanopowders – A Brief Denition and Applications of Nanopowders.”[137] Fritsch, “Operating Manual Planetary Micro Mill Pulverisette 7 Premium Line,” tech.rep.[138] ASTM, “ASTM C1444 - 00 Standard Test Method for Measuring the Angle of Repose ofFree-Flowing Mold Powders.”[139] E. A. Olevsky, “Theory of sintering: from discrete to continuum,” Materials Science andEngineering: R: Reports, vol. 23, no. 2, pp. 41–100, 1998.[140] L. Chen, P. Fleming, V. Morris, J. D. Holmes, and M. A. Morris, “Size-Related LatticeParameter Changes and Surface Defects in Ceria Nanocrystals,” The Journal of PhysicalChemistry C, vol. 114, pp. 12909–12919, August 2010.[141] S. Somiya, F. Aldinger, N. Claussen, R. M. Spriggs, K.Uchino, K.Koumoto, andM.Kaneno, Handbook of Advanced Ceramics Volume II: Processing and Their Applica-tions. 2003.109Bibliography[142] C. C. Koch, Nanostructured Materials, 2nd Edition: Processing, Properties and Applica-tions. 2 ed., 2007.[143] B. D. Cullity, Elements Of X Ray Diraction. 1956.[144] Bordia, R. K. and Camacho-Montes, H, “Sintering - fundamentals and practice,” 2012.[145] K. K. Bodla and S. V. Garimella, “Simulated Microstructural Evolution and Design ofPorous Sintered Wicks,” Journal of Heat Transfer, vol. 136, p. 072601, March 2014.[146] T. Rivera, “Synthesis and Thermoluminescent Characterization of Ceramics Materials,”in Advances in Ceramics - Synthesis and Characterization, Processing and Specic Appli-cations, InTech, August 2011.[147] D. U. H. Palmour III, R.M. Spriggs, Science of Sintering. 1989.[148] S. S. Moosath, J. Abraham, and T. V. Swaminathan, “Thermal Decomposition of RareEarth Metal Oxalates. IV. Oxalates of Cerium and Thorium,” Zeitschrift for anorganischeund allgemeine Chemie, vol. 324, pp. 103–105, August 1963.[149] C. Padeste, N. W. Cant, and D. L. Trimm, “The inuence of water on the reduction andreoxidation of ceria,” Catalysis Letters, vol. 18, no. 3, pp. 305–316, 1993.[150] A. Hadi, I. I. Yaacob, and L. S. Ling, “Mechanochemical Synthesis of NanocrystallineCeO2: The Eect of Annealing Temperatures on the Particle Size,” Materials ScienceForum, vol. 517, pp. 252–256, 2006.[151] S. El-Houte and M. El-Sayed Ali, “Thermal decomposition of cerium oxalate and mixedcerium-gadolinium oxalates,” Journal of Thermal Analysis, vol. 37, pp. 907–913, May1991.[152] W. C. Chueh and S. M. Haile, “A thermochemical study of ceria: exploiting an old ma-terial for new modes of energy conversion and CO2 mitigation,” Philosophical Transac-110Bibliographytions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences,vol. 368, no. 1923, 2010.[153] D. Marrocchelli, N. H. Perry, and S. R. Bishop, “Understanding chemical expansionin perovskite-structured oxides.,” Physical chemistry chemical physics : PCCP, vol. 17,pp. 10028–39, April 2015.[154] H.-L. Lee, H.-J. Lim, S. Kim, and H.-B. Lee, “Thermomechanical Properties of beta-Sialon Synthesized from Kaolin,” Journal of the American Ceramic Society, vol. 72,pp. 1458–1461, August 1989.[155] A. Jha and S. J. Yoon, “Formation of titanium carbonitride phases via the reduction ofTiO2 with carbon in the presence of nitrogen,” Journal of Materials Science, vol. 34, no. 2,pp. 307–322, 1999.[156] M. I. El-Guindy and W. G. Davenport, “Kinetics and mechanism of llmenite reductionwith graphite,” Metallurgical Transactions, vol. 1, pp. 1729–1734, June 1970.[157] R. Wang and M. Fang, “Improved low-temperature reducibility in ceria zirconiananoparticles by redox treatment,” Journal of Materials Chemistry, vol. 22, pp. 1770–1773, 2012.[158] S. R. Bishop, D. Marrocchelli, N. H. Perry, H. L. Tuller, G. Watson, B. Yildiz, K. Amezawa,and J. A. Kilner, “Chemical Expansion in SOFC Materials: Ramications, Origins, andMitigation,” ECS Transactions, vol. 57, pp. 643–648, October 2013.[159] N. Frage, S. Cohen, S. Meir, S. Kalabukhov, and M. P. Dariel, “Spark plasma sintering(SPS) of transparent magnesium-aluminate spinel,” Journal of Materials Science, vol. 42,pp. 3273–3275, May 2007.[160] J. P. Ackerman and T. D. Claar, “Materials for high-temperature fuel cells,” Journal ofMaterials for Energy Systems, vol. 8, pp. 284–290, December 1986.111[161] G. Ranga Rao, P. Fornasiero, R. Di Monte, J. Kašpar, G. Vlaic, G. Balducci, S. Meriani,G. Gubitosa, A. Cremona, and M. Graziani, “Reduction of NO over Partially ReducedMetal-Loaded CeO2–ZrO2Solid Solutions,” Journal of Catalysis, vol. 162, pp. 1–9, Au-gust 1996.[162] K. J. Kim and G. M. Choi, “Phase stability and oxygen non-stoichiometry of Gd-dopedceria during sintering in reducing atmosphere,” Journal of Electroceramics, vol. 35,pp. 68–74, December 2015.[163] K. Vanmeensel, A. Laptev, J. Hennicke, J. Vleugels, and O. Van der Biest, “Modellingof the temperature distribution during eld assisted sintering,” Acta Materialia, vol. 53,no. 16, pp. 4379–4388, 2005.[164] M. Choël, K. Deboudt, J. Osán, P. Flament, and R. Van Grieken, “Quantitative deter-mination of low Z elements in single atmospheric particles on Boron substrates byautomated Scanning Electron MicroscopyEnergy-Dispersive X-ray Spectrometry,” An-alytical Chemistry, vol. 77, pp. 5686–5692, September 2005.112Appendix Aa bc de fFigure A.1: As-sintered CeO2 pellets: a) 1200◦C, 50MPa, 5minutes, b) 1500◦C, 50MPa,25minutes, c) 1200◦C, 50MPa, 5minutes, d) 1100◦C, 65MPa, 5minutes, e) 1150◦C, 65MPa,5minutes, f) 1150◦C, 65MPa, 2minutes.113Appendix Aa bc de fg hFigure A.2: As-sintered CeO2 pellets: a) 1150◦C, 65MPa, 5minutes, b) 1:3 Nano-Micro blend,1200◦C, 50MPa, 5minutes, c) 1:1 Nano-Micro blend, 1200◦C, 50MPa, 5minutes, d) 3:1 Nano-Micro blend, 1200◦C, 50MPa, 5minutes, e) Nano-Micro blend, 1062◦C, 70MPa, 5minutes, f)Inframat® nano CeO2, 1062◦C, 70MPa, 5minutes, g) 1050◦C, 70MPa, 5minutes, h) 1050◦C,70MPa, 5minutes. 114Appendix Aa bc de fg hFigure A.3: As-sintered CeO2 pellets: a) 1000◦C, 50MPa, 5minutes, b) 1062◦C, 70MPa, 5min-utes, c) 1250◦C, 50MPa, 5minutes, d) 1062◦C, 70MPa, 5minutes, e) 1:3 Nano-Micro blend,1062◦C, 70MPa, 5minutes, f) Inframat® nano CeO2, 1062◦C, 70MPa, 5minutes, g) 1500◦C,50MPa, 25minutes, h) 1200◦C, 50MPa, 5minutes.115Appendix Aa bc dFigure A.4: As-sintered CeO2 pellets: a) 1062◦C, 70MPa, 5minutes, b) 1200◦C, 60MPa, 5min-utes, c) 1200◦C, 60MPa, 10minutes, d) 1200◦C, 60MPa, 5minutes.116Appendix BFigure B.1: Elemental map of 10 wt% SiC-CeO2 blend.Figure B.2: Elemental map of 15 wt% SiC-CeO2 blend.117Appendix BFigure B.3: Microstructure of of 15 wt% SiC-CeO2 as-sintered pellet.Figure B.4: Elemental map of 15 wt% SiC-CeO2 as-sintered pellet.118Appendix Ca bFigure C.1: Scanning Electron Micrograph of 1:1 Nano-Micro CeO2 blend powder.119

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0355271/manifest

Comment

Related Items