Processing of Uranium Dioxide Nuclear Fuel Pellets Using Spark Plasma Sintering

MISSING IMAGE

Material Information

Title:
Processing of Uranium Dioxide Nuclear Fuel Pellets Using Spark Plasma Sintering
Physical Description:
1 online resource (159 p.)
Language:
english
Creator:
Ge, Lihao
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Mechanical Engineering, Mechanical and Aerospace Engineering
Committee Chair:
SUBHASH,GHATU
Committee Co-Chair:
ARAKERE,NAGARAJ KESHAVAMURTHY
Committee Members:
YANG,YONG
TULENKO,JAMES S

Subjects

Subjects / Keywords:
densification -- mechanical -- nuclear -- sintering -- sps -- thermal -- uo2
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre:
Mechanical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Uranium dioxide (UO2), one of the most common nuclear fuels, has been applied in most of the nuclear plant these days for electricity generation. The main objective of this research is to introduce a novel method for UO2 processing using spark plasma sintering technique (SPS). Firstly, an investigation into the influence of processing parameters on densification of UO2 powder during SPS is presented. A broad range of sintering temperatures, hold time and heating rates have been systematically varied to investigate their influence on the sintered pellet densification process. The results revealed that up to 96% theoretical density (TD) pellets can be obtained at a sintering temperature of 1050C for 30s hold time and a total run time of only 10 minutes. A systematic study was performed by varying the sintering temperature between 750C to 1450C and hold time between 0.5min to 20min to obtain UO2 pellets with a range of densities and grain sizes. The microstructure development in terms of grain size, density and porosity distribution was investigated. The Oxygen/Uranium (O/U) ratio of the resulting pellets was found to decrease after SPS. The mechanical and thermal properties of UO2 are evaluated. For comparable density and grain size, Vickers hardness and Youngs modulus are in agreement with the literature value. The thermal conductivity of UO2 increased with the density but the grain size in the investigated range had no significant influence. Overall, the mechanical and thermal properties of UO2 are comparable with the one made using conventional sintering methods. Lastly, the influence of chromium dioxide (Cr2O3) and zirconium diboride (ZrB2) on the grain size of doped UO2 fuel pellet was performed to investigate the feasibility of producing large-grain-size nuclear fuel using SPS. The benefits of using SPS over the conventional sintering of UO2 are summarized. The future work of designing macro-porous UO2 pellet and thorium dioxide (ThO2) cored UO2 pellet is proposed.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Lihao Ge.
Thesis:
Thesis (Ph.D.)--University of Florida, 2014.
Local:
Adviser: SUBHASH,GHATU.
Local:
Co-adviser: ARAKERE,NAGARAJ KESHAVAMURTHY.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2014
System ID:
UFE0046627:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 PROCESSING OF URANIUM DIOXIDE NUCLEAR FUEL PELLETS USING SPARK PLASMA SINTERING By LIHAO GE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE D EGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2014

PAGE 2

2 2014 Lihao Ge

PAGE 3

3 To my parents, Cao Ge and fengchun Xu, for their love and support

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my appreciation to D r. Ghatu Subhash my advisor and the supervisory committee chair During my four years of work Dr Subhash has influenced me with his abundant knowledge and his enthusiasm in scientific discoveries. Without his valuable guidance and support, this dissertat ion would not have been possible. I would like to thank Professor Jam es Tulenko and Dr. Nagaraj K. Arakere my supervisory committee member s and research mentor s Professor Tulenko makes significant contributions on building my knowledge in the nuclear a rea. Dr Arakere provides important advices on my project a s a respectful scie ntist in solid mechanics. I am also thankful for Dr Yong Yang who agreed to be my supervisory committee member. Dr Yang instructs me in the field of nuclear materials and has gi ven me significant feedback on my work I also acknowledge Dr. Ronald Baney for previously bei ng on my supervisory committee. For three years, Dr Baney has helped me with his expertise in material science. I hope he will enjoy his retirement. Special than ks to the researchers in Major Analytical Instrumentation Center for their help on instrument usage and advices ; t o Kenneth McCleelan at Los Alamos National Laboratory for his professional advice on nuclear material processing ; a nd to my colleagues in Labo ratory for the Development of Advanced Nuclear Fuels and Materials and Laboratory for Dynamic Response of Advanced Materials; I would also like to thank my friends for their company with me in these years Last, but not least, I need to t hank my father Cao Ge, my mother Fengchun Xu, and my grandmother Lu Si. I cannot express how much support from my family has meant to me. It is very lucky for me to have you in my life. I love you.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 12 ABSTRACT ................................ ................................ ................................ ................................ ... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 15 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 24 Introduction of Sp ark Plasma Sintering ................................ ................................ .................. 24 Mechanism Development of Spark Plasma Sintering ................................ ............................ 25 Future and Challenge of Spark Plasma Sintering ................................ ................................ ... 30 3 DENSIFICATION OF URANAIUM DIOXIDE PREPARED BY SPARK PLASMA SINTERING ................................ ................................ ................................ ........................... 43 Introduction ................................ ................................ ................................ ............................. 43 Experimental Procedure ................................ ................................ ................................ .......... 46 Starting Powder ................................ ................................ ................................ ............... 46 SPS Sintering ................................ ................................ ................................ ................... 46 Characterization Methods ................................ ................................ ................................ 48 Result and Discussion ................................ ................................ ................................ ............. 49 Starting P owder R esults ................................ ................................ ................................ .. 49 Densification ................................ ................................ ................................ .................... 49 Microstructure ................................ ................................ ................................ ................. 55 Conclusions ................................ ................................ ................................ ............................. 57 4 MICROSTRCTURE DEVELOPMENT OF URANIUM DIOXIDE PREPARED BY SPARK PLASMA SINTERING ................................ ................................ ............................ 71 Introduction ................................ ................................ ................................ ............................. 71 Experimental Procedure ................................ ................................ ................................ .......... 72 Starting P owder ................................ ................................ ................................ ............... 72 SPS P rocessing C onditions ................................ ................................ .............................. 72 Characterization Methods ................................ ................................ ................................ 73 Result and Discussion ................................ ................................ ................................ ............. 74 Density and G rain S ize ................................ ................................ ................................ .... 74

PAGE 6

6 Microstructure D evelopment ................................ ................................ ........................... 76 O/U R atio ................................ ................................ ................................ ......................... 78 Conclusions ................................ ................................ ................................ ............................. 81 5 HARDNESS, YOUNGS MODULUS AND THERMAL CONDUCTIVITY OF URANIUM DIOXIDE PREPARED BY SPARK PLASMA SINTERING .......................... 93 Introduction ................................ ................................ ................................ ............................. 93 Experimental Procedure ................................ ................................ ................................ .......... 93 Sample Preparation ................................ ................................ ................................ .......... 93 Mechanical Property Measurement ................................ ................................ ................. 94 Thermal Conductivity Measurement ................................ ................................ ............... 94 Result and Discussion ................................ ................................ ................................ ............. 95 Mechanical Property ................................ ................................ ................................ ........ 95 Thermal Property ................................ ................................ ................................ ............. 96 Conclusion ................................ ................................ ................................ .............................. 99 6 GRAIN GROWTH OF DOPED URA NIUM DIOXIDE PREPARED BY SPARK PLASMA SINTERING ................................ ................................ ................................ ........ 109 Introduction ................................ ................................ ................................ ........................... 109 Experimental Procedure ................................ ................................ ................................ ........ 111 Result and Discussion ................................ ................................ ................................ ........... 113 Density ................................ ................................ ................................ ........................... 113 Dopant Dispersion ................................ ................................ ................................ ......... 113 Grain Size Distribution ................................ ................................ ................................ .. 114 Conclusion ................................ ................................ ................................ ............................ 121 7 CONCLUSION AND FUTURE WORK ................................ ................................ ............. 137 Conclusion ................................ ................................ ................................ ............................ 137 Future Work ................................ ................................ ................................ .......................... 139 Macro porous UO 2 Pellet ................................ ................................ .............................. 139 ThO 2 cored UO 2 Pellet ................................ ................................ ................................ .. 140 LIST OF REFERENCES ................................ ................................ ................................ ............. 150 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 159

PAGE 7

7 LIST OF TABLES Table page 2 1 List of spark plasma sintering manufacturer ................................ ................................ ...... 31 3 1 The pro cessing conditions and the resultant densities for the SPSed pellets ..................... 58 4 1 SPS processing parameters and the resulting properties of UO 2 pellets ............................ 82 4 2 XRD peak positions of the phases on the surface of the as sintered pellet ........................ 83 6 1 The processing conditions in SPS and the resulting density of different pellets .............. 123 7 1 Comparison of SPS and conventional sintering techniques ................................ ............. 142

PAGE 8

8 LIST OF FIGURES Figure page 1 1 World CO 2 emissions.. ................................ ................................ ................................ ....... 20 1 2 Shares of different energy source on the electricity generation in 2009 ............................ 21 1 3 The percentage of nuclear electricity generation in different countries in 2012. ............... 21 1 4 The schematic of UO 2 fluorite structure ................................ ................................ ............. 22 1 5 The schematic of UO 2 nuclear fuel pellet ................................ ................................ ........... 22 1 6 Steps of UO 2 pellet manufacturing ................................ ................................ ..................... 23 1 7 Spark plasma sintering system (Dr. Sinter SPS 1030) in Particle Engineering Research Center (PERC). ................................ ................................ ................................ ... 23 2 1 Scientific publication of SPS. ................................ ................................ ............................. 32 2 2 Sintered SPS Samples. ................................ ................................ ................................ ........ 33 2 3 Schematic drawing illustrating the features of an SPS apparatus ................................ ..... 34 2 4 Effect of current on assisting densification of powder in SPS ................................ ........... 35 2 5 Image of polyethylene fiber revealing the effect of electrical discharge. .......................... 36 2 6 TEM micrograph of FAST consolidated AlN. ................................ ................................ ... 37 2 7 Schematic of sample of sphere to plate sintering geometry ................................ .............. 38 2 8 Time dependence of neck growth between copper spheres and copper plates at 900 o C under different currents. ................................ ................................ ................................ 39 2 9 Relative density an d grain size versus the SPS temperature of nc YAG powder at 100MPa for 3min. ................................ ................................ ................................ .............. 40 2 10 Densification map for aluminum powder in SPS ................................ ............................... 41 2 11 Contribution to shrinkage rate from different mechanisms of mass transport for an alumina powder ................................ ................................ ................................ ................. 42 3 1 Schemetic of die assembly and sintering chamber in SPS. ................................ ................ 59 3 2 SEM image of the starting powder (UO 2.11 ) ................................ ................................ ....... 60 3 3 SPS parameter profiles during a sintering run. ................................ ................................ ... 61

PAGE 9

9 3 4 SPS sintered UO 2 Pellets of diameter 12.5 mm X 6 mm thickness ................................ ... 62 3 5 XRD pattern of the starting UO 2.11 powder, reduced UO 2.00 powder an d the spark plasma sintered (SPSed), polished and reduced compact. ................................ ................. 63 3 6 Densification profile for various heating rates during sintering. ................................ ........ 64 3 7 Densities of UO 2 pellets versus sintering temperature for various hold times, pressures and heating rates. ................................ ................................ ................................ 65 3 8 Densification rate versus temperature during sintering for two heating rates, different hold time and different maximum temperature. ................................ ................................ 66 3 9 Densification (thick line) and densification rate (thin line) versus temperature during sin tering. ................................ ................................ ................................ ............................. 67 3 10 Microstructure of sintered pellets at 1150 o C for 5min, and 96.3% TD. ............................. 68 3 1 1 Micrographs revealing grain g rowth over different isothermal hold time at 1500 o C. ....... 69 3 1 2 Plot of average grain size with hold time revealing increased rate of grain growth with hold time. ................................ ................................ ................................ ................... 70 4 1 Typical UO 2 images. ................................ ................................ ................................ ........... 84 4 2 XRD patterns of the starting powder (UO 2.16 ) and the reduced powder (UO 2.00 ). ............. 85 4 3 The influence of hold time on the density of the sintered pellets at different maximum sintering temperatures. ................................ ................................ ................................ ....... 86 4 4 The influence of hold time a nd maximum sintering temperature on the average grain size of the sintered pellets. ................................ ................................ ................................ 87 4 5 The evolution of average grain size with theoretical density of the samples sintered using a hea ting rate of 200 /min. ................................ ................................ ..................... 88 4 6 Images of selected pellets revealing the grain size density relationship. ........................... 89 4 7 SEM ima ge revealing the formation of the intragranular pores d uring densification. ....... 90 4 8 The influence of hold time and maximum sintering temperature on the resulting O/U ratio of the sintered pellets. ................................ ................................ ................................ 91 4 9 Comparison of XRD patterns of the surface of the as sintered (1450 0.5min) pellet. ................................ ................................ ................................ ................................ .. 92 5 1 Image of laser flash machine (Anter Flashline TM 3000 ) ................................ ................... 101 5 2 Influence of density on Vicker s hardness of UO 2 pellet in SPS. ................................ ...... 102

PAGE 10

10 5 3 Plot of Vickers hardness versus inverse of the grain size revealing the conformity with the Hall Petch relationship. ................................ ................................ ...................... 103 5 4 literature ................................ ................................ ................................ ........................... 104 5 5 The measured thermal diffusivity of the pe llets at 100 500 and 900 ................. 105 5 6 The thermal conductivity of the pellets at the temperatures of 100 500 and 900 ................................ ................................ ................................ ............................... 106 5 7 The thermal conductivity of the low density (<90% of the TD) pellets at the temperatures of 100 500 and 900 ................................ ................................ ....... 107 5 8 The thermal conductivity of the high density (95% 97% of the TD) pellets versus the average grain size at the temperatures of 100 500 and 900 ............................... 108 6 1 SEM image of starting powders. ................................ ................................ ..................... 124 6 2 The schematic of how the pellet was half cut. The blue area showed the examined cross section. ................................ ................................ ................................ .................... 125 6 3 Influence of doping concentration on final densities of doped UO 2 pellets. .................... 126 6 4 WDS scan across the cross section from pellet surface to its interior. ............................. 127 6 5 EMPA of the polis hed cross section of 2000 ppm Cr 2 O 3 UO 2 pellet. ............................. 128 6 6 Grain size distribution of pure UO 2 from pellet surface to the interior. ........................... 129 6 7 Influence of Cr 2 O 3 doping concentration of pure, 1000 ppm, 1500 ppm and 2000 ppm on grain size distribution of Cr 2 O 3 UO 2 ................................ ................................ ......... 130 6 8 Grain size distribution of 1500ppm ZrB 2 UO 2 pellet of 6 mm thickness. ....................... 131 6 9 Grain size distribution of 10 mm ZrB 2 UO 2 pellet. ................................ .......................... 132 6 10 Fracture SEM image of 10 mm ZrB 2 UO 2 pellet. ................................ ............................ 133 6 1 1 Current flow during sintering UO 2 in SPS. ................................ ................................ ...... 134 6 1 2 Fracture image of ZrB 2 UO 2 pellet. ................................ ................................ .................. 135 6 1 3 Summary of grain size distribution of pure UO 2 Cr 2 O 3 UO 2 and ZrB 2 UO 2 pellets. ..... 136 7 1 Near net shape UO 2 fuel pe llet made by SPS. ................................ ................................ 143

PAGE 11

11 7 2 Comparison of processing efficiency between SPS and other sintering methods to sinter high density UO 2 related nuclear fuel pellets ................................ ....................... 144 7 3 Macro pores of a sintered body by using PMMA fore former. ................................ ........ 145 7 4 PMMA beads with mean diameter of 3.6 m. ................................ ................................ 146 7 5 The morphology of PMMA UO 2 mixture. ................................ ................................ ....... 147 7 6 The microstructure of sintered PMMA UO 2 pellet. ................................ ......................... 148 7 7 ThO 2 cored UO 2 pellet sintered at 1150 o C for 5 min. ................................ ...................... 149

PAGE 12

12 LIST OF ABBREVIATIONS ASTM A merican S ociety for T esting and M aterials DC Direct current EPMA E lectron probe microanalysis FAST Field as sisted s intering t echnique GWe G iga watt electrical HREM/ARM H igh and atomic resolution electron microscopy IEA In ternational Energy Agency LWR L ight water reactor O/U Oxygen/Uranium OM Optical microscope PHWR Heavy water reactor SPS Spark plasma sintering SEM Scanning electron microscopy TD Theoretical d ensity TEM Transmission electron microscopy UHTC U ltra high temperature ceramic WNA World Nuclear Association XRD X ray diffraction

PAGE 13

13 Abstract of Dis sertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PROCESSING OF URANIUM DIOXIDE NUCLEAR FUEL PELLETS USING SPARK PLASMA SINTERING By Lihao Ge M ay 2014 Chair: Ghatu Subhash Major: Mechanical Engineering Uranium dioxide (UO 2 ), one of the most common nuclear fuels, has been applied in most of the nuclear plant these days for electricity generation. The main objective of this research is to introduce a novel method for UO 2 processing using spark plasma sintering technique (SPS). Firstly, a n investigation into the influence of processing parameters on densification of UO 2 powder during SPS is presented. A broad range of sintering tempe ratures, hold time and heating rates have been systematically varied to investigate their influence on the sintered pellet densification process The results revealed that up to 96% theoretical density (TD) pellets can be obtained at a sintering temperatur e of 1050 o C for 30s hold time and a total run time of only 10 minutes. A systematic study is performed by varying the sintering temperature between 750 to 1450 and hold time between 0.5 min to 20 min to obtain UO 2 pellets with a range of densities and grain sizes. The microstructure development in terms of g rain size, density and porosity distribution is investigated. The Oxygen/Uranium (O/U) ratio of the resulting pellets is found to decrease after SPS.

PAGE 14

14 T h e mechanical and thermal properties of UO 2 are evaluated. For comparable density and grain size, Vickers hardness and Young s modulus are in agreement with the literature value. The thermal conductivity of UO 2 increase s with t he density but the grain size in the investigated range ha s no significant in fluence. Overall, the mechanical and thermal properties of UO 2 are comparable with the one made using conventional sintering methods. Lastly, the influence of chromium dioxide (Cr 2 O 3 ) and zirconium diboride (ZrB 2 ) on the grain size of doped UO 2 fuel pellet is performed to investigate the feasibility of producing large grain size nuclear fuel using SPS. The benefits of using SPS over the conventional sintering of UO 2 are summarized. The future work of designing macro porous UO 2 pellet and thorium dioxide (Th O 2 ) cored UO 2 pellet is also proposed.

PAGE 15

15 CHAPTER 1 INTRODUCTION Along with the development of human society, the demand for energy has increase d substantially over the centuries. Energy normally comes from natural resources such as coal, natural gas and petroleum. However, due to the limited amount of these resources and the emission of greenhouse gases that cause environmental problems, it is of great impo rtance to find alternatives of these traditional resources. Hydrogen energy solar energy and wind e nergy are considered as the next generation of clean energy source s In effect, g reat efforts have been paid over the decades to improve the power efficiency of these energy sources. However, up to now these resources only take up a limited amount of the total energy demand. Nuclear e nergy has been used for power in society for decades. Although safety issues like radiation and waste disposal are still a concern, n uclear energy has many unique benefits First, l arge reserves and high heat value of nuclear fuels makes it a sustainable resource available for centuries. Based on the data provided in the World Nuclear Association (WNA) [ 1 ] at the current consumption rate, the supply of uranium will be available for more than 190 years. Another merit of nuclea r energy is its ability to provide electricity with zero emission of greenhouse gases. Nowaday s, owing to human activit y more greenhouse gases, especially carbon dioxide (CO 2 ), are emitted into the natural environment. According to the In ternational Energ y Agency (IEA) [ 2 ] coal, oil and natural gas are the three major sources of emission and respo nsible for 99.6% of the CO 2 emissions in 2009 as seen in Figure 1 1 (a). T he total amount of CO 2 emissions has also significantly increased from 14000 million ton s in 1971 to 29000 million ton s in 2009, as see n Figure 1 1 (b). The p ublic is now aware of the dangers of global warming and climate change by the excessive consumption of these fossil fuels. However, besides these fossil fuels, nuc lear energy

PAGE 16

16 is the only readily available large scale alternative to decrease the greenhouse emission gases while k eeping a continuous and adequate power supply for society. The importance of nuclear energy is revealed from the data of electricity generation in IEA. Figure 1 2 demonstrated the contributio n of different energy source for electricity genera tion in 2009 I t is noted that c oal and natural gas ranked the highest and they together provide d 62% of the total electricity. However, these fuels emit enormous CO 2 Although h ydropower contribute d 16.2% of ener gy the development of hydropower is limited due to the e nvironmental conditions. It is now still challenging to scale up electricity generation by hydropower without harming the environment. N uclear energy accounts for 13.4% of the electricity and rank s 4th n ext to hydropower which is much higher than oil and other powering sources including geothermal, solar and wind, etc This makes it a good candidate for the major provider of green energy. Currently, nuclear energy development has been emphasized in a lot of count r ies throughout the world. The proportion of electricity generated by nuclear energy [ 3 ] in 2012 in different countries is shown in Figure 1 3. It is seen that for the world s average level, 11% of the electricity is generated using nuclear energy while most western countries and South Korea are above the average level. There are m ore than 10 European countries that have the proportion more than twice of the average value. In France, more than 75% of the electricity is based on nuclear energy. In developing countries, nuclear deve lopment is still behind the world s pace where most countries have the proportion of nuclear energy below the world s average level. However, some countries like China and India, have realized the importance of nuclear energy since recent decades and have set up ambitious goals to build more nuclear plants. In 2007, China s state Council set a goal of increasing its nuclear power capacity from 7 G iga watt electrical (GWe) to 40 Gwe by 2020 and the current trends even impl y an actual more than 70

PAGE 17

17 Gwe by 20 20. In the long term, up to 2050, the total capacity is anticipated to be 300 500 Gwe [ 4 ] India, e xpected to have 14.6 GWe nuclear capacity line by 2020 and aimed to supply 25% of electricity usin g nuclear energy by 2050 [ 5 ] In the United Sta tes, although only 20% of electricity is currently generated by nuclear power, Department of Energy (DoE) has invested over $233 million for nuclear research projects in addition to infrastructure investment since 2008 to promote a sustainable nuclear indu stry in the U.S. Therefore, it is becoming a worldwide trend to develop nuclear energy in these days. In nuclear power plant s the heat is generated by burning nuclear fuels through nuclear fission. The most common fissile nuclear fuel is Uranium 235 ( 235 U ). Despite different forms of nuclear fuels, e.g. uranium nitride and uranium carbi de, the most commonly used fuel is uranium dioxide (UO 2 ). UO 2 has a calcium fluori t e crystal structure with ionic bonding between U 4+ and O 2 The structure belongs to the F m3m group with U 4+ occupying fcc positions and O 2 taking the tetrahedral sites, see Figure 1 4. Ideally, the unit cell parameter for stoichiometric UO 2 is 0.547 nm [ 6 ] However, a large non stoichiometric range from UO 1.65 UO 2.25 is achievable under differen t oxygen potentials [ 7 ] When exposed in the oxidizing atmosphere, oxygen is easily dissolved into the fluorite structure and transforms to the hyper stoichiometric phase UO 2+ x and the lattice parameter ( a ) is cont racted subject to the equation a = 0.132 x + 5.4705 by Teske, et al [ 8 ] On the other hand, the hypo stoichiometric phase UO 2 x is also achievable in a reducing atmosphere [ 9 ] The bonding of UO 2 makes it a high temperature material with a melting point around 2400 2800 o C in 1 standard atmosphere [ 10 ] which also makes it a perfect nuclear fuel for operating at high temperatures (900 1000 o C). The harmonic thermal vibration of the fcc structure provides UO 2 better thermal stability than other types of fuels. UO 2 also has better chemical stability which can be easily compati ble with different cladding materials such

PAGE 18

18 as zirc alloy and stainless steels. However, there are still some drawbacks of using the current UO 2 fuel in the nuclear reactor. For example, the low thermal conductivity of UO 2 creates a great radial thermal gra dient and the centerline temperature c an be as high as 1350 o C [ 11 ] which would shorten the cycle life of nuclear fuel. Also, the brittle ceramic form of the fuel enables the easy formation and propagat ion of cracks during the manufacturing, transportation and service, which may cause the pellet crumbling, fission gas release and eventually may threaten the safe operation of the nuclear reactor Generally, there are three mai n stages involved in the manu facturing of UO 2 pellet s used in light water reactor s (LWR) and heavy wat er reactors (PHWR) [ 12 ] 1. Producing pure UO 2 fro m UF 6 or UO 3 2. Producing high density accurately shaped ceramic UO 2 pellets. 3. Producing the rigid metal framework for the fuel assembly (mainly zirconium alloy). Of these the critical stage that determine s the fuel performance is stage 2. In another words, the fabrication of UO 2 pellet influences the final life cycle of fuel in the nuclear reactor. To reach the mass production scale, a general p rocessing procedure is introduced as follows [ 12 ] : First, the powder needs to be blended to ensure its uniformity in terms of particle size distribution and specific surface area. Then, additives such as U 3 O 8 [ 13 ] and Cr 2 O 3 [ 14 ] may be added to either enhance the sinterability or modify the microstructure for bett er performance, such as enhanced fission gas retention and higher plasticity. Other ingredients, such as lubricants may also need to be included. After the powder preparation, the UO 2 powder is then fed into a die and biaxial ly pressed into a cylinder to f orm a green pellet with a density that is around 50% of the theoretical density (% TD). Then, these green pellets are sintered in a furnace at around 1750 o C for 3 4 hours [ 9 ] with a controlled reducing atmosphere (argon hydrogen). After sinterin g the pellets are m achined to about 1cm in height and diameter. In addition, for some

PAGE 19

19 reactors, dimples and chamfers and also shaped on both end s of the pellet to cancel the effect of fission gas swelling and thermal expansion, see Figure 1 5. The flow ch art of production of UO 2 fuel is provided in Figure 1 6. As mentioned in the above paragraph, using t he current processing procedure to produce nuclear fuel pellets, multiple steps are taken and hours of processing time are required In this research, a no vel fabrication method, spark plasma sintering (SPS) see Figure 1 7, is used to process UO 2 nuclear fuel pellets. The objective of this research is to evaluate the feasibility of applyi ng SPS on processing UO 2 pellet s Firstly, an investigation into the influence of processing parameters on densification of UO 2 powder during SPS is presented. The microstructure development in terms of grain size, densi ty and porosity distribution is then investigated. The properties, such as ulus, and thermal conductivity are also evaluated. F inally s everal different concept s of UO 2 composite fuel s are investigated and the benefits of using SPS over the conventional sintering of UO 2 are summarized.

PAGE 20

20 Figure 1 1 World CO 2 emissions [ 2 ] A ) 2009 fuel shares of CO 2 emissions. B ) World CO 2 emission fr om 1971 to 2009 A B

PAGE 21

21 Figure 1 2 S h ares of different energy source on the electricity generation in 2009 [ 2 ] Figure 1 3 The per centage of nuclear electricity generation in different countries in 2012 [ 3 ]

PAGE 22

22 Figure 1 4 T he schematic of UO 2 fluorite structure Figure 1 5 T he schematic of UO 2 nuclear fuel pellet [ 15 ]

PAGE 23

23 Figure 1 6 S teps of UO 2 pellet manufacturing Figure 1 7 S park p lasma s intering system (Dr. Sinter SPS 1030) in Particle Engineering Research Center (PERC).

PAGE 24

24 CHAPTER 2 LITERATURE REVIEW Introduction of S park P lasma S interi ng As for a c omparatively novel technique for sintering, SPS provides the possibility to sinter the material with better properties than conventional sintering methods and is capable of sintering what cannot be sintered in conventional methods [ 20 ] U sing the SPS, the finer grain sizes as wel l as the higher density can be achieved with the rapid heating rate. With such incredible features, great attentions have been drawn on SPS for both laboratory and industry application s Figure 2 1 reveals the publications (on the left) and citations (on t he right) of SPS listed from 1995 to 2014 in web of knowledge citation index website It is shown that within the past 18 years, the number of the publication has grown tremendous from 2 to more than 550 per year. And more than 7000 citations are record ed by the end of 2013. This rapid growing numbers of paper indicates the popularity of SPS among the scientific researchers. In industry there are three major vendors supplying commercial SPS systems as listed in Table 2 1. Among them, t here are at least tw o major vendors of SPS that have disclosed their ambition of making SPS system [ 21 ] that is capable of manufacturin g mass ive products, large size sample, near net shape sample with complex shape [ 22 ] as seen in Figure 2 2 All of these indicate a high demand of SPS in the industrial world. T h e sintering mechanism in SPS is similar to the hot pressing and the green body is located in the graphite die with the uniaxial pressure applied to the powder using the grap hite punch, as shown in Figure 2 3 Distinctively, instead of using external heating method, there is

PAGE 25

25 pulsed direct curren t ( DC ) flows through the punches as well as the die, which means that the powders is heated from both outside and inside through Joule heating if the sample is electronically conductive With the addition of the vacuum system, the whole set can provide th e sinter period within 5 to 20 minutes and the sintering temperature 200 to 500 o C lower than the conventional counterparts hot pressing. [ 23 ] Thus, it is of great importance for us to understand the intrinsic mechanism of SPS M echanism D evelopment of S park P lasma S intering The mechanism of SPS was first described by M. Tokita [ 24 ] in 1993. In his paper he introduced four major effects caused by large current pulse that result in the sintering characteristics, that is, spark plasma, spark impa ct pressure, Joule heating and the electrical field diffusion effect. T o illustrate all these effect, how the pulse current flows through the particles in SPS is schematized in Figure 2 4

PAGE 26

26 M. Tokita didn t provide the evidence to prove the mechanism he claimed. Mamoru Omori [ 25 ] used SPS for the etching of organic fibers to claim the existence of plasma. In th e experiment, discharge is applied to t he bers in the graphite die for 3s with t he atmosphere choosing from air, N 2 Ar, and vacuum The SEM displayed in Figure 2 5 By contrast, the point s on the surface of the polyethy lene implied the generation of the plasma and t he etched area s are localized. The points where s park plasma is generated are near the place of contact ing other particles and there is no new bonds ma d e in the process that connect s By illustrating he said that it is seemly low energy plasma occurred with the energy higher than ultraviolet in the sintering by providing evidences that plasma cuts C C bonds without generating carbon [ 26 ] In the research o f Joanna R. Groza [ 27 ] the plasma hypothesis is proved by electrical discharge effects. She advocated that whereas the discharge ha sn disclosed distinct surface effects of the current discharges have been noticed in FAST consolidated specimens in YBCO. And by sintering the AlN alloy and Si 3 N 4 powder [ 28 ] she also directly observed the clean grain boundaries with direct grain to grain contact and concentrated Al 2 O 3 pockets in AlN, see Figure 2 6 and she claimed that e ither a low temperature gas plasma state or of powder particle s and there is critical voltage for the plasma generati on. The effect of the plasma cleaning particle surface is also investigated [ 29 ] by K.R. Anderson and Groza The transmission electron microscopy ( TEM ) observation of the FAST sintered NiAl shows that there are no surface oxide layers and the h igh and atomic resolution electron microscopy ( HREM/ ARM ) of the pure tungsten powder indicated the clean boundaries both of which implied the cleaning function of the pulsed electrical filed.

PAGE 27

27 Meanwhile, some researchers disagree with the concept of plasma formation in the pulsed discharge sintering process Dustin M. Hulbert [ 30 31 ] in UC Davis conducted a series of experiments in SPS using several different powders e.g. Al, Al 2 O 3 NaCl, etc. to prove there is no plasma occurred during the sintering process. He adopted three major methods in his investigation: 1. I n situ atomic emission spectroscopy 2. D irect visual observation 3. U ltrafast in situ voltage measurement In the spectroscopy, he claimed no characteristic photons were detected from the result. A nd d ue to the sensitivity issue of the AES device and human eyes and to make it more persuasive, he did the additional voltage measurement, and still, no voltage anomalies for the evidence of plasma were observed T hus, none of the experimental methods employed detecting the generation of the plasma or any sparking or arcing present in the whole stages of SPS process While Dr Hulbert made the comprehensive experiments to exclude the plasma generated between the part that have not been well considered like the extent of the pressure and the size of the particles or there may occurs the arc in some small parts but not the whole sample t hat haven t been detected. Thus, it seems premature for Dr Hulbert s conclusion. Apart from the direct discussion of the plasma generation, the researches turn ed to the current effect then. For the better understanding of role current played in the SPS, Um berto Anselmi s team composed a series of f undamental investigations First, by studying the effect of DC pulsing on the reactivity [ 32 ] of Mo/Si system, he concluded that, the RMS(root mean square) of current value between the die is steady when changing the pulse pattern,

PAGE 28

28 indicating RMS is the dominant condition to the power and temperature. In addition, direction has no effect on the reactivity as well as the different on off pulse pattern I n the successive research, t he modeling of current and heat distribution is established by utilizing the conductive and non conductive materials as samples. [ 33 ] The current effect on the reactivity was studied and basing on the previous experimental data of SPS for Mo/Si system, h e found that when under the current, MoSi 2 layer grows significantly faster than that without the current. And the presence of the current does not alter the original reaction mechanisms, [ 34 ] suggesting that the pulsing of DC, responsible for the activation of plasma, had no effect on the mass transport. F or answering how the role of current on the mass transport, James M. Frei [ 35 ] sintered copper spheres to the copper plates by puls ed electric current method, see Figure 2 7 He said that volume diffusion mechanism is the dominant factor in absence of current on the neck growth. And under the current, the enhanced neck growth see Figure 2 8 arises for the electro migration with voi d formation in the high density areas of the current. In addition, the formation and increase in defect mobility under the influence of a current was also proved by Javier E. Garay [ 36 ] in his research on Ni 3 Ti intermetallics, who indicated that changes in the concentration of point defects or mobility would result in the changes of growth mechanism As for the effect of the pressure to the rapid den sification rate, J. Reis and R. Chaim [ 37 ] made effort by using a HIP model to construct the densification maps for SPS of nano MgO. In this model p lastic yield and diffusion process are the dominat ing factors of SPS and factors of particle coarsening and grain growth are added to the HIP model as well. The result shows that the densification rates are too slow when applying HIP model to descri be the SPS experimental data, which indicat es th e additional faster kinetics in mechanisms in SPS implying the current effects. Chaim [ 38 ] also introduced the effect of the pa rticle size in

PAGE 29

29 SPS T he nano sized particles may possibly activate discharge due to the narrow gap and high surface area. He claimed that, morphological and material dependently, nanoparticles ena ble high electric charge accumulati ng even under low voltages in SPS, which may result in the surface plasma formation and particle surface heating, the densification driving factors. The example is, when sintering yttrium aluminum garnet ( YAG ) under 360MP a at 1785 o C which has the high yield stress in high temperature, the plastic deformation caused by the pressure cannot account for the rapid densification rate, see Figure 2 9 while the only reason is the nano size of the particles, on which the heated particle surface may become liquid and the liquid can aid for the grain rotation and sliding, accelerating the sintering process. Besides, Chaim and Zhijian Shen [ 39 ] also studied to control the g rain size by external pressure application regime during SPS of YAG, concluding that appl ying particles in SPS is for the suppression of further grain growth in the dens ification process. By using the numerical methods, Eugene A. Olevsky [ 23 ] considered two major factor s of den tion which contribut e to SPS mass transfer : grain boundary di usion and power law creep In these material transport factor s three driv ing sources are considered : externally applied load sintering stress (surface tension) and steady state electro mi gration while omit ting a number of factors including possible plasma effect. From above, he derived the constitutive model and the densification map for aluminum, see Figure 2 10 using a series of equations for the total shrinkage rate in SPS by combining the grain boundary diffusion and power law creep mechanisms. Additionally, to broaden the scope of mechanisms in the modeling framework, he also studied the thermal factors [ 40 41 ] in SPS that enhance the densification d uring SPS and introduced the temperature gradient driven thermal diffusion into account as well. By combining

PAGE 30

30 the constitutive model to t he shrinkage rate, the equation has been obtained and the result for the contributions of different mass transport mechanisms to the overall shrinkage rate are also calculated using the alumina powders, which is shown in the below figure 9 and which also c orresponds well with the experimental data. In addition, the shape effect of the die in the role of SPS is also concerned now by Olevsky [ 42 ] By considering the different geomet ry of the die, different thermal and electrical distribution would occur and the additional model should be established to comply with the growing demand of the industry. Future and C hallenge of S park P lasma S intering Whereas it has shown clearly improveme nt from processing condition to final product in SPS, there are still some challenges prevent ing it from substituting conventional sintering methods. First, the problem of reduction and carbon contamination due to the low oxygen partial pressure in the SPS chamber causes side effect s on the final product. It is documented in the literature that the properties of some sintered product are changed due to the carbon contamination [ 43 44 ] In our work, when sintering UO 2 at 1 45 0 o C in SPS a thin layer of uranium carbide is detected which resulting in the crumbles of the resulting compact in most cases Solvin g this problem may involve the improvement of the processing condition or changing the die materials. Noudem et al., [ 43 ] has managed t o sinter the oxides under air atmosphere by using stainless steel/tungsten carbide dies. However, this modification can be only applied at low temperatures (<1000 o C). Alternative solution is yet to be developed for materials to sinter at high temperatures in SPS Another challenge for SPS is to fabricate large scale, complicated shape products. With the increase of the sample dimension, higher temperature gradient may occur within the sample. Also, the temperature distribution may become complicated when t he complexity of the shape is increased. Both of them may have detrimental effect on the final product. As is mentioned in

PAGE 31

31 [ 45 ] some efforts have been made to improve the situation. Additional heat ing element can be applied surrounding the die to decrease the temperature gradient Also, a careful design of the die set makes it possible to sinter a relatively complex shape with identical microstructure [ 46 ] Table 2 1 List of spark plasma sintering manufacturer Manufacture Max. Load[kN] Max. Cur rent[A] Country Reference Sumitomo Coal Mining Company 50 3000 1000 20000 Japan http://sps.fdc.co.jp/ Thermal Technology 100 2500 3000 60000 U.S http://www.thermaltechnology.com / FCT Systeme GmbH 50 4000 3000 48000 Europe http://www.fct systeme.de/

PAGE 32

32 Figure 2 1. Scientific publication of SPS. A) Papers of SPS published from 1995 to 2014. B) C itations of SPS from 1995 to 2014. A B

PAGE 33

33 Figure 2 2 Sintered SPS Samples [ 22 ] A) SPS sample with diameter up to 400 m m B) C omplex near net shape parts A B

PAGE 34

34 Figure 2 3 Schematic drawing illustrating the features of an SPS apparatus [ 47 ]

PAGE 35

35 Figure 2 4 E ffect of current on assisting densification of powder in SPS [ 16 ]

PAGE 36

36 Figure 2 5 Image of polyethylene fiber revealing the effect of electrical discharge [ 25 ] A) SEM image of polyethylene fiber B) SEM image of the polyethylene fiber exposed to electrical discharge in air A B

PAGE 37

37 Figure 2 6 TEM micrograph of FAST consolidated AlN. AlN polyt ypes and Al 2 O 3 pockets are indicated by arrows [ 28 ]

PAGE 38

38 Figure 2 7 Schematic of sample of sphere to plate sintering geometry [ 48 ]

PAGE 39

39 Figure 2 8 Time dependence of neck growth between copper spheres and copper plates at 900 o C under different currents. The neck size at zero time refers to the value obtained during ramp up to temperature [ 48 ]

PAGE 40

40 Figure 2 9 Relative density and grain size versus the SPS temperature of nc YAG powder at 100M Pa for 3min [ 38 ]

PAGE 41

41 Figure 2 1 0 Densification map for aluminum powder in SPS T = 673 K Pressure=2.83MPa [ 23 ]

PAGE 42

42 Figure 2 11 Con tribution to shrinkage rate from different mechanisms of mass transport for an alumina powder applied stress 30MPa, porosity 0.3 heating rate 200 K/min, grain size 0.5 m. [ 40 ]

PAGE 43

43 CHAPTER 3 DENSIFICATION OF URANAIUM DIOXIDE PREPARED BY SPARK PLASMA SINTERI NG Introduction Uranium dioxide (UO 2 ) has been widely used as nuclear fuel in water cooled reactors since 1960s due to its excellent corrosion resistance in water steam and satisfactory compatibility with the claddings [ 49 ] However, UO 2 is a refractory oxide with a melting point in excess of 2800 o C [ 50 51 ] and therefore, requires high sintering temperature around 1700 o C in a hydrogen atmosphere for several hours using conventional sintering methods [ 52 ] Numerous efforts have been made over decades to lower the sintering temperature of UO 2 Williams et al., [ 53 ] first studied the effect of hyperstoichiometry and atmosphere on sinterability of UO 2 and claimed that high density pellets could be obtained at 1100 1400 o C in neutral atmosphere by sintering the oxide with an oxygen (O) to metal (M) ratio greater than 2.06 and then reducing in a h ydrogen atmosphere. Carrea [ 52 ] achieved low temperature sintering by increasing the BET specific surface area of UO 2 during sintering. A two step heating cycle was set up to sinter UO 2 at 1200 o C for 2 hours to achieve 97.6% theoretical density (TD) and the total run time was 8 hours. The significance of enhanced uranium diffusion coefficient which accounts for the improved sinterability in hyperstoichiometric UO 2+x was discus sed by Lay and Carter [ 54 ] Stuart and Adams [ 55 ] sintered UO 2.20~2.28 in H 2 N 2 H 2 O gas mixture at 1300 o C and achieved 96% TD in a single stage process. However, the total run time was more than 9 hours. Chevrel et al ., [ 13 ] introduced hyperstoichiometric UO 2 sintering by mixing U 3 O 8 with UO 2.08 to achieve O/U ratio of 2.25 and obtained 96.5% TD under a combined argon and oxygen atmosphere at 1100 o C for 3 hours Kutty et al., [ 9 ] studied shrinkage behavior of UO 2 under six different atmospheres, from which they concluded that shrinkage may occur at a temperature 300 400 o C lower in an oxidiz ing atmosphere than in others, but the typical total run time was several hours. Although

PAGE 44

44 progress has been achieved in improving sinterability of UO 2 sintering by these methods was still a diffusion controlled process, which means long processing time an d high temperature are required. S intering by non traditional methods has also been pursued by many researchers Amato et al., [ 56 ] utilized i nductive hot pressing while sintering reactor grade UO 2 to achieve more than 9 5% TD at 1200 o C and 8000 psi (55MPa). However, the total processing cycle and quality of sintered pellet were not reported and hence, the validation for inductive hot pressing of UO 2 remains questionable. In recent years, increasing interest has been show n in sintering UO 2 by alternative sintering techniques. Microwave sintering of UO 2 has been attempted by Yang et al., [ 57 ] who obtained 96.4% TD at 1600 o C under H 2 atmosphere for more than 1 hour. However rapid sintering has not been achieved due to formation of cracks at high heating rates (20 30K/min). Later, Yang et al., [ 58 ] introduced pressureless induction heating for rapid UO 2 sintering within 5 minutes to produce crack free 96% TD pellets. However, many critical issues such as crack formation during high er heating rate s (442K/min) and density inhomogeneity continue to be major challenges in the field. Since 1990s [ 16 ] spark plasma sintering (SPS), also known as plasma activated sintering (PAS) [ 17 ] or field assisted sintering technique (FAST) [ 29 ] has become a popular sintering method for consolidation of pow ders in various fields [ 19 ] The merits of SPS include rapid processing, homogeneity of final product, low energy consumption, e tc [ 17 ] In SPS, the green body is placed in a graphite die and uniaxial pressur e is applied to the powder through graphite punches, as shown schematically in Figure 3 1. Low voltage (3 5V) and high amperage (600 800A) pulse current are applied to the powder compact via the graphite punches. The current flows through the punches and t he die, causing rapid heating. Numerous mechanisms for

PAGE 45

45 sintering have been proposed, including resistive heating by dies and punches, direct Joule heating if the powder is electrically conductive, enhanced diffusion and electron migration by the electronic field [ 23 35 ] particle surface cleaning by the plasma [ 29 ] surface plasma formation and particle surface heating [ 38 ] spark plasma and spark impact pressure [ 16 27 31 59 ] power law creep mechanism by external pressure [ 23 ] temperature gradient driven thermal diffusion [ 40 41 ] etc There has been an intense debate on whether or not a plasma is created in the SPS process [ 31 ] Regardless, it is now well accepted that pulsed current causes joule heating at the inter particle contact areas and upon application of pressure, these particles fuse to form the final compact. In the case of dielectric materials, the microscopic level discharge caused by the electrical field is expected to occur and favor densification [ 47 ] Typical sintering times are on the order of 1 to 10 minutes at significantly lower processing temperatures than those in conve ntional sintering method. More details of SPS process are available in reference reports [ 19 23 27 35 38 47 59 60 ] Although a variety of refractory materials have been successfully processed in SPS [ 19 ] to the authors knowledge, literature on processing of nuclear fuels by SPS is limited. In this manuscript, a s ystematic study of densification during processing of UO 2 in SPS is described. The main objectives of the investigation were (i) to successfully sinter UO 2 powder in to pellets with 96 % theoretical density and (ii) to reduce the maximum sintering temperatu re and sintering duration. Pellets were produced under different heating rates, hold times, and maximum sintering temperature s The microstructure evolution and densification process under each of these processing conditions are investigated.

PAGE 46

46 Experimental Procedure Starting Powder The uranium dioxide powder was supplied by Areva Fuel System, Charlotte, NC The powder was reported to have a bulk density of 2.3g/cm 3 tap density of 2.65g/cm 3 mean particle diameter of 2.4m, and a BET surface area of 3.11m 2 /g The grain size was determined using high resolution SEM to be between 100 400 nm, see Figure 3 2. The O/U ratio for the starting powder was determined to be 2.11 by measuring the weight change before and after reducing the powder into stoichiometric UO 2 using ASTM equilibration method (C1430 07). Note that in conventional oxidative sintering [ 13 ] hyperstoichiometric powder UO 2.25 ( U 3 O 8 +UO 2.11 in 30:70 wt. ratio ) is often used. In the current SPS, this step was completely eliminated by using the as received powders. SPS Sintering

PAGE 47

47

PAGE 48

48 Characterization Methods After sintering, the chamber was allowed to cool for 1 hour and the graphite die with sintered compact wa s t aken out of the SPS chamber The pellets were ground using 240 grit sandpaper to remove the residual graphite foil and aero gel on the surface. The pellets were then reduced to stoichiometric UO 2 following the procedure described in ASTM C1430 07. For the

PAGE 49

49 micro indentation hardness measurements, the pellet surfaces were ground and polished down to 0.05 used to determine the density of each pellet. The pellet surfaces were p olished and thermally etched at 1400 o C for 1 hour to reveal the grain boundaries. X ray Diffraction (Philips APD 3720) was conducted to detect the possible presence/absence of residual carbon and formation of carbides (or intermetallic) after the sinterin g and reduction processes. S canning electron microscopy (SEM) was conducted using an accelerating voltage of 15KV and electron beam current of 10 12 A to image microstructural features. Grain size was measured by the line intercept method using ASTM E112 method covering over 100 grains in each sample. The styles used in this document are called paragraph styles. Paragraph styles are used to format the entire text within a paragraph. To apply a style, follow these instructions: Result and Discussion Startin g P owder R esults The XRD results in Figure 3 5 show that due to the hyper stoichiometry of the starting UO 2.11 powder, small peaks in addition to UO 2 peaks are revealed. The reduction process completely reduces the starting powder into the stoichiometric U O 2 powder. Also, the spark plasma sintered, polished and then reduced UO 2 compact reveals identical peaks to those of reduced UO 2.00 powder. This result implied that no residual carbon or formation of carbides (or intermetallic) occurs after the whole proc ess. Densification As stated earlier, our objective was to reduce both the maximum sintering temperature, as well as the sintering time, while achieving the desired 95% TD. Several combinations of process parameters were employed during the sintering of va rious pellets to investigate the densification process. The results are listed in Table 3 1. The densification was measured by the reduction in

PAGE 50

50 the length of the green body pellet during sintering. Figure 3 6 shows few selected plots of the die surface tem perature and densification ( o ) of UO 2 compacts for two heating rates, different hold times, and various maximum temperatures during sintering. is the displacement of the lower punch and L o is the original green body thickness before sintering. It is seen in Figure 3 6 that the densification process is different for different process conditions. In general, an s shape profile was revealed in all cases except for the 850 o C plot. When the furnace temperature is below 720 o C, only limited densification wa s observed. As the temperature increased beyond 720 o C, there was a rapid increase in densification until 1000 o C was reached This temperature range, 720 o C 1000 o C is referred to as subsequent discussions. Further increase s in the temperature did not result in further increase in densification, as revealed by the plateau of the s shaped curve. These phenomena are in good agreement with the three primary sintering stages in a conventional sintering process [ 62 ] Further densification is now possible only by the application of controlled pressure (40 MPa), as revealed by the steep rise at the end of each curve in Figure 3 6 When the temperature wa s increased beyond 1350 o C, even with the application of the controlled pressure, a slight decrease in densification was noted. This behavior may imply that the thermal expansion of the punch was greater than the shrinkage of the compact in this high temper ature regime. Thus, for all the above cases where the maximum sintering temperature was above 1000 o C, the densification behavior followed the s shaped curve. However, when the maximum sintering temperature was only 850 o C, the densification behavior was dif ferent. There was no plateau at the final sintering temperature. This behavior will be discussed in more detail in Figure 3 9 Therefore, t he effect of the maximum sintering temperature on densification can be separated into two regimes: w hen the set sinte ring temperature was below 1000 o C, say 850 o C, no s shape densification curve was

PAGE 51

51 observed. Once the set maximum sintering temperature was above 1000 o C, say 1050 o C, an s shape curve was observed and further increases in temperature would only extend the len gth of the plateau. However, no effect of the heating rate and hold time was revealed in this plot. Regardless of the heating rate (200 o C/min or 100 o C/min), hold time (0.5 20 min) and maximum set temperature (1050 o C 1500 o C), the densification range in the s curve remained the same and all sintering runs yielded the same degree of densification ( L/L o ) and almost the same final density except for the case of sintering at the lowest set temperature (850 o C), which would be further discussed in Figure 3 9 Figure 3 7 illustrates the plot of pellet density as a function of maximum sintering temperatur e under various processing conditions. By adopting a rapid heating rate (200 o C/min) and a short hold time (0.5min), when sintering was conducted at 850 o C with the non controlled pressure, the maximum density achieved in the pellet was only 78.4%. Upon appl ication of the controlled pressure (40 MPa) at the same temperature, the density increased to 86.5%. Further increase in pellet density was achieved by increasing the maximum sintering temperature and application of the controlled pressure. At a maximum si ntering temperature of 950 o C, 89.4% TD was achieved and when the maximum sintering temperature was raised to 1050 o C, 96.3% TD was achieved. Beyond this temperature, varying the processing conditions revealed little influence on the final pellet density. T he highest densities of 97% 97.6% were achieved by further increasing the maximum sintering temperature to 15 00 15 25 o C with the application of controlled pressure starting from 1350 o C. To further understand the effect of the controlled pressure on the fi nal density at these higher temperatures, another pellet was sintered at 1525 o C without the application of the controlled pressure. There was only marginal increase in the density (from 96% to 97%) under these conditions. It is obvious that the effect of p ressure is

PAGE 52

52 more significant when the controlled pressure was applied at the low temperature (850 o C) than at the high temperature (1 350 o C). Recall in Figure 3 6 that the major densification started from 720 o C and ended around 1000 o C. When controlled pressur e is applied at the low temperature say 850 o C a num ber of the powder particles were still undergoing particle adhesion and rearrangement which is described as the first stage of sintering [ 62 ] With the help of the controlled pressure, more particles may contact each other and the neck formation was much easier between the neighboring particles thus contributing to the densification. If the controlled pressure was only appl ied at the high temperature, say 1350 o C, when the final stage of sintering was attained and the densification was mainly controlled by diffusion, the effect of the controlled pressure becomes insignificant. To better reveal and compare the effects of vario us sintering parameters on densification, the densification rate d( o )/dt was plotted in Figure 3 8 as a function of die surface temperature during the sintering process at two heating rates (100 o C/min and 200 o C/min), different hold times and maximum temperatures. It can be seen that the densification rate increased ra pidly at the temperature above 700 o C, and reached the maximum rate between 800 90 0 o C. When the maximum set sintering temperature was only 850 o C, which is below the upper o C in Figure 3 6 ), the densification rate decreased rapidly to zero after the hold time. For higher sintering temperatures, the densification rate fell gradually to zero and there was little effect of varying the set maximum sintering temperature o, when the maximum sintering temperature was limited to 850 o C, the final density of a sintered pellet would be less than 87 % as shown in Figure 3 7 Therefore, a further increase in the sintering temperature is essential for continued densification and a positive densification rate as observed in Figure 3 8 The effect of the heating

PAGE 53

53 rate on the densification rate can be clearly seen in this Figure Increasing the heating rate to 200 o C/min, compared to 100 o C/min, caused almost the same initial densificatio n rate. However, a significant increase in the maximum densification rate was observed for 200 o C/min. As is shown in Fig s 3 6 and 3 7 the final density of the pellets remained almost the same in both cases. The application of controlled pressure close to the end of the sintering phase (see Figure 3 6 ) provided the additional densification rate which could be seen as a hump at the end of each curve. Also, no effect on the densification rate was seen for various hold times. Finally, regardless of the differe nces in the above process conditions, the temperature range where the positive densification rate occurs is between 720 o C and 1000 o C, which is in good agreement Figure 3 6 The effect of the controlled pressure on densification and densification rate is demonstrated in Figure 3 9 by choosing the densification plot corresponding to a pellet sintered at a maximum temperature of 850 o C. This plot was selected because at 850 o C, the UO 2 compact was still in the major d ensification phase ( Figure 3 6 ) with high densification rate ( Figure 3 8 ) and the effect of controlled pressure can be better illustrated than at higher set sintering temperatures. In Figure 3 9 the die surface temperature versus the densification and the densification rate during sintering are presented. Recall that the non controlled pressure provided the uniaxial pressure between 14 24MPa during the whole processing period while the controlled pressure (set by the user) was applied at 800 o C and then rea ched a peak pressure of 40 MPa at 850 o C. The plot reveals that without the controlled pressure (dashed line), the densification increased from around 700 o C up to the maximum sintering temperature and continued to increase even after 0.5 minutes hold time elapsed after which the temperature dropped to around 825 o C due to cooling. The densification rate curve also revealed the same phenomena. The steep

PAGE 54

54 drop of the curve at 850 o C indicates that the densification has stopped. On the other hand, by applying add itional controlled pressure (solid line), both higher densification and higher densification rate were achieved. The resulting final density was 86.5% with the controlled pressure and only 78.4% with non controlled pressure as shown in Figure 3 7 In conve ntional or oxidative sintering methods [ 9 13 52 58 ] the densification of UO 2 was controlled by the diffusion of uranium ions [ 9 ] at high sintering temperature and under long hold time (more than 3 hours) [ 9 13 52 54 ] Chevrel et al., [ 13 ] studied the effect of initial O/U ratio on densification, in which the onset of shrinkage of all compositions with different O/U ratio were observed from 800 900 o C in vacuum (10ppm of O 2 ). The highest theore tical density was reported as 96.5% for sintering temperatures up to 1100 o C at the total duration of more than three hours and with initial O to U ratio of 2.22. The influence of the atmosphere on the densification of UO 2 was studied by Kutty et al., [ 9 ] where it was reported that the lowest temperature for densification in oxidizing atmosphere (300 1000ppm of O 2 ) started around 700 o C and reached 90% TD at 1350 o C. The total run time was also above three hours. While in reducing atmosphere, the o nset temperature for densification was between 1100 o C and 1150 o C, and a maximum sintering temperature of 1600 o C was required to attain 90%TD. In our present work, the densification started at a temperature as low as 720 o C and 96.3% TD was achieved at 1050 o C for hold time of only 0.5 minutes at the maximum temperature (as shown Figure 3 7 ). The required total run time for the whole process in our research was only 10 minutes. This result is a significant improvement over the conventional sintering method [ 52 ] as well as other methods [ 9 1 3 ] mentioned above. The result also implies that diffusion of uranium is not a major factor in densification because the entire sintering cycle was only 10 minutes and the duration above 720 o C during sintering was only around 4 minutes. This result is co nsistent with the well

PAGE 55

55 documented facts about SPS that the uniaxial pressure [ 38 39 47 63 ] and the pulsed current [ 16 27 29 38 47 ] are the key factors favoring rapid densification of powders. The effec t of the uniaxial pressure was apparent: by applying the controlled pressure, both densification and densification rate have been increased. The pulsed current also contributed to the densification. Although there is still uncertainty on the formation of t he plasma, due to the dielectric nature of the oxide and only limited current passing through the particles, it is well accepted [ 47 ] that a microscopic discharge occurs on the surfaces of the particles which promotes the densification. Microstructure To evaluate the quality of the densification as well as the effect of sintering parameters, it is important to investigate the resulting microstructure of the sintered compacts. The microstructure of a sintered pellet at 1150 o C for 5 minutes is shown in Figure 3 10 where Figure 3 10 (a) shows the fracture surface and Figure 3 10 (b) reveals the polished surface after thermal etching at 1400 o C for 30 minutes. Both micrographs reveal significant inter and intra granular pores. The density of the pellet was 96.3% and the average grain size was 2.9 0.3um. Comparison of the two micrographs reveals that the thermal etching process did not cause significant grain growth. Comparing the grain size to the starting particle size (2.4m), it is noted that there was limited grain growth during the SPS process which is mainly due to the rapid heating rate and short hold time. With the help of rapid heating, coarsening dominated region in the lower temperature range (below 700 o C) has been bypassed rapidly [ 62 ] and the densification is favored in a very short time at higher temperature (above 700 o C). Interestingly, the images in Figure 3 10 (a) and (b) reveal a large number of intra granular pores. The appe arance of these intra granular pores is often a measure of over sintering in the traditional sintering process [ 62 ] where high temperatures or long sintering times are employed Under these conditions, inter granular pores break away from grain boundaries and migrate to the interior of the grains. When

PAGE 56

56 the pores were isolated from the grain boundary, it is difficult to eliminate them due to the low lattice diffusion rate [ 62 ] However, in SPS, the powder compact was unlikely to get over sintered within such a short time and at low temperature. An alternative explanation may be the effect of the high d ensification rate. It is likely that due to the high heating rate and application of the uniaxial pressure at the peak sintering temperature, neck formation occurs along the inter particle contacts and closed pores are formed before the gas between the par ticles can escape. Inter granular pores are more likely to diminish due to the grain boundary diffusion during the neck formation while most of the intra granular pores remained locked in. A similar on sintering alu mina using SPS [ 47 ] where the porosity (both inter and intra granular porosity) was significantly increased with the increase in heating rate from 50 o C/min to 580 o C/min. This residual porosity may explain the limitation in final density achieved in the UO 2 pelle ts. It is seen in Figure 3 7 that the maximum density was 97%. The remaining 3% porosity is mostly intra granular porosity and thus it is difficult to reduce the porosit y over a short sintering time. Compared to the inter granular porosity, the presence of the intra granular porosity helps the fission gas retention, which may be beneficial during the in pile operation. [ 64 ] In addition, as is mentioned in the reference [ 65 ] isolated pores and intra granular fission gas bubbles have less thermal resistance compared with the pores with the mutual interference for the same amount of the porosity. In our case, the SPS sintered pellets have a higher level of intra granular pores compared to inter granular pores and may provide a higher thermal conductivity. The hold time is shown to have little effect on densification as discussed previously. However, the influence of the hold time on the grain growth can be significant. The isothermal gr ain growth of UO 2 at 1500 o C is shown in Figure 3 11 for various hold times. As the

PAGE 57

57 temperature was held constant, the average grain sizes increased with an increase in hold time. It is seen that when the hold time was increased from 1min to 5min, there was only limited increase (13%) in grain size, while from 5min to 10min, a significant increase (53%) in grain size was observed, as seen in Figure 3 1 2 Similar phenomena have also been observed by researchers while sintering other ceramics in SPS [ 63 ] where a critical temperature or hold time existed for activation of the rapid grain growth. This phenomenon can be explained by the elimination of inter granular pores during hold time. With less pores along the grain boundaries, the force retarding the grain boundary migration is reduced, and hence a higher rate of grain growth is expected [ 62 ] However, the increase in density over a 10 minutes period was not significant. Conclusions The investigation of the influence of processing parameters during spark plasma sintering of UO 2 powder revealed that 96.3% theoretical density can be achieved in SPS at a maximum sintering temperature of 1050 o C for only 0.5 minutes hold time. As long as the sintering conditions are above these value, the variations in parameters such as the maximum sinterin g temperature (up to 1500 o C), heating rate (200 o C or 100 o C/min) and hold time (0.5 20 min) exhibit little influence on the densification behavior when moderate pressure (40 MPa) is applied. Most densification occurs in the temperature range of 720 o C 1000 o C Application of the controlled uniaxial pressure promotes the densification as well as the densification rate. The increase in hold time increases the grain size and reduces porosity.

PAGE 58

58 Table 3 1 T he processing conditions and the resultant densities f or the SPSed pellets Sintering Temperature( o C) Hold time(min) Heating Rate ( o C/min) Controlled Pressure Density (% TD) 1525 1 100 N/A 96.81% 1525 1 100 N/A 96.00.8% 1525 1 100 Y 97.20.7% 1500 1 100 Y 971.1% 1500 5 100 Y 97.10.6% 1500 10 100 Y 97 .40.4% 1500 20 100 Y 97.60.6% 1350 5 100 Y 96.40.7% 1150 5 100 Y 96.31.4% 1050 5 100 Y 96.91.1% 1150 5 200 Y 96.70.9% 1150 0.5 200 Y 95.40.9% 1050 0.5 200 Y 96.31.1% 950 0.5 200 Y 89.42.7% 850 0.5 200 Y 86.50.9% 850 0.5 200 N/A 78.40.4 %

PAGE 59

59 Figure 3 1 Schemetic of die assembly and sintering chamber in SPS.

PAGE 60

60 Figure 3 2 SEM image of the starting powder (UO 2.11 )

PAGE 61

61 Figure 3 3 SPS parameter profiles during a sintering run. The grey arrows indicate the direction of axes labels.

PAGE 62

62 Figure 3 4 SPS sintered UO 2 Pellets of diameter 12.5mm X 6mm thickness

PAGE 63

63 Figure 3 5 XRD pattern of the starting UO 2.11 powder, reduced UO 2.00 powder and the spark plasma sintered (SPSed), polished and reduced compact.

PAGE 64

64 Figure 3 6 Densification profile for various heating rates during sintering. It is seen that the majority of the densification occurs between 720 o C 1000 o C and is shown in the plot end of the si ntering phase. The arrows in the plot indicate the sharp increase in densification caused by the controlled uniaxial pressure of 40 MPa. The final density of each pellet is shown next to the sintering parameters.

PAGE 65

65 Figure 3 7 Densities of UO 2 pellets vers us sintering temperature for various hold times, pressures and heating rates. Note that all pellets were sintered with the controlled n controlled sintered with non controlled pressure.

PAGE 66

66 Figure 3 8 Densification rate versus temperature during sintering for two heating rates, different hold time and different maximum temperature. Note that the controlled pressure of 40 MPa was applied towards the end of sintering in all cases and the arrow s indicate the associated increase in densification rate.

PAGE 67

67 Figure 3 9 Densification (thick line) and densification rate (thin line) versus temperature during sintering with controlled pressure (40 MPa) and non controlled pressure (14 24 MPa) for maxim um sintering temperature of 850 o C and hold time of 0.5 minutes.

PAGE 68

68 Figure 3 10 Microstructure of sintered pellets at 1150 o C for 5min, and 96.3% TD revealing intra granular pores A) Microstructure of fracture surface before thermal etching B) Microst ructure of polished surface after thermal etching.

PAGE 69

69 Figure 3 11 Micrographs revealing grain growth over different isothermal hold time at 1500 o C. The hold time, relative density and average grain size are listed on each image.

PAGE 70

70 Figure 3 12 Plot o f average grain size with hold time revealing increased rate of grain growth with hold time. All pellets had almost the same (97% 97.4%) density.

PAGE 71

71 CHAPTER 4 MICROSTRCTURE DEVELOPMENT OF URANIUM DIOXIDE PREPARED BY SPARK PLASMA SINTERING Introduction The m icrostructur al properties, such as porosity and grain size, of UO 2 pellets play an important role in their performance as a n effective nuclear reactor fuel. The creep response yield strength, thermal conductivity, fission ga s retention, and swelling effec t are all known to be influenced by the microstructure [ 49 ] Thus, it is essential to control the microstructure of the nuclear fuel during the fabrication process. T he influences of microstructure on the properties of conventionally sint ered UO 2 have been studied extensively [ 49 50 66 ] In the conventional methods, a preformed UO 2 powder compact is sintered at 1600 1700 for several hours without the application of any pressure. It usually takes more than 10 hours to fabricate a UO 2 pellet [ 67 ] This time includes the rise time (2 4 hrs) to desired high temperature, hold time (2 4 hrs) at that temperature to cause sintering and cooling time (2 4 hrs) to room temperature. Sintering atmosphere and powder characteristics are the key parameters that influence the sinterability of UO 2 in conventional sintering methods. By modifying the surface conditions of the staring powder [ 13 67 ] or changing the sintering condition into a properly oxidized atmosphere [ 9 13 ] lower sintering temperature (1200 1400 ) could be achieved. However, in both methods, the enhanced sinterability is still controlled by the diffusion process, which means that the long processing time (several hours) is needed to achieve the desired high density. Recently, a new technique called spark plasma sintering (SPS) or field assisted sintering technique (FAST) has been gaining populari ty in the fabrication of UO 2 nuclear fuels [ 68 69 ] It has been shown that high density (>95% TD) UO 2 fuel pellets can be produced at a significantly low er sintering temperature (1050 ) within a hold time of only 30 seconds [ 69 ] D ue to th is short processing time the microstructure of the spark plasma sintered UO 2 pellets is quite different from the

PAGE 72

72 pellets made by conventional sintering methods For example, t he grain size in SPS may be smalle r and may consequently influence the thermal properties of the resulting fuel pellets. Systematic studies on the influence of processing conditions on microstructure evolution and the thermal properties are still needed to evaluate the suitability of pellets made by SPS process in nuclear reactor en vironment. I n this manuscript, the O/U stoichiometry and microstructure evolution during SPS of UO 2 pellets were investigated by varying the processing condition s such as hold time heating rate, and maximum sintering temperature. Experimental Procedure St arting P owder The uranium dioxide powder was supplied by Areva Fuel System, Hanford, WA The powder was reported to have a bulk density of 2.3 g/cm 3 tap density of 2.65 g/cm 3 and BET surface area of 3.11 m 2 /g. The powder particle grain size was determine d using high resolution SEM to be around 100 400nm as shown in Figure 4 1 (a) The stoichiometry of the starting powder was determined to be UO 2.16 by measuring the weight change before and after reducing the powder to stoichiometric UO 2 using ASTM equilibr ation method (C1430 07). SPS P rocessing C onditions Sintering was performed using a Dr. Sinter SPS 1030 system. The starting powder was loaded in to a 12.5mm inner diameter graphite die with a die wall thickness of 9.5mm. The die was covered with a graphite felt to limit the heat loss from the die. A small window was cut in the felt to focus the pyrometer for monitoring the temperature of the die surface during the sintering process. The above die assembly was placed in the sintering chamber of the SPS which was depressurized to 10 Pa. Two heating rates of 50 and 200 /min were used and a uniaxial pressure of 40 MPa was applied when the maximum sintering temperature was reached and held for desired duration of time to achieve different grain sizes in the microstructure s The detailed

PAGE 73

73 sintering cycle was described in a previous publication [ 69 ] The maximum sintering temperature was varied from 750 to 1450 and the hold time was varied from 0.5min to 20min. Pellets of size 12.5mm in diameter and 6mm in height were fabricated. Characterization Methods A ll the sintered pellets were reduced to UO 2.00 as described in [ 68 ] The O/U ratio of the resulting pellets were estimated by mea suring the weight change before and after the reduction process The density of the reduced pellets was measured using th e Archimedes method by immersing the pellets into the distilled water. X ray Diffraction (XRD) was con ducte d on the starting powders and each sintered pellet to measure the lattice parameter of UO 2 and detect the possible formation of intermetallics and ot her reaction products. The diffraction patterns were recorded at a temperature of 25 (Cu =1.5406 ). The 2 angle was scanned from 20 80 o with a step size of 0.0082 o and a n The soft ware X'Pert HighScore Plus was used to fit the peak profile. The resulting diffractogram of the starting powder UO 2.16 and the reduced powder UO 2.00 is plotted in Figure 4 2. It is seen that both UO 2.00 and UO 2.16 show typical peaks of the cubic crystal s tructure. The lattice parameter of UO 2.00 is 5.4720.0096, calculated by averaging the all the peaks in the plot. This value is consistent with the 5.47110.001 reported in literature [ 6 70 ] The lattice parameter for UO 2.16 is 5.4550.008, which shows contraction of the lattice. The inset in the plot shows that with a higher O/U ratio, the 111 peak of UO 2.16 shifted to higher diffraction angle indicating a smaller lattice parameter. The contraction of the lattice parameter of UO 2+ x can be attributed to the formation of U 5+ and U 6+ which have a smaller ionic radius and a higher

PAGE 74

74 specific charge. Detailed discussion on the influence of oxygen on the lattice parameter of UO 2 can be obtained in the literature [ 71 ] A f ield e mission s canni ng electron microscopy was used to image microstructural features. Grain size w as measured from several micrographs of the polished pellet surface using the mean line ar intercept method and observation of the fracture surfaces in SEM. Result and Discussion Density and G rain S ize A typical pellet sintered at 1050 0.5min hold time, and 40MPa pressure is shown in Figure 4 1 (b). The pellet was 96.3% of the TD. A fracture surface micrograph is shown in Figure 4 1 (c) to reveal the typical microstructure. The estimated average grain size is about 3 m. The influence of hold time at different maximum sintering temperature s on the density of the sintered pellets is plotted in Figure 4 3. It is seen that below 95% of the TD, densification can be enhanced either by increasing the maximum sintering temperature or hold time. At 750 the pellet reached only 76% of the TD when the hold time was 0.5 min. By increasing the hold time to 20 min, the density increased to 95% of the TD. The density can also be increased to 96% of the TD by increasing the temperature to 1050 with only a 0. 5 min hold time. Thus, one can increase the density of the pellet s either by increasing the hold time at a lower sintering temperature or by increasing the maximum sintering temperature but for a shorter hold time. However, after reaching 95% of the TD, in creasing either the maximum sintering temperature or hold time will result in little further densification. As a result, all the densities achieved in this study were below 98% of the TD.

PAGE 75

75 The influence of hold time and maximum sintering temperature on grai n size is plotted in Figure 4 4. At 750 with a hold time of 0.5 min, the resulting average grain size is only 0.2 m which is the same range as the starting powder (0.1 0.4 m). Even after increasing the hold time to 20 min at this temperature the averag e grain size is increased to only 0.9 m. On the other hand by increasing the maximum sintering temperature to 1050 for a 0.5 min hold time, the average grain size increases to 3 m and with further increase in maximum sintering temperature to 1450 t he average grain size increased significantly to 6.3 m with a hold time of only 0.5 min. However at a slightly lower temperature of 1350 t he maximum average grain size of 7 m wa s achieved with a hold time of 20 min. Th us the effect of maximum sintering te mperature and the hold time are clearly evident in this figure, i.e., at a given maximum sintering temperature, an increase in hold time has a marginal effect on the grain size. On the other hand, for a given hold time, an increase in sintering temperature has a dramatic effect. Further increase in the maximum sintering temperature or hold time beyond 1450 causes large cracks and eventual crumbling of pellets This may be due to the excessive thermal expansion and potential chemical interaction between the graphite dies and UO 2 powder. A summary of processing conditions and the resulting density and grain size of the pellets is given in Table 4 1. The correlation between the grain size and densi ty of the sinte red UO 2 pellets for all the sintering runs with different hold times and maximum sintering temperatures is plotted in Figure 4 5 It is seen that the grain size of UO 2 appears to be a function of the pellet theoretical density, regardless of the hold time and maximum sintering temperature. The plot reveals that during the early stage of densification until around 90% of the TD, there is only limited gr ain growth; the average grain size remains below 0.6 m until the density reaches 95% of the TD where it increase s to 0.9 m. Beyond this density, the grain size increase s dramatically to 3 m while

PAGE 76

76 there is only a slight increase in th e density. The grain s ize reaches to almost 7 m when the density reaches close to 97% of the TD. Thus, most of the dramatic increase in grain size occurs between 95% to 97% of the TD. Th e possible mechanism associated with this delay in the grain growth and densification proce ss can be explained with the help of microstructure evolution at various processing conditions as discussed below Microstructure D evelopment From Figure 4 5, it appears that t he final microstructure of the sintered pellet is a function of the level of d en sification and grain growth behavior during the sintering process. Also the grain size appears to be only a function of the density regardless of the hold time and maximum sintering temperature. This grain size density relationship was also observed in ot her materials [ 72 74 ] sintered in conventional sintering method s as well as in SPS which implies that the microstructure development of UO 2 in SPS is similar to th ose of other materials sintered using conventional methods. The lack of the grain growth at low density (<95% of the TD) can be rationalized based on the intergranular porosity that exists in the starting powder compact. As the sintering process continues under the applied pressure, the powder particles get closer but the grain growth is limited by the pinning effect of the intergranular pores present along the grain boundaries as revealed in Figure 4 6 (Aa Ac). T he se microstructure s show that the grain siz e is constant at about 0.4 m but the density values increase from 77% to 90 % of the T D Large porosity levels are clearly seen in these low densit y pellets. The pores surrounding each grain inhibit the grain boundary migration thus limiting the densifica tion process. With increasing temperature or hold time along with the applied pressure grains get closer and the densification process continues where most pores shr i nk due to the grain boundary diffusion Some pores remain attached to the grain boundary with the boundary migration and grain growth As the

PAGE 77

77 grain gets larger, a small portion of the pores are left w ithin the grain, forming intra granular pores as shown in Fig ure 4 6 (Ba B c). Once the density reaches around 95% of the TD, the grain size increa se s rapidly. The residual porosity now mainly consist ed of intra granular porosity as seen in the high magnification images on Ba B c of Fig ure 4 6 The elimination of intra granular porosity is possible by lattice diffusion which requires longer processing t ime [ 75 ] T hus, the presence of intragranular porosity limit s the final density of the pellet to 97% of the TD under the current processing condition s The mechanism for formation of intra granular porosity during densification can be clearly observed in the high mag nification SEM image of a pellet, shown in Fig ure 4 7, sintered at 750 with a density of 77% of the TD As indicated by the arrows in the image, the sequence of mechanisms which are operat ive during grain growth and subsequent densification are shown Ini tially, neck formation occurs between two adjacent grains (Step#1) With the surrounding grain s simultaneously forming similar neck s (Step#2) grain growth occurs and inter granular pores were formed (Step#3) As the densification and grain growth induced g rain boundary migration continues some pores shrank and closed while some others separated from the grain boundar ies were le ft inside the grain, forming intra granular pore s (Step#4) Based on these sequence of observations, it is now possible to rationali ze the relationship between the grai n size and densification in Figure 4 5. During the early stage, the starting powder particles cannot grow into large grains as they are farther apart. With application of the pressure and increase in temperature, the gra ins get closer and start to grow due to the neck formation. This process eventually reduces porosity and allows densification to occur beyond 95% of the T D with simultaneous rapid gra in growth as illustrated in Figure 4 5. Thereafter, a majority of the int ragranular porosity remains due to the short duration of the sintering period.

PAGE 78

78 The ability to fabricate pellets with a controlled intragranular porosity is of significant value during reactor operation. The released fission gases can be captured in these intragranular pores. On the other hand, if only intergranular pores are present, the fission gas trapped in these pores can cause grain boundary cracking which eventually leads to pellet cracking. Thus, SPS offers the benefit of control of the porosity in the UO 2 pellet. O/U R atio Recall that the starting O/U ratio of the UO 2 pow der is 2.16 However, it has been observed in our experiments that, depending on the process conditions, the stoichiometry of the sintered pellet changed, see Fig ure 4 8. The O/U r atio decreased moderately with an increase in hold time but more severely with the maximum sintering temperature. At a low sintering temperature of 750 only a moderate decrease in O/U ratio is obtained when extending the hold time from 0.5min to 20 min. However, with an increase in the maximum sintering temperature to 850 the O/U ratio dropped more rapidly, and at 1450 only a 0.5 min hold time was needed for the O/U ratio to reach the desired 2.00. No further decrease in O/U ratio was observed i n the range of processing conditions. These results implied th at chemical reaction s may occur during sintering to reduce the oxygen level in the powder. Sto ichiometry of UO 2 ( i.e. O/U ratio) plays a critical role in thermal con ductivity. A slight deviation in the O/U ratio from 2.00 may result in significant decrease in thermal conductivity [ 76 ] Thus, for efficient operation of UO 2 pellet in reactor environment, the O/U ratio of 2.00 must be maintained in a pellet. In conventional sintering methods, there are two major ways t o maintain this optimal O /U ratio [ 13 67 ] One method is to sinter the starting powder in a H 2 atmosphere, which results in the O/U ratio of the final pelle t at 2.00. However, due to t he low sinterability of UO 2.00 high temperature in the range of 1600 1700 is required

PAGE 79

79 for sintering high density pellets [ 67 ] Another way is to sinter the hyper stoichiometric powder (O/U=2.25) to promote the si nterability [ 13 ] and thus decrease the sintering temperature to 1200 1400 How ever, this method usually requires the pre sintering step o f mixing UO 2 and U 3 O 8 powder to get O/U of 2.25 and the subsequent post sintering step, i.e. reduction in H 2 atmosphere as per ASTM (C1430 07) to reduce the pellet to UO 2 .00 On the contrary, the pellets produced in SPS revealed an O/U ratio lower than th a t of the starting powder see Figure 4 8. Additionally, it is noted that, at a maximum sintering temperature o f 1350 and hold time of 5 min, the O/U ratio of 2.00 is automatically achieved in the pellet. Th is result indicated that by manipulating the processing conditions in SPS, the O/U ratio of the starting powder can be reduced into the stoichiometric form of UO 2.00 without any need for post sintering reduction step The reduction of UO 2 may occur from the chemical reaction between graphite punch and the powder. XRD was p erformed to analyze the reaction products in the sintered pellet The reaction between graphite a nd UO 2 is usually used for the preparation of uranium carbides. Depending on the atmosphere and starting powder conditions, the reaction begins at 1000 1300 [ 77 79 ] The most common reaction products are UC, UC 2 [ 78 ] or UC 1 x O x [ 80 ] depending on the amount of the residua l oxygen in the UC lattice. According to a mass spectrometry study carried by Gosse et al [ 79 ] for hyper stoichiometric UO 2 powder, the reaction starting temperature could be as low as ~700 and in this reaction, UO 2+ x is reduced into UO 2.00 without the formation of carbides. In our work, UO 2 powder was enclosed in a graphite die and heated up to 1450 which allowed the reaction between graphite and UO 2 to form carbides. The decrease of th e O/U ratio even at 750 in our study also confirms the result by Gosse et al [ 79 ] The carbi de was detected by XRD on the surface of the as sin tered pellet. It is seen in Figure 4 9 that the formation of UC was revealed on the surface of the pellet sintered at 1450 and 0.5min hold time. The six peaks clearly reveal the

PAGE 80

80 cubic structure of UC and all the peak positions match the UC reference based on Powder Diffraction File (PDF), the official database for phase identification. The slight derivation from the reference may be due to the residual oxygen in the UC lattice or thermally induced residual stress. The graphite detected at around 26 o is attributed to the residual graphite foil on the pellet surface. The rest of the peaks remained unknown. These unknown peaks may be attributed to the tetragonal UC 2 phase. However, due to the low intensity of these peaks, no exact matches could be detected. All the identified peaks in Figure 4 9 are listed in Table 4 2. After hand grinding the surface layer on the pellet with 400 grit SiC paper for two min utes only the UO 2 peaks were detected. This result sug gests that only a layer of reaction product between the graphite punch and UO 2 powder was formed on the pellet surface. The reaction between graphite and UO 2 is more likely to happen on the surface because of the higher temperature on this region. During S PS, graphite punches act as a heating source where a large amount of Joule heat is generated by the high density of electric current flowing through the punch [ 33 ] Thus, the following sequence of reactions [ 79 81 ] are possible in the surface layer. UO 2+ x + x C UO 2 + x C O UO 2 + 4C UC 2 + 2CO 3UC 2 +UO 2 4UC +2CO However, once the surface layer is removed, carbon can no longer be detected in the UO 2 pellet when observed in SEM. Due to the trace amount of carbon and limited resolution in SEM, no evidence of carbon was found in EDS either. WDS analysis was not successful due to the overlapping peak positions between uranium and carbon. Therefore, it is assumed that the reaction products between graphite and UO 2 only appeared on the surface of the pellet. Thus, SPS offers the benefit of reducing the stoichiometry of UO 2 powder into 2.00 and eliminating the additional post sintering steps by employing suitable processing conditions.

PAGE 81

81 Conclusions High density UO 2 pellets with grain sizes between 0.9 9 m were fabricate d by spark plasma sintering. It was found that high density pellets (up to 97% of the TD) can be fabricated either at moderate sintering temperature with short hold time (1 050 0.5min) or at low s intering temperature with relatively long hold time (750 20 min). An increase in sintering temperature has a significant influence on grain growth than an increase in hold time. The O/U ratio of the resulting pellets wa s found to decrease under SPS which is possibly due to a chemical reaction between graphite punch and the UO 2 powder. The mechanisms of densification and grain growth are similar to those found in traditional oxidative sintering process. Two main phases of densification and grain growth ar e observed. When the density of pellet is low (<95% of the TD), the grain growth is marginal and the porosity is mainly intergranular. When the density is high (>95% of the TD), grain growth occurs rapidly and intragranular porosity evolves. The ability to control the intragranular porosity in SPS can benefit in reactor performance by trapping the fission gases and limit pellet cracking.

PAGE 82

82 Table 4 1 SPS process ing parameters and the resulting properties of UO 2 pellets Max.Temp. ( ) Heating rate ( /min) Ho ld time (min) Density (% of TD) Grain size 750 200 0.5 77.6+/ 2.0 0.2 +/ 0. 1 750 200 5 84.5+/ 0.3 0.4 +/ 0. 1 750 200 10 90.0+/ 0.3 0.6 +/ 0. 1 750 200 20 95.1+/ 0.4 0.9 +/ 0. 1 850 200 0.5 86.1 +/ 0.9 n/a 850 200 1 94.7 +/ 0.6 n/a 850 200 5 96.4 +/ 0.1 n/a 850 200 20 95.7 +/ 1.4 2+/ 0.5 1050 200 0.5 96.3+/ 1.1 3.0+/ 0.7 1050 200 5 96.3+/ 0.9 3.4+/ 0.4 1050 200 20 95.9+/ 0.5 4.2+ / 0.5 1150 200 0.5 95.4+/ 0.9 3.9+/ 0.8 1150 200 5 95.6+/ 0.5 4.7+ / 0.9 1350 200 0.5 96.5+/ 0.1 5.0+/ 0.7 1350 200 5 96.0 +/ 0.1 5.6+ / 1.4 1350 200 20 96.6+/ 0.5 6.9+ / 1.8 1450 200 0.5 95.8+/ 1.0 6.3+/ 1.4 850 50 20 96.9+/ 0.2 2.5+/ 0.8 1350 50 20 96.9+/ 1.0 8.9+/ 1.4

PAGE 83

83 Table 4 2 XRD peak positions of the phases on the surface of the as sintered (1450 0.5min) pellet Phase hkl Peak position in this study ( o ) Reference peak position ( o ) PDF code Graphite 002 26.4 7 26.4 3 00 001 0646 UC 111 31.1 6 31.1 4 00 009 0214 200 36.1 5 36.2 4 220 52.1 3 52.1 7 311 62.0 6 61.9 8 222 65.0 2 65.0 3 400 76.7 1 76.8 1 UO 2 111 28.22 28.23 03 065 0285 200 32.71 32.71 220 46.93 46.94 311 55.68 55.68 222 58.38 58.38 400 68.56 68.55 331 75.73 75.72 420 78.06 78.05

PAGE 84

84 Fig ure 4 1 Typical UO 2 images. A ) Typical UO 2 starting powder p articles B ) UO 2 bulk pellet sintered at 105 0 for 0.5min C ) F racture surface of the sintered pellet revealing its typical micro structure and grain size.

PAGE 85

85 Fig ure 4 2 XRD pattern s of the starting powder (UO 2.16 ) and the reduced powder (UO 2.00 ) indicate the fcc structure of UO 2 and its peak shift to higher 2 region due to the excessive oxygen.

PAGE 86

86 Fig ure 4 3 The influence of hold time on the density of the sintered pellets at different maximum sintering temperature s. Note that above 1050 even 30 sec ond hold time can provide a pellet with 95% of the TD.

PAGE 87

87 Fig ure 4 4 T he influence of hold time and maximum sintering temperature on t he average grain size of the s intered pellets Note that the influence of maximum sintering tempe rature at a given hold time on the grain size is more dramatic than the influence of hold time at a given temperature.

PAGE 88

88 Fig ure 4 5 The evolution of average grain size with theoretical density of the samples sintered using a heating rate of 2 00 / min. Until almost 95% of the TD is reached, there is only marginal change in grain size and then it increases rapidly.

PAGE 89

89 Fi gure 4 6 Images of selected pellets revealing the grain size density relationship. Images Aa Ac reveal densification with limited grain growth and high porosity, whereas images Ba Bc reveal grain growth at high densities.

PAGE 90

90 Fig ure 4 7 SEM image revea ling the f ormation of the intra granular pores during densification. The density of the pellet is 77% of the TD The arrows indicate d th e steps that lead to intra granular pores during the sintering process

PAGE 91

91 Fig ure 4 8 The influence of hold time and maximum sintering temperature on the resulting O/U ratio of the sintered pellets. The upper dash line indicates the initial O/U ratio of 2.16. Note that at higher maximum sintering temperatures, the O/U ratio can be decreased to 2.00 within 30 seconds.

PAGE 92

92 Fig ure 4 9 Comparison of XRD pattern s of the surface of the as sintered ( 1450 0.5min) pellet before/after surface grinding indicates the formation of UC. The lines show the reference peak position and intensity for different planes of UC.

PAGE 93

93 CHAPTER 5 HARDNESS, YOUNGS MODULUS AND THERMAL CONDUCTIVITY OF URANIUM DIOXIDE PREPARED BY SPARK PLASMA SINTERING Introduction T he properties of sintered uranium dioxide (UO 2 ) nuclear fuel pellets play an important role in the in the reactor. T h e mechanical properties determine the stress induced crack formation during service and the thermal properties influences the temperature g radient inside the fuel pellet. Sound mechanical and thermal properties of fuels provide the reactor with a higher safety level and longer life of the nuclear fuel In literature, s everal publications have been reported regarding to mechanical and thermal response of UO 2 by using different techniques [ 61 82 85 ] However, little literature has been reported by applying mechanical and thermal tests on UO 2 sintered by spark plasma sintering (SPS). The objective of this study is to evaluate the mechanical and therm al properties of UO 2 processed by SPS and compare the results to the literature values of UO 2 processed using conventional sintering methods. In this study, UO 2 was sintered in SPS under different processing conditions to achieve different microstructures ; Vickers indentation was made to measure the hardness regarding to different densities and grain sizes; Ultrasonic measurement was made to measure the elastic properties of UO 2 and detect the possible internal cracks; Flash method was used to measure the t hermal diffusivity of UO 2 and investigate the influence of temperature densities, and grain size on the thermal properties of UO 2 Experimental Procedure Sample P reparation The uranium dioxide powder was supplied by Areva Fuel System, Charlotte, NC. The powder was reported to have a bulk density of 2.3g/cm 3 tap density of 2.65g/cm 3 mean particle diameter of 2.4m, and a BET surface area of 3.11m 2 /g. The grain size was deter mined using

PAGE 94

9 4 high resolution SEM The density is measured by Archimedes method. Th e details of density and grain size measurement s can be referred to the e xperimental part in Chapter 3 or Chapter 4. The processing conditions and resulting density and grain size of UO 2 can be referred to Table 3 1 and Table 4 1. Mechanical Property M easu rement Indentation hardness measures the resistance of a sample to deformation due to a constant compression load without destructing the sample. Ultrasonic measurement is a non destructive method by applying ultrasonic waves to the sample body and measuri ng modulus a measure of the stiffness of a material. It is also easy to detect the internal defects of the material by u ltrasonic measurement e.g. porosity and internal cracks. Thermal C onductivity M easurement T hermal conductivity is a key factor that influences UO 2 fu el performance in a reactor. With higher thermal conductivity, the thermal gradient can be reduced inside the fuel pellet to prevent pellet crack ing due to high thermal stress es T hermal conductivity measurement was calculated using the relationship k = C p where k is thermal conductivity (W/m K) C p is constant pressure specific heat (J/kg K) i s density (g/c m 3 ) a nd i s thermal diffusivity (cm 2 /s) The thermal diffusivity wa s measured at three temperatures 100 500 and 900 under N 2

PAGE 95

95 atmosphere us ing laser flash method (Anter Flashline TM 3000) see Figure 5 1 Before the measurement, the sintered pellets were sectioned into disks of 3mm thickness. Both surfaces of the disks were coated with a colloidal graphite spray to ensure constant heat absorpt ion during the measurement. The laser flash method utilizes a xenon pulse to generate heat on the front surface of the disc specimen and the temperature rise o n the rear surface is recorded [ 86 ] The thermal diffusivity ( ) is calculated by measuring the specimen thickness ( ) and the time ( ) for the temperature of the rear face of the disk to rise to half of its maximum value ( = ). Due to the difficulty in directly measuring the specific he at, the theoretical specific heat for UO 2 is used for the calculation, which is 258 ( J / kg K), 305 ( J / kg K) and 314 ( J / kg K) at 100 500 and 900 respectively [ 61 ] Result and Discussion Mechanical P ropert y Sound mechanical properties of UO 2 are essential for successful application of nuclear fuels. A s discussed in Chapter 4, the microstructure of UO 2 in SPS can be divided into densification region and grain growth region. W h en the density is below 9 5 %TD, densification dominates and there is only limited grain growth while when the density is higher t han 9 5 % TD, grain growth is dominant and densification is limited. Thus, t he results of the Vickers hardness measurements are investigated by separating the effect of porosity ( or density) and grain size. Figure 5 2 reveals the influence of porosity on the hardness below 9 5 % TD when the grain size of all the sample is around 0.4 0.9 m. It is seen that with increasing density, the hardness is increased significantly. The increase of hardness is mainly due to the decrease of the porosity in the bulk and the s ample is more resistant to the deformation when the density increased. Vickers hardness versus the inverse of average grain size is plotted in Fi gure 5 3 All of the pellets

PAGE 96

96 measured in this plot have the comparable density of 96 %TD. It is seen that the h ardness values increased linearly with decrease in the grain size and this correlation is in good agreement with the Hall Petch relation [ 62 ] The average hardness value was ar ound 6.40.4GPa which is in agreement with reference value 6.40.5GPa at comparative load level [ 82 87 ] Th E were determined using ultrasonic measurement. The correlation between the longitudinal velocity shear velocity and density are given by [ 83 ] : equations (1) and (2) and the value are plotted in Fig ure 5 4 which revealed a linear relationship a ffected by the porosity (or density) of the pellets [ 83 ] In Fig ure 5 4 the results of Gatt et al., [ 85 ] and Padel and Novion [ 88 ] are also plotted. These results show good agreement with our 20418GPa, which is in good agreement with the reference data [ 83 85 88 ] This result reveals that using SPS, intact high density pellet can be sintered without detectable crack formation. Thermal P ropert y The results of the thermal diffusivity measurements of the sintered pellets are plotted in Fig ure 5 5 at three different temperatures. For all the pellets, the thermal diffusivity decrease d with the increase in operating temperature. However, a significant difference is seen among the pellets with different processing conditions at each operati ng temperature. The pellets sintered at (1) ( 2 )

PAGE 97

97 750 for hold times of 0.5min and 5min have the lo west diffusivity values of only 0.02cm 2 /s at 100 while the one sintered at 1350 for 20min hold time showed a diffusivity of 0.033cm 2 /s, an increase of 65%. With the increase in operating temperature, this difference tended to decrease. At 900 the lowes t value was 0.008 cm 2 /s and the highest value was 0.011 cm 2 /s, an increase of only 38%. For comparison, the diffusivity values of conventionally sintered UO 2 are referenced from a review paper by Fink [ 61 ] By comparing the literature value with the ones in this study, it is noted that the pellets with the comparable densities showed similar diffusivity values, indicating the validity of the c urrent measurements Using the se values presented in Fig ure 5 5 we calculated thermal conductivity and plotted the se value s in Fig ure 5 6 as a function of temperat ure. The thermal conductivity versus temperature revealed a trend similar to that of diffusivity in Fig ure 5 5 In general, a higher sintering temperature and a lo nger hold time appear to be more favorable to yield a higher thermal diffusivity and conducti vity in a pellet. This result may be due to the improved microstructure development (i.e., higher density and larger grain size). Thus, the microstructure plays an important role in the resulting thermal properties. The low thermal conductivity of UO 2 has been a major concern while operating as a nuclear fuel in a reactor environment. From this perspective, it is essential to understand the dependence of thermal conductivity on properties such as density, grain size and other microstructural features such a s porosity distribu tions as well as the operating temperature. From Fig ure 5 5 and Fig ure 5 6 it is clear that operating temperature influences the thermal conductivity of UO 2 The heat transport in UO 2 at the low temperature (<1700 ) ge nerally occurs through the motion of phonons in the crystal lattice [ 61 ] With the increase in temperature, the mean free path of phonon is decreased because of the stronger phonon phonon

PAGE 98

98 scattering. The higher level of scattering leads to a higher resistance to heat transfer and thus the thermal c onductivity is reduced at the higher temperatures. The microstructure also has an important role in influencing the thermal conductivity. Recall from Figure 4 5 that there are two distinct mechanisms that are operative during the sintering process: densifi cation dominated regime below 9 5 % TD and grain growth dominated regime above 95% TD The pellets with the same grain size in the densification phase (<95% TD) are chosen to study the influence of density on thermal conductivity. Fig ure 5 7 shows the therma l conductivity of pellets as a function of density. The conductivity value increased with increase in density at all three operating temperatures. Due to the existence of a high level of porosity at these densities in the microstructure, the phonon scatter ing is much more severe at the gas solid interfaces. The poor thermal conductivity of the gas inside the pores further impedes the heat transfer in the solid. This relationship between density and thermal conductivity is also in agreement with the reported literature [ 61 66 ] for conventionally sintered UO 2 pellets. The relationship between grain size and thermal conductivity for all the pellets is plotted in Fig ure 5 8 In order to get a more comprehensive understanding of the influence of the grain size additional SPS runs with slower heating rate ( 50 /min) have also b een conducted and the processing conditions for these pellets are also listed in Table 4 1. As seen in Fig ure 5 8 by varying the grain size from 2 to 9 m no significant difference in thermal conductivity was noted at all three temperatures. Although the grain size in our study is as small as 2 m compared with 10 15 m in conventional sintering methods [ 64 ] the thermal conductivity of pellets prepared by SPS appears to be either in the upper range or slightly above the reported values in literature as indicated. Also at larger grain size, a higher value of thermal conductivity is seen, especially at lower temperatures (~100 ). Th is dependence of thermal conductivity on grain size can be

PAGE 99

99 explained from the interfacial resistance, which is measured in terms of Kapit za length [ 89 ] It is well noted that when the grain size is smaller than or comparable to the Kapitza length, which is ~100nm at 10 0 for UO 2 [ 89 ] the grain itself offers the same thermal resistance as the interface. Thus, the thermal conductivity is mainly dominated by grain size in this regime. Further decrease of the grain size may strongly decrease the thermal conductivity. On the other hand, when the grain size is larger than the Kapit za length, only the interfacial resistance at the grain boundaries affects the thermal conductivity. With larger grain size, there is less volume of grain boundaries that can act as the barrier to heat transport. Thus, higher thermal conductivity is expect ed for larger grain size. In our case, the typical grain size is more than 10 times larger than the Kapitza length, which indicates that the increase in thermal conductivity is mainly due to the decreased volume of grain boundaries (or larger grain size). According to the simulation work by Watanabe et al [ 89 ] there is only a marginal increase in thermal conductivity of UO 2 at 300K when the grain size is larger than 1 m This conclusion is in agreement with the result presented in our study. In addition, the average values of thermal conductivity for these pelle ts are 8.2W/mK, 4.7W/mK and 3.4W/mK at three operating temperatures as indicated by the dashed lines in Fig.13, which is also comparable with the values reported in the literature for conventionally sintered UO 2 pellets [ 61 ] as indicated by the shaded areas. Conclusion The resulting UO 2 pellets had an a modulus of 20418GPa, which are in excellent agreement with value s reported in literature for UO 2 processed by other methods. T h e Vickers hardness is increased with density for pellet with density below 95% TD and Hall Petch relationship is revealed for the pellet with density (~96% TD). No internal crack formation is detected for all UO 2 pellets. The t hermal conductivity of UO 2 pellets increased with density but the grain size in the investigated range ha d no significant

PAGE 100

100 influence. The measured thermal conductivity values up to 900 were consistent with the reported literature for conventionally sintered UO 2 pellets This research reveals that the mechanical and thermal properties of UO 2 are comparable with the literature and SPS is a feasible tool to process UO 2 fuel pellets with r eliable properties

PAGE 101

101 Figure 5 1 Image of laser flash machine (Anter Flashline TM 3000 ) located in New Engineering Building (NEB)

PAGE 102

102 Figure 5 2 Influence of density on Vickers hardness of UO 2 pellet in SPS T h e grain sizes of all measured pellets are b etween 0.4 0.9 m.

PAGE 103

103 Fig ure 5 3 Plot of Vickers hardness versus inverse of the grain size revealing the conformity with the Hall Petch relationship. Note that the data presents the mean grain size and standard deviation measured at all loads of 0.2kg, 0.5 kg and 1kg.

PAGE 104

104 Fig ure 5 4 literature

PAGE 105

105 Fig ure 5 5 The measured thermal diffusivity of the pellets at 100 500 and 900 The processing conditions and densities of the pellets are listed in the legend.

PAGE 106

106 Fig ure 5 6 The thermal conductivity of the pellets at the temperatu res of 100 500 and 900 The processing conditions and the resulting densities are listed in the legend.

PAGE 107

107 Fig ure 5 7 The thermal conductivity of the low density (<90% of the TD) pellets at the temperatures of 100 500 and 900 With increasing density, there is a significant increase in thermal conductivity, especially at the l ow temperature.

PAGE 108

108 Fig ure 5 8 The thermal conductivity of the high density (95% 97% of the TD) pellets versus the average grain size at the temperatures of 100 500 and 900 The dashed lines indicate the average value of thermal conductivity i n our wo rk while the shaded areas indicate the literature values by Fink [ 61 ]

PAGE 109

109 CHA PTER 6 GRAIN GROWTH OF DOPED URANIUM DIOXIDE PREPARED BY SPARK PLASMA SINTERING Introduction C onsiderable attention ha s been drawn on the development of large grain size UO 2 fuel pellets in nuclear industry. UO 2 with large grain sizes (e.g. 50 m) is expected to exhibit better in pile performance including high fission gas retention capacity, high creep resistance, delay of high burn up structure formation, and increased thermal conductivity [ 90 ] These benefits may reduce the fuel cycle costs and increase the accidence tolerance during operation. Currently, doping small amount of additives (usually less than ~1 wt%) ha s been proven an effective way to increase the grain sizes [ 91 ] Among which, Cr 2 O 3 is been most intensively investigated. It is revealed that by using conventional sintering methods, the grain size of C r 2 O 3 doped UO 2 can be successfully increased by 700% from 10um to 73um [ 14 ] Although the mechanism is not thoroughly understood, some experimental [ 14 92 ] and simulation [ 93 ] works indicated that Chromium prefers staying in the grain boundary of UO 2 and form ing (Cr, U, O) solid solution with low melting temperature The high mobility of the liquid phase facilitates the mass transport between the two grains an d achieves the large grain size. However, in order to achieve the large grain size, at least 6 hour hold time is required to fulfill the grain growth kinetics [ 14 ] Rec ently, spark plasma sintering (SPS) has been applied in sintering UO 2 fuel pellets [ 68 69 ] The merits of SPS include rapid processing cycle and reduced energy consumption. It is revealed that only 10 minutes total processing time is required to sinter high density UO 2 pellets via SPS [ 69 ] However, the grain size of UO 2 made in SPS is usually from 1~10 m [ 94 ] In literature, SPS is generally used for processing ma terials with fine grain size [ 95 ] Very limited publications are considering of developing a microstructure with large grain size. In order to d evelop large grain

PAGE 110

110 size UO 2 pellets using SPS one strategy is to follow the traditional processing method by doping Cr 2 O 3 into UO 2 and process the mixed powder by SPS. Another strategy is to change the electrical properties of UO 2 by adding Zirconim dibor ide (ZrB 2 ) into the starting powder. It is known that during SPS, electric current has great effect in mass transport by electromigration and Joule heating [ 23 48 ] ZrB 2 is a promising ultra high temperature ceramic (UHTC) with its high melting point of 3200 o C [ 96 ] excellent corrosion resista nce [ 97 ] as well as high electrical (9.210 6 [ 98 ] and thermal conductivities (60 140 Wm 1 K 1 ) [ 98 ] By adding ZrB 2 into UO 2 a better electrical conductivity and power dissipation in UO 2 powder is expected in SPS which may possibl y enhance the mass transport in UO 2 during grain growth. In addition, ZrB 2 has also bee n applied in nuclear industry for years. ZrB 2 has been used as an integral fuel burnable absorber (IFBA) by applying it on the out er surface of UO 2 fuel as a thin coating layer [ 99 ] During the operation of a nuclear reactor boron compensate s the reactivity effect of fuel burn up. A flatte n ing of reactivity to time curve can be then achieved by balancing the reactivity loss due to the burn up of fuel and the reactivity gain due to the burn up of the fuel poison boron [ 100 ] Thus, it is reasonable to study the effect of ZrB 2 on UO 2 grain size. The present work describes the result of a preliminary study on enhancing the grain growth of UO 2 in SPS in a short period of time by adding Cr 2 O 3 and ZrB 2 into UO 2 matrix In this study, Cr 2 O 3 wa s doped in UO 2 (Cr 2 O 3 UO 2 ) with different concentrations and ZrB 2 wa s doped in UO 2 (ZrB 2 UO 2 ) with one concentration in SPS. The grain size distributions were examined for all the doped pellets. To better understand the grain growth development of UO 2 in

PAGE 111

111 S PS pure UO 2 pellet wa s also sint ered at the same sintering temperature with different hold times f or comparison. Experimental Procedur e The grain size of the powder particle was determined using high resolution SEM as shown in Figure 6 1. The UO 2 and Cr 2 O 3 powder s had the grain size around 3 00 6 00nm and the grain size of ZrB 2 powder was about 3 5 m T h e UO 2 were blended with 1000 ppm (1000 g Cr 2 O 3 / 1g UO 2 ), 1500 ppm and 2000 ppm Cr 2 O 3 1500 ppm ZrB 2 with the aid of 2,3 Dihydroperfluoropentane in a SPEX 8 000 shaker for 1 hour. After mixing, the residual contamination was eliminated by evaporation in a fume hood. This process resulted in homogeneous dispersion of Cr 2 O 3 or ZrB 2 powder in UO 2 matrix as will be discussed in the following section Sintering was performed using a Dr. Sinter SPS 1030 system. The mixed powder was loaded in a graphite die so that a UO 2 pellet of about 6 mm in thickness could be achieved after SPS The heating rate was fixed at 200 o C/min and the uni axial pressure of 40 MPa was appli ed at the sintering temperature for all the pellets The maximum sintering temperature s of 1450 o C and 1600 o C and the hold time s of 0.5min and 40min were set for different pellets the detailed sintering procedures can be referred to our earlier publication s [ 68 69 ] It is known that sintering temperature and hold time control the activation and development of grain growth during sintering Thus, to achieve large grain size, the highest possible temperature 1600 o C and longest practical hold time of 20 min were used. To prevent the chemical reaction between graphite punch and UO 2 powder above 1450 o C [ 94 ] 160 0 o C was used only for 0.5 min hold time and at 145 0 o C, 20 min hold time was used so as to obtain the intact pellets. For pure UO 2 hold time s of 0.5 min, 10 min and 20 min were used to reveal the dev elopment of grain growth at 1600 o C. All

PAGE 112

112 pure UO 2 Cr 2 O 3 UO 2 and ZrB 2 UO 2 pellets were produced into the length of 6 mm and diameter of 12.7 mm In addition, a full size ZrB 2 UO 2 pellet of 10 mm thickness was also made to reveal the influence of thickness o n grain size distribution in SPS. The details of processing conditions in SPS was revealed in Table 6 1. After sintering, residual graphite foil was polished off the pellets and Archimedes method was used to determine the density of the pellets. The result s are listed in Table 6 1. The pellets were then longitudinally cross sectioned and polished see Figure 6 2 Thermal etching was conducted using CO 2 gas at 1290 o C for 10 30 minutes to reveal the grain size of the polished pellets while minimizing the grai n growth Optical microscope (OM) and scanning electron microscope (SEM) were used to observe the grain size The grain size distribution was measured by observing the polished surface of the pellet in OM over a sequence of steps Linear intercept method w as applie d to determine the grain size. Homogeneity of the dopant in the UO 2 matrix wa s determined by electron probe microanalysis (EPMA) using JEOL Superprobe 733 The elemen tal composition of the pellets wa s measured by wavelength dispersive X ray analys is (WD S ). P eaks for Cr K and U M were measured using Lithium Fluoride ( LIF ) crystal and Pentaerythritol ( PET ) crystal respectively on each pellet. The chromium and uranium contents were determined by scan ning the polished cr oss section of the pellet see F igure 6 2. Two scan mode s were used during analysis : the c oarse scan mode with a 20 m diameter electron beam and a 100 m step size and the fine scan mode with 1 m diameter electron beam and a 1 m step size T he examined area was scanned from the surface to the interior of the pellet as indicated in Figure 6 2. z procedure wa s applied t o convert the measured intensities to concentrations.

PAGE 113

113 Result and Discussion Density The densities of all sintered pellets were revealed in Table 6 1. It wa s seen that by sintering either at 1450 o C for 20 min or 1600 o C for 0.5 min high density (>95%TD) wa s achieved for all UO 2 Cr 2 O 3 UO 2 and ZrB 2 UO 2 pellets. For pure UO 2 pellets, increasing hold time significantly increase d the final density. When hold time wa s equal or more than 10 min, ultra high density of 99 % wa s reached. Ho wever, due to the longer hold time at 1600 o C, these pellets came out crumbled. Therefore, only 0.5 min hold time was applied at 1600 o C to process the rest of the pellets. Meanwhile, for the doped pellet, t he effect of doping concentration on the resulting density wa s revealed in Figure 6 3. It wa s seen that for all concentrations, the pellet s sintered at 1450 o C for 20 min ha d densities slightly higher than 1600 o C for 0.5 min. However in both processing conditions, the density wa s decreased by increasing the doping amount. This result m ight due to the excessive addition of Cr 2 O 3 segregated on the grain boundar ies [ 14 ] and prevent ed the densification of the UO 2 The undisso lved and segregated Cr 2 O 3 wa s also revealed in EMPA which would be discussed in the following section. For ZrB 2 UO 2 pellet, a slight ly lower density wa s revealed on the 10 mm pellet than 6 mm pellet Dopant D ispersion To realize the homogeneity of dopant mixing procedure EMPA wa s used on the polished 2000 ppm Cr 2 O 3 UO 2 pellet. The concentration of uranium and chromium were plotted as a function of the depth from surface to the interior of the pellet. The depth wa s normalized as the ratio of the total pell et length. It wa s seen in Figure 6 4 that the measured average concentration of chromium (0.16 wt%) wa s close to the calculated theoretical concentration (0.14 wt%). In m ost depths, chromium concentration s were relatively consistent. N o significant chromiu m concentration gradient wa s revealed from the pellet surface to the center. However, chromium

PAGE 114

114 peaks of high concentration were shown in the interior of the pellet, especially in the depth between 0.35 and 0.42 of the total length. This high intensity peak stands for the Cr 2 O 3 particles that 2 matrix As seen in Fig ure 6 5 a fine WDS s can with a spot size of 1 m was done across this region to better reveal the segregation of chromium Note that Figure 6 5 ( A ) i s a secondary ele ctron image showing the pores and segregated Cr 2 O 3 particle as black spots in the image Figure 6 5 ( B ) is a n X ray mapping image in which chromium wa s distinguished from the pores on the surface as the white spot in the figure. By applying WDS line scan (red dash line) on the white spot in Figure 6 5 ( B ), a high concentration (14 wt%) of Cr wa s revealed see Figure 6 5 ( C ). This indicate d a large volume of undissolved chromium with the size about 2 3 m Grain S ize D istribution The grain size distribution of pure UO 2 under different hold time s wa s revealed in Figure 6 6. A gainst the co nventional grain growth, the large grain region of UO 2 at 1600 o C was expanded from the surface to the interior of the pellet. There exist ed an exaggerated grain growth region originate d from the surface of the pellet. As seen in Figure 6 6 (A), when the hold time was 0.5 min, large grains (around 40um) only existed at 0.01 (1%) of the thickness. Below this region, the grain size was evenly distributed throughout the depth and the average grain size was below 10 m. When the hold time was extended to 10 min, the exaggerated grain growth region with grain size larger than 60 m expanded to the depth 25% of the thickness. Below this region, the grain size sharply decreased to around 25 m at the depth 35% of the thi ckness. This transformation of the grain size was also revealed in the OM image in Figure 6 6 (B) It was seen that a region containing the exaggerated large grains of 60 80 were on the top half of the image while on the bottom half, the grain size was only less than 20 However, in spite of the

PAGE 115

115 large grain size a wide gap crack propagated along several grain boundaries was also revealed in the exaggerated grain growth region. This crack might be a cause of the pellet crumbling. Between these two reg ions, an interface region was marked by two white dash lines in the image. In this region, the intermediate grain size of 30 40 was shown within the width of only about 80 which implied a rapid grain size transformation during SPS. With the increase of the hold time, the exaggerated grain growth region continually expanded to the interior area. W h en the hold time reach ed 20 min, the exaggerated grain growth region expanded to the entire half of the pellet ( 4 5% of the thickness) with the grain size of around 45 60 and there was no normal grain growth region throu ghout the pellet cross section. The grain size distribution of Cr 2 O 3 UO 2 wa s revealed in Figure 6 7. It wa s seen that for both processing conditions and all concentrations, largest grain size appeared on the surface of all pellets. Below the surface, different concentrations exhibited different influences on the grain size distribution. For 2000 ppm Cr 2 O 3 UO 2 the distribution was comparable with pure UO 2 In both processing conditions, the grain size was sharply decreased to less than 10 on the depth 10% of the thickness. Below that, the grain size maintained the same and averaged about 10 m in the rest of the interior areas However, 1000 and 1500 ppm Cr 2 O 3 increased the grain size of UO 2 In the condition of 1450 o C and 20 min, the gr ain size of 1000 ppm pellet was larger than pure UO 2 until it decreased to around 10 2 0% of the thickness see Figure 6 7 (A) Similar distribution was also in 1500 ppm pellet. The grain size was larger than the pure UO 2 in all the depths u ntil it reaching 43% of the thickness In another word, 1000 ppm and 1500 ppm Cr 2 O 3 helped increasing the grain size of UO 2 in a way that the grain size was larger on the surface and decreased with a lower rate than pure UO 2 from the surface to the interi or of the pellet. The grain size distribution of pellets made in

PAGE 116

116 1600 o C and 0.5 min was also consistent with these results It was seen in Figure 6 7 (B) that 2000 ppm pellet exhibited the same grain size distribution with pure UO 2 The grain size in 1000 ppm pellet was decreased with a slower rate than pure UO 2 and in 1500 ppm pellet, an even much slower rate was observed. Additionally, processing conditions also influenced the grain size distributions. For example, at 1600 o C, the grain size in the interi or area of the pellet was larger than 1450 o C. At 1450 o C, the grain size of 1000 ppm pellet reached 10 at the depth 20% of the thickness while at 1600 o C it was at the depth 40% of the thickness. In the case of 1500 ppm, the smallest grain size at 40% of the thickness was 17 at 1600 o C, which is much larger than the one at 1450 o C. Difference of hold tim es also accounted for the difference of the grain size on the surface of the pellet. By holding 20 min, large variation of the surface grain size was observed 1000 and 1500 ppm Cr increased the grain size to about 50 while in pure and 2000 ppm pellets, the surface grain size was only 30 By contrast, using 0.5 min hold time with an even higher temperature of 1600 o C showed almost no difference of surface grain size. Instead, c onsistent grain size of around 35 40 for all concentration s The grain size distribution of 1500 ppm ZrB 2 UO 2 wa s revealed from one surface to the other surface (bottom) of the pellet as seen in Figure 6 8. Unlike pure UO 2 and Cr 2 O 3 UO 2 pellet s for ZrB 2 UO 2 pellets, the surface grain size was not the largest throu ghout the thickness. Instead, lowest grain size of around 25 28 was revealed on both surface and bottom of the pellet. The highest grain size of around 31 36 appeared on the depth 20% or 79% of the thickness from both ends, forming a double hump distribution through the entire thickness. The average grain size of around 30 was indicated by a dash line in the plot.

PAGE 117

117 In nuclear reactor, a real fuel pellet has the length of around 10 mm. Thus, a full size ZrB 2 UO 2 pellet of 10 mm thickness was made by SPS to compare its grain size distribution with the 6 mm thick p ellet. Figure 6 9 (A) showed its distribution in the radial direction from one edge of the pellet to the other. The highest grain size of around 27 appear ed at the center of the pellet and was gradually deceased when the position was derivate from the c enter The smallest grain size of around 18 was reached on the one edge of the pellet. The average grain size was around 23 and was indicated by a dash line in the plot. Th is result indicated a thermal gradient from the pellet center to the edge s cau sing the distribution of the grain size where the die work ed as a heat sink and temperature on the edge of the pellet wa s therefore lower than on the center I n the longitudinal direction, see Figure 6 9 ( B ), the grain size was very similar with the 6 mm pellet in Figure 6 8 A double hump grain size distribution was revealed with the smallest grains of on the surface, center (32% of the thickness) and the bottom and the largest gains on the depths of around 17% or 83% of the thickness. The highest grain s ize was about 47 m while the lowest was about 25 m. The average grain size was about 33 m. Comparing with 6 mm and 10 mm ZrB 2 doped pellets with other UO 2 pellets, it is easy to conclude that 1500 ppm ZrB 2 increased the average grain size of UO 2 to the scale of 30 m in both cases. Distinguished from pure UO 2 and Cr 2 O 3 UO 2 pellets, 6 mm and 10 mm ZrB 2 UO 2 pellets ha d the same longitudinal double hump grain size distribution with largest grain size on the certain depth underneath the surface and the small est grains on the surfaces However, l arger difference between the largest and smallest grain size was revealed in 10 mm pellet which may be possibly due to the larger thermal gradient within the 10 mm pellet. The detailed microstructure of 10 mm ZrB 2 UO 2 was revealed in Figure 6 10.

PAGE 118

118 This wa s known in SPS, if the powder in the die wa s electrical ly i n solat ed, the electrical current would bypass the powder and only flow s through the die which was indicated as the blue arrows Figure 6 1 1 (A). The heat used t o sinter the pellet was generated through Joule heating and transferred to the powder from the graphite punch [ 33 94 ] In the case of pure UO 2 sintered at 1600 o C for 0.5 min, it was revealed that there was a thin layer of grains possessed the large grain size of around 36 m on the pellet surface. Meanwhile, after a closer observation, the exaggerated grain growth was also found at the four corners of the pellet as seen in Figure 6 1 1 (B). It was seen that, in the corner, this affected area with the exaggerated grain growth had the grain size of around 40 60 m. Below the corner, the grain size became unaffected and remained 5 10 m. Thus, a model describes the affected areas was presented in Figure 6 1 1 (C) where the exaggerated grain growth regions were located on the four corners and both surfaces of the pellet as well. Comparing with the electrical current flow path and the location of the affected regions, it appears that the rapid grain growth was affected by the flow of electrical current close to the UO 2 powder, which was indicated as the red lines in Figure 6 1 1 (A). Most of the affected regions were in contact with the punches and the die where the current was flowing through. The following effects of the current might facilitate the grain growth: 1) Increased Joule heating due to the contact resistance between the punch, die and the powder helped increased the local temperature and thus accelerated the diffusion process in grain boundary migration in the local area. 2) Increased ion mobility as well as defect densiti es due to the current flow and the electrical field between the electrodes (punches), which also promoted the mass transport in these regions. UO 2 has very low thermal and electrical conductivities. Due to a lack of power dissipation, excessive input power enabled the localized exaggerated grain growth, forming the affected area as indicated in Figure 6 1 1 (C). By increasing the hold time to 10 and 20 min, the

PAGE 119

119 localized power gradually dissipated into the further depth of the pellet and thus, the exaggerate d grain growth region expanded into the interior until reaching the entire pellet. The effect of Cr 2 O 3 is generally considered to be the formation of solid solutions with high mobility on the UO 2 grain boundaries This solid solution has a lower melting po int and forms a thin film of liquid phase wetting the grain boundaries, which accelerate the mass transport between the grain boundaries [ 14 92 93 ] However, no trace of liquid phase was observed and there was only limited acceleration of grain growth on certain depths of the pellet. This result might be attribut ed to the shorter processing time, which prevents the complete formation of (Cr, U, O) solid solution Thus, only moderate increase of the grain size was observed in 1000 ppm and 1500 ppm Cr 2 O 3 UO 2 pellets No influence of 2000 ppm Cr 2 O 3 was revealed in th is research. This may because of the excessive doping materials segregating on the UO 2 grain boundary, which impaired the grain boundary migration of UO 2 However, by doping ZrB 2 into U O 2 pellet, there was no exaggerated grain growth region on the surfaces Instead, a double hump grain size distribution was revealed across the longitudinal direction of the pellet. The smallest grains were on both surfaces and the largest were on certain depths underneath the surface of the pellet. However, on all four corne rs of the pellet, large grain size regions were still observed by examining its entire cross section. Figure 6 1 2 (A) showed the image of the fracture cross section of a ZrB 2 UO 2 sintered at 1600 o C with 0.5 min hold time. It was seen that on all four corne rs of the pellet, the color turned from black into silver. The color change of these areas implied the affected regions where the exaggerated grain growth happened during SPS. The grain size in these affected regions was much bigger than on the rest of the regions, thus reflecting as sliver rather than black under the light. Figure 6 1 2 (B) was a fracture SEM image of the location pointed out with a white dash rectangular in Figure 6

PAGE 120

120 1 2 (A). It was clearly seen in the image that there was no abrupt grain si ze decrease from the surface to the interior of the pellet. The grain size on the surface was around 20 30 which was even slightly smaller than the interior grain size (top left part of the image). This result was consistent with the plot in Figure 6 8 The grain size on the corner of the pellet was revealed in a fracture SEM image in Figure 6 1 2 (C). As seen in the image, the grain size on the pellet was about 60 80 which was much larger than the grains in unaffected areas as seen in Figure 6 1 2 (B) and could be viewed as the exaggerated grain growth region. Comparing with the grain size distribution in pure UO 2 the disappearance of the surface exaggerated grain growth region and the enhanced interior grain size in ZrB 2 UO 2 is much likely attributed to the unique properties of ZrB 2 Apart from the high electrical conductivity of ZrB 2 itself, boron anion is also possible to increase the electrical conductivity of UO 2 As a semiconductor, doping boron with the concentration of 10 1 9 atom /cc by ion impla ntation can increase the electrical conductivity of UO 2 four orders of magnitude larger than the undoped single crystal UO 2 from 1.5*10 3 S/cm to 6 S/cm [ 101 ] It is also known that the electrical conductivity of amorphous carbon is 10 20 S/cm [ 102 ] which means the electrical conductivity of boron doped UO 2 could be comparable with amorphous carbon. In our work, assuming boron is evenly distributed in the UO 2 matrix, 1500 ppm ZrB 2 could be equivalent with the concentration of 2*10 19 atom /cc. Thus, an increasing amount of electrical current is expected to directly flow through the UO 2 powder during SPS. This increase in electrical c urrent density could lead to the grain growth in UO 2 pellet. The effect of current density on the microstructure development wa s also studied in the field of flash sintering. In some recent papers, it were clearly revealed that a higher electric current de nsity enhanced both densification and grain growth of ceramic powders [ 103 104 ] Under the electrical field and the electric current, Frenkel

PAGE 121

121 defect pairs (vacancy and interstitial pairs) can be created [ 103 ] These increased defect s promote d the diffusion process. Meanwhile, the mobility of these charged carriers were also promoted under the electrical field, leading to a higher rate of mass transport and grain growth. Additionally, due to the increased electrical conductivity, the contact resistance between the punch and pellet surface was decreased, the input power wa s thus better dissipated into the interior of the pellet, eliminating the exaggerated grain growth area on the surface of the pellet. Therefore the result of grain s ize distribution of pure UO 2 Cr 2 O 3 UO 2 and ZrB 2 UO 2 were summarized in Figure 6 1 3 Although pure UO 2 sintered at 1600 o C with 20 min hold time provide d the highest grain size, the pellet was crumble d after SPS. Therefore, d oping ZrB 2 into UO 2 provides the most feasible way of rapidly increasing the grain size of UO 2 in SPS. However, some issues, such as the reason for the double hump longitudinal grain size distribution, have still not been fully understood yet. To further investigate, emphasis on optimizi ng the dopant concentration and SPS processing conditions are still undergoing to better understand the effect of ZrB 2 on UO 2 microstructure development Conclusion Grain growth development wa s investigated in pure UO 2 An exaggerated grain growth region with large grains (>50 ) was observed on the surface s and corners of the pellet sintered at 1600 o C By extending the hold time from 0.5min to 20min, this exaggerated grain growth region expanded from the surface to the entire pellet interior. Large grain pellet wa s partially achieved in Cr 2 O 3 UO 2 pellets. The grain size distribution of 2000ppm Cr 2 O 3 UO 2 pellet was similar with pure UO 2 1000ppm and 1500ppm Cr 2 O 3 UO 2 increased the grain size in a way th at grain size decrease d with a lower rate than pure UO 2 from surface (~40 ) to the interior (~15 ) of the pellet

PAGE 122

122 Large grain pellet with average grain size of around 30 wa s achieved in ZrB 2 UO 2 pellets. N o exaggerated grain growth region was appeared on the surface of ZrB 2 UO 2 pellet s. Radially, the grain size wa s increased from edge to the center of the pellet. L ongitudinal ly, 6 mm and 10 mm pellets exhibit similar grain size distribution where a double hump grain size distribution wa s revealed in both pellet s

PAGE 123

123 Table 6 1 T he processing conditions in SPS and the resultin g density of different pellets Doping type Maximum sintering temperature ( o C) Hold time (min) Density (% TD) Pure 1600 0.5 97.6 0.4 1600 10 99.4 0.3 (Crumbled) 1600 20 99.1 0.3 (Crumbled) 1450 20 98.7 0.4 1000 ppm Cr 2 O 3 1600 0.5 97.5 0.3 1450 20 97.5 0.1 1500 ppm Cr 2 O 3 1600 0.5 96.7 0.2 1450 20 97.3 0.1 2000 ppm Cr 2 O 3 1600 0.5 95.8 0.4 1450 20 97.3 0.2 1500 ppm ZrB 2 (6 mm in length) 1600 0.5 98.2 0.4 1500 ppm ZrB 2 (10 mm in length) 1600 0.5 97.9 0.4

PAGE 124

124 Fig ure 6 1 SE M image of starting powders. A) UO 2 B) Cr 2 O 3 C) ZrB 2

PAGE 125

125 Figure 6 2 The schematic of how the pellet wa s half cut. The blue area show ed the examined cross section. The arrow point ed out the line scan direction in EMPA

PAGE 126

126 Figure 6 3 I nfluence of dop ing concentration on final densities of doped UO 2 pellets

PAGE 127

127 Fig ure 6 4 WDS scan across the cross section from pellet surface to its interior It wa s seen that the measured average Cr concentration wa s consistent with the theoret ical concentration (2000 ppm or 0.14 wt %)

PAGE 128

128 Fig ure 6 5 EMPA of the polished cross section of 2000 ppm Cr 2 O 3 UO 2 pellet. A ) Secondary electron image revealing the dark spots on the examined cross section. B ) X ray mapping image of the same location re vealing the spots were undissolved chromium The red dash line demonstrated a WDS line scan path. C ) The plot of the fine WDS line scan quantitatively indicated 1 4 wt % of Cr which was revealed as a white spot in (B)

PAGE 129

129 Figure 6 6 G rain size distribution of pure UO 2 from pellet surface to the interior A) Grain size distribution of pellets sintered at 1600 o C with 0.5min, 10min, and 20min hold times. B) OM image of pellet with 10min hold time indicating the normal grain growth region and exaggerated grain growth region on the depth 25% of the thickness The dash lines on indicated the interface between two regions. The schematic of examined cross section of 3mm thickness is shown on the right top position in the plot.

PAGE 130

130 Figure 6 7 I nfluence of Cr 2 O 3 dopi ng concentration of pure, 1000 ppm, 1500 ppm and 2000 ppm on grain size distribution of Cr 2 O 3 UO 2 A) Grain size distribution of Cr 2 O 3 UO 2 sintered at 1450 o C with 20min hold time. B) Grain size distribution of Cr 2 O 3 UO 2 sintered at 1600 o C with 0.5min hold time

PAGE 131

131 Figure 6 8. G rain size distribution of 1500ppm ZrB 2 UO 2 pellet of 6 mm thickness. The schematic of examined cross section is shown on the right top position. The white dash line represents the OM position from pellet surface to bottom

PAGE 132

132 Figure 6 9 G rain size distribution of 10 mm ZrB 2 UO 2 pellet The schematics of the examined cross section are shown on the right top positions. The white dash line represents the OM positions. A ) R adial direction. B ) L ongitudinal direction.

PAGE 133

133 Figure 6 1 0 Fract ure SEM image of 10 mm ZrB 2 UO 2 pellet A) Position 0.1 in Figure 6 9 (A). B) Position 0.5 in Figure 6 9 (A). C) Position 0.9 in Figure 6 9 (A). D) Position 0.1 in Figure 6 9 (B). E) Position 0.5 in Figure 6 9 (B). F) Position 0.9 in Figure 6 9 (B).

PAGE 134

134 Fig ure 6 1 1 Current flow during sintering UO 2 in SPS. A) Schematic of current flow in UO 2 loaded die in SPS. B) OM image of pure UO 2 indicating the affected and unaffected areas for the exaggerated grain growth. C) Schematic of affected (red shadowed) and un affected area (non shadowed) of a pellet cross section for the exaggerated grain growth.

PAGE 135

135 Figure 6 1 2 Fracture image of ZrB 2 UO 2 pellet. A) Image of the fracture cross section. B) The fracture SEM image of pellet surface pointed out with white dash rect angular in (A) indicating the surface grain size distribution. C) The fracture SEM image on the pellet corner circled with red dash triangle in (A) indicating the grain size on the corner of the pellet.

PAGE 136

136 Figure 6 1 3 Summary of grain size distribution o f pure UO 2 Cr 2 O 3 UO 2 and ZrB 2 UO 2 pellets.

PAGE 137

137 CHARPTER 7 CONCLUSION AND FUTURE WORK Conclusion The investigation of processing parameters during spark plasma sintering of UO 2 powder revealed that high density pellets (up to 97% of the TD) can be fabricat ed either at moderate sintering temperature with short hold time (1 050 0.5 min) or at low sintering temperature with relatively long hold time (750 20 min). Most densification occur red in the temperatures between 720 o C and 1000 o C. Application of the controlled uniaxial pressure promote d the densification as well as the d ensification rate. The increase in hold time increase d the grain size and reduce d porosity. The mechanisms of densification and grain growth were similar with those found in the traditional oxidative sintering process. The O/U ratio of the resulting pellet s wa s found to decrease under SPS which wa s possibly due to a chemical reaction between graphite punch and the UO 2 powder. The resulting UO 2 pellets had an average Vickers hardness of 6.40.4 modulus of 20418 GPa, which were in excellent agreement with value s reported in literature for UO 2 processed by other methods. T h e Vickers hardness wa s increased with density and Hall Petch relationship is revealed for the high density pellet s indicated that n o interna l crack formation wa s detected for all UO 2 pellets. The t hermal conductivity of UO 2 pellets increased with density but the grain size in the investigated range ha d no significant influence. The measured thermal conductivity up to 900 were consistent with the reported literature for conventionally sintered UO 2 pellets This research reveal ed that the mechanical and thermal properties of UO 2 are comparable with the literature and SPS is a feasible tool to process UO 2 fuel pellets with reliable mechanical and thermal properties.

PAGE 138

138 Grain growth development wa s also investigated in pure UO 2 It wa s found that an exaggerated grain growth region with large grains (>50 m ) appear ed on the surface of the pellet sintered at 1600 o C By extending the hold time from 0.5 min to 20 min, this exaggerated grain growth region expanded from the surface to the entire pellet interior. Large grain size pellet wa s partially achieved in Cr 2 O 3 UO 2 pellets. In the case of 2000 ppm Cr 2 O 3 UO 2 pellet, the grain si ze distribution was similar with pure UO 2 pellet when sintered either at 1600 o C for 0.5 min or 1 45 0 o C for 20 min. In 1000 ppm and 1500 ppm Cr 2 O 3 UO 2 the grain size was increased in a way that grain size decreased with a lower rate than pure UO 2 from surfa ce ( 40 interior ( Large grain size pel let with average grain size of 30 m wa s achieved in ZrB 2 UO 2 pellets. There wa s no exaggerated grain growth region on the surface of the ZrB 2 UO 2 pellet s. Radially, the grain size wa s increased from edge to the center of the pellet. L ongitudinal ly, 6 mm and 10 mm pellets exhibit ed similar grain size distribution where a double hump grain size distribution wa s revealed in both pellet s Thus, processing of UO 2 using spark plasma sinterin g (SPS) offers numerous advantages: significantly reduced processing time, lower sintering temperature, good control of porosity, and stoichiometry, as well as good mechanical and thermal conductivity similar to those of conventionally sintered pellets Be sides, by doping Cr 2 O 3 and ZrB 2 full size (10 mm) pellets with large grain size (30 40 m) is achievable. The benefits of using SPS extend into all three stages of processing: pre sintering, sintering and post sintering. In the pre sintering stage, there is no need for hyperstoichiometric mixture of the starting powders and preparation of green compact. In the sintering step, the heating rate is faster (up to 200 /min compared to 5 10 /min in conventional sintering), hold time is shorter (minutes versus ho urs), and grain growth is better

PAGE 139

139 controlled. Recent literature also revealed that SPS is capable of fabricating difficult to sinter UO 2 composites such as UO 2 SiC [ 105 106 ] UO 2 Diamond and UO 2 CNT [ 107 ] Finally, in the post sintering, SPS pellets do not need additional step of reduction to obtain desired stoichiometry (UO 2.00 ). Also, due to the fact that the SPS pellet is made in a die ca vity, every pellet comes out with the same dimension and hence dimensional accuracy is well controlled from pellet to pellet. By designing special sintering punches, near net shape pellets can be also fabricated in SPS see Figure 7 1 [ 105 ] The efficiency of rapid processing the investigated nuclear fuel s in our work are compared with other processing techniques in Figure 7 2 By comparing with different sintering technique s the required hold time and total processing time in SPS is dramatically decreased by several magnitude s T h ese advantages of SPS over conventional sintering are further summarized in T able 7 1 The features are expected to yield significant economic benefit if large scale manufacturing using SPS can be impleme nted. Such efforts are currently underw ay [ 21 ] Nevertheless, further investigations on better understanding of pr ocessing microstructure property relationship are needed for continued progress in this field. Future W ork Macro porous UO 2 P ellet The demand of fission gas capture has arisen the need for development of macro porous UO 2 fuel pellet. By engineering the m acro pores inside the pellet, there would be less fission gas swelling and leakage and thus the operation safety of nuclear reactor could be enhanced. Macro porous UO 2 pellet has been successfully produced by using conventional sintering technique, see Fig ure 7 3 [ 108 ] However, in those techniques, complicated steps of processing prevents it scaling up the productivity into mass production. SPS is proven to be successfully process

PAGE 140

140 nuclear mater ials. The objective of processing macro porous UO 2 pellet in SPS is to shorten the processing time while engineering the pore size in SPS The macro porous UO 2 pellet can be achieved by incorporating burnable pore former with low evaporating temperature, e .g. Poly Methyl Methacrylate (PMMA) see Figure 7 4, supplied by Bangs Laboratories, IN, USA. PMMA is incorporated with UO 2 matrix by dispersing it through dry mixing for 1 h. The result of mixing can be seen in Figure 7 5 It is seen that before mixing, P MMA is addicted to the surface of large UO 2 particles (Figure 7 5 (a)) while after mixing, due to the smaller grain size of UO 2 PMMA is wrapped by small UO 2 grains which can be clearly seen in the high magnification window in Figure 7 5 (b). The mixed powde r is then sintered in SPS following the regular procedure at 1600 o C for 0.5min with 20 MPa uniaxial pressure. The resulting pellet has the density of 93 %TD with the grain size less than 1 m. The microstructure, see Figure 7 6 reveals that PMMA delayed t he densification as well as grain growth of UO 2 The future work of this study is to investigate the optimal the processing condition for the survival of macro pores It is seen in Figure 7 6 that macro pores was not able to form. Instead, only intra gran ular pores with regular shape appears. Thus, t he effect of maximum sintering temperature, hold time, heating rate and uniaxial pressure in SPS on the formation and elimination of pores is to be examined to better understand the mechanism of processing macr o porous UO 2 pellet in SPS ThO 2 cored UO 2 P ellet The idea of ThO 2 cored UO 2 fuel pellet originated from the high center line temperature in current UO 2 fuel. ThO 2 has higher thermal conductivity than UO 2 By incorporating ThO 2 into the core of UO 2 lower center line temperature is expected to be achieved. The use of ThO 2 as a nuclear fuel can be achieved by mixing it with certain amount of enriched UO 2 until it produces

PAGE 141

141 enough 233 U to sustain itself [ 109 ] SPS is proved to be a feasible technique to process nuclear materials with faster sintering rate and shorter processing time. Due to the complicated manufacturing steps that may occur during the fabricati on of ThO 2 cored UO 2 pellet in conventional sintering methods, the objective of processing ThO 2 cored UO 2 pellet in SPS is to shorten the processing time while keeping the integrity of the composite pellet. Making ThO 2 cored UO 2 green pellet in SPS die can be achieved by simultaneously loading the ThO 2 powder in the center and UO 2 powder in the peripheral areas into a die All the other procedures can be the similar with the regular SPS pro cedure s Additionally, d ue to the thermal coefficients mismatch, preliminary study would be focused on tailoring the processing condition as well as the ratio between ThO 2 and UO 2 to achieve the intact pellet without the crack generation. The demonstrated ThO 2 cored UO 2 pellet is seen in Figure 7 7 The maximum sintering temperature is 1150 o C with 5 min hold time. As seen in the figure, cracks surrounding the interface between ThO 2 and UO 2 indicate the thermal stress induced by thermal coefficient mismatch.

PAGE 142

142 Table 7 1 Comparison of SPS and conventional sintering techniques Sintering stage Feature SPS Conventional Pre sintering Modification to starting powder Not required Required Cold compaction of green body Not required Required Sintering Tempe rature ramp rate 50 200 /min 2 10 /min Maximum sintering temperature 750 1450 1600 1700 (H 2 ) 1200 1400 (Oxidative) Hold time 0.5 20 min 1 10 hrs Total sintering run time <1hr ~15 hrs. Pressure 20 80 MPa No Sintering environment Vacuum (~10Pa) Ga seous environment Dimensional control Yes Limited Pellet stoichiometry during sintering Changed Unchanged Control of Grain growth High Low Ability to produce near net shape pellets Yes L imited Ability to sinter difficult to sinter materials Yes L imited Post sintering Requires reduction of sintered pellet to desired stoichiometry No (if initial powder is at the right stoichiometry) Yes Pellet requires additional machining to obtain desired final dimensions May not be required Yes

PAGE 143

143 Figu re 7 1 Near net shape UO 2 fuel pellet made by SPS

PAGE 144

144 Figure 7 2 Comparison of processing efficiency between SPS and other sintering methods to sinter high density UO 2 related nuclear fuel pellets A) Comparison of t ot al processing time B) Comparison of hold time at maximum sintering temperature .. A B

PAGE 145

145 Fi gure 7 3 M acro pores of a sintered body by using PMMA fore former [ 108 ]

PAGE 146

146 Figure 7 4 PMMA beads with mean diameter o f 3.6 m

PAGE 147

147 Figure 7 5 T he morphology of PMMA UO 2 mixture A) B efore mixing. B ) A fter mixing. The window on the right sid e indicates the high magnification picture of PMMA beads.

PAGE 148

148 Figure 7 6 T he microstructure of sintered PMMA UO 2 pellet.

PAGE 149

149 Figure 7 7 ThO 2 cored UO 2 pell et sintered at 1150 o C for 5 min. Note that the white is ThO 2 and black is UO 2

PAGE 150

150 LIST OF REFERENCES 1. "World e nergy n eeds and n uclear p ower", World Nuclear Association, January 2014 [ http://www.world nuclear.org/info/Current and Future Generation/World Energy Needs and Nuclear Power/ ] 2. IEA: Key w orld e nergy s tatistics 2011 : OECD Publishing. [ https://www.iea.org/publi cations/freepublications/publication/key_world_energy_stats 1.pdf ] 3. "World n uclear p ower r eactors & u ranium r equirements", World Nuclear Association, January 2014 [ http://www.world nuclear.org/info/Facts and Figures/World Nuclear Power Reactors and Uranium Requirements/ ] 4. Cao B, Wu J, Yang T, Ma X, Chen Y: Preliminary Study on Nuclear Fuel Cycle Scenarios of China before 2050 Ene rgy Procedia 2013, 39 (0):294 299. 5. "Nuclear p ower in India", World Nuc lear Assosication, January 2014 [ http://www.world nuclear.org/info/Country Profiles/Countries G N/India/ ] 6. Cardinaels T, Govers K, Vos B, Van den Berghe S, Verwerft M, de Tollenaere L, Maier G, Delafoy C: Chromia doped UO2 fuel: Investigation of the lattice parameter Journal of Nuclear Materials 2012, 424 (1 3):252 260. 7. Moore E, Guneau C, Cr ocombette J P: Diffusion model of the non stoichiometric uranium dioxide Journal of Solid State Chemistry 2013, 203 (0):145 153. 8. Teske K, Ullmann H, Rettig D: Investigation of the oxygen activity of oxide fuels and fuel fission product systems by solid electrolyte techniques. Part I: Qualification and limitations of the method Journal of Nuclear Materials 1983, 116 (2 3):260 266. 9. Kutty TRG, Hegde PV, Khan KB, Basak U, Pillai SN, Sengupta AK, Jain GC, Majumdar S, Kamath HS, Purushotham DSC: Densifica tion behaviour of UO2 in six different atmospheres Journal of Nuclear Materials 2002, 305 (2 3):159 168. 10. Hausner H: Determination of the melting point of uranium dioxide Journal of Nuclear Materials 1965, 15 (3):179 183. 11. Wang J: Developing a high thermal conductivity nuclear fuel with silicon carbide additives University of F l orida; 2008. 12. "Nuclear f uel f abrication", World Nuclear Association, October 2013 [ http://www.world nuclear.org/info/Nuclear Fuel Cycle/Conversion Enrichment and Fabrication/Fuel Fabrication/ ]

PAGE 151

151 13. Chevrel H, Dehaudt P, Francois B, Baumard JF: Influence of surface phenomena during sintering of overs toichiometric uranium dioxide UO2 + x Journal of Nuclear Materials 1992, 189 (2):175 182. 14. Bourgeois L, Dehaudt P, Lemaignan C, Hammou A: Factors governing microstructure development of Cr2O3 doped UO2 during sintering Journal of Nuclear Materials 200 1, 297 (3):313 326. 15. D. Pramanik MR, G.V.S.H. Rao, R.N. Jayaraj: Innovative p rocess t echniques to o ptimize q uality and m icrostructure of UO2 f uel for PHWRs in India In: International Atomic Energy Agency Technical Committee Meeting (IAEA TECDOC 1654); Villigen 2009. 16. Tokita M: Mechanism of spark plasma sintering, Proceedings of the International Symposium on Microwave, Plasma and Thermochemical Processing of Advanced Materials, ed S. Miyake and M. Samandi, JWRI, Osaka Universities, Japan In ; 199 7: 69 76. 17. Yamazaki K, Risbud SH, Aoyama H, Shoda K: PAS (Plasma activated sintering): Transient sintering process control for rapid consolidation of powders Journal of Materials Processing Technology 1996, 56 (1 4):955 965. 18. Anderson KR, Groza JR, Fendorf M, Echer CJ: Surface oxide debonding in field assisted powder sintering Materials Science and Engineering: A 1999, 270 (2):278 282. 19. Orr R, Licheri R, Locci AM, Cincotti A, Cao G: Consolidation/synthesis of materials by electric current activa ted/assisted sintering Materials Science and Engineering: R: Reports 2009, 63 (4 6):127 287. 20. Munir Z, Anselmi Tamburini U, Ohyanagi M: The effect of electric field and pressure on the synthesis and consolidation of materials: A review of the spark pla sma sintering method Journal of Materials Science 2006, 41 (3):763 777. 21. Aalund R: Unveiling s park p lasma s intering h igh t hroughput p rocessing In: Processing and Properties of Advanced Ceramics and Composites II. John Wiley & Sons, Inc.; 2010: 1 10. 22. H. U. Kessel JH, R. Kirchner, T. Kessel: Rapid sintering of novel materials by FAST/SPS Future development to the point of an industrial production process with high cost efficiency [ http://www.fct systeme.de / ] 23. Olevsky E, Froyen L: Constitutive modeling of spark plasma sintering of conductive materials Scripta materialia 2006, 55 (12):1175 1178.

PAGE 152

152 24. Tokita M: mechanism of spark plasma sintering In : Sumltomo Coal Ming Company, Ltd. [ http://xa.yimg.com/kq/groups/3862917/2054596553/name/SUMITOMO%2BREVIEW Spark Plasma Sintering.pdf / ] 25. Omori M: Sintering, consolidation, reaction and crystal growth by the spark plasma system (SPS) Materials Science and Engine ering: A 2000, 287 (2):183 188. 26. Omori M: Basic r esearch and i ndustrial p roduction u sing the s park p lasma s ystem (SPS) Advances in Microwave and Radio Frequency Processing 2006:745 754. 27. Groza JR, Zavaliangos A: Sintering activation by external ele ctrical field Materials Science and Engineering A 2000, 287 (2):171 177. 28. Groza JR, Garcia M, Schneider JA: Surface effects in field assisted sintering Journal of Materials Research 2001, 16 (01):286 292. 29. K.R. Anderson JRG, M. Fendorf, C.J. Echer: powder sintering Materials Science and Engineering A 1999, 270 (1999):278 282. 30. Hulbert DM: The a bsence of p lasma in s park p lasma s intering" Journal of Applied Physics 2008 104 ( 033305 ) 31. Hulbert DM, Anders A, Anders son J, Lavernia EJ, Mukherjee AK: A discussion on the absence of plasma in spark plasma sintering Scripta Materialia 2009, 60 (10):835 838. 32. Chen W, Anselmi Tamburini U, Garay JE, Groza JR, Munir ZA: Fundamental investigations on the spark plasma sinte ring/synthesis process:: I. Effect of dc pulsing on reactivity Materials Science and Engineering A 2005, 394 (1 2):132 138. 33. Anselmi Tamburini U, Gennari S, Garay JE, Munir ZA: Fundamental investigations on the spark plasma sintering/synthesis process: II. Modeling of current and temperature distributions Materials Science and Engineering A 2005, 394 (1 2):139 148. 34. Anselmi Tamburini U, Garay JE, Munir ZA: Fundamental investigations on the spark plasma sintering/synthesis process:: III. Current effe ct on reactivity Materials Science and Engineering: A 2005, 407 (1 2):24 30. 35. Frei JM, Anselmi Tamburini U, Munir ZA: Current effects on neck growth in the sintering of copper spheres to copper plates by the pulsed electric current method Journal of a pplied physics 2007, 101 :114914. 36. Garay JE, Anselmi Tamburini U, Munir ZA: Enhanced growth of intermetallic phases in the Ni Ti system by current effects Acta materialia 2003, 51 (15):4487 4495.

PAGE 153

153 37. Chaim JRR: ing of nanocrystalline MgO ceramics: Particle coarsening and grain growth effects Materials Science and Engineering A 2008, 491 :356 363. 38. Chaim R: Densification mechanisms in spark plasma sintering of nanocrystalline ceramics Materials Science and En gineering: A 2007, 443 (1 2):25 32. 39. Chaim R, Shen Z: Grain size control by pressure application regime during spark plasma sintering of Nd YAG nanopowders Journal of Materials Science 2008, 43 (14):5023 5027. 40. Olevsky EA, Froyen L: Impact of therma l diffusion on densification during SPS Journal of the American Ceramic Society 2009, 92 :S122 S132. 41. Olevsky EA, Kandukuri S, Froyen L: Consolidation enhancement in spark plasma sintering: Impact of high heating rates Journal of applied physics 2007, 102 (11):114913 114913 114912. 42. Olevsky E, Khaleghi E, Garcia C, Bradbury W: Fundamentals of s park p lasma s intering: a pplications to n et s haping of h igh s trength t emperature r esistant c omponents Trans Tech Publ 2010 : 412 415. 43. Noudem JG, Quet el Weben S, Retoux R, Chevallier G, Estourns C: Thermoelectric atmosphere Scripta Materialia 2013, 68 (12):949 952. 44. Bernard Granger G, Benameur N, Guizard C, Nygren M: Influence of graphite contamination on the optical propert ies of transparent spinel obtained by spark plasma sintering Scripta Materialia 2009, 60 (3):164 167. 45. Eriksson M, Radwan M, Shen Z: Spark plasma sintering of WC, cemented carbide and functional graded materials International Journal of Refractory Met als and Hard Materials 2013, 36 (0):31 37. 46. Voisin T, Durand L, Karnatak N, Le Gallet S, Thomas M, Le Berre Y, Castagn J F, Couret A: Temperature control during Spark Plasma Sintering and application to up scaling and complex shaping Journal of Materi als Processing Technology 2013, 213 (2):269 278. 47. Shen Z, Johnsson M, Zhao Z, Nygren M: Spark p lasma s intering of a lumina Journal of the American Ceramic Society 2002, 85 (8):1921 1927. 48. Frei JM, Anselmi Tamburini U, Munir ZA: Current effects on nec k growth in the sintering of copper spheres to copper plates by the pulsed electric current method Journal of Applied Physics 2007, 101 (11):114914.

PAGE 154

154 49. Stehle H, Assmann H, Wunderlich F: Uranium dioxide properties for LWR fuel rods Nuclear Engineering a nd Design 1975, 33 (2):230 260. 50. Lyons MF, Boyle RF, Davies JH, Hazel VE, Rowland TC: UO2 properties affecting performance Nuclear Engineering and Design 1972, 21 (2):167 199. 51. H H: Determination of the melting point of uranium dioxide Journal of N uclear Materials 1965, 15 (3):179 183. 52. Carrea AJ: Sinering of uranium dioxide in an atmosphere of controlled hydrogen content Journal of Nuclear Materials 1963, 8 (2):275 277. 53. Williams J, Barnes E, Scott R, Hall A: Sintering of uranium oxides of c omposition UO2 to U3O8 in various atmospheres Journal of Nuclear Materials 1959, 1 (1):28 38. 54. Lay KW, Carter BE: Role of the O/U ratio on the sintering of UO2 Journal of Nuclear Materials 1969, 30 (1 2):74 87. 55. Stuart WI, Adams RB: Effect of sinte ring atmosphere on the density of uranium dioxide pellets Journal of Nuclear Materials 1975, 58 (2):201 204. 56. Amato I, Colombo RL, Balzari AMP: Hot pressing of uranium dioxide Journal of Nuclear Materials 1966, 20 (2):210 214. 57. Yang JH, Song KW, Le e YW, Kim JH, Kang KW, Kim KS, Jung YH: Microwave process for sintering of uranium dioxide Journal of Nuclear Materials 2004, 325 (2 3):210 216. 58. Yang JH, Kim Y W, Kim JH, Kim D J, Kang KW, Rhee YW, Kim KS, Song KW: Pressureless Rapid Sintering of UO2 Assisted by High frequency Induction Heating Process Journal of the American Ceramic Society 2008, 91 (10):3202 3206. 59. Omori M: Sintering, consolidation, reaction and crystal growth by the spark plasma system (SPS) Materials Science and Engineering A 2000, 287 (2):183 188. 60. Zamora V, Ortiz AL, Guiberteau F, Nygren M: Crystal size dependence of the spark plasma sintering kinetics of ZrB2 ultra high temperature ceramics J Eur Ceram Soc 2012, 32 (2):271 276. 61. Fink JK: Thermophysical properties of u ranium dioxide Journal of Nuclear Materials 2000, 279 (1):1 18. 62. German RM : Sintering Therory and Practice ; New York NY, Wiley; 1996 : 95 100

PAGE 155

155 63. U. Anselmi Tamburini JEG, Z.A. Munir, A. Tacca, F. Maglia and G. Spinolo: Spark plasma sintering and ch aracterization of bulk nanostructured fully stabilized zirconia: Part I. Densification studies Journal of Materials Research 2004, 19 :3255 3262. 64. Ronchi C, Sari C: Properties of lenticular pores in UO2, (U, Pu) O2 and PuO2 Journal of Nuclear Material s 1974, 50 (1):91 97. 65. Lucuta PG, Matzke H, Hastings IJ: A pragmatic approach to modelling thermal conductivity of irradiated UO2 fuel: Review and recommendations Journal of Nuclear Materials 1996, 232 (2 3):166 180. 66. Goldsmith LA, Douglas JAM: Meas urements of the thermal conductivity of uranium dioxide at 670 1270 k Journal of Nuclear Materials 1973, 47 (1):31 42. 67. Carrea AJ: Sintering of uranium dioxide in an atmosphere of controlled hydrogen content Journal of Nuclear Materials 1963, 8 (2):275 277. 68. Yeo S, McKenna E, Baney R, Subhash G, Tulenko J: Enhanced thermal conductivity of uranium dioxide silicon carbide composite fuel pellets prepared by Spark Plasma Sintering (SPS) Journal of Nuclear Materials 2013, 433 (1 3):66 73. 69. Ge L, Subh ash G, Baney RH, Tulenko JS, McKenna E: Densification of uranium dioxide fuel pellets prepared by spark plasma sintering (SPS) Journal of Nuclear Materials 2013, 435 (1 3):1 9. 70. Grnvold F: High temperature X ray study of uranium oxides in the UO2 U3O8 region Journal of Inorganic and Nuclear Chemistry 1955, 1 (6):357 370. 71. Yakub E, Ronchi C, Staicu D: Computer simulation of defects formation and equilibrium in non stoichiometric uranium dioxide Journal of Nuclear Materials 2009, 389 (1):119 126. 72 Aman Y, Garnier V, Djurado E: Spark p lasma s intering k inetics of p Alumina Journal of the American Ceramic Society 2011, 94 (9):2825 2833. 73. Santanach JG, Weibel A, Estourns C, Yang Q, Laurent C, Peigney A: Spark plasma sintering of alumina: Stu dy of parameters, formal sintering analysis and hypotheses on the mechanism(s) involved in densification and grain growth Acta Materialia 2011, 59 (4):1400 1408. 74. Su H, Johnson DL: Master s intering c urve: a p ractical a pproach to s intering Journal of t he American Ceramic Society 1996, 79 (12):3211 3217. 75. German RM: Sintering theory and practice New York NY, Wiley; 1996 : 95.

PAGE 156

156 76. Nichenko S, Staicu D: Molecular d ynamics study of the effects of non stoichiometry and oxygen Frenkel pairs on the therma l conductivity of uranium dioxide Journal of Nuclear Materials 2013, 433 (1 3):297 304. 77. Hennecke JFA, Scherff HL: Carbon monoxide equilibrium pressures and phase relations during the carbothermic reduction of uranium dioxide Journal of Nuclear Materi als 1971, 38 (3):285 291. 78. Biasetto L, Zanonato P, Carturan S, Bernardo PD, Colombo P, Andrighetto A, Prete G: Developing uranium dicarbide graphite porous materials for the SPES project Journal of Nuclear Materials 2010, 404 (1):68 76. 79. Goss S, Gu neau C, Alpettaz T, Chatain S, Chatillon C, Le Guyadec F: Kinetic study of the UO2/C interaction by high temperature mass spectrometry Nuclear Engineering and Design 2008, 238 (11):2866 2876. 80. Carvajal Nuez U, Martel L, Prieur D, Lopez Honorato E, El oirdi R, Farnan I, Vitova T, Somers J: Coupling XRD, EXAFS, and 13C NMR to Study the Effect of the Carbon Stoichiometry on the Local Structure of UC1x Inorganic Chemistry 2013, 52 (19):11669 11676. 81. Goss S, Guneau C, Chatillon C, Chatain S: Critical review of carbon monoxide pressure measurements in the uranium carbon oxygen ternary system Journal of Nuclear Materials 2006, 352 (1 3):13 21. 82. Kurosaki K, Saito Y, Muta H, Uno M, Yamanaka S: Nanoindentation studies of UO2 and (U,Ce)O2 J Alloy Compd 2004, 381 (1 2):240 244. 83. Phani KK, Sanyal D, Sengupta AK: Estimation of elastic properties of nuclear fuel material using longitudinal ultrasonic velocity A new approach Journal of Nuclear Materials 2007, 366 (1 2):129 136. 84. Sengupta AK, Basak C B, Jarvis T, Bhagat RK, Pandey VD, Majumdar S: Effect of titania addition on hot hardness of UO2 Journal of Nuclear Materials 2004, 325 (2 3):141 147. 85. Gatt JM, Monerie Y, Laux D, Baron D: Elastic behavior of porous ceramics: application to nuclear fue l materials Journal of Nuclear Materials 2005, 336 (2 3):145 155. 86. Taylor R: Construction of apparatus for heat pulse thermal diffusivity measurements from 300 3000K Journal of Physics E: Scientific Instruments 1980, 13 (11):1193 1199. 87. Yamada K, Y amanaka S, Katsura M: Mechanical properties of (U,Ce)O2 J Alloy Compd 1998, 271 273 (0):697 701.

PAGE 157

157 88. Padel A, De Novion C: Constantes elastiques des carbures, nitrures et oxydes d'uranium et de plutonium Journal of Nuclear Materials 1969, 33 (1):40 51. 8 9. Watanabe T, Sinnott SB, Tulenko JS, Grimes RW, Schelling PK, Phillpot SR: Thermal transport properties of uranium dioxide by molecular dynamics simulations Journal of Nuclear Materials 2008, 375 (3):388 396. 90. Very h igh b urn ups in l ight w ater r eacto rs NEA No 6224 OECD 2006. [ http:// www.oecd.org/newsroom/36511265.pdf ] 91. Kashibe S, Une K: Effect of additives (Cr2O3, Al2O3, SiO2, MgO) on diffusional release of 133Xe from UO2 fuels Journal of Nuclear Materials 1998, 254 (2 3):234 242. 92. Yang JH, Kim KS, Nam IH, Oh JS, Kim D J, Rhee YW, Kim JH: Effect of step wise variation of oxygen potential during the isothermal sintering on the grain growth behavior in Cr2O3 doped UO2 pellets Journal of Nuclear Materials 2012, 429 (1 3):25 33. 93. Hong M, Uberuaga BP, Andersson DA, Stanek CR, Phillpot SR Sinnott SB: Role of Computational Materials Science 2013, 78 (0):29 33. 94. Ge L, Subhash G, Baney RH, Tulenko JS: Influence of processing parameters on thermal conductivity of uranium dioxide pellets prepared by spark plasma sintering Journal of the European Ceramic Society 2014 34 (7 ) : 1791 1801 95. Maglia F, Tredici IG, Anselmi Tamburini U: Densification and properties of bulk nanocrystalline functional ceramics with grain size below 50nm Journal of the European Ceramic Society 2013, 33 (6):1045 1066. 96. Ghosh D, Subhash G, Orlovskaya N: Measurement of scratch induced residual stress within SiC grains in ZrB2 SiC composite using micro Raman spectroscopy Acta Materialia 2008, 56 (18):5345 5354. 97. Mallik M, Kailath AJ, Ray KK, Mitra R: Electrical and thermophysical properties of ZrB2 and HfB2 based composites Journal of the European Ceramic Society 2012, 32 (10):2545 2555. 98. Yuan H, Li J, Shen Q, Zhang L: Preparation and thermal condu ctivity characterization of ZrB2 porous ceramics fabricated by spark plasma sintering International Journal of Refractory Metals and Hard Materials 2013, 36 (0):225 231.

PAGE 158

158 99. Fuel with advanced burnable absorbers design for the IRIS reactor core: Combined Erbia and IFBA Annals of Nuclear Energy 2009, 36 (8):1201 1207. 100. van Dam H: Long term control of excess reactivity by burnable particles Annals of Nuclear Energy 2000, 27 (8):733 743. 101. Meek TT, von Roedern B: Semiconductor devices fabricated from actinide oxides Vacuum 2008, 83 (1):226 228. 102. Y. Pauleau PBB, P. B. Barna: Protective coatings and thin films: synthesis, characterizatio n, and applications : Springer; 1997. 103. Yoshida H, Sakka Y, Yamamoto T, Lebrun J M, Raj R: Densification behaviour and microstructural development in undoped yttria prepared by flash sintering Journal of the European Ceramic Society 2014, 34 (4):991 1000. 104. Steil MC, Marinha D, Aman Y, Gomes JRC, Kleitz M: From conventional ac flash sintering of YSZ to hyper flash and double flash Journal of the European Ceramic Society 2013, 33 (11):2093 2101. 105. G. Subhash LG, S. Yeo, J. Tulenco, R. B aney: Comparison of spark plasma sintering versus conventional sintering of UO2 and UO2 composite fuel pellets LWR Fuel Performance Meeting/TopFuel 2013, September 15 19, Charlotte, NC 2013. 106. S. Yeo RB, G. Subhash, J. Tulenko: The influence of SiC pa rticle size and volume fraction on the thermal conductivity of spark plasma sintered UO2 SiC composites Journal of Nuclear Materials 2013, 442 :245 252. 107. Andrew Cartas DP, Lihao Ge, Sunghwan Yeo, James Tulenko, Ronald Baney, Ghatu Subhash: Fabricating the next generation of nuclear fuel utilizing spark plasma sintering American Nuclear Society 2013 Student Conference Massachusetts Institute of Technology, Boston, Massachusetts, USA, April 4 6, 2013 108. Santa Cruz H, Spino J, Grathwohl G: Nanocrystalline ZrO2 ceramics with idealized macropores Journal of the European Ceramic Society 2008, 28 (9):1783 1791. 109. Banerjee J, Ray A, Kumar A, Banerjee S: Studies on sintering kinetics of ThO2 UO2 pellets using master sintering curve approach Journal of Nuclear Materials 2013, 443 (1 3):467 478.

PAGE 159

159 BIOGRAPHICAL SKETCH Lihao Ge was born in Nantong City, Jiangsu Province, People s Republic of China in 1988. His parents are Cao Ge and Fengch un Xu. Lihao graduated with a Bachelor of Science in material science and engineering from Nanjing University of Science and Technology in July 2010. After that, he enrolled in the master s program of materials science and engineering at University of Flor ida in August of 20 10 In May of 2011, he transferred to the D epartment of M echanical and A erospace E ngineering as a PhD student at University of Florida During his PhD studies he published three peer reviewed jou r nal papers had several conference publi cations and was a co inventor of a patent. He received a non thesis Master of Science degree in December 2013 and received his Doctor of Philosophy degree from the University o f Florida in the spring of 2014