<%BANNER%>

Potential Inert Matrix Materials

Permanent Link: http://ufdc.ufl.edu/UFE0024359/00001

Material Information

Title: Potential Inert Matrix Materials Materials Synthesis and Evaluation of In-Service Engineering Parameters
Physical Description: 1 online resource (203 p.)
Language: english
Creator: Xu, Peng
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: ceramics, composite, corrosion, imf, inert, irradiation, nuclear, pyrochlore, reprocessing, spinel
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Containing no fertile materials, inert matrix fuel (IMF) has been introduced as a potential transmutation solution for the increasing inventory of both weapon grade and reactor grade plutonium (Pu). In the present work, the MgO-pyrochlore (Nd2Zr2O7) composites and spinel magnesium stannate (Mg2SnO4) were selected as potential inert matrix (IM) materials. A comprehensive investigation was conducted on evaluation of the engineering parameters of the potential IM materials. The MgO-Nd2Zr2O7 composites and Mg2SnO4 were fabricated through conventional solid state processing. The crystal structure and microstructure of the synthesized composites and Mg2SnO4 were studied. The irradiation tolerance of the potential IM materials was first assessed. The resistance of Mg2SnO4 against irradiation induced amorphization was assessed experimentally using in situ TEM technique. The critical amorphization doses for Mg2SnO4 irradiated by 1 MeV Kr2+ ions were determined to be 5.5 dpa at 50 K and 11.0 dpa at 150 K, respectively. The obtained results were compared with other spinels especially MgAl2O4, and the radiation tolerance of spinels were discussed. The next evaluation was water corrosion resistance of the potential IM materials. Homogeneous MgO-Nd2Zr2O7 composites exhibited an improved hydrothermal corrosion resistance than inhomogeneous composites and pure MgO. Even though spinel Mg2SnO4 was not stable in water at 300?C and saturation pressure, the corrosion was limited only to the surface, and the volume and mass changes were less than 1 % after 720 h corrosion. Feasibility of aqueous reprocessing was evaluated by studying the dissolution behavior of the potential IM materials in acidic solutions, with an emphasis on nitric acid. Dissolution of the MgO-Nd2Zr2O7 composites in HNO3 resulted in a selective dissolution of MgO. Mechanical agitation such as magnetic bar stirring was necessary to achieve a completed dissolution of MgO and disintegration of porous Nd2Zr2O7 matrix. It was demonstrated that Nd2Zr2O7 could be successfully digested in boiling concentrated H2SO4. Similarly, dissolution of Mg2SnO4 in HNO3 also resulted in a selective leaching of Mg2+ from the matrix. The undissolved substance consisted of SnO2 and substantial amorphous materials. Final evaluation will be performed by irradiating the potential IM materials in the Advanced Testing Reactor at Idaho National Lab.
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 Peng Xu.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Nino, Juan C.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024359:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024359/00001

Material Information

Title: Potential Inert Matrix Materials Materials Synthesis and Evaluation of In-Service Engineering Parameters
Physical Description: 1 online resource (203 p.)
Language: english
Creator: Xu, Peng
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: ceramics, composite, corrosion, imf, inert, irradiation, nuclear, pyrochlore, reprocessing, spinel
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Containing no fertile materials, inert matrix fuel (IMF) has been introduced as a potential transmutation solution for the increasing inventory of both weapon grade and reactor grade plutonium (Pu). In the present work, the MgO-pyrochlore (Nd2Zr2O7) composites and spinel magnesium stannate (Mg2SnO4) were selected as potential inert matrix (IM) materials. A comprehensive investigation was conducted on evaluation of the engineering parameters of the potential IM materials. The MgO-Nd2Zr2O7 composites and Mg2SnO4 were fabricated through conventional solid state processing. The crystal structure and microstructure of the synthesized composites and Mg2SnO4 were studied. The irradiation tolerance of the potential IM materials was first assessed. The resistance of Mg2SnO4 against irradiation induced amorphization was assessed experimentally using in situ TEM technique. The critical amorphization doses for Mg2SnO4 irradiated by 1 MeV Kr2+ ions were determined to be 5.5 dpa at 50 K and 11.0 dpa at 150 K, respectively. The obtained results were compared with other spinels especially MgAl2O4, and the radiation tolerance of spinels were discussed. The next evaluation was water corrosion resistance of the potential IM materials. Homogeneous MgO-Nd2Zr2O7 composites exhibited an improved hydrothermal corrosion resistance than inhomogeneous composites and pure MgO. Even though spinel Mg2SnO4 was not stable in water at 300?C and saturation pressure, the corrosion was limited only to the surface, and the volume and mass changes were less than 1 % after 720 h corrosion. Feasibility of aqueous reprocessing was evaluated by studying the dissolution behavior of the potential IM materials in acidic solutions, with an emphasis on nitric acid. Dissolution of the MgO-Nd2Zr2O7 composites in HNO3 resulted in a selective dissolution of MgO. Mechanical agitation such as magnetic bar stirring was necessary to achieve a completed dissolution of MgO and disintegration of porous Nd2Zr2O7 matrix. It was demonstrated that Nd2Zr2O7 could be successfully digested in boiling concentrated H2SO4. Similarly, dissolution of Mg2SnO4 in HNO3 also resulted in a selective leaching of Mg2+ from the matrix. The undissolved substance consisted of SnO2 and substantial amorphous materials. Final evaluation will be performed by irradiating the potential IM materials in the Advanced Testing Reactor at Idaho National Lab.
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 Peng Xu.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Nino, Juan C.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024359:00001


This item has the following downloads:


Full Text

PAGE 1

POTENTIAL INERT MATRIX MATERI ALS: MATERIALS SYNTHESIS AND EVALUATION OF IN-SERVICE ENGINEERING PARAMETERS By PENG XU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Peng Xu 2

PAGE 3

With my deepest love to my entire family 3

PAGE 4

ACKNOWLEDGMENTS First of all, I acknowledge my advisor Dr. Juan C. Nino, for his support and guidance. His knowledge, care optimism, diligence and aspira tions inspired me to work hard and expand my potential. I express deepest gratitude to my other committee members (Dr. Wolfgang Sigmund, Dr. Simon R. Phillpot and Dr. Ronald Baney and Prof. James S. Tulenko) for t heir time and guidance. I acknowledge my mentors Dr. Mitchell K. Meyer and Dr. Pavel G. Medvedev for their help and guidance during the five and half month internship at the Idaho National Lab (INL). I want to thank Dr. Paul E. Murray, who taug ht me step by step using Abaqus to perform thermal and sa fety analysis for the UF project. I want to give my special thanks to the people wh o were actively involved and helped me at INL, Julie A. Foster, Gregg W. Wachs, Robert D Hergeshei mer, Joseph W. Nielsen, Dr. James R. Parry, and Dr. Frances M. Marshall. I thank the group members: Lu Cai, Donald Moore, Sa mantha Yates, Wei Qiu, Mohammed Elshennawy, Satyajit Phadke, Laurel Wucherer, Marta Giachino, Kevin Tierney, Shobit Omar, and etc., for providing me excellent research environment and being helpful. Last but not least, I am deeply indebted to my parents and their c ontinuous support. Their loves have encouraged me to go through the hard time. I c annot imagine myself having any of my progress without my familys strong support behind me. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ..................................................................................................4 LIST OF TABLES ..........................................................................................................10 LIST OF FIGURES ........................................................................................................11 LIST OF ABBREVIATIONS ...........................................................................................15 CHAPTER 1. INTRO DUCTION .......................................................................................................19 1.1 Statement of Problem and Motivation .................................................................19 1.2 Scientific Approach .............................................................................................21 1.3 Organization of Dissertation ................................................................................24 1.4 Contributions to the Field ....................................................................................25 2. BACKG ROUND .........................................................................................................27 2.1 Nuclear Energy and Fuel Cycle ..........................................................................27 2.2 Crystal Structures of Potential IM Materials ........................................................28 2.2.1 Rocksalt Structure .....................................................................................29 2.2.2 Pyrochlore Structure ..................................................................................31 2.2.3 Spinel Structure .........................................................................................33 2.2.3.1 Detailed structure description .......................................................35 2.2.3.2 Cation arrangement .....................................................................37 2.2.3.3 Bonds length and bonds angles ...................................................38 2.2.3.4 Stability of spinel and global instability index (GII) .......................39 2.3 Radiation Damage in Materials ...........................................................................40 2.3.1 Radiation Damage Mechanisms ................................................................41 2.3.2 Radiation Damage in Nuclear Fuels ..........................................................46 2.3.3 Irradiation Induced Amorphization and Crystal St ructure Considerations .47 2.3.4 Stopping and Range of Ions in Matter (SRIM) ...........................................48 2.4 Chemical Dissolution of Metal Oxides .................................................................50 2.4.1 Dissolution Mechanism .............................................................................50 2.4.2 Dissolution Kinetics ...................................................................................53 5

PAGE 6

2.4.2.1 Kinetics models ............................................................................54 2.4.2.2 Effect of temperature ...................................................................55 2.4.2.3 Dissolution rate determining factors .............................................55 2.5 Description of the Advanced Test Reactor (ATR) ...............................................56 3. MATERIALS SELECTION, SYNTHES IS AND CHARAC TERIZATION.....................58 3.1 Materials Selection ..............................................................................................58 3.1.1 MgO Based Composites ...........................................................................58 3.1.2 Single Phase Spinel Compound ................................................................59 3.2 Experimental Procedures ....................................................................................62 3.2.1 Fabrication of MgO-Nd2Zr2O7 Composites ................................................63 3.2.1.1 Solid state synthesis of Nd2Zr2O7.................................................63 3.2.1.2 Sol-gel Synthesis of Nd2Zr2O7......................................................64 3.2.1.3 Synthesis of MgO .........................................................................66 3.2.1.4 Powder mixing .............................................................................66 3.2.1.5 Pellet formation ............................................................................67 3.2.2 Fabrication of Single Phase Spinel Mg2SnO4............................................68 3.2.2.1 Solid state synthesis of Mg2SnO4................................................68 3.2.2.2 Pellet formation ............................................................................69 3.2.3 Characterization ........................................................................................70 3.2.3.1 X-ray diffraction ............................................................................70 3.2.3.2 Rietveld refinement ......................................................................70 3.2.3.3 Particle size measurement ...........................................................71 3.2.3.4 Transmission elec tron microscopy (TEM) ....................................71 3.2.3.5 Scanning electron microscopy (SEM) ..........................................71 3.3 Results and Discussion .......................................................................................72 3.3.1 MgO-Nd2Zr2O7 Composites .......................................................................72 3.3.1.1 Pyrochlore Phase formation .........................................................72 3.3.1.2 Powder characterization ..............................................................74 3.3.1.3 Microstructure analysis ................................................................76 3.3.2 Single Phase Spinel Mg2SnO4..................................................................78 3.3.2.1 Phase formation ...........................................................................78 3.3.2.2 Structure analysis ........................................................................78 6

PAGE 7

3.3.2.3 Microstructure analysis ................................................................82 3.4 Summary and Conclusions .................................................................................84 4. EVALUATION OF RADI ATION TOLE RANCE..........................................................86 4.1 Introduction .........................................................................................................86 4.2 Literature Review ................................................................................................86 4.2.1 Irradiation Stability of MgO ........................................................................86 4.2.2 Irradiation Stability of Nd2Zr2O7.................................................................87 4.2.3 Irradiation Stability of MgAl2O4..................................................................88 4.3 Experimental Procedure ......................................................................................89 4.3.1 Specimen Preparation ...............................................................................89 4.3.2 In situ Ion Irradiation and Characterization ................................................89 4.4 Results ................................................................................................................90 4.4.1 Transport of Ions in Ma tter (TRIM) Based Calculation ..............................90 4.4.2 Irradiation of Mg2SnO4 at 50 K ..................................................................91 4.4.3 Irradiation of Mg2SnO4 at 150 K ................................................................97 4.4.4 Irradiation Damage Mechanism of Mg2SnO4 by 1 MeV Kr2+.....................99 4.4.5 Irradiation Tolerance of Mg2SnO4............................................................100 4.5 Summary and Conclusions ...............................................................................104 5. EVALUATION OF HYDR ATION RESIST ANCE......................................................106 5.1 Introduction .......................................................................................................106 5.1.1 Hydration of MgO ....................................................................................106 5.1.2 Hydration of Polycrystalline MgO ............................................................109 5.2 Experimental Procedure ....................................................................................110 5.2.1 Hydrothermal Corrosion Testing Setup ...................................................110 5.2.2 Sample Preparation for Characterization ................................................111 5.2.3 Composite Microstr ucture Quantification .................................................111 5.3 Results and Discussion .....................................................................................113 5.3.1 Hydrothermal Co rrosion of the MgO-Nd2Zr2O7 Composites ....................113 5.3.1.1 Preliminary results .....................................................................113 5.3.1.2 Microstructure characterization ..................................................116 5.3.1.3 Hydrothermal corrosion of t he composites made by ball milling 118 7

PAGE 8

5.3.1.4 Normalized mass loss rate and temperature dependence .........121 5.3.1.5 Coprecipit ation and gel-casting ..................................................125 5.3.2 Hydrothermal Corrosion Resistance of Mg2SnO4....................................132 5.4 Summary and Conclusions ...............................................................................136 6. EVALUATION OF AQEUOUS RE PROCESSING F EASIBILITY .............................138 6.1 Introduction .......................................................................................................138 6.2 Experimental Procedure ....................................................................................140 6.2.1 Dissolution Test Setup ............................................................................140 6.2.2 Structural Characterization ......................................................................141 6.3 Results and Discussion .....................................................................................141 6.3.1 Dissolution of the MgO-Nd2Zr2O7 Composites in HNO3..........................141 6.3.1.1 Dynamic dissolution and dissolution rate of MgO ......................142 6.3.1.2 Effect of porosity and MgO content ............................................145 6.3.1.3 Characterization of the co mposites before and after dissolution 147 6.3.1.4 Static and ultrasonic dissolution .................................................149 6.3.2 Dissolution the MgO-Nd2Zr2O7 Composites in H2SO4.............................151 6.3.3 Dissolution of Mg2SnO4 in HNO3.............................................................154 6.4 Summary and Conclusions ...............................................................................160 7. REACTOR T ESTING..............................................................................................162 7.1 Introduction .......................................................................................................162 7.2 Objectives .........................................................................................................163 7.3 Description of the Process ................................................................................163 7.4 Summary of Internship Activities .......................................................................166 8. SUMMARY AND FU TURE WO RK..........................................................................167 8.1 Summary ...........................................................................................................167 8.2 Future work .......................................................................................................172 8.2.1 Processing and Water Corrosion Resistance ..........................................172 8.2.2 Radiation Tolerance ................................................................................172 8.2.3 Aqueous Reprocessing ...........................................................................173 8.2.4 Reactor Testing and Post Irradiation Examination (PIE) .........................174 8.2.5 Other In-service E ngineering Parameters for Potential IM Materials .......174 8

PAGE 9

APPENDIX A. NEUTRONIC PROPERTIES OF POTENTIAL IM MA TERIALS..............................176 A.1 Introduct ion.......................................................................................................178 A.2 Neutronic Properties of MgO-Nd2Zr2O7 Composite and Mg2SnO4....................179 B. CONCEPTUAL hARDWARE DESIGN FOR REACTOR TESTING........................179 B.1 Introduct ion.......................................................................................................181 B.2 Conceptual Design...........................................................................................181 C. THERMAL AND SAFETY ANALYSIS FOR REACTOR TESTING.........................181 C.1 Introduct ion.......................................................................................................183 C.2 Experimental Procedur e...................................................................................184 C.3 Results and Discussion ....................................................................................185 C.3.1 Steady-State Operation at Nominal Cycl e Power ...................................183 C.3.2 Pump Coast-down and Reactivity Insertion Acci dent (RIA)....................186 C.3.3 Passive Cool ing in Ai r.............................................................................189 LIST OF REFE RENCES.............................................................................................191 BIOGRAPHICAL SKETCH ..........................................................................................203 9

PAGE 10

LIST OF TABLES Table page 2 Rocksalt (AX) structure....................................................................................30 2 Pyrochlore (A2B2X6X) structure data a fter Subram anian.................................33 B 2 Spinel (AB2X4) structure data afte r Sickafus and Wills .....................................35 2 Spinel bond length and selected lattice distances, as a function of lattice parameter ( a ) and anion parameter ( u )............................................................38 3 X-ray Diffraction (XRD) reflections of spinel Mg2SnO4.....................................80 3 Refined structural param eters from x-ray diffraction data for the spinel Mg2SnO4 phase...............................................................................................81 3 Bond length and bond angl e obtained from refined Mg2SnO4 structure...........82 3 Bond valence calculations for Mg2SnO4..........................................................82 5 Microstructure analysi s results for the MgO-Nd2Zr2O7 composites................116 5 Normalized mass loss (NML) ra te and apparent activation energy (Ea).........124 5 Zeta potential and pH of the slu rries..............................................................130 10

PAGE 11

LIST OF FIGURES Figure page 1-1 Methodology for Inert Matrix Fuel (IMF) development .....................................20 1-2 Requirements on material properties for multi-cycling Inert Matrix (IM)...........21 2-1 Nuclear fuel cycl e.............................................................................................28 2-2 Cubic rocksalt stru cture of m agnesia. ..............................................................29 2-3 Cation polyhedra a rrangement in the rocksa lt structure...................................30 2-4 Type I and type II subcells of pyrochlo re structure derived from the fluorite structur e...........................................................................................................31 2-5 Two octants of the pyrochlore structure filled with Type I and Type II subcells ............................................................................................................32 2-6 Unit cell of the spinel stru cture.........................................................................34 2-7 Cation polyhedra a rrangement in one spi nel unit ce ll......................................36 2-8 Layer sequence along the z-axis ([001]) for spinel structure............................37 2-9 Bond angles in spinel crystal structure as a f unction of anion parameter ( u )...39 2-10 Energy deposition mechanisms of inci dent ions in targ et materials.................41 2-11 Illustration of the proce ss of Coulomb explosi on..............................................43 2-12 Ion track formation in Nd2Zr2O7 irradiated by 120 MeV I ions to a fluence of 5.6 1010 cm-2..................................................................................................44 2-13 Frenkel pair formation result ed from displace ment da mage............................45 2-14 Transport of Ions in Matter (TRI M) calculation parameters window (V 2006.02). ..........................................................................................................49 2-15 Illustration of nucleophi lic and electroph ilic atta ck...........................................51 2-16 Advanced Test Reactor Co re...........................................................................57 3-1 Neutron adsorption cro ss section of element s.................................................60 3-2 Solid state synthesis of Nd2Zr2O7.....................................................................64 3-3 Sol gel synthesis of Nd2Zr2O7..........................................................................66 11

PAGE 12

3-4 Processing flowchart for fabrication of composite pellets through three different mixi ng methods ..................................................................................68 3-5 Processing flowchart for solid state synthesis of Mg2SnO4..............................69 3-6 X-ray Diffraction profiles of Nd2Zr2O7 made from sol-gel processing after calcination at 1000 C, 1150 C and 1200 C for 6 hours................................73 3-7 X-ray Diffraction profiles of calcined MgO, Nd2Zr2O7, and the sintered composite pellet...............................................................................................74 3-8 Particle size and size distribution of sol-gel process derived and solid state process derived Nd2Zr2O7 powder...................................................................75 3-9 Sol-gel derived Nd2Zr2O7 powder ....................................................................76 3-10 Microstructures of the MgO-Nd2Zr2O7 composit es...........................................77 3-11 X-ray Diffraction profiles of starting materials SnO2 and MgO, and obtained Mg2SnO4 by calcination at 1200 C for 12 h.....................................................78 3-12 The calculated observed X-ray Diffr action profiles and the difference for Mg2SO4............................................................................................................79 3-13 Bond lengths and bond angles for Mg2SnO4....................................................81 3-14 Microstructure Mg2SnO4..................................................................................83 4-1 Total target atom displacementsin dpa and the concentration of implanted Kr ions in Mg2SnO4 spinel as a functi on of target depth..................................91 4-2 Focused Ion Beam (FIB) prepared Mg2SnO4 specim en..................................93 4-3 Mg2SnO4 sample irradiated by 1 MeV Kr2+ at a fluence of 5 1019 Kr2+ ions/m2 at 50 K and structure was thermally recovered at room temperature..95 4-4 Selected area electron diffrac tion (SAED) pattern recorded on the segregation layer.............................................................................................96 4-5 Selected area electron diffracti on (SAED) patterns of the crushed Mg2SnO4 grains irradiated at 150 K.................................................................................98 4-6 Electronic ((dE/dx)e) and nuclear ((dE/dx)n) stopping powers for 1 MeV Kr2+ ions in the Mg2SnO4 target .............................................................................100 5-1 Crystal of Mg(OH)2.........................................................................................107 5-2 Characterization of the hy drated composite with inhomogeneous microstructu re................................................................................................115 12

PAGE 13

5-3 Microstructure of MgO-Nd2Zr2O7 composites made by ball milling process...117 5-4 Characterization of the hy drated composit e with homogeneous microstructu re................................................................................................119 5-5 Microstructure of the interface between the composite and the hydration product layer at high magnificati on................................................................120 5-6 Quantitative anal ysis result s..........................................................................123 5-7 Flowchart of the copr ecipitation process for fabrication of the MgONd2Zr2O7 composit es.....................................................................................126 5-8 Microstructure of the composites made by the coprecip itation method..........127 5-9 Schematics of ex periment des ign..................................................................128 5-10 Particle size dist ribution of the MgO and Nd2Zr2O7 slurry..............................129 5-11 Morphology of the particles in the mixed slurry..............................................130 5-12 Stability of slurries with and without polyele ctrolytes......................................131 5-13 Mass and volume change as a f unction of corrosion time for Mg2SnO4........132 5-14 X-ray Diffraction pr ofile of the corroded Mg2SnO4 surface after 120 h exposure to H2O at 300 C and saturation pr essure.......................................133 5-15 Morphology of the Mg2SnO4 pellet after exposure to water at 300 C and saturation pressure for 120 h.........................................................................134 5-16 Electron Dispersive Spectroscopy (EDS) spectra. .........................................135 6-1 Dynamic dissolution of the composite with 70 vol% of MgO in 11 M HNO3 at 60 C...........................................................................................................142 6-2 Dissolution of MgO-Nd2Zr2O7 composites in HNO3........................................146 6-3 Structural char acterizati on.............................................................................148 6-4 X-ray Diffraction profile of the sintered composite, surface of the porous matrix, and the residual powder collected from the flask after dissolution test.................................................................................................................149 6-5 Dissolution of the MgO-Nd2Zr2O7 composites 11 M HNO3 at 60 C................150 6-6 Dissolution of t he composites in H2SO4.........................................................152 6-7 Dissolution of Mg2SnO4 in 11 M HNO3 at 60 C with magnetic bar stirring.....155 13

PAGE 14

6-8 Microstructure and chemical analysis of dissolved Mg2SnO4.........................156 6-9 Characterization of dissolution re sidue..........................................................158 7-1 Roadmap for experiment irradiat ion...............................................................164 A-1 Neutron multiplication factor as a function of equiva lent burn up...................178 B-1 Conceptual design of the Transmissi on Electron Microscopy (TEM) sample holder. ............................................................................................................180 B-2 Capsule as sembly..........................................................................................180 C-1 Reactor power transient in condition 4 Reactivity Insertion Accident (RIA)....182 C-2 Finite element mesh of Capsule C a ssembly.................................................183 C-3 Temperature c ontour plot s.............................................................................184 C-4 Temperature of bottom end pl ug....................................................................185 C-5 Temperature of se lected specim ens..............................................................186 C-6 Pump coast-dow n analysi s............................................................................187 C-7 The Reactivity Insertion Accident (RIA) transient. ..........................................188 C-8 Temperature of Caps ule B and C assembly with sleeve and basket in air after 5 h dec ay...............................................................................................189 14

PAGE 15

LIST OF ABBREVIATIONS ANL Argonne National Lab ATR Advanced Test Reactor BF bright field ccp cubic close packing CN coordination number CSA composite spheres assemblage CVD chemical vapor deposition DI deionized DNBR departure from nucleate boiling ratio dpa displacement per atom EDS energy dispersive spectroscopy EMC Electron Microscopy Center FIB focused ion beam FIR flow instability ratio HLW high level nuclear waste IM inert matrix IMF inert matrix fuel INL Idaho National Lab JCPDS Joint Committee on Powder Diffraction Standards LANL Los Alamos National Lab LOCA loss of coolant accident LWR light water reactor MAIC Major Analytical Instrumentation Center MD molecular dynamics 15

PAGE 16

MOX mixed oxide fuel NDR normalized dissolution rate NML normalized mass loss NSUF National Scientific User Facility ORNL Oak Ridge National Lab PAA ammonium polyacr ylate dispersant PDAC Poly diallyldimethylammonium chloride PIE post irradiation examination PLZT Lanthanum modified lead zirconate titanate PSS Poly sodium 4-styrenesulfonate rpm round per minute SAED selected area electron diffraction SEM scanning electron microscopy SRIM Stopping and Range of Ions in Matter TAD temperature accelerated dynamics TEM transmission electron microscopy TRIM Transport of Ions in Matter YSZ ytrria stabilized zirconia 16

PAGE 17

Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy POTENTIAL INERT MATRIX MATERI ALS: MATERIALS SYNTHESIS AND EVALUATION OF IN-SERVICE ENGINEERING PARAMETERS By Peng Xu May 2009 Chair: Juan C. Nino Major: Materials Science and Engineering Containing no fertile materi als, inert matrix fuel (IMF) has been introduced as a potential transmutation solution for the incr easing inventory of both weapon grade and reactor grade plutonium (P u). In the present work the MgO-pyrochlore (Nd 2 Zr 2 O 7 ) composites and spinel magnesium stannate (Mg 2 SnO 4 ) were selected as potential inert matrix (IM) materials. A comprehensive in vestigation was conduct ed on evaluation of the engineering parameters of the potential IM materials. The MgO-Nd 2 Zr 2 O 7 composites and Mg 2 SnO 4 were fabricated through conventional solid state processing. The crystal struct ure and microstructure of the synthesized composites and Mg 2 SnO 4 were studied. The irradiati on tolerance of the potential IM materials was first assess ed. The resistance of Mg 2 SnO 4 against irradiation induced amorphization was assessed experimentally using in situ TEM technique. The critical amorphization doses for Mg 2 SnO 4 irradiated by 1 MeV Kr 2+ ions were determined to be 5.5 dpa at 50 K and 11.0 dpa at 150 K, res pectively. The obtained results were compared with other spin els especially MgAl 2 O 4 and the radiation tolerance of spinels were discussed. 17

PAGE 18

The next evaluation was water corrosion resi stance of the potential IM materials. Homogeneous MgO-Nd 2 Zr 2 O 7 composites exhibited an improved hydrothermal corrosion resistance than inhomogeneous co mposites and pure MgO. Even though spinel Mg 2 SnO 4 was not stable in water at 300 C and saturation pre ssure, the corrosion was limited only to the surface, and the volume and mass changes were less than 1 % after 720 h corrosion. Feasibility of aqueous repr ocessing was evaluated by studying the dissolution behavior of the potential IM materials in acidic solutions, with an emphasis on nitric acid. Dissolution of the MgO-Nd 2 Zr 2 O 7 composites in HNO 3 resulted in a selective dissolution of MgO. Mechanical agitation such as magnet ic bar stirring was necessary to achieve a completed dissolution of MgO and disintegration of porous Nd 2 Zr 2 O 7 matrix. It was demonstrated that Nd 2 Zr 2 O 7 could be successfully diges ted in boiling concentrated H 2 SO 4 Similarly, dissolution of Mg 2 SnO 4 in HNO 3 also resulted in a selective leaching of Mg 2+ from the matrix. The undissolv ed substance consisted of SnO 2 and substantial amorphous materials. Final ev aluation will be performed by irradiating the potential IM materials in the Advanced Testin g Reactor at Idaho National Lab. 18

PAGE 19

CHAPTER 1 INTRODUCTION 1.1 Statement of Problem and Motivation Excess plutonium (Pu) is a global problem and potentially a major threat related to proliferation and environm ental safety. Current excess st ockpiles of Pu mainly came from dissembled nuclear weapons after the Cold War and from production of reprocessed spent nuclear fuels from nuclear power plants. It has been reported that the total inventory of Pu has reached the order of 200 t ons of weapon grade plutonium and 1000 tons of civilian plutonium, respec tively at the end of last century. 1 Besides Pu, disposal of minor actinides such as americ ium (Am) and curium (Cm) which are present in spent nuclear fuels is also a major conc ern for environmental safety because of their radiotoxicity and decaying heat generation. Most major and minor actinides are long lived radio nuclides. For example, 239 Pu has a half-life of 24,000 years and 243 Am has a half-lif e of 7360 years. Therefore, geological disposal of these actinides is a major challenge for the long-term integrity of storage facilities and the materials that ar e used for immobilization of nuclear waste should be chemically inert and irradiation-tolerant for hundreds and thousands years. 2 A different approach is to incinerate Pu and minor actinides in nuclear reactors. It is a safe, effective, and economic solution to reduce the high level nuclear waste (HLW) and produce electricity. Pu can be burned in the light water reactors (LWRs) in a fuel type called mixed uranium-plutonium oxide (MOX). The MOX fuel is made of a mixture of UO 2 and PuO 2 so it contains 238 U which is a fertile material that adsorbs thermal neutrons and converts into fission material 239 Pu by irradiation in nuclear reactors. As a 19

PAGE 20

result, burning MOX fuel in a LWR is inefficient and does not allow a rapid reduction of stocked Pu. Therefor e, replacement of UO 2 by a neutron transparent matrix which contains less or no fertile materials is desirab le and the concept of inert matrix fuel (IMF) has been proposed. Similar to 238 U in conventional low enr iched uranium (LEU) or MOX fuel, the inert matrix (IM) plays an important role as a diluent for the fissile phase to achieve the desired volume concentrations to serve as nuclear fuel in nuclear reactors. Figure 1-1 shows the methodology used to develop and qualify IMF in LWRs. 3,4 In the flowchart, materials selection, synthesis, char acterization, and irradiation testing are the first few steps which serve as the basis for the development of IMF. A completed evaluation on the pro perties and performance of potential IM materials is not only essential for the development of IMF, but also important for understanding the structureproperty-performance relationships for nuclear ma terials. In the present work, research is focused on selecting, synthesizing, and evaluating potential IM materials for transmutation of Pu and minor actinides in LWRs. Screening Studies Neutronic/ Thermodynamics Fabrication Characterization Irradiation tests Accelerators Reactors Modeling for LWR Fuel qualification IMF fueling in a Commercial reactor Spent fuel Inventory Actinides & Fission products Stability Retention Natural analogue Geological disposal Dissolution of spent fuel Aqueous separation Figure 1-1. Methodology for Inert Matrix Fuel (IMF) development. 3,4 20

PAGE 21

1.2 Scientific Approach The high temperature and high irradiation dose environment in nuclear reactors is extremely hostile, so there are several cr itical requirements for IM materials. Figure 1-2 summarizes the requirements for IM in terms of materials properties. The specific requirements are listed below, 1 1. Low neutron adsorption cro ss section, which is economically important to sustain reactivity and achieve high burn-up for nuclear fuels; 2. Large safety margin provided by good t hermophysical properties, such as high melting point and good thermal conductivity; 3. Good chemical stability such as compatibil ity with cladding (Zircaloy, stainless steel) and coolant (water, Na); 4. Good irradiation stability agains t thermal and fast neutrons, -decay and fission fragments; 5. Good mechanical stability during irradiat ion and interaction with cladding, such as suitable elastic constants and hardness; 6. Good solubility in nitric acid if designed for reprocessing. Gap Fuel Pellet Cladding Cladding Breach High T, High P Water Corrosion ~100 MeV heavy fission fragments FCCI (Chemical Resistance) FMCI (Irradiation Resistance) Neutron Economy Steep Temperature Gradient High Melting Point, Good Thermal Conductivity Low Neutron Adsorption Cross Section Good Irradiation Stability Good Hydrothermal Corrosion Resistance Dissolution Behavior Feasible Aqueous Reprocessing Figure 1-2. Requirements on material properties for multi-cycling Inert Matrix (IM). 1 21

PAGE 22

Several IM candidate materials have been proposed and investigated in the past few decades. Some of the most studied ma terials are: magnesium aluminate (MgAl 2 O 4 ), zirconium silicate (ZrSiO 4 ), zirconia (ZrO 2 ), ceria (CeO 2 ), magnesia (MgO), carbide materials such as silicon carbide (SiC), and nitride materials such as silicon nitride (Si 3 N 4 ). All of those materials have advant ageous properties and drawbacks compared to UO 2 or the MOX fuel. For example, MgAl 2 O 4 is susceptible to fission fragments damage 5-7 ZrSiO 4 dissociates at relati ve low temperature (1690 C 8 ) and is not resistant to irradiation 9 ZrO 2 has low thermal conductivity 10 MgO hydrates easily when exposed to water 11 (coolant in the primary coolant system in LWRs), SiC and Si 3 N 4 are difficult to dissolve in acidic solutions posing a challenge for fuel reprocessing. 12 Therefore, improvements are neede d to overcome the drawbacks of these potential IM materials, and searchi ng for new potential IM materials should continue. Recently, a composite concept was propos ed to improve the sh ortcomings of the current candidate materials. For instance, to improve the hydration resistance of MgO, Medvedev 11 and coworkers investigated the introducti on of a second phase that acts as a hydration barrier. An MgO-ZrO 2 composite was specifical ly studied and the results show that the composite exhibited improved hydration resistance compared to pure MgO. However, ZrO 2 was insoluble in HNO 3 which is undesirable for fuel reprocessing. Moreover, the thermal conductivity of ZrO 2 was low, typically less than 3 Wm -1 K -1 at 1000 C. 10 Therefore, further improvement on MgO based composites is possible and worth investigation. 22

PAGE 23

Even though MgAl 2 O 4 exhibited large swelling during in pile testing, the spinel structure itself was found to be irradiation resistant in general. 13 Other spinel compounds may exhibit better irr adiation tolerance than MgAl 2 O 4 and may be qualified as IM. Therefore, a natura l step forward is exploration and evaluation of other potential spinel compounds. Based on the above considerations, the se lection will be narrowed down to MgObased composites and spinel compounds. The materials selection will be conducted based on literature survey. After the candidate materials are selected, materials will be synthesized in house and followed by char acterization. Different techniques and research tools will be used to characterize t he materials. The microstructure analysis will be performed using scanning electron micr oscopy (SEM) and transmission electron microscopy (TEM). X-ray diffraction (XRD) will be used to identify the crystal structures. The structural analysis will be performed using Rietveld refinement method. After potential IM materials are fabricated and characterized, some of the key engineering parameters fo r IMF will be evaluated. The irradiation tolerance will be first investigated. Ideally, the irradiation tests should be conducted in a testing reactor. However, it is a time consuming proce ss and usually takes years from planning an experiment to obtaining the final results, and thus reactor testing should be the final evaluation of the potential IM materials. In order to assess the radiation resistance in a timely manner, preliminary tests using an ion beam accelerator will be performed as an alternative for a screening type of study. Th e compatibility of the potential IM materials with coolant water will be evaluated in an autoclave which provides a hydrothermal condition similar to the primary coolant system in LWRs. The suitability of aqueous 23

PAGE 24

reprocessing will be assessed by performing aqueous dissolution tests for the potential IM materials. Finally, reac tor testing in the Advanced Test Reactor (ATR) at Idaho National Lab (INL) will be performed to evaluate the overall performance of the potential IM materials with an emphasis on the thermophysical properties. 1.3 Organization of Dissertation Chapter 2 provides a backgr ound that covers some of the fundamentals on nuclear and material science. The purpose of this chapter is to assist readers who are not familiar with this research field allowing them to gain a better understanding of the following chapters. This chapter covers th e following contents: the nuclear fuel cycle, crystal structures of the potential IM materi als, radiation damage, TR IM calculation, and chemical dissolution of metal oxides. Chapter 3 discusses material selection, synthesis and characterization. The process of selecting potential IM materials is presented. Detailed experimental procedures on materials synthesis and characterization are described and results are presented. Results on structure analysis of spinel Mg 2 SnO 4 are presented. The relationships between microstructure and processing are discussed. Chapter 4 focuses on the irradiation behav ior of investigated materials. The irradiation tolerance of MgO, pyrochlores, and spinels is reviewed. The experimental results on in situ ion irradiation of Mg 2 SnO 4 are presented and discussed. In Chapter 5, the hydrothermal corro sion behavior of the MgO-pyrochlore composites and Mg 2 SnO 4 is assessed. Quantitative analysis on the contiguity and homogeneity of the composites is presented. The relati onship between the corrosion resistance and the composite microstructure is discussed. 24

PAGE 25

Chapter 6 discusses the aqueous dissolut ion behavior of MgO-pyrochlore composites and Mg 2 SnO 4 Results and discussion are focused on nitric acid digestion as an assessment for the aqueous reprocessing feasibility. Chapter 7 briefly describes the research work and activities for the reactor testing of potential IM materials in the ATR. Finally in Chapter 8, a summary of the dissertation is pr esented and the future work in the relevant research areas is discussed. At the end of the thesis, th ree appendices are provided. Appendix A presents the neutronic properties of the potential IM Mate rials; Appendix B pres ents the conceptual hardware design for reactor testing, and A ppendix C presents a det ailed thermal and safety analysis for the ATR test. 1.4 Contributions to the Field The main contributions of this dissertation to the develop ment of IM materials are summarized below: 1. Two new IM candidate materials, the MgO-Nd 2 Zr 2 O 7 composites and the single phase spinel Mg 2 SnO 4 were proposed, and some of the in-service engineering parameters such as irradiation stability water corrosion resistance, and aqueous dissolution behavior were evaluated. 2. The structural information of Mg 2 SnO 4 were studied using Rietveld refinement based on powder XRD pattern, and the lattice parameter and the oxygen dilation parameter were reported. 3. The irradiation behavior of Mg 2 SnO 4 was studied. The critic al amorphization dose at low temperatures 50 and 150 K were determined. The thermal annealing effect was assessed. Since Mg 2 SnO 4 is an inverse spinel, the preliminary experimental results in the present work were in agr eement with the recent atomic simulation studies. 4. The water corrosion resistance of the MgO-Nd 2 Zr 2 O 7 composites and Mg 2 SnO 4 was evaluated. The microstructure dependence on corrosion resistance for the composites was studied. The relationshi ps between MgO volume fraction, water 25

PAGE 26

temperature and mass loss rate were studied. The desired microstructure for enhanced water corrosion resistance was proposed. 5. The aqueous dissolution behavior of MgO-Nd 2 Zr 2 O 7 composites and Mg 2 SnO 4 in HNO 3 was studied. The dissolution rate of MgO was determined at varied conditions. It was suggested that mechanica l agitation such as magnetic bar stirring is important to enhance the dissolution rate and disintegrate undissolved substance, and is necessary to achieve a complet ed dissolution of MgO. Dissolution of Nd 2 Zr 2 O 7 was achieved in boiling concentrated H 2 SO 4 The experimental results indicate that aqueous reprocessing of such IMF is possible. The dissolution mechanism for Mg 2 SnO 4 as a complex oxide in HNO 3 was proposed. 6. The present study resulted in a research op portunity for performing irradiation test in the ATR. In collaboration with INL techni cians and staff, the author acted as the main participant and completed the irradiation test preparation. Target materials are now being irradiated in the ATR. 26

PAGE 27

CHAPTER 2 BACKGROUND The present chapter briefly summarizes some of the theoretical background and fundamentals required for understanding the res earch work discussed in the following chapters. 2.1 Nuclear Energy and Fuel Cycle Nuclear energy is used worldwide now. As of January 5, 2009, there were 436 nuclear power reactors in operation th roughout the world, generating over 372 gigawatts (GW) of electrical energy according to the World Nuclear Association. 14 The United States operates 104 reactors that produce approximately 20% of total electricity for the country and the government is heavily involved in the operations of the nuclear industry. The nuclear fuel cycle refers to all of the activities of handling fissile materials as main fuels in nuclear reactors. It starts with the extraction of U ore from the ground and terminates with the disposal of radioactive wastes. If the spent nucl ear fuels are subject to direct geological disposal, the fuel cycle is referred to as an open fuel cycle or a once-through fuel cycle. If the spent nuclear fuels are s ubject to reprocessing, the nuclear fuel cycle is referred to as a closed f uel cycle. Details on eac h single step in the fuel cycle will not be discusse d here; instead, a summary of the closed fuel cycle is illustrated in the diagram shown in Figure 2-1 For the closed fuel cycle, the reprocessing plant is an important link t hat connects spent nuc lear fuels, waste management, and fuel fabrication toget her as shown in the figure. 27

PAGE 28

Figure 2-1. Nuclear fuel cycle. (Diagram after Japan Nuclea r Fuel Limited, Rokkasho, Aomori, Japan). 15 2.2 Crystal Structures of Potential IM Materials One of the most important fundamentals in materials science is crystallography because the structure of a material is correla ted to its properties and performance. The investigated materials in this thesis relate to three different types of crystal structures, from the simple rocksalt structure to the co mplex pyrochlore and spinel structures. The three crystal structures are briefly review ed here, with an emphasis on spinel crystal structure. 28

PAGE 29

2.2.1 Rocksalt Structure The general formula for the rocksalt structur e is AX, where A is a cation and X is an anion. A unit cell of rocksalt structure consists of four molecules (Z = 4) and there are 4 anions and 4 cations in total. This struct ure is named after NaCl, and there are over 400 compounds with this type of structure. Most alkali-earth metal oxides such as magnesium oxide (MgO) form the rocksalt stru cture. The space group for the rocksalt structure is mFm 3 ( mmF /23/4 No. 225 in the International Tables 5 hO 16 ). In general, materials with a rocksalt crystal structure ar e highly ionically bonded. Large anions are arranged in cubic close packing (ccp) and all the octahedral interstitial positions are filled with cations. The tetrahedral sites in the structure are all empt y. The coordination number (CN) for both cation and anion is 6. The unit cell of the rocksalt structure is shown in Figure 2-2 using MgO as an example. T he lattice parameter for rocksalt oxides is on the order of ~ 5 Figure 2-2. Cubic rocksalt structure of magnesia. 29

PAGE 30

Figure 2-3 shows cation polyhedra arrangement in the rocksalt structure. The cation octahedra share edges with the nearby ca tion octahedra in this structure. Figure 2-3. Cation polyhedra arrangem ent in the rocksalt structure. The location of the atoms, site symmetry, and atomic coordinates for rocksalt structure are given in Table 2 In the table, the X anion is chosen as the origin to describe the atomic positions as the unit cell shown in Figure 2-2 The bond length of A-X is half the lattice parameter ( a /2), and the bond angles fo r A-X-A and X-A-X are both 90 Table 2. Rocksalt (AX) structure. Ion Location Site Symmetry Coordinates (0,0,0; 0,1/2,1/2; 1/ 2,0,1/2; 1/2,1/2,0)+ 4A 4a O h 0,0,0 4X 4b O h 1/2,1/2,1/2 30

PAGE 31

2.2.2 Pyrochlore Structure The general formula of pyroch lore can be written as A 2 B B 2 X 7 which constitutes two different kinds of cations and one kind of ani on that is usually oxy gen. The space group for pyrochlore structure is mFd3 ( mdF /23/41 No. 227 in the International Tables ). A unit cell of pyrochlore structure consists of eight molecules (Z = 8) and there are 56 anions and 32 cations in total. Pyrochlore can be seen as an anion deficient fluorite superstructur e. There are two types of subcells in the pyrochlore structure, and their relationship wi th fluorite structure is shown in 7 hO 16 Figure 2-4 Fluorite (AX2) A X Type I Cell Type II Cell Vacant Site X X Figure 2-4. Type I and type II s ubcells of pyrochlore structur e derived from the fluorite structure. As shown in the Type I and II cells in Figure 2-4 the A and B cations are located at the corner and face-centered positions. The anions sit in the tetrahedral positions coordinated with A or B cations inside the cube. In the type I cells, the A cations sit on 31

PAGE 32

face diagonals originating from an A ion at the upper righthand corner and the anion at the lower left-hand corner is missing. In the type II cells, the A cations are positioned on face diagonals originating from the lower left-hand corner and the oxygen opposite this corner is missing. There are two types of anions in the subcells The regular anions are coordinated with two A cations and two B cations, and there are six regular anions in every sub cell. The other type of anion is special and coor dinated with four A cations. There is only one special anion lo cated in every sub cell shown in Figure 2-4 as X. A unit cell of the pyrochlore structure consis ts of four type-I s ubcells and four type-II subcells, and each of them sits in an octant wh ich is one eighth of the unit cell. These subcells are arranged in a way such that the same type cubes are diagonally opposite one another, and only different type cubes are next to each other. Figure 2-5 shows the arrangement of the subcells in a simplified unit cell of pyro chlore filled with only two octants on the diagonal directions. Due to this superstructur e, the lattice parameter of pyrochlore is typically ~ 10 twice that of fluorites. Figure 2-5. Two octants of the pyrochlore structure filled with Type I and Type II subcells. 32

PAGE 33

The location of the atoms, site symmetr y, and atomic coordinates are given in Table 2 In the table, the B cation is chosen as the origin for describing the atomic positions in the pyrochlore structure. The perfect octahedra (x = 5/16) or cubic coordination polyhedra (x = 3/8) cannot be simultaneously satisfied. Thus t he value of oxygen x parameter depends on the specif ic chemical composition and differs from one to another, which leads to multiple local struct ure variations. The oxygen parameter can be determined by refining X-ray or neutron dat a obtained from synt hesized materials. Table 2. Pyrochlore (A 2 B B 2 X 6 X) structure data after Subramanian 17 Ion Location Site Symmetry Coordinates (0,0,0; 0,1/2,1/2; 1/ 2,0,1/2; 1/2,1/2,0)+ 16A 16d D 3d 1/2,1/2,1/2; 1/2,1/4, 1/4; 1/4,1/2,1/4; 1/4,1/4,1/2 16B 16c D 3d 0,0,0; 0,1/4,1/4; 1/ 4,0,1/4; 1/4,1/4,0 48X 48f C 2V x,1/8,1/8; -x,7/8,7 /8; 1/4-x,1/8,1/8; 3/4+x,7/8,7/8; 1/8,x,1/8; 7/8,-x,7 /8; 1/8,1/4-x,1/8; 7/8,3/4+x,7/8; 1/8,1/8,x; 7/8,7/8, -x; 1/8,1/8,1/4-x; 7/8,7/8,3/4+x 8X 8b T d 3/8,3/8,3/8; 5/8,5/8,5/8 x for regular octahedra: 0.3125 x for regular cube: 0.375 2.2.3 Spinel Structure The general formula of spinel can be written as AB 2 X 4 which also contains two different types of cations and one type of ani on that is usually oxygen. The space group for spinel structure is mFd 3 ( mdF /23/41 No. 227 in the International Tables 7 hO 16 ). A unit cell of spinel structure consists of ei ght molecules (Z = 8) and there are 32 anions and 24 cations in total. The two types of ca tions are distinguished by their coordination numbers: one is called the A-type cation and occupies a tetrahedral site with 33

PAGE 34

coordination of four; the other one is called the B-type cation and occupies an octahedral site with co ordination of six. The A-type ca tion site is commonly chosen as the origin for the spinel struct ure. The Bravais lattice of the unit cell is a face-centered cubic (fcc), and the basis consists of tw o formula units. M any compounds with the spinel structure have import ant technological applications including use as electronic materials, magnetic materials, refrac tory, and high temperature ceramics for applications in radiation env ironments. In addition, s ilicate spinels are important constituents in the earths mantle. 18 Therefore, the physical and chemical properties of spinels are of general interest in fields ranging from materials physics to geophysics. 19 The structure is named after a natural mineral, magnesium aluminate (MgAl 2 O 4 ). The lattice parameter for spinel is ~ 8-9 slightly smaller than a pyrochlore. The lattice parameter for natural spinel MgAl 2 O 4 is 8.0898(9) 20 A-atom tetrahedral site B-atom octahedral site Figure 2-6. Unit cell of the spinel structure. 34

PAGE 35

2.2.3.1 Detailed structure description Description of spinel depends on the choi ce of setting for the origin in the mFd3 space group and, as ment ioned before, the A-type cation site is taken as the origin here. The best way to see the spinel structure is from the cutaway view in Figure 2-6 As the figure shows, the unit cell is divided equally into eight octants, and each octant contains either an A-site tetrahedron formed by one A-ty pe cation in the center and four anions in the corners, or a distorted cube formed by f our B-type cations and four anions in the corners. The atomic location, the site symmetry, and the atomic coordinates are given in Table 2 The coordinates of the anions loca ted at 32e vary from one composition to another, and thus a parameter u is introduced to identify t he anion positions similar to the parameter x in pyrochlore. For a perfect cubic close-packed (ccp) anion arrangement u = 3/8 (0.375); however, anions in re al spinel structures are usually dilated away from their ideal ccp positions. Th is dilation is a very important factor that may induce some changes to the crystal and influence the structure stability. Table 2. Spinel (AB 2 X 4 ) structure data after Sickafus and Wills 21 Ion Location Site Symmetry Coordinates (0,0,0; 0,1/2,1/2; 1/2,0,1/2; 1/2,1/2,0)+ 8A 8a m 34 0,0,0; 1/4,1/4,1/4 16B 16d m 3 5/8,5/8,5/8; 5/8,7/8, 7/8; 7/8,5/8,7/8; 7/8,7/8,5/8 32X 32e 3m u,u,u; u,-u,-u; -u,u,-u; -u,-u,u; 1/4-u,1/4-u,1/4-u; 1/4+u,1/4+u,1/4-u; 1/4+u,1/4-u,1/4+u; 1/4-u,1/4+u,1/4+u The nearly perfect ccp stacking array of the oxygens is along the [111] direction. The octahedra are joined along edges to form rows and planes parallel to (111) of the 35

PAGE 36

structure, and the tetrahedra provide cross links between layers of octahedra. The polyhedral structure of spinel is illustrated in Figure 2-7 A-type cations occupy tetrahedral-shaped cavities within the anion fr amework (shown in green), whereas the octahedral shaped cavities are occupied by B-type cations (shown in purple). The Atype cation tetrahedra do not share corners or edges with each other, but the B-type cation octahedra do share one edge with nearby octahedra. Besides eight A-type tetrahedra and sixteen Btype octahedra, there are 56 te trahedral vacancies and 16 octahedral vacancies which ar e not shown in the figure. Figure 2-7. Cation polyhedra arrangement in one spinel unit cell. A-type tetrahedral sites are shown in green and B-type oc tahedral sites are shown in purple. 36

PAGE 37

A more convenient way to explain a crystal stru cture is to show t he structure by layer sequence along one crystallographic directi on. The layer sequence along [001] direction for spinel structure is shown in Figure 2-8 The different cation and anion lattice sites are marked and labeled in the diagram. Z=0 Z=1/8 Z=1/4 Z=3/8 Z=1/2 Z=5/8 Z=3/4 Z=7/8 A type cation B type cation Anion Figure 2-8. Layer sequence along the zaxis ([001]) for spinel structure. 2.2.3.2 Cation arrangement Many spinel compounds can accommodate significant amounts of cation disorder and therefore the designations normal spinel, disordered spinel and inversed spinel are introduced for the different configurat ions of cations in spinel compounds. 22 The normal spinel refers to these spinel co mpounds that keep all A-type cations in tetrahedral sites and all B-type cations in octahedral sites. Natural spinel MgAl 2 O 4 is a normal spinel but all synthetic MgAl 2 O 4 has certain degree of cation disorder. The inversed spinel stands for the spinel compounds that retain all the A-type cations in 37

PAGE 38

octahedral sites while one half of the B-type cations in tetrahedral sites and the other half in octahedral sites. The disordered spi nel refers to those compounds whose cation orders are located somewher e between the extremes of the normal and the inversed spinel. To quantify the degree of cation diso rder in disordered spinel compounds, the inversion parameter i is introduced and the structure form ula for spinel with composition M(1)M(2) 2 X 4 can be expressed as follows: 21 4 VI 2 2 i 2 i-2 IV ii-1X]M(1) [M(2)]M(2) [M(1) (2-1) Therefore, for normal spinel i = 0, for inverse spinel i = 1, and for disordered spinel i is between 0 and 1. Many factors can influence the cation inversion such as temperature, cationic r adii, and cationic charge. 23 Following the charge neutrality principle, the combination for cation charges c an be 2-3, 4-2 and 6-1, w here , and refer to the charge of cation. 2.2.3.3 Bonds length and bonds angles The bond lengths and selected lattice distances for A-A, A-B, B-B, A-X, B-X and X-X in spinel structure are listed in Table 2 Table 2. Spinel bond length and selected lattice distances, as a function of lattice parameter ( a ) and anion parameter ( u ). Bond type Bond length (generic) A-A ( 4/3 )a=0.433013a A-B ( 8/11 )a=0.414578a B-B ( 4/2 )a=0.353553a A-X 2 a(u-1/4) B-X a[2(u-3/8) 2 +(5/8-u) 2 ] 1/2 X-X 2 2 a[1/2-u] (shared anions) 38

PAGE 39

The bond angles for the spinel crystal stru cture are independent of lattice parameter ( a ), but vary with anion parameter ( u ), except that the X-AX bond angle is fixed to be 109.47 The relationship is summarized in a plot shown in Figure 2-9 0.360.370.380.390.40 70 80 90 100 110 120 130 A-X-A X-B-X X-A-X A-X-B B-X-B Interbond Angles (degrees)u (anion parameter) Figure 2-9. Bond angles in spin el crystal structure as a function of anion parameter ( u ). 21 2.2.3.4 Stability of spinel an d global instability index (GII) Global Instability Index (GII) evaluates the extent to which the valence sum rule is violated and can be used as an indication for the stability of crystals. Brown 24 has pointed out that values of GII larger than 0. 05 are indicative of stress which produces intrinsic strain in the structure (lattice induced strain). Cryst als with GII >> 0.2 are generally unstable. The value of GII is calculated as follows: N VV GIIN i calcioxi 1 2 ,,) ( (2-2) 39

PAGE 40

where N is the number of atoms in an asymmetric unit, V i,ox is the oxidat ion state and V i,calc is the summation of bond valence shown below: j ij calcisV, (2-3) where j is the number of bonds attached to the atom and equals the coordination number of the atom. The bond valences s ij are calculated by using the equation proposed by Brown and Altermatt 25 : ) exp( B rR sijij ij (2-4) R ij is the bond valence parameter, r ij is the actual distanc e between bonded atoms and B is a universal constant that equals to 0.37 The calculated GII for spinel MgAl 2 O 4 with perfect ccp anion arrangement ( u = 0.375) is 0.94, which is much great er than 0.2. Therefore, spinel with perfect ccp anion arrangement is an extremely unstable structure. As u increases, oxygen is displaced along the [111] direction re sulting in an enlarged tetr ahedral site and a reduced octahedral site. The shift of oxygen ions releases lattice strain which stabilizes the spinel structure and the GII value is low. The reported GII value for MgAl 2 O 4 in literature is only 0.038. 26 This value is even less than 0.05 indicating that the structure of MgAl 2 O 4 is stable and the lattice stain is inconsiderable. 2.3 Radiation Damage in Materials Radiation damage is impor tant for nuclear materials, es pecially nuclear fuels. This section discusses radiation damage mechanisms and the responses of materials to radiation. 40

PAGE 41

2.3.1 Radiation Damage Mechanisms When an energetic ion passes through a material it interacts with the material and losses its energy in several ways. The f our main mechanisms of ion energy deposition in a solid are summar ized schematically in Figure 2-10 Nuclear transmutation is a process in which a nucleus captures an incident ion and forms daughter nuclei by nucleus decay. Nuclear scattering is a proce ss in which an ion loses energy due to elastic scattering by atomic nuclei in target materials. Electron scatte ring is similar to nuclear scattering but the energy loss is due to inelastic scattering by electrons in the target. An ion can also travel through a material without losing energy by having no interaction with the target material. (A) ion capture and transmutation reaction (C) scattering by target electron eFlux of energetic ions (energy E0, atomic no. Z0, mass M0)Target material (atomic density NT, No. ZT, mass MT) (E1, Z1, M1) (E2, Z2, M2) (ET, ZT, MT) (E0-dE/dxlnucleardx, Z0, M0) (E0-dE/dxlelectronicdx,Z0,M0) (E0,Z0,M0) (B) Scattering by target atom (D) no scattering eeFlux of energetic ions (energy E0, atomic no. Z0, mass M0)Target material (atomic density NT, No. ZT, mass MT) (E1, Z1, M1) (E2, Z2, M2) (ET, ZT, MT) (E0-dE/dxlnucleardx, Z0, M0) (E0-dE/dxlelectronicdx,Z0,M0) (E0,Z0,M0) (B) Scattering by target atom (D) no scattering Figure 2-10. Energy deposition mechanisms of incident ions in target materials. 27 41

PAGE 42

The probability for each mechanism to occur depends on the ion-solid interaction parameters described in Figure 2-10 The formula used to calculate the probability can be written as follows: dxNPT (2-5) where P is the probability of i on target interaction, is the interaction cross section, N T is the atomic density of the target, and dx is the thickness of the target. The standard unit for in nuclear science is the barn, and one barn equals to 10 -28 m 2 Different energy deposition mechanism result s in different levels of damage in materials. The transmutation mechanism anni hilates the incident ions and substitutes them with transmutation products, such as fi ssion fragments. This mechanism can not only produce energetic ions that cause significant structural damage, but also change the materials chemistry. It also leads to activation of target materials due to the formation of radionuclides. Therefore, transmutation is an important damage mechanism. Nevertheless, the for transmutation is typically small so the probability for such event to oc cur is rare. For IM materials, this damage mechanism is even less important due to high neutron transpar ency. In a real scenario, the for nuclear and electronic scattering is much larger than t he transmutation in almost all irradiation conditions. As a consequence, the nuclear scattering and electronic scattering should be considered the main damage mechanisms for ion solid interactions. The characteristics of damages resulti ng from nuclear scatt ering and electron scattering are also different. The partitioning of the energy loss of energetic ions in target materials due to el ectronic scattering and nuclear scattering determines which mechanism dominates the process. 28 When the electronic scattering events dominate the energy deposition and the depos ition rate is larger than the threshold values 42

PAGE 43

(typically ranging from ~1 to 20 keV/ nm for insulators depending on materials 29 ), the radiation damage effects are mainly due to atomic redistributions associated with electronic excitation. There is still co ntroversy regarding the irradiation damage mechanisms in this energy deposition range. 30 Among these proposed models, the Coulomb explosion model 31,32 and the thermal spike model 33-35 are generally accepted. Figure 2-11 schematically shows the concept of Coulomb explosion. When a swift heavy ion such as a fission fragment impinges in a target material, it ionizes the atoms within a column along the incident ions path. As a result, these ionized target atoms are all positively charged and subsequently repel one another. The Coulomb force is large enough to knock these atoms out of its lattice sites and create a damage track along the ion trajectory. The damage track can be amor phous or remains certain crystallinity. The electron scattering is the main damage mechanism which is responsible for fission fragm ents damage due to their high kinetic energy (~ 70 MeV 100 MeV). e+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + e-e-e-e-e-e-e-eFigure 2-11. Illustrati on of the process of Coulomb explosion. 27 43

PAGE 44

Figure 2-12 shows typical TEM micrographs of i on tracks formed in ceramic materials by swift heavy ion irradiation. 36 The thermal spike model has also been proposed to explain the formation of ion track. Due to large elec tronic energy deposition rate, a high-temperature region can be formed along t he trajectory of an energetic ion and may exceed the melting point of the material. As a result, it melts within the trajectory and subsequently cools down at a high cooling rate of 10 14 10 15 K/s. The high quenching rate results in structure damage which can lead to amorphization. A B Figure 2-12. Ion trac k formation in Nd 2 Zr 2 O 7 irradiated by 120 MeV I ions to a fluence of 5.6 10 10 cm -2 at A) normal electron beam condition, and B) tilted condition. (Image after Lutique 36 ). If the nuclear stopping power exceeds the electronic stopping power and dominates the process, the target atoms are displaced from their original lattices primarily by ballistic collisions. When an atom leaves its original atomic si te it goes to an interstitial position. The Frenkel pair defects are produced in this way which is shown schematically in Figure 2-13 44

PAGE 45

Vacancy Interstitial Figure 2-13. Frenkel pair formation resulted from displacement damage. The minimum energy required to create a stabl e interstitial-vacancy pair is called the displacement threshold energy (E d ), and the value depends on target material. When an incident particle (charged or neut ral) traverses in target mate rials, it transfers kinetic energy to a target atom and knocks it out from it s lattice site. The ejected target atom is known as a primary knock-on at om or PKA. The PKA travels in the target material and may collide with other target atoms which are on the way, and these knock-on atoms are referred to second knock-on atoms or SKAs As long as the energy of these moving particles is greater than E d the process continues and more vacancies and interstitials are generated, forming a displacement casc ade. The density of defects usually increases quickly to a peak value in about 10 -15 s and then drops of f to an equilibrium level during the next 10 -12 10 -11 s (equivalent to 10 100 lattice vibrations), which is 45

PAGE 46

due to spontaneously annihilation by in terstitial vacancy recombination. 27 The defects left behind after the cascade are referred to re sidual defects. For the most part, these defects are responsible for the structure an d property evolution of irradiated materials and need to be characterized. To quantify the atomic displacement induced by ion irradiation, the displacement per atom (dpa) unit is the most common unit of dose used for discussing radiation damage under nuclear stopping regime. One dpa equals the ion fluenc e at which every target atom has been knocked from its lattice si te one time on average. The dpa unit can be converted from the unit of (number of i-v pairs)/nm/ion using the equation below: ][] # [ ]/[ ]/[ ] /10 1 ][ # [] # [3 2 7dpa atom cmatomsN cmions nmcm ionnmionnmT (2-6) where # is number of interstitial-vacancy pairs, is the ion fluence, and N T is the atomic density. The number of vacancies generated during irradiation can be calculated by computer programs, such as the Transport of Ions in Matter (TRIM) developed by Ziegler 37 2.3.2 Radiation Damage in Nuclear Fuels Nuclear fuels are subjected to irradiation of nearly all kinds of origins, such as thermal (< 0.1 MeV) and fast neutrons (> 0.1 MeV) from fission reactions, -particles (~ 6 MeV) and heavy recoil atoms (100 keV) fr om natural actinides-decay, and fission fragments. These energetic parti cles lead to different damage processes, small or large displacement cascades, and electronic intera ctions. Neutron irradiation causes transmutation damage and ballistic displacement damage; -particles and recoil atoms mainly lead to ballistic displacement dam age; fission fragments lead to formation of fission tracks probably due to coul omb explosion and thermal spik e. It is important to 46

PAGE 47

understand the different damage mechanism s and distinguish them because the response of materials differs from one mechanism to another. 2.3.3 Irradiation Induced Amorphization and Crystal Structure Considerations Radiation tolerance of a material refers to the ability to resist any type of radiation damage, such as amorphization, defects, ma croscopic swelling, phase transformation, etc.. 38 This section focuses on one particular criterion which is the resistance to radiation induced amorphization. The response to radiation damage of various oxides is considered to be related to their crystal structures. 38 Even though great efforts have been made on investigating ion-beam-induced amorphization of ceramic oxide materials, a universal model for predicting the susceptibility of oxides to irradiationinduced-amorphiza tion has not been developed so far. 2,9 In general, oxide compounds with simple structures such as rocksalt and fluorite are relatively irradiation resistant, and compounds with complex structures are susceptible to irradiation-induced amorphiza tion. However, there are some exceptions such as some spinel and pyrochlore compounds, which have been found highly resistant to r adiation induced amorphization. 39 The mechanisms behind the enhanced irradiation tolerance for th ese complex oxides will be discussed in Chapter 3 and 4. The relative susceptib ility of crystalline ceramic phases to amorphization depends on not only crystal structure, but also on physicochemical factors such as bond strength, and bond type. The critical am orphization dose is the ion dose at which a crystalline material is comple tely amorphized, and it is commonly used to evaluate the susceptibility of a material to amorphization. 47

PAGE 48

2.3.4 Stopping and Range of Ions in Matter (SRIM) SRIM is a set of programs which utilize a quantum mechanical treat ment of ion-atom collisions to calculate the st opping and range of ions from 10 eV up to 2 GeV into matter. This section focuses on one of the useful pr ograms, which is call ed TRIM (the Transport of Ions in Matter). 37 TRIM is a program based on M onte Carlo methods. It provi des results of the final 3D distribution of the ions and the kinetic phenomena associated with the ion's energy loss, such as target damage, sputtering, ionization, and phonon production. It can handle complex target materials with up to eight different layers. A limitation to this calculation is that t here is no temperature effect include d in TRIM, so "self-annealing" or thermal annealing is not considered. 37 In order to run TRIM, seve ral parameters regarding the incident ions and the target materials have to be defined in the program. Figure 2-14 shows the input window where the initial parameters required by the pr ogram are entered (example is based on irradiation of Mg 2 SnO 4 by 1 MeV Kr 2+ ). The incident ion is defined by atomic number, ion mass (usually the most abundant isotope), ion energy (units = keV), and incident angle with respect to the target surface. If the ion is perpendicular to the target surface, the incident angle is defined as 0 The number of ions can be selected up to 9999999, but adequate statistics can be obtained from thousands of io ns or even less. The parameters for target material s include chemical compositi on, target thickness, and target density. The displacement threshold energy (E d ), surface binding energy (SBE), and lattice binding energy (LBE) for element s are defined in TRIM by default. These 48

PAGE 49

parameters are crucial for obtaining representative results. In most cases, users have to find out specific values of these energies based on specif ic materials and structure. Figure 2-14. Transport of Ions in Matter (TRIM) calculation parameters window (V 2006.02). Several output files can be generated from TRIM calculatio n. The RANGE.TXT file provides information to calculate the implanted ion concentration in a target material. The COLLISON.TXT file keeps a re cord of the detailed results of collision cascades. It contains comprehensive information about the i on atom interaction, such as number of displacement collisions, number of generated vacancies, electronic stopping power, and recoil energy. The VACANCY.TXT file provi des the number of vacancies generated by 49

PAGE 50

incident ions and recoil ions. With these values, the dpa can be calculated for the target material using the method described in section 2.3.1 The number of vacancies can be converted into the energy which is required to create these vacancies. In the IONIZ.TXT file, the energy losses of ions and recoils due to electron excitation (inelastic electronic scattering) are provided and can be used to calculate the electron stopping powder. The PHONON.TXT file lists the number of phonons generated as a function of depth. If the phonon energy is known, the number of phonons can be converted into phonon energy. The total nuclear stopping power is simply a summation of this phonon energy and the vacancy production energy which can be obtained from the VACANCY.TXT file. 2.4 Chemical Dissolution of Metal Oxides The dissolution behavior and kinetics of meta l oxides in acidic solutions has been extensively studied and the knowledge has been appl ied in various fields such as metal etching, extraction of ores, removal of deposits from thermal power equipment, and nuclear fuel reprocessing. 40-49 Since dissolution of spent nuc lear fuels is always the first step for fuel reprocessing, the dissolution behavior is of special interest and worth investigating. The dissoluti on behavior of metal oxides in aqueous solutions is highly dependent on its structural characteristics su ch as crystal structure, the bond type and energy, coordination, and t he electronic structure. 50 2.4.1 Dissolution Mechanism The dissolution of metal oxides in aqueou s solutions can be simply viewed as destruction of metal-oxygen bonds and subs equent separation and stabilization in solutions. In general, the dissolution mechanis ms can be classified into two categories: 50

PAGE 51

dissolution involving charge transfer and di ssolution without charge transfer. For insulating metal oxides, i.e. primarily ioni c oxides and covalent oxides, the dissolution does not involve charge transfer. The dissolu tion of insulating metal oxides is mainly governed by electrophilic or nucleophilic attack, or both. Figure 2-15 illustrates the two dissolution mechanisms. M denotes a meta l cation, O represents an oxygen ion, and HX is an acid in solution. The electrophilic attack refers to the process that surface oxygen atoms are attacked by positively ch arges ions from solution. Hydrogen ion (H + or presents as H 3 O + ) is an effective reagent to achiev e this attack. The nucleophilic attack occurs on the surface metal ions wh ich attract negatively charged ions from solution. A typical nucleophilic anion is OH Some other anions are more powerful than OH such as fluoride (F ), phosphate ion ((PO 4 ) 3), sulfate ion ((SO 4 ) 2), and complexation ligands, such as ethyl enediaminetetraacetic acid (EDTA). +HX M M O M M O M M O M M O X H + HX M M O M M O H X M M X OH M M X OH Electrophilic attack Nucleophilic attack Figure 2-15. Illustration of nucleophilic and electrophilic attack. 51

PAGE 52

The dissolution behavior of insulating oxi des depends on their bond type. Most ionic oxides such as compounds with rocksalt struct ure (i.e. MgO) can be readily dissolved in water or acids. In covalent oxides, the affinity of oxygen ions for protons decreases and the oxides become more acidic. The disso lution behavior of these oxides differs depending on the acid str ength. Strong acidic oxides such as CrO 3 and Mn 2 O 7 can be dissolved in aqueous solutions easily, while the so lubility of weakly acidic oxides such as Al 2 O 3 SnO 2 and ZrO 2 is very limited. Usually in a practical attempt to achieve dissolution of weakly acidic covalent oxides, high pH enchants such as strong bases are more effective than low pH enchants su ch as strong acids, because OH ions are able to provide an adequate thermody namic driving force to break the strong covalent bonds in a reasonable kinetic regime. Clearly, stronger anions such as F can be used to achieve a more effective and efficient dissolution. The dissolution behavior of semiconducting oxides and transition metal oxides is more complicated than insulating oxides, bec ause charge transfer may be involved in the dissolution reaction. The dissolution accompanying with the charge transfer is called redox dissolution. C harge transfer does not lead to dissolution, and requires the presence of at least one of the nucleophilic and el ectrophilic attack at t he critical stage. Thus, charge transfer is an addit ional, parallel pathway that may lead metal ions to be more susceptible to nucleophili c or electrophilic attack. In many cases, the dissolution kinetics can be altered dramatically and the dissolution rate may be several orders of magnitude faster. 51 An exception is dissolution of the transition metal oxides with metal ions in high oxidation states, i.e. TiO 2 and ZrO 2 These oxides share the same characteristics as 52

PAGE 53

the covalent oxides in the way they in teract with aqueous s pecies, such as Al 2 O 3 SiO 2 and SnO 2 Since complex metal oxides contain at le ast two cations in the structure, their dissolution behavior can be characterized as ei ther congruent dissolution or incongruent dissolution. If an oxide is dissolved congr uently, the composition of the oxide and the dissolved solute stoichiometrically match. By contrast, if an oxide is dissolved incongruently, the composition of the solute in solution does not match that of the solid. The dissolution can be completely selective, meaning some phases can be dissolved completely but others cannot be dissolved at a ll. In this case, the soluble phases may be leached out from the insoluble phases whic h are intact during dissolution, or it may be a congruent dissolution followed by precipitat ion of insoluble phases. An example of selective dissolution is leaching perovskite CaTiO 3 : only Ca 2+ ions can be dissolved in acidic solutions but dissolution of Ti 4+ is unnoticeable. Nevertheless, the interpenetrating TiO 6 is attacked and followed by a form ation of an amorphous film and subsequent recrystallization of brookite and anatase. 52,53 2.4.2 Dissolution Kinetics Dissolution of metal oxide in acids in volves production and transfer of multiple species, and is a heterogeneou s process. It consists of a few consecutive steps, including (1) diffusion of the r eactant to the surface, (2) adsor ption of reactant molecules, (3) chemical reaction, (4) desorption of re action products, and (5) diffusion of reaction products into the solution. 54 Steps (1) and (5) are determined by diffusion kinetics, and steps (2), (3), and (4) are usually characteri zed as chemical processes. The overall dissolution rate is determined by the slowest step. In general, the rate of dissolution of 53

PAGE 54

a solute in a solvent depends on the in trinsic mass transfer coefficient k s1 the surface area S, and the driving force of concentration difference (C eq C) as shown below: )(1 teq sCCSkR (2-7) where C t is the ion concentration in solution at time t 2.4.2.1 Kinetics models The kinetics for dissolution of metal oxides in the acidic media can be linked with the well-known shrinking core models, which are classified into three categories: dissolution controlled by diffusion through liquid film, disso lution controlled by surface reaction, and dissolution controlled by diffusion through build-up product layers. 55 The rate controlling mechanism depends on the type of reaction and reaction conditions such as temperature, viscosity of solvent, concent rations of reactants and products, and fluid dynamics. Models have been derived for thr ee simple geometries which are sphere, infinite flat plate, and infinite cylinder. Taking spheres as an example for the following reaction: both) (or products solid or fluid bB(solid) A(fluid) (2-8) where b is the stoichiometric coefficient. The three kinet ics models can be expressed as follows: tkxF for film diffusion control (2-9) tkx xD2 3 2)1(2)1(31 for product layer diffusion control (2-10) tkxS3 1)1(1 for surface reaction control (2-11) where x is the reacted fraction at time t and k F k S and k D are apparent rate constants given in the following equations: 03 r bkC kF (2-12) 54

PAGE 55

2 06 r bDC kD (2-13) 0r bkC kn s (2-14) where is the density of t he reacted particle, r 0 is the initial particle radius, k is the intrinsic mass transfer coefficient between fluid and solid, D is effective diffusivity, C is the bulk concentration of reactant, and n is the order of reaction. 55 2.4.2.2 Effect of temperature The temperature effect on dissolution rate of metal oxide can be expressed in principle, by the empi rical Arrhenius equation: 51 g aR E T R )/1( ln (2-15) where R is dissolution rate, T is temperature in Kelvin, E a is apparent activation energy, and R g is gas constant. The difference of activation energy values between a diffusion controlled process and a chemical reaction c ontrolled process is typically large enough to distinguish these two mechanisms. The ac tivation energies for diffusion in water are in the range of 4 to 12 kJ/mo l, so that experimental E a values near or below ~15 kJ/mol indicate a significant contri bution from diffusion kinetics. 56 Dissolution reaction with an activation energy that is higher than the value is usually considered a reaction controlled process. 56 2.4.2.3 Dissolution rate determining factors The effect of temperature on dissolution rate is discussed above, however, there are many other rate determining factors which hav e not been covered, ye t. As a summary, the most important factor s are provided in the fo llowing list for reference: 57 55

PAGE 56

1. reaction temperature 2. pH of the solution 3. stoichiometry of metal oxide 4. redox potential of the solution 5. concentrations of oxidizing and reducing component in a redox system 6. presence of complexation ligand 7. interfacial characteristics, such as cr ystallographic orientati on and density of defects 8. presence of catalysts 9. doping of metal oxides. 2.5 Description of the Advanced Test Reactor (ATR) Aiming at developing nuclear energy and promoting collaborat ion between academia, industry and federal government the U.S. Department of Energy (DOE) designated the ATR at the Idaho National Laboratory (INL) as a National Sci entific User Facility (NSUF) in April 2007. The ATR is located at the AT R Complex on the INL site. It is a water cooled thermal test reactor with a maximum thermal power of 250 MW th and a maximum thermal neutron flux of 10 15 n/cm 2 s. The ATR is cooled by light wa ter pressurized at ~2.5 MPa. When operating at full powder, the inlet and outlet water te mperatures ar e about 52C and 71C, respectively. As a research based test reactor, the ATR offers versatile irradiation conditions with different neutron fl uxes and heating rates at different locations. The configuration of the ATR is show in Figure 2-16 Two types of experiments can be performed in the ATR: one is called static drop-in test and the other one is called instrumented-lead test. In the static test, testing materials are dropped into the reactor with no dynamic control of the irradiati on environment. The instrumented-lead test offers more flexibility and controls on the irradiation conditions such as varying temperature, gas mixture, c oolant and coolant flow, but it costs much more than the static test and requires a much longer preparation time. 56

PAGE 57

Figure 2-16. Advanced Test Reactor Core. 58 57

PAGE 58

CHAPTER 3 MATERIALS SELECTION, SYNTHESIS AND CHARACTERIZATION 3.1 Materials Selection As mentioned previously in Chapter 1, in the present work material selection is narrowed down to MgO based composites and a single phase spinel compound. The selection process is discussed in this section. 3.1.1 MgO Based Composites As a potential IM material, MgO has seve ral favorable properties such as high melting point (2827C) 59 high thermal conductivity (13 W/Km at 1000C) 59 low neutron adsorption cross section, and good irradiation tolerance. 60,61 To overcome the poor hydration resistance of MgO, the concept of MgO based composites has been proposed and thus the secondary phase should be selected. Very recently, there has been consider able interest in utilizing pyrochlore compounds for immobilization of nuclear wast e, mainly due to the good irradiation and chemical stability. Series of compounds wi th pyrochlore structures have been studied, such as titanate pyrochlores 62 zirconate pyrochlores 63 and stannate pyrochlores 64 Lanthanide titanate pyrochlore s are generally susceptible to radiation-induced amorphization. 62 The lanthanide stannate pyrochlores show greater variation in their response to ion irradiation: La, Nd, and Gd stannates can be amorphized at relatively low dpa and their critical amor phization temperatures are correspondingly high, while Y and Er stannates cannot be amorphized even at 25 K. 64 Most zirconate pyrochlores are irradiation-resistant. 63,65 Gd 2 Zr 2 O 7 cannot be amorphized even at doses as high as 58

PAGE 59

~100 dpa at room temperature, showing excellent irradiation stability. However, Gd has extremely high neutron ads orption cross section, which ex cludes it from IM applications. Another zirconate pyrochlore Nd 2 Zr 2 O 7 also shows excellent ir radiation stability. It has been reported that Nd 2 Zr 2 O 7 transferred to a defect fluorite structure at ~7 dpa and 25 K under 1.5 MeV Xe + irradiation but complete amorphization could not be achieved. 63,65 Moreover, the resistance of Nd 2 Zr 2 O 7 against fission fragment impact has also been studied. 36 Furthermore, an adequate ther mal conductivity of 6 Wm -1 K -1 at 1000 C for the MgO-Nd 2 Zr 2 O 7 composite with 90 vol% MgO has been reported. 36,66 Last but not least, Nd 2 Zr 2 O 7 is chemically inert in general and is expected to be compatible with coolant water and cladding at 300 C. Based on thes e considerations, Nd 2 Zr 2 O 7 was selected as the additive to MgO, and the performance of MgO-Nd 2 Zr 2 O 7 composites as potential IM materials was evaluated thereafter. 3.1.2 Single Phase Spinel Compound Similar to pyrochlore, spinel has a remarkable range of compos itions and over 200 compounds with spinel crystal structure have been reported. 21,67 Unlike pyrochlore of which the irradiation stability has been systematically studied, the irradiation behavior of spinel compounds is not well investigated except for MgAl 2 O 4 As discussed in Chapter 2, spinel is a complex metal crystal structur e that is resistant to radiation damage. Even though MgAl 2 O 4 shows poor irradiation resist ance to fission fragments damage, other spinel compounds may exhibit good irradiation resistance to both ballistic displacement and fissi on fragment damage. Spinel selection was bas ed on a screening type study. The neutron adsorption cross section ( n ) was the first selection criteri on. A summary plot based on the 59

PAGE 60

periodic table for the n of elements is shown in Figure 3-1 to assist material selection. One criterion for IM materials is that n should be less than 2.7 barns, so the elements in green and yellow ar e highly interested. 68 In order to expand th e selection and explore more possibilities, the elements in gray and blue could also be considered. Elements in red were not considered as potential IM c onstituents for spinel compounds due to their unfavorably large n Figure 3-1. Neutron adsorpti on cross section of elements. 69 As discussed in Chapter 2, the spinel compounds can be classified into three categories based on cation charge: 6-1 spinel, 4-2 spinel, and 2-3 spinel (number refers to cation charge). Spinel solid solutions we re not the scope of th e current study. The most common 2-3 type spinels ar e aluminate spinels, chromite spinels, gallate spinels, ferrite spinels, and indium oxide spi nels. The A-site cation can be Ni 2+ Cu 2+ (most are not thermodynamically stable), Mg 2+ Co 2+ Zn 2+ Fe 2+ Mn 2+ and Cd 2+ Mn, In, and Cd have large n and their spinel compounds were not considered in this study. Among 60

PAGE 61

other 2-3 type spinels, the i rradiation tolerance of MgCr 2 O 4 ZnAl 2 O 4 MgGa 2 O 4 ZnFe 2 O 4 MgFe 2 O 4 NiFe 2 O 4 and Fe 3 O 4 has been studied either experimentally or theoretically using computer simulation. 70-77 However, none of them can be concluded to possess better irradiation tolerance than MgAl 2 O 4 Computer simulation suggests that the 2-3 type spinels with less cation di sordering exhibit better ability to suppress defects formation. Therefor e, aluminate spinels and chromi te spinels are worth further investigation, particularly NiAl 2 O 4 NiCr 2 O 4 ZnAl 2 O 4 and ZnCr 2 O 4 The 6-1 type spinels are rare and not stable at high temperatur es. Their melting temperatures are typically low (< 1000 C), so the 6-1 type spinels were not considered. 78,79 The 4-2 type spinels consist of s ilicate spinels, germanate spinels, titanate spinels and stannate spinels. The irradiation behavior of Mg 2 SiO 4 has been studied but the results show it is not a radiation resistant material. 80 Other silicate spinels such as Co 2 SiO 4 81 Zn 2 SiO 4 82 Fe 2 SiO 4 83 and Mn 2 SiO 4 84 have low melting temperatures less than 1512 C. The irradiation behavior of germanate spinels, titanate spinels and stannate spinels has not been reported in open literature. Based on literature survey, MgAl 2 O 4 still seems to be the most irradiation tolerant spinel among the current inve stigated spinel compounds. As a result, selection was focused on the spinel compounds of which t he irradiation behavior is not known instead of continuing to study a spinel material that has been investigated. Due to little knowledge about the irradiation behavior of 4-2 type spinels, it is of significance to investigate 4-2 type spinels for screening study. Selection was first limited to stannates due to the low n of Sn, and then finalized at magnesium stannate (Mg 2 SnO 4 ) since Mg also has low n 61

PAGE 62

Although the thermodynamic stability of the stannate spinel s has been referred to by several authors, the results fo r the enthalpy of formation of the oxides are disputable. 8588 There is relatively limited amount of research availa ble on the Mg-Sn-O system, and no reliable temperature-compos ition phase diagram exists. Nevertheless, the existence of the stable phase Mg 2 SnO 4 has been found, and dense polycrystalline Mg 2 SnO 4 can be achieved by sintering at 1600 C. 86,88-90 Mg 2 SnO 4 has a similar crystal structure to MgAl 2 O 4 with the same space group No. 227 ( m3Fd ). The tetravalent Sn ions are located only at octahedral sites; while one half of the divalent Mg ions occupy octahedral sites, and the ot her half occupy tetrahedral sites. Sn has low n (0.626 10 -28 m 2 ) which is desirable for IM materials. Mo reover, the atomic weight of Sn is 118.710 g/mol and the ionic radius of Sn 4+ (VI) is 0.69 which makes Sn a much heavier and larger ion than Al 3+ (VI). 91 Since the main barrier to t he motion of target atoms out of their original lattice positions and result in damage tracks (primary damage phenomenon in fission fragment damage) is the existence of neighbor ing target atoms which are simply in the way, the compound with the same crystal structure but bigger and heavier atoms may be more resistant to swift heavy ion damage. The inverse spinel structure also makes Mg 2 SnO 4 interesting to study because very little is know about the irradiation stability of inverse spinel experimentally. 3.2 Experimental Procedures All the testing samples were synthesized fr om raw materials in the electroceramics processing lab in the Department of Materials Science and E ngineering at the University of Florida. 62

PAGE 63

3.2.1 Fabrication of MgO-Nd 2 Zr 2 O 7 Composites 3.2.1.1 Solid state synthesis of Nd 2 Zr 2 O 7 Solid state processing is an important synthesis method fo r ceramic materials. In solid state processing, high purity oxi de powders of the co mponents are added to a vessel that contains an optimized size and volu me of milling media. A solvent, such as water or ethanol, is added to improve the effe ctiveness of the process due to the higher energy utilization of the mill, to reduce the wear of t he milling media and any sample contamination that may result from such wear. A dispersant is added to the slurry to prevent the milled powder fr om agglomeration as the parti cle size decreases during milling. The vessel will then be rolled for a specific period of time, during which the media will both mix and crush the oxid e powders, producing a homogeneous mixture and narrow particle size distribution. The c haracteristics of the powder after milling depends on the particle size of the starting sample, sample volume, media type and distribution, mill time, and mill s peed. An ideal solid state pr ocess would have all of the parameters optimized to produce a consistent particle si ze that would minimize the variations in calcination and final grain size in a sintered pellet. To synthesize pyrochlore Nd 2 Zr 2 O 7 equimolar ratios of Nd 2 O 3 (Alfa Aesar 99.9%) and ZrO 2 (Alfa Aesar 99.7%) were weighed to 0.0005 g and added to the milling media in a 102 mm diameter Te flon ball mill jar with 70 ml of deionized (DI) water and 2 wt% of ammonium polyacrylate dispersant (PAA, Darvan 821A). The milling media contained 35 g of 10 mm YSZ spheres and 70 g of 3 mm YSZ spheres. The slurry was milled for 24 hours on the ball mill at 85 rp m, poured onto a 0.0254 mm thick Teflon sheet lining a 305 mm square glass dish, and co vered with aluminum foil punctured with small holes. The slurry was dried overnight in an oven at 120 C. The dried powder was 63

PAGE 64

then ground with a corundum mortar and pestle and sieved through a 212 m mesh. The powder was then placed in an al umina crucible and calcined at 1350 C in air for 12 h. After calcination, the Nd 2 Zr 2 O 7 was added to the prepared media in the Teflon ball mill jar with 70 ml of DI water and 2 wt% of PAA dispersant. The slurry was milled for another 24 h on the ball mill, and dried overnight at 120 C. The dried powder was then ground with the corundum mortar and pestle and sieved through the 212 m mesh. Figure 3-2 shows the flowchart for the solid state synthesis of Nd 2 Zr 2 O 7 Slurry Preparation Ball Milling Drying Sieving 212 m Calcination Nd2Zr2O7Powder Sieving 212 m Grinding Grinding Nd2O3 ZrO2 24 h 24 h 12 h 1350 C 120 C Oversize Oversize Ball Milling 24 h Figure 3-2. Solid st ate synthesis of Nd 2 Zr 2 O 7 3.2.1.2 Sol-gel synthesis of Nd 2 Zr 2 O 7 The objective of using so l-gel method to make Nd 2 Zr 2 O 7 powder is to reduce the particle size and subsequently achieve dist inct microstructure from solid state processing. Since Nd 2 Zr 2 O 7 is added as a secondary phas e to MgO to serve as a 64

PAGE 65

hydration barrier, it is envisioned t hat a homogeneous dis persion of small Nd 2 Zr 2 O 7 grains in the composites would lead to high thermal c onductivity and adequate hydration resistance. The sol-gel process is a wet chemical r oute for making ceramic and glass materials by polymerization reactions. In the solgel process, organo-metallic precursors are used and react in liquid solutions through hydrolysis and condensation reactions leading to a new phase formation. The powder produced through so l-gel process has several advantages compared with the powder produced in the solid state reaction such as higher purity, better homogeneity, smaller parti cle size, and narrower size distribution. 92 To prepare Nd 2 Zr 2 O 7 powder by the sol-gel method, a 1 M solution was prepared by dissolving 0.02 mol of Nd(NO 3 ) 3 H 2 O (99.9%, Alfa Aesar) in 20 ml of Acetic acid (99+%, Alfa Aesar). The solution was stirred at 300 rpm and heated to 1 05 C for 5-10 min to evaporate the water. The solution was then cooled to room temperature, and zirconium (IV) n-propoxide (70% w/w in n-propanol, Alfa Aesar) was added to the solution in an equimolar ratio. The solution was stirr ed for another 15 min, and 5 ml of deionized water was then added drop by drop while stirri ng. The sol was dried at 120 C in oven for 24 h and then pre-heated to 600 C for 2 h to burnout th e organic materials. The obtained powder was ground and subsequently calcined at 1200 C for 6 h to achieve the pyrochlore phase. The calcined pow der was ground again a nd sieved through 45 m mesh. Figure 3-3 summarizes the sol gel synthesis of Nd 2 Zr 2 O 7 in a flowchart. 65

PAGE 66

Grinding Precursors Nd(NO3)3Zr[O(CH2)2CH3]4 Sol Preparation Acetic AcidDI Water Drying 24 h 120 C Organic Burnout2 h 600 C Nd2Zr2O7Powder Sieving 45 m Grinding Oversize 1200 C 6 h Calcination Figure 3-3. Sol gel synthesis of Nd 2 Zr 2 O 7 3.2.1.3 Synthesis of MgO Magnesium carbonate (Mg 2 CO 3 Alfa Aesar ~40-43.5% MgO) was loosely packed into an alumina crucible and calcined at 600 C for 2 h, and then the temperature was increased to 900 C for another 2 h. 3.2.1.4 Powder mixing Cercer composites with MgO volume fr action varying from 40 % to 70 % were fabricated. The powders were mixed in th ree different ways, in cluding mortar and pestle mixing, water magnetic bar stirri ng and ball milling to produce different microstructures. In the mortar and pestle mixing method, MgO and Nd 2 Zr 2 O 7 were added to an alumina mortar and pestle and ground for 5 mi nutes to combine the oxide constituents. 66

PAGE 67

In the water magnetic bar stirring method, MgO and Nd 2 Zr 2 O 7 were added to a beaker filled with 100 ml of DI water. The sl urry was then stirred wi th a magnetic bar for 6 h at 200 rpm and dried overnight in an oven at 120 C. The dried powder was ground and sieved through 300 m mesh screen before being pressed into pellets. Yates 93 has found completed hydration of MgO after water magnetic bar stirring and subsequent drying, and applied an additional calcination step at 1000 C for 2 h. However, in the present study, the additional calcination was not used. In the ball milling method, MgO and Nd 2 Zr 2 O 7 were added to YSZ milling media in a Teflon ball mill jar with 70 ml of anhydrous ethanol (99.5%, Acros Organics) and 2 wt% of PAA dispersant. The slurry was mill ed for 24 h and dried in hood at ambient condition overnight. The dried pow der was then ground and sieved through 45 m mesh screen before being pressed into pellets. 3.2.1.5 Pellet formation The flowchart for fabricati on of the composites through the different mixing methods is shown in Figure 3-4 To synthesize the pellets, the powder mixture and ~2 wt% of binder (Celvol 103 Polyvinyl Alcohol, PVA) were added to the mortar and pestle and ground to combine thoroughly. The powder was dried in oven at 120 C for 5 min. Approximately 0.2 0.5 g of the powder was added to a 7 13 mm punch and die set cleaned with acetone, lubricated with WD40, and pressed with 200 MPa on a Carver press. The pellet was removed from the di e and examined for cracks and surface finish. A geometric green density was ca lculated, and the pellets that met or exceeded the 50% of theoretical density were sinter ed in air at 1400 1650 C. The sintered pellets were checked for cracks and surface finish. The weight and dimension of the pellets 67

PAGE 68

were measured using a digital balance and a digital caliper, and the density of the pellets was determined first geometrically and then using Archimedes method. The density was then compared to the theoretical values to get sample porosity. Weigh Nd2Zr2O7and MgO Powders Add Nd2Zr2O7and MgO to the Mortar Prepare the Slurry Combine Using the Pestle Sieve the Mixed Nd2Zr2O7 and MgO Add Nd2Zr2O7and MgO to Water Stir at 200 rpm Ball Milling Dry the Slurry Grind and Sieve Nd2Zr2O7and MgO Pellet Formation Sintering Method 1 Method 2 Method 3 Figure 3-4. Processing flowchart for fabric ation of composite pellets through three different mixing methods. 3.2.2 Fabrication of Single Phase Spinel Mg 2 SnO 4 3.2.2.1 Solid stat e synthesis of Mg 2 SnO 4 Mg 2 SnO 4 samples were synthesized through solid state processing. Stoichiometric ratios of MgO (Cerac Inc. 99.95% or obtained from ca lcination of MgCO 3 ) and SnO 2 (Alfa Aesar 99.9%) were added to YSZ milli ng media in a Teflon jar with 70 ml of anhydrous ethanol and 2 wt% PAA dispersant. The slurry was milled for 24 h on the ball mill at 85 rpm and dried in hood at ambient condition ov ernight. The dried powder was ground, sieved and calcined at 1200 C in air for 12 h to achieve the spinel phase. 68

PAGE 69

After calcination, the Mg 2 SnO 4 powder was added again to the YSZ media in the Teflon ball mill jar with 70 ml of deionized water and 2 wt% PAA dispersant. The slurry was milled for another 72 h on the ball mill, and dried overnight in an oven at 120 C. The dried powder was finally ground and sieved through a 212 m mesh. 3.2.2.2 Pellet formation The overall processing for Mg 2 SnO 4 is shown in Figure 3-5 The green pellets of Mg 2 SnO 4 were fabricated in the same way as the composites described in section 3.2.1.5 The green pellets were si ntered in air at 1400 1550 C for different periods of time. The density of pellets was measur ed in the same way as the composites. 2MgO + SnO2 Ball milling for 24 h Calcining at 1200 C for 12 h Forming pellets Burning out binder at 450 oC for 2 h Grinding and sieving through 212 m mesh Ethanol Dispersant Binder Drying for 16 h in air in fume hood Sintering at 1400 -1550 oC Grinding and sieving through 212 m mesh Ball milling for 72 h Drying for 16 h in air in fume hood Deionized water Dispersant 2MgO + SnO2 Ball milling for 24 h Calcining at 1200 C for 12 h Forming pellets Burning out binder at 450 oC for 2 h Grinding and sieving through 212 m mesh Ethanol Dispersant Binder Drying for 16 h in air in fume hood Deionized water Dispersant Ball milling for 72 h Drying for 16 h in air in fume hood Grinding and sieving through 212 m mesh Sintering at 1400 -1550 oC Figure 3-5. Processing flowchart for solid state synthesis of Mg 2 SnO 4 69

PAGE 70

3.2.3 Characterization 3.2.3.1 X-ray diffraction The crystal structures of raw materials, calcined powders and sintered pellets were characterized by X-ray diffractometry (XRD, Philips APD 3720). The XRD was conducted using CuK radiation and the operation powe r for the generator was set to 45 KV and 40 mA. 3.2.3.2 Rietveld refinement Rietveld refinement is a technique which utilizes mathematical methods to obtain crystal structural information by fitting neutron or X-ray diffraction patterns. 94 The least square approach was used to refine the com puted profile until t he best fit to the theoretical profile was achieved. Valuabl e crystallographic information was obtained from Rietveld refinement, such as latti ce parameter, atomic position, and thermal parameter. The goodness of the fit was evaluat ed by R values. The weighted profile R value ( R WP ) was used here and the formula is given as: 2 1 2 2 WP} ))(( )]()([ {R i ii i i iiobsyW calyobsyW (3-1) where y i (obs) is the observed intensity at step i y i (calc) is the calculated intensity at step i and W i is the weight. The best possible R WP quantity is called expected R ( R exp ), which can be expressed in the following equation: 2/1 2 exp])(/)[(N i iiobsyWpNR (3-2) Where N is number of observations, p is number of parameters. R exp represents the quality of data. 70

PAGE 71

The structure refinement was perform ed based on XRD patterns using software MAUD V2.058 (Material Analysis usi ng Diffraction) developed by Lutterotti 95 3.2.3.3 Particle size measurement The particle size and size distribution of ceramic powders was characterized using laser scattering (Beckman Coulter LS 13320). The samples collected for particle size measurement were suspended in deionized water and placed in an ultrasonic bath for 30 s before analysis. 3.2.3.4 Transmission elect ron microscopy (TEM) The TEM analysis was conduced on JOEL 200CX operated at 200 keV. Ceramic powder was dispersed in et hanol and the suspension was subsequently dropped onto a copper grid for TEM observation. 3.2.3.5 Scanning electr on microscopy (SEM) The microstructure of the sintered cerami c pellets was characterized using scanning electron microscopy (SEM, JEOL 6335F). The sintered ceramic samples were mechanically polished to a mirror finish usin g alumina dispersed polymer grinding discs of different grit sizes from 8 down to 1 m. After polishing, ceramic samples were cleaned in ethanol in an ultr asonic cleaning bath for 5 m. The cleaned samples were thermally etched at a temperature 50 C below the respective si ntering temperature for 0.5 h. The heating and c ooling ramp rates were 400 C/h. Some ceramic samples were not polished or thermally etched and were s ubject to a direct SEM examination. All SEM samples were sputtered with a carbon f ilm with thickness of ~20 nm prior to examination. The SEM was operated with an accelerating voltage of 15 kV, a probe current of 8 A, and a working distance of approximately 15 mm. 71

PAGE 72

3.3 Results and Discussion In this section, characterization of the MgO-Nd 2 Zr 2 O 7 composites and the spinel Mg 2 SnO 4 is discussed separately. The composit e characterization is discussed first, focusing on pyrochlore phase formation and mi crostructure of the composites. The spinel characterization focuses on structur al analysis and microstructure development. 3.3.1 MgO-Nd 2 Zr 2 O 7 Composites 3.3.1.1 Pyrochlore phase formation The XRD patterns for the sol-gel derived Nd 2 Zr 2 O 7 powder calcined at 1000 C are shown in Figure 3-6 Phase pure pyrochlore could not be formed until the calcination temperature reached 1200 C. At 1000 C, the formed phase shows cubic fluorite like structure (J CPDS PDF 49-1642) and the corresponding peaks are labeled as C in the figure. At 1150 C, formation of the Nd 2 O 3 phase was observed (JCPDS PDF 43-1023) and the corresponding peaks ar e labeled as N in the figure. The calcination temperature at wh ich pure pyrochlore phase can be achieved as indicated by the figure is 1200 C, which is ~150 C lower than the solid state reaction (pure pyrochlore phase was obtained by calc ination of the solid-state derived Nd 2 Zr 2 O 7 powder at 1350 C for 12 h and the XRD pattern is not shown here 96 ). The lower calcination temperature fo r the sol-gel derived Nd 2 Zr 2 O 7 powder can be attributed to the homogeneous mixing of the raw mate rials at a molecular level. Structure analysis for Nd 2 Zr 2 O 7 was not performed here bec ause the structure of Nd 2 Zr 2 O 7 has been well studied both by x-ray and neutron diffraction. 97,98 The lattice parameter ( a ) and oxygen parameter ( x ) determined for Nd 2 Zr 2 O 7 by neutron diffraction are 10.67587(3) and 0.33481(3) respectively. 98 72

PAGE 73

10203040506070 C C N N N N C C C C (711) (444) (531) (511)(111)(551)1200 oC 6 h(222) (400) (331) (440) (622)(311) Intensity (Arb. Units)2(Degrees)1000 C 6 h 1150 C 6 h C C C C Figure 3-6. X-ray Diffraction profiles of Nd 2 Zr 2 O 7 made from sol-gel processing after calcination at 1000 C, 1150 C and 1200 C for 6 hours. Letters indicate phases. C = cubic fluorite like phase, N = Nd 2 O 3 phase. Pyrochlore phase is indexed. The XRD profiles for the obtained MgO powder, sol-gel derived Nd 2 Zr 2 O 7 powder and a sintered MgO-Nd 2 Zr 2 O 7 composite are shown in Figure 3-7 All the peaks for the sintered composite can be identifi ed as the two phases MgO and Nd 2 Zr 2 O 7 and no other phase is detected. The peak positions for Nd 2 Zr 2 O 7 and MgO are also found to be identical to these for the pow ders, indicating that the tw o phases in the composite are stable and do not react with each other at sintering temperature. 73

PAGE 74

10203040506070 (551) (711)(111)(311) (222) (400) (511) (440) (531) (622) (444) (331)Calcined Nd2Zr2O7at 1200 oC for 6 h (111)Intensity (Arb. Units)2Degrees)(200) (222)Calcined MgO at 900oC for 2 h Pellet Sintered at 1550 oC for 4 h Figure 3-7. X-ray Diffraction profiles of calcined MgO, Nd 2 Zr 2 O 7 and the sintered composite pellet. 3.3.1.2 Powder characterization The particle size and size distribution of the Nd 2 Zr 2 O 7 powder synthesized through the solid state route and the sol-gel route are compared in Figure 3-8 It can be seen that the sol-gel processing can significantly reduce the mean particle size. This is mainly attributed to the molecular level mixing and the lower calcination temperature achieved in the sol-gel processing. Ne vertheless, particles aggregation is still a problem for the sol-gel derived powder, as indicated by the small peak between 1 and 10 m. Even though the calcination temperature required for the sol-gel derived powder is 150 C lower than the solid state derived pow der, it is still high enough to cause particle aggregation. 74

PAGE 75

0.1110 0 1 2 3 4 5 6 Solid state routeVolume %Particle Diameter (m) Sol-gel route Figure 3-8. Particle size and size distributi on of sol-gel process derived and solid state process derived Nd 2 Zr 2 O 7 powder. The sol-gel derived Nd 2 Zr 2 O 7 powder was further examined in TEM. Figure 3-9 shows a micrograph image and a selected area di ffraction pattern (SAED). The bright field (BF) image shows two large particle aggregates (~1-2 m) and several single crystalline particles (~100-200 nm). Submic ron size particles can be seen inside the aggregates, indicating that particles coalesc ed during calcination. The SAED was taken for a single particle, and the pattern can be indexed as Nd 2 Zr 2 O 7 along the zone axis [001]. 75

PAGE 76

XZA: [001] -2-20 -220 XZA: [001] -2-20 -220 A B Figure 3-9. Sol-gel derived Nd 2 Zr 2 O 7 powder. A) BF image, and B) SAED taken on zone [001]. 3.3.1.3 Microstructure analysis The microstructures of the composites m ade by different mixing methods are shown in Figure 3-10 All composites contain 50 vol% of MgO. The images were taken using the secondary electron detector, so the dark grains on the SEM ar e MgO and the light grains are Nd 2 Zr 2 O 7 because Nd 2 Zr 2 O 7 has a higher atomic mass than MgO. The microstructures of the composites are found highly dependent on processing methods. As shown in Figure 3-10 (A) and (B), the microstructu re produced by the mortar and pestle mixing and water magnetic bar st irring are not homogeneous. Both microstructures contain agglomerates of MgO and Nd 2 Zr 2 O 7 The Nd 2 Zr 2 O 7 in these two composites were made th rough the solid state processi ng. The heterogeneity of the microstructure obtained by the mortar and pestle mixing could be attributed to a poor mechanical mixing of the two ceramic powders. The water magnetic bar mixing didnt achieve a homogeneous mixt ure either. This could be attributed to a poor dispersion of the ceramic powders in water because no dispersant was used and the 76

PAGE 77

two powders have relatively large particle sizes (~1-10 m). Moreover, MgO dissolved in water and precipitated as Mg(OH) 2 which could cause the mixing of the two phases to be even less predictable. So it is not surprising that wa ter magnetic bar stirring could not produce a homogeneous microstructure. A relatively homogeneous microstructure was achieved by ball milling as shown in Figure 3-10 (C). The Nd 2 Zr 2 O 7 in the composites mixed by ball milling was synt hesized through the sol-gel processing. The grain sizes of both MgO and Nd 2 Zr 2 O 7 are ~1 m, which is smaller compared to the first two methods due to the continuous mechanica l grinding. A deta iled microstructure analysis and quantification on the ball milling produced microstructure will be discussed in Chapter 5. A B C Figure 3-10. Microstruc tures of the MgO-Nd 2 Zr 2 O 7 composites. A) Mortar and pestle mixing, B) water magnetic bar stirring, and C) ball milling. 77

PAGE 78

3.3.2 Single Phase Spinel Mg 2 SnO 4 3.3.2.1 Phase formation The XRD profiles for the starting powders and obtained Mg 2 SnO 4 after calcination are shown in Figure 3-11 The results indicate t hat a pure spinel phase of Mg 2 SnO 4 (JCPDS PDF 24-0723) can be obtained by calc ining the ball milled powder mixture at 1200 C for 12 h. 10203040506070 Mg2SnO4(220) (111) (311) (222) (400) (331) (422) (511) (440) (531) MgO Intensity (Arb. Units)2 (Degrees) SnO2 Figure 3-11. X-ray Diffraction pr ofiles of starting materials SnO 2 and MgO, and obtained Mg 2 SnO 4 by calcination at 1200 C for 12 h. 3.3.2.2 Structure analysis It is known that Mg 2 SnO 4 is an inverse type spinel ( i = 1) with the same space group as MgAl 2 O 4 (No. 227, m3Fd ). The lattice parameter ( a ) reported in liter ature is 8.60 99 Other than that, ther e is limited information on the crystal st ructure of Mg 2 SnO 4 In 78

PAGE 79

order to better understand the structure of Mg 2 SnO 4 Rietveld refinement was performed based on the obtained X-ray diffract ion pattern for the synthesized Mg 2 SnO 4 to obtain structural information, espec ially the lattice parameter ( a ) and oxygen dilation parameter ( u ). 203040506070 Observed Calculated 2(Degrees)Intensity (Arb. Unit)Difference (220) (311) (400) (222) (331) (422) (511) (440) (531) Figure 3-12. The calculated observed X-ray Diffraction profiles and the difference for Mg 2 SO 4 Lattice parameter (a) is 8.6065(4) R w (%) = 6.74% and R exp (%) = 4.32%. The XRD profile and the fitting by t he Rietveld analysis are shown in Figure 3-12 The powder X-ray diffraction profile for Mg 2 SnO 4 was obtained at 25 C. The calculation and observed profiles are shown as the red solid line and soli d blue dots, respectively. The difference between the calc ulated and observed intensitie s is plotted on the bottom of the figure. The R val ues for the refinement are R wp (%) = 6.74% and R exp (%) = 4.32%. Table 3 shows the calculated peak position and d-spacing for Mg 2 SnO 4 The refined structural param eters are listed in Table 3 The refined lattice parameter (a) 79

PAGE 80

for Mg 2 SnO 4 is 8.6065(4) and the ox ygen dilation parameter ( u ) is 0.3834(4). The obtained lattice parameter is consistent with the va lue reported in literature. 99 Since Xrays are not sensitive for the measurement of oxygen in th e presence of heavy ions, the thermal parameter for oxygen was not refined. 98 Table 3. X-ray Diffraction (XRD) reflections of spinel Mg 2 SnO 4 Data were obtained from Maud V2.058. h K l 2 cal () d cal () Relative Intensity 1 1 1 17.836 4.96897 79.76 2 2 0 29.328 3.04286 6.71 3 1 1 34.536 2.59496 100 2 2 2 36.124 2.48448 20.11 4 0 0 41.956 2.15162 65.61 3 3 1 45.925 1.97447 12.67 4 2 2 52.012 1.75679 2.08 5 1 1 55.429 1.65632 31.77 3 3 3 55.429 1.65632 6.06 4 4 0 60.835 1.52143 57.55 5 3 1 63.943 1.45476 14.82 4 4 2 64.961 1.43442 0.01 6 2 0 68.952 1.36081 0.65 Based on electrostatic potential consider ations, the inverse 4-2 type spinels are more stable compared to the normal 4-2 spinels when u > 0.381. 21,100 The obtained u value for Mg 2 SnO 4 agrees with the fact that Mg 2 SnO 4 is an inverse spinel. The calculated electrostatic contribution to the lattice energy for inverse spinels indicates that cations at B-sites tends to be orderly distributed when u < 0.385 and disordered when u > 0.385. 100 Therefore, the obtained u value suggests that Mg 2 SnO 4 may have an ordered distribution of Mg 2+ and Sn 4+ at B-sites. However, this may not be the case; it has been suggested that Mg 2 SnO 4 is entropy stabilized and a major contribution to its 80

PAGE 81

entropy comes from the cation disordering at B-site. 101 An ordered arrangement at Bsite may lead to a less stable structure. It is worth noting that the boundaries defined based on the electrostatic potential calculati ons may not be exact, mainly because the parameter values for the 4-2 type spinels are not very accurate. 100 Therefore, the cation ordering at B-site in the structure of Mg 2 SnO 4 requires further investigation and clarification, i.e. doing synchrotron X-rays or neutron diffraction studies. Table 3. Refined structural parameters from x-ray diffr action data for the spinel Mg 2 SnO 4 phase. The atomic positions x, y, z are presented. Lattice parameter (a) is 8.6065(4) R w (%) = 6.74% and R exp (%) = 4.32%. Atom Position Occupancy x y z U 100() Mg 8a 1 0 0 0 1.5(2) Mg 16d 0.5 0.625 0.625 0.625 0.3(19) Sn 16d 0.5 0.625 0.625 0.625 0.329(5) O 32e 1 0.3834(4) 0. 3834(4) 0.3843(4) N/A The selected bond lengths and bond angles for Mg 2 SnO 4 were obtained from Crystal Maker with the refined structure para meters, and the results are listed in Table 3 and also shown in Figure 3-13 1.9886 109.471o 1.9886 109.471o 2.0818 94.047 85.953 2.0818 94.047 85.953 A B Figure 3-13. Bond lengt hs and bond angles for Mg 2 SnO 4 are shown in A) tetrahedra site, and B) octahedra site. 81

PAGE 82

Table 3. Bond length and bond angle obtained from refined Mg 2 SnO 4 structure. A site Bond length () B site Bond length () Mg-O () 1.9886 Mg-O () 2.0818 Sn-O () 2.0818 Bond angle ( ) Bond angle ( ) O-Mg-O 109.471 O-Mg(Sn)-O 94.047 O-Mg(Sn)-O 85.953 GII was calculated to evaluate the structure stability of Mg 2 SnO 4 The R values for Mg II -O and Sn IV -O were taken as 1.693 a nd 1.905 from Breses paper. 26 The bond length of A-X and B-X in Mg 2 SnO 4 were obtained from Table 3 The calculation procedure was described pr eviously in section 2.2.3.4 The calculated GII for Mg 2 SnO 4 was 0.153. As discussed previously in Chapte r 2, the GII values gr eater than ~0.05 are indicative of stress, which produces intrinsic st rain in the structure. Structures with GII >> 0.2 are generally unstable. The Mg 2+ ions at A-site and the Sn 4+ ions at B-site are both underbonded as indicated in Table 3 and the GII value for Mg 2 SnO 4 is greater than 0.05. These results indicate that the structure of Mg 2 SnO 4 is stable but subject to some lattice strain. Table 3. Bond valence calculations for Mg 2 SnO 4 Site Bond valence Bond valence sum Oxidation state Mg(IV) 0.450 1.799 2 Mg(VI) 0.350 2.098 2 Sn(VI) 0.620 3.721 4 3.3.2.3 Microstructure analysis The microstructures of the Mg 2 SnO 4 pellets sintered at different temperatures and soak time are shown in Figure 3-14 Images (A) and (B) were taken for one sample sintered at 1400 C for 6 h at two different magnifica tions, image (C) and (D) were taken 82

PAGE 83

for one sample sintered at 1500 C for 6h, and image (E) and (F) were taken for one sample sintered at 1500 C for 24 h. The Mg 2 SnO 4 pellets sintered at 1400 C for 6 h was highly porous and the grain size ranged from submicron to ~4 m. The average density of the sample was ~82%, which is lower than the value r eported at the same sintering condition in literature (~89%). 86,88 However, by increasing the sintering temperature to 1500 C and extending the sintering time to 6 h, the porosity of the Mg 2 SnO 4 pellet was significantly reduced. Pellets with density above 90% were fabricated. A comparison between figure (B ) and (D) indicates increasing sintering temperature by 100 C resulted in not only densification but also grain growth. In order to examine the surface morphol ogy, the surfaces of these pe llets were not polished or thermally etched. Some of the grains showed growth terraces on the grain surface, which is probably an indication of enhanced surface diffusion or an early stage of sublimation. 102 By extending the soak time from 6 h to 24 h, the dens ification of the samples was further improved along wi th a significant grain growth. A B Figure 3-14. Microstructure Mg 2 SnO 4 sintered at A) and B) 1400 C for 6 h ( = ~82%), C) and D) 1500 C for 6 h ( = ~92%), E) and F) 1500 C for 24 h ( = ~96%). 83

PAGE 84

C D E F Figure 3-14. Continued. 3.4 Summary and Conclusions Based on literature su rvey and engineering par ameters, the MgO-Nd 2 Zr 2 O 7 composites and single phase spinel compound Mg 2 SnO 4 were selected as the potential IM materials for further evaluation. Pyrochlore Nd 2 Zr 2 O 7 was synthesized through the conventional solid state processing and the sol-gel processing. As the calcination temper ature increased, the sol-gel derived Nd 2 Zr 2 O 7 powder first formed fluorit e like structure at 1000 C, and Nd 2 O 3 subsequently precipitated out at 1150 C. A calcination temperature of 1200 C was 84

PAGE 85

required to achieve pure pyrochlore phase. The sol-gel processing can significantly reduce the main particle size from a fe w microns down to a few hundred nanometers, however, some particles aggregated and grew as large as microns. Distinct microstructures of the composites were obtained from three processing methods, including mortar and pestle mixing, water m agnetic bar stirring, and ball milling. The ball milling produced a relatively homogeneous microstructure compared with the other two processing methods. The spinel phase Mg 2 SnO 4 was achieved by calcination of ball milled MgO and SnO 2 powder mixture in the stoichiometric ratio at 1200 C for 12 h. Rietveld refinement was performed for spinel Mg 2 SnO 4 and the obtained lattice parameter ( a ) and oxygen dilation parameter ( u ) were 8.6065(4) and 0.3834(4) The electrostatic potential calculation suggests that Mg 2 SnO 4 has an ordered inverse spin el structure; however, the thermodynamic calculations suggest that the B type cati ons are mixed and disordered. The calculated GII for Mg 2 SnO 4 was 0.15, indicating t hat the structure is stable, even though combined with some lattice strain. Dense Mg 2 SnO 4 pellets were fabricated by sintering at 1500 C for 24 h. 85

PAGE 86

CHAPTER 4 EVALUATION OF RADIATION TOLERANCE 4.1 Introduction As discussed in Chapter 2, nuclear fuels and IM materials are subject to various types of radiation damage. Th e objective of this chapter is to investigate and assess the radiation tolerance of the potential IM materials. T he radiation tolerance of rocksalt MgO, pyrochlore Nd 2 Zr 2 O 7 and spinel MgAl 2 O 4 are briefly summarized based on literature review. Since the radiation behavior of Mg 2 SnO 4 has not been studied, an insitu irradiation test was performed for Mg 2 SnO 4 and the resistance to irradiation induced amorphization was evaluated. 4.2 Literature Review In this section, the irradi ation tolerance of MgO, Nd 2 Zr 2 O 7 and MgAl 2 O 4 is briefly reviewed. The radiation tolerance is mainly characterized by resistance to radiation induced amorphization, microstructure and stru ctural evolution. Neutron irradiation, heavy ion irradiation and fissi on fragments damage are discussed. 4.2.1 Irradiation Stability of MgO It has been shown that MgO has excellent irradiation stability against radiation induced amorphization. Similar to most fluorite oxides such as ZrO 2 and UO 2 MgO cannot be amorphized even at 20 K. 39 The behavior of MgO under neutron irradi ation has been extensively studied. 103-107 Even though MgO cannot be amor phized, low temperature neutron irradiation may result in significant swelling. It has been reported that the swelling is ~3.0 vol% for polycrystalline MgO irradiated by fast neutron to an estimated 30 dpa at 430 K. 103 The 86

PAGE 87

swelling can be attributed to defec t formation, more specifically the elongated interstitial loops and dislocation clusters. 103 The thermophysical pr operties and mechanical properties have been found degrading as a consequence of neutron radiation. 108 In addition to the displacement damage descr ibed above, the nuclei of Mg and O are subjects to transmutation by fast neutron fluence, which produces He and Ne in MgO and could result in gas bubble formation upon thermal annealing. 109,110 It has been found that MgO is resistant to fission fragments induced damage. Minimum swelling was observed for MgO com pared with several other compounds such as ZrO 2 and SiC after being irradi ated to an ion dose of 5 10 18 fission fragments/m 2 111 Swift heavy ion irradiation tests conducted by Beauvy 112 indicate that MgO exhibits best radiation toleranc e compared to MgAl 2 O 4 and Al 2 O 3 4.2.2 Irradiation Stability of Nd 2 Zr 2 O 7 As discussed earlier in Chapter 3, Nd 2 Zr 2 O 7 shows excellent irradiation stability against ballistic displacement damage and complete amorphization could not be achieved. 63,65 Lutique 36 uses swift heavy ion (120 Me V Iodine ion) irradiating Nd 2 Zr 2 O 7 to study the fission fragment damage. TEM observation s hows formation of ion tracks with a diameter of ~ 10 nm. These ions tracks do not become amorphize, but transform into an unknown structure. Nevertheless, amorphization is achiev ed by overlapping of these damage tracks. They compare th e results with Wangs studies 113 on ZrO 2 and conclude that the radiation resistance of Nd 2 Zr 2 O 7 against swift heavy ions is not as good as ZrO 2 The radiation behavior of Nd 2 Zr 2 O 7 or other pyrochlore compounds under neutron irradiation has not been studied, yet. Ba sed on the studies on ion irradiation of 87

PAGE 88

pyrochlores, it is expected that Nd 2 Zr 2 O 7 should possess excellent resistance against neutron induced displacem ent damage. The transmutation damage and associated chemical effects are hardl y known and requires further studies. The influence of irradiation damage on thermophysica l and mechanical properties of Nd 2 Zr 2 O 7 has not been investigated yet. 4.2.3 Irradiation Stability of MgAl 2 O 4 As mentioned earlier in Chapter 3, the irradiation behavior of MgAl 2 O 4 has been extensively studied. MgAl 2 O 4 exhibits excellent resi stance to radiation induced amorphization caused by neutron or low energy i on irradiation. 103,114-117 It has been reported that the amor phization dose of MgAl 2 O 4 under irradiation of 1.5 MeV Xe + ions at 30 K is equivalent to 35-40 dpa. 118 Single crystal MgAl 2 O 4 exhibits zero swelling after fast neutron irradiatio n to ~23 dpa at 925 K. 103 The mechanisms underlying the radiation resistance of the spinel stru cture have been determined as cation disorder (anti-site defects 65 ) and interstitial-vacancy (i-v) recombination. 119 Moreover, MgAl 2 O 4 is a complex oxide compound with two types of cations in the structure. It has been suggested that this multi-component chemistry suppresses the nucleation and growth of interstitial dislocation loops during irradiation. 103 Nevertheless, polycrystalline MgAl 2 O 4 does swell due to the presence of grain boundar ies. It has been observed that voids can be formed near grain boundar ies, indicating that these boundaries are effective sinks for interstitials. 103 The swelling is determined to be ~ 1.6 vol% for polycrystalline MgAl 2 O 4 irradiated by neutr on to ~23 dpa at 1100 K. 103 Regardless of swelling of polycrystalline MgAl 2 O 4 under neutron irradiation, the key problem of MgAl 2 O 4 is that it swells significantly dur ing in-pile irradiation tests, which 88

PAGE 89

excludes it from utilizat ion as an IM material. Recent studies suggest that the swelling may be attributed to fission fragment induced damage and amorphization, but the mechanisms are not well understood. 5 Very limited research work has been con ducted on investigat ing the irradiation behavior of the 4-2 type spinel compounds. To further explore spinel compounds as potential IM materials, an inverse 4-2 type spinel Mg 2 SnO 4 was selected for studying. The radiation stability of Mg 2 SnO 4 was first evaluated by the resistance to irradiation induced amorphization. The following part of this chapter pr esents the performed irradiation experiment, re sults, and discussion for Mg 2 SnO 4 4.3 Experimental Procedure 4.3.1 Specimen Preparation The TEM specimens for in situ ion irradiation were pre pared by crushing the ceramic samples or using focused ion beam (FIB). Using the former method, the calcined Mg 2 SnO 4 powder was crushed in ethanol using a small agate mortar and pestle, and the suspension was subsequent ly dropped onto a carbon coated copper grid. Using the latter method, an electron transparent lamella was prepared with an FEI Strata DB 235 FIB/SEM dual-beam system. An auto-FIB script for TEM sa mple preparation was used and the sample was ion-milled until the desir ed thickness (~200 nm) was achieved. The specimen was then cut free and lifted out ex situ using an optical microscope and micromanipulators. 4.3.2 In situ Ion Irradiation and Characterization The irradiation tests and characterizati on were carried out at the IVEM-Tandem facility in the Electron Microscopy Center (EMC) at Argonne National Laboratory. Kr 89

PAGE 90

ions were used as the irradiation source, and were double-charged and accelerated to 1 MeV using a 650 kV NEC Ion Implanter. The irradiati on tests were performed at cryogenic temperatures (50K and 150 K), using a liquid-heliu m-cooled cold stage. The temperature was monitored using a thermo couple attached to the sample holder near the sample position. In situ characterization was per formed on a HITACHI H-9000NAR electron microscope operated at 300 kV. The electron beam traveled vertically down through the specimen and the irradiating ion beam was incident on the specimen at 30 to the vertical. TEM samples were irradiated at a dose rate between 10 14 and 10 16 Kr 2+ ions/m 2 /s. The maximum ion fluence obtained in this work was 10 20 Kr 2+ ions/m 2 Periodically during the irradi ation, the ion beam was shut off and the specimens were inspected in situ so that bright field (BF) images and selected-area electron diffraction (SAED) patterns could be recorded. The c hemical compositions of TEM specimens were determined ex situ at room temperatur e using energy dispersive spectroscopy (EDS) on JOEL 2010F TEM in the Major Analyt ical Instrumentation Center (MAIC) at the University of Florida. 4.4 Results 4.4.1 Transport of Ions in Ma tter (TRIM) Based Calculation The average penetrati on depth of 1 MeV Kr 2+ in Mg 2 SnO 4 was calculated using the program TRIM-2008. 37 The projected range of 1 MeV Kr 2+ ions in Mg 2 SnO 4 was about 370 nm, with a longitudinal str aggling (defined as the square root of the variance, which is an average of the square of the deviations of the ion ranges from the mean projected range) of about 103 nm. Only the grains with grain size at or less than 200 nm were chosen to be characterized for irradiation st ability. The irradiat ion damage profile and 90

PAGE 91

ion implantation concentration as a func tion of target depth are plotted in Figure 4-1 The TRIM calculation results indicate t hat at sample depths between 100 and 300 nm, the atomic displacement dam age ranges from about 1.0 displa cement per atom (dpa) to 1.2 dpa per 10 19 Kr 2+ ions/m 2 averaged over the sublattices of spinel. A threshold energy for displacement was assumed to be 40 eV for Mg, Sn, and O based on values published in literature. 120 As indicated in Figure 4-1 the concentration of implanted Kr ions at depth of 200 nm was about 0.0135 at.% at ion fluence of 10 19 Kr 2+ ions/m 2 For the maximum fluence used in this experi ment, the implanted Kr concentration was about 0.135 at.% at depth of 200 nm. 0.00 0.01 0.02 0.03 0.04 0.05 02004006008001000 at.% Kr per 1019 Kr2+ ions/m2 Target Depth (nm) Total dpa per 1019 Kr2+ ions/m2 0.0 0.3 0.6 0.9 1.2 Figure 4-1. Total target atom displaceme nts (averaged over all sublattices of the material) in dpa, and the concentra tion of implanted Kr ions, in Mg 2 SnO 4 spinel as a function of target dept h, for irradiations using 1 MeV Kr 2+ ions. Values on the ordinates are normalized to a fluence of 1 10 19 ions/m 2 TRIM simulations based on 10000 ions. 4.4.2 Irradiation of Mg 2 SnO 4 at 50 K The FIB prepared Mg 2 SnO 4 specimen was irradiated by 1 MeV Kr 2+ ions to a maximum fluence of 5 19 ions/m 2 at 50 K. A BF image of the entire specimen before 91

PAGE 92

irradiation is shown in Figure 4-2 (A). The specimen was a 15 m 6 m electron transparent lamella with a thickness of ~ 200 nm. The dark strip on one side of this lamella was Pt, a foreign materi al that was coated to protec t the surface of the sample from ion damage. The hole in the lamella originated from a pore inside the sintered Mg 2 SnO 4 pellet. Two SAED patterns were recorded for areas b (0.1 dpa) and c (0 dpa) in the BF image and are shown in Figure 4-2 (B) and (C). The specimen was slightly tilted so that the zone axis was [114] for area b and [125] for area c. Figure 4-2 (D) shows a BF image of the specimen after irradiation by 1 MeV Kr 2+ ions at an ion fluence of 5 19 ions/m 2 It can be seen that the amor phous carbon film was extensively damaged by ion irradiation. Figure 4-2 (E) and (F) show the SAED patterns for areas e and f in the BF image after irradiation, wh ich are the same locations as areas b and c in Figure 4-2 (A). As shown in Figure 4-2 (E) and (F), no diffraction spots can be seen, which indicates that the specimen was comp letely amorphisized. The BF image shows less contrast than the unirradiated samp le, with the grain boundaries fading or disappearing. While the SAED pattern s recorded at lower ion doses (10 18 ions/m 2 2 18 ions/m 2 5 18 ions/m 2 10 19 ions/m 2 and 2 19 ions/m 2 ) are not shown here, all of them show diffracted spots or diffused spots, suggesting t hat the specimen was not completely amorphisized at those ion fluences. Besides the two monitored areas, several grains of the specimen were examined at the ion influence of 5 19 ions/m 2 and none of them showed diffraction spots. Therefore, the critical amorphization dose for Mg 2 SnO 4 at the cryogenic temperature 50 K can be concluded as 5 19 ions/m 2 which corresponds to an atomic displaceme nt dose of approximat ely 5.5 dpa at a depth of ~200 nm. 92

PAGE 93

A B C D E F Figure 4-2. Focused Ion Beam (FIB) prepared Mg 2 SnO 4 specimen: A) BF image of the specimen before irradiation, B) SAED pattern for area b C) SAED pattern for area c, D) BF image of the specimen irradiated by 1 MeV Kr 2+ to fluence of 5 10 19 Kr 2+ ions/m 2 (5.5 dpa), E) SAED pattern for area e and F) SAED pattern for area f 93

PAGE 94

The liquid helium was released after irr adiation and the specimen was warmed up to room temperature. The specimen was examined agai n using TEM on the next day in order to study the ther mal annealing effects. Figure 4-3 (A) shows a BF image of the irradiated specimen at room te mperature. The specimen was tilted in order to align the electron beam near zone axis for areas b and c in the BF image, which matched the same locations as shown in Figure 4-3 The SAED patterns were recorded and are shown in Figure 4-3 (B) and (C), which can be indexed as spinel Mg 2 SnO 4 along [001] zone axis. The results indicate that the spinel crystal structur e was formed at these regions, suggesting that annealin g at room temperature is effective in thermally recovering the spinel crystalline phase from the fully amorphous Mg 2 SnO 4 The emergence of grain boundaries in the BF image confirms that the annealed specimen is a polycrystalline material. Nevertheless, a thin layer of ~600 nm in width was observed on the side of the sample (shown in Figure 4-3 (A)) after the irradiati on and annealing. A BF image recorded for the area d (see Figure 4-3 (D)) shows a heterogeneous microstructure. Chemical analysis was performed using EDS and the collected spectra for areas e and f are shown in Figure 4-3 (E) and (F). The ion source of the FIB, Ga was detected throughout the sample. Ga ion implantati on and structural dam age caused by FIB in the TEM sample preparation is a typically problem even though Ga ions travel at glancing angles of < 1 on sample surface. 121 Semi-quantitative EDS analysis reveals that area e was a Sn depleted region with an atomic ratio of Mg to Sn about 20, and the dark precipitate at area f was a Sn rich region, with an atomic ratio of Mg to Sn about 1.7 (slightly lower than the expected 2:1 stoichiometry). 94

PAGE 95

b c d 5.5 dpa annealed Thin layer ~600 nm A B 45 nm f e 45 nm f e segregation layer C D 0510 SnLGaKGaKCuKSnLMgKOKCK Intensity (Arb.)keV e. EDS of point b EDS of area e 0510 GaKGaKSnLCuKCuKSnLMgKOK Intensity (Arb.)keV f. EDS of point c EDS of area f E F Figure 4-3. Mg 2 SnO 4 sample irradiated by 1 MeV Kr 2+ at a fluence of 5 10 19 Kr 2+ ions/m 2 at 50 K and structure was thermally recovered at room temperature: A) BF image, B) SAED pattern for area b C) SAED pattern for area c, D) BF image of area d, E) EDS spectra of area e and F) EDS spectra for area f 95

PAGE 96

A SAED pattern recorded at the area d is shown in Figure 4-4 where a polycrystalline ring pattern was obtained and c an be indexed as MgO. Some diffraction spots cannot be identified as MgO, indicati ng that other crystalline phases were also present. The specimen was also annealed at 573 K for 0.5 h in a conventional oven but no significant change was observed. The formati on of this segregation layer is not clear at this moment. Wang and his co-workers 13 have found spinel -SiFe 2 O 4 first becomes amorphous and then crystallizes into Fe 3 O 4 and SiO 2 under continued Kr + irradiation at 873 K, and they attributed it to t he high pressure metastable phase of -SiFe 2 O 4 However, this is unlikely the case for Mg 2 SnO 4 because the thermal recovery of spinel structure was achieved elsewhere and c onfirmed by SAED patterns as shown in Figure 4-3 This segregation layer could be the result of surface contamination by implanted Ga + ions during FIB preparation. It may also be formed during irradiat ion as a result of differential Mg and Sn sputtering followed by non-stoichiometric cr ystallization at room temperature. Further inve stigation is needed to eluci date the formation of this segregation layer. Figure 4-4. Selected area electron diff raction (SAED) pattern recorded on the segregation layer. 96

PAGE 97

4.4.3 Irradiation of Mg 2 SnO 4 at 150 K Crushed Mg 2 SnO 4 grains were irradiated at 150 K and six randomly orientated grains or grain clusters were chosen and m onitored. The random orientations were chosen to avoid the potential effects of ion channeling. The SAED patterns were recorded at ion fluences of 0, 10 19 5 10 19 and 10 20 Kr ions/m 2 and a series of patterns for a grain cluster are shown in Figure 4-5 The orientation of grains usually changed after ion irradiation, which was probably due to buckling of the carbon film as a result of ion-induced breakdown of formvar whic h came from the film fabrication. 122 The presence of diffracted spots in the SAED patterns in Figure 4-5 (B)-(C) indica tes that the specimen was not completely amorphized at those ion fluences. The diffraction spots vanished finally at an ion fluence of 10 20 ions/m 2 which was equivalent to an atomic displacement damage of about 11.0 dpa for grai ns with a diameter of ~200 nm. The specimen was translated and seve ral other grains were exam ined using SAED after the final ion dose. It was observed that appr oximately one out of five grains was not amorphous and still retained crysta llinity. A similar case wa s also reported by Smith and his co-workers 122 ; and the grains that cannot be amorphisized were referred to as outliers. Several possible explanat ions were given in Smiths paper 122 for the existence of outliers, but the most likely scenario in this case was that the elevated temperatures due to poor thermal connection between the outliers and the carbon film caused annealing of irradiation damage. 80 Therefore, the outliers were not considered in determining the amorphizati on dose, which was then concluded to be 11.0 dpa for Mg 2 SnO 4 at 150 K. 97

PAGE 98

A B C D Figure 4-5. Selected area el ectron diffraction (SAED) patterns of the crushed Mg 2 SnO 4 grains irradiated at 150 K: (A) prior to ion irr adiation and (B)-(D) following exposure to 1.0 MeV Kr 2+ ions to fluences of 10 19 (1.1 dpa), 5 10 19 (5.5 dpa), and 10 20 (11.0 dpa) Kr 2+ ions/m 2 respectively. Since defects created by ion irradiation can be annihilated by the thermal annealing process, the irradiation stab ility of a material is expect ed to be improved at elevated temperatures. 123 As a result, the amorphization dos e should increase with irradiation temperatures because the t hermal annealing becomes more efficient in restoring crystallinity. The temper ature dependence of amorphizati on is a result of the competition between amorphization and recovery pr ocesses. It is worth noting that the geometries of the samples i rradiated at 50 K and 150 K are different: the one irradiated at 50 K is a thin slab cut out from a sintered polycrystalline Mg 2 SnO 4 using FIB and the 98

PAGE 99

one irradiated at 150 K are a bundle of cr ushed particles. Nevertheless, the amorphization dose for Mg 2 SnO 4 increases from 5.5 dpa to 11.0 dpa as the irradiation temperature increases from 50 K to 150 K, regardless the geometry of the specimens. 4.4.4 Irradiation Dama ge Mechanism of Mg 2 SnO 4 by 1 MeV Kr 2+ As discussed previously in Chapter 2, the energy loss of energetic ions in target atoms is mainly due to electroni c scattering and nuclear scattering. 28 The partitioning of energy transferred into electronic scattering and nuclear scattering is an important process controlling the effect of radiation. 2 In order to study the irradiation damage mechanisms of 1 MeV Kr 2+ ions in the inverse spinel Mg 2 SnO 4 the electronic and nuclear stopping powers were calculated us ing TRIM simulations and the results are presented in Figure 4-6 It is shown that t he electron stopping power, (dE/dX) e (the peak value ~1.5 keV/nm/ion at the samp le depth of 200 nm), exceeded the nuclear stopping, (dE/dX) n (the peak value ~1.0 keV/nm/ion at the sample depth of 300 nm), except at the end of range of ions. The resu lts indicate that both nuclear (ballistic) scattering events and electronic (ionization) scattering events played important roles on the ion-solid interaction. Since fission tr acks were not observed and the deposition rate for the electronic scattering was only above 1 keV/nm, the Coulomb explosion and thermal spike were not likely to be t he dominant damage mechanisms even though the electronic stopping power exceeded the nuclear stopping power. 29,31 As a result, the major damage mechanism was the atomic displacement induced defect accumulation. 99

PAGE 100

0200400600800 0.0 0.4 0.8 1.2 1.6 0.0 0.4 0.8 1.2 1.6 dE/dx (keV/atom) ( for fluence =1019 Kr2+ ions/m2 ) (dE/dX)ndE/dx (keV/nm/ion)Penetration Depth in Target (nm) NMg2SnO4T = 8.8 x 1028 atoms/m3 (dE/dX)e Figure 4-6. Electronic ((dE/dx) e ) and nuclear ((dE/dx) n ) stopping powers for 1 MeV Kr 2+ ions in the Mg 2 SnO 4 target. The figure is plotted in two types of units: [keV/nm/ion] and [eV/atom], the la tter for an arbitrary fluence of = 10 19 Kr 2+ ions/m 2 4.4.5 Irradiatio n Tolerance of Mg 2 SnO 4 Since most studies have focused on the irradiation behavior of MgAl 2 O 4 the critical amorphization doses for other spinel co mpounds are not well known. It has been reported that the amorphiza tion dose is ~4 dpa for FeCr 2 O 4 at 20 K and 0.2 dpa for SiFe 2 O 4 below 723 K. 13 Mg 2 SnO 4 exhibits a moderate irradi ation resistance against ion damage at intermediate energies (typically a few MeV), but the ability to annihilate atomic displacement induced defects is not as good as the normal spinel MgAl 2 O 4 13,118,124 100

PAGE 101

Some considerations are given here to explain the irradiation tolerance of Mg 2 SnO 4 Mg 2 SnO 4 is an inverse 4-2 type spinel. The obtai ned experimental results indicate that inverse spinels may be less irradiation tolerant compared to norma l ones in the nuclear stopping regime. Recently, the effect of ca tion inversion on irradiation tolerance of spinel compounds has been investigated using at omic simulations. More specifically, MgAl 2 O 4 MgGa 2 O 4 and MgIn 2 O 4 have been studies. Bacorisen and co-workers 73 have used molecular dynamics (MD) to simu late collision casca des and found that the irradiation induced damage to the structure is more ext ensive for the fully inverse MgIn 2 O 4 compared with the half-inverse MgGa 2 O 4 and normal MgAl 2 O 4 Uberuaga and co-workers 74 have used temperature accelerated dynamics (TAD) to simulate and characterize the kinetics of defects for the sa me three spinel oxides in order to study the cation ordering effects. It was concluded that the cation disorder greatly complicated and inhibited the motion of point defects through the spinel structure, which leads to defect accumulation and less i rradiation tolerance. Nevertheless, it might be misleading to corre late the irradiation behaviors purely with inversion because many other factors (such as ionic radius and bond type) may also have large effects on irradiation tolerance. In addition, the inverse spinel compound in t he atomic simulation studies is the 2-3 type MgIn 2 O 4 which differs from the 4-2 type Mg 2 SnO 4 in chemistry. Therefore, an atomic simulation on Mg 2 SnO 4 would still be necessary to understand the effect of cati on disordering on radiati on damage in this material. The irradiation tolerance of oxide com pounds may be correlated to the chemical bond. 125-127 It has been found that the irradiation stability is governed by the ionicity of bonding in general, with higher ionicity leading to more stable materials under irradiation. 101

PAGE 102

The theory is in good agreement with experim ental results for simple oxide compounds. For instance, the radiat ion stability of SnO 2 TiO 2 and ZrO 2 increases because of greater ionicity. 125 The ionicity of a chemical bond can be estimated using Paulings electronegativity and indicated by the difference of elec tronegativity between a cation and an anion. 128 The relative electronegativities det ermined by Pauling for Mg, Sn (IV), Al and O were 1.31, 1.96, 1.61 and 3.44, respectively. The electronegativities were further modified by Batanov 129 for crystalline compounds with a consideration of the valence state and the coordinatio n of atoms, and the obtained va lues for Mg, Sn (IV), Al and O were 0.8, 2.0, 1.4 and 3.2, respectively. The di fference of electronegativity between Sn and O is smaller than Al and O and also smaller than Mg and O, suggesting that the covalen cy of bond is higher than and , and Mg 2 SnO 4 is more covalently bonded than MgAl 2 O 4 Similarly in pyrochlore stannate compounds, high covalency of bond has also been found bot h experimentally using neutron and x-ray diffraction and theoretic ally using density functional theory (DFT) computer simulation. 64,130-132 Therefore, the potentially high covalency of bond may lead Mg 2 SnO 4 more susceptible to radi ation damage compared to MgAl 2 O 4 It is worth noting that it is relative ly difficult to predict the i rradiation tolerance of complex oxides purely based on ionicity, because subtle structural changes such as bond length may change the bond type of ca tions in the structure. 64 It has been mentioned earlier in this chapter that the stannate pyrochlores show greater variation in irradiation stability and not all stannate pyrochlores are susceptib le to radiation damage. Actually, Y 2 Sn 2 O 7 and Er 2 Sn 2 O 7 have excellent irradiation stabilit y and cannot be amorphized even at 25 K. Therefore, It is useful to further study the irradiat ion tolerance of other stannate 102

PAGE 103

spinels, such as Zn 2 SnO 4 Mn 2 SnO 4 and Co 2 SnO 4 and to investigate if there are subtle changes in bond type as found in stannate pyrochlores and how they are correlated to irradiation stability. 64 As mentioned previously, the irradiat ion tolerance of pyrochlores has been intensively studied both experimen tally and theoretically. The contour plot of anti-site formation energy in pyrochlores generated by atomic simulation is in agreement with the experimental results in general. Similarly, ant i-site formation is al so an important factor that determines irradiation tolerance of spinels. Sickafus and co-workers 133 have observed a significant amount of cation disorder in the normal spinel MgAl 2 O 4 upon neutron exposure in excess of 50 dpa at 670 K. The obtained experimental results in the present study suggest that the barri er for the anti-site formation in Mg 2 SnO 4 may be higher than MgAl 2 O 4 and result in less irradiation tolerance. Even though the anti-site formation energy for spinels has not been calc ulated by atomic simulations, one may link the anti-site formation energy to cati on radii difference and analyze the anti site formation barrier qualitativel y. In normal spinel MgAl 2 O 4 the ionic radii of Mg 2+ at tetrahedral sites and Al 3+ at octahedral sites are 0.57 and 0.535 respectively. In the inverse spinel Mg 2 SnO 4 the ionic radii of Mg 2+ at tetrahedral sites and Sn 4+ at octahedral sites are 0.57 and 0.69 respectively. All ionic radii are obtained from Shannon. 91 The ionic radius ratio of r B /r A for MgAl 2 O 4 and Mg 2 SnO 4 are calculated to be 0.94 and 1.21, which indicates that the cation radii difference in Mg 2 SnO 4 is larger than in MgAl 2 O 4 The larger cation radii difference in Mg 2 SnO 4 may cause higher anti site formation barrier. Moreover, Sn 4+ is a tetravalent ion and Al 3+ is a trivalent ion. 103

PAGE 104

Swapping between Sn 4+ and Mg 2+ will most likely cause more lattice instability or distortion due to the greater charge difference. The spinel structure is also well known for its abundant structur al vacant sites on both tetrahedral positions ( 87.5 % are vacant) and octahedr al positions (50 % are vacant). The large fraction of vacancy sites in the structure is re sponsible for effective interstitial-vacancy (i-v) recombination, which is also an important mechanism for annihilation of point defec ts created by irradiatio n. Nevertheless, MgAl 2 O 4 and Mg 2 SnO 4 both have equal numbers of vacancy sites in their structures, and it is not likely to become an important factor that results in different irradiation tolerances. Overall, Mg 2 SnO 4 is not as resistant as MgAl 2 O 4 against ballistic displacements under intermediate energy ion irradiation. Since the stopping power for swift heavy ions traversing in solids is mainly electroni c stopping, which is a different damage mechanism, the irradiation tolerance of Mg 2 SnO 4 in the electronic stopping region is still not known. Therefore, it is useful to conduct high energy ion irradiation to further evaluate the irradiation stability of Mg 2 SnO 4 and compare it with MgAl 2 O 4 4.5 Summary and Conclusions Irradiation tolerance of MgO, Nd 2 Zr 2 O 7 and prototype spinel MgAl 2 O 4 has been briefly summarized. All the three materials exhibit excellent resistance to irradiation induced amorphization. MgO has good stabili ty against fission fragment damage, while MgAl 2 O 4 exhibit large swelling in in-pile irradi ation tests, which is most likely due to fission fragment damage. It has been observed that Nd 2 Zr 2 O 7 could be amorphized by overlapping fission tracks under swift heavy ion irradiation. MgO exhibits significant 104

PAGE 105

swelling under neutron i rradiation due to defects format ion, while single crystal MgAl 2 O 4 shows nearly zero swelling under neutron irradiation. A preliminary study wa s conducted to evaluate the resistance of Mg 2 SnO 4 to radiation induced amorphization. Samp les were irradiated with 1.0 MeV Kr 2+ ions at 50 K and 150 K to a maximum fluence of 5 10 19 Kr 2+ ions/m 2 and 10 20 Kr 2+ ions/m 2 respectively. Microstructure and crystal structure evolutions were monitored and recorded in-situ by BF images and SAED patterns. The amorphization doses for Mg 2 SnO 4 irradiated by 1.0 MeV Kr 2+ ions at 50 K and 150 K were determined to be 5 10 19 Kr ions/m 2 and 10 20 Kr ions/m 2 which corresponded to an at omic displacement of 5.5 dpa and 11.0 dpa, respectively. The spinel crystalline structure was thermally recovered at room temper ature from the amorphous Mg 2 SnO 4 which was irradiated at 50 K. The electronic stopping power exce eded the nuclear stopping power except the end of range of ions, but was only ~1.5 keV/ nm. This suggests t hat the amorphization phenomenon observed in this study was mainly due to atomic displacement induced defect accumulation. Mg 2 SnO 4 showed less irradiation resistance compared with MgAl 2 O 4 against ballistic displacement damage, wh ich can be attributed to its inverse structure, higher covalency of the bond, larger ioni c size and charge difference between Mg 2+ and Sn 4+ 105

PAGE 106

CHAPTER 5 EVALUATION OF HYDRATION RESISTANCE 5.1 Introduction A very important requirement for IM materials is the com patibility with coolant. In LWRs, the coolant in the prim ary coolant system is water. The fuel materials should exhibit high resistance to hot water corrosion at ~300 C. This chapter focuses on evaluation of the chemical stability of the potential IM materials at hydrothermal conditions. The hydration behavior of MgO is briefly reviewed. The hydrothermal corrosion behavior of the MgO-Nd 2 Zr 2 O 7 composites and Mg 2 SnO 4 are then discussed. 5.1.1 Hydration of MgO The hydration of MgO has been extensively studied and is briefly reviewed here. Water molecules can be adsorbed onto the Mg O surface readily and react with MgO. 134 The hydration product of MgO is Mg(OH) 2 or brucite, which is a stable mineral crystallizing in a layer ed structure. Mg(OH) 2 forms a nearly close-packed array with a hexagonal crystal structure, and space group No. 164 ( 13 mP ). Figure 5-1 (A) shows unit cell of Mg(OH) 2 The lattice parameters of Mg(OH) 2 are a = 0.3142 nm and c = 0.4766 nm. 135 Figure 5-1 (B) shows a layered structure of Mg(OH) 2 The Mg ions are coordinated by six O-H groups with the hydrogen pointing to the next layer. As seen from the figure, only hy drogen bonds exist between the two layers, even though the distance between hydrogen and the nearest oxygen in the nearby layer is relatively large for hydrogen bonds (~2.5 ). Neve rtheless, these weak hydrogen bonds seem the only possible explanation for the stability of the structure. 136 The unit cell of MgO consists of four Mg ions and the volume of the unit cell is 0.075 nm 3 The unit cell of 106

PAGE 107

Mg(OH) 2 consists of one Mg ion and the volume of the unit cell is 0.041 nm 3 Therefore, the volume expansion for the transform ation of equal molar MgO to Mg(OH) 2 is calculated to be 118%. The hydration reaction can be described as follows: 2 2)( OHMgOHMgO (5-1) O H Mg A B Figure 5-1. Crystal of Mg(OH) 2 A) Unit cell, and B) laye r structure showing hydrogen bonds. 107

PAGE 108

It has been well established t hat hydration of MgO is th e result of MgO dissolution followed by precipitation of Mg(OH) 2 137,138 The overall hydration process can be summarized as (1) water molecules are adsorbed onto the MgO surface and dissociate into protons and hydroxyl ions, (2) the surf ace of MgO is reconstructed and dissolves into Mg 2+ and OH and (3) nucleation of Mg(OH) 2 occurs. 137,139-141 It has been suggested that the hydrati on reaction is mainly governed by dissolution of MgO. 137 The hydration of MgO by water vapor involves one more step which is the adsorption and condensation of initial water v apor onto the MgO surface to form a liquid layer, and the remaining steps are the same as in liquid water. 142,143 The hydration rate of MgO depends on the reactivi ty of MgO, which is relat ed to material history and production processes, such as calcination te mperature and the raw materials which it is made from. 138,144,145 For example, the dead-bur ned MgO obtained at a higher calcination temperature (~1200 C or above) exhibits higher hydration resistance than light burned MgO obtained at a lowe r calcination temperature (~900-1200 C). 139 The difference in hydration resistance of t hese MgO samples can be attributed to the surface chemistry. The dissolution of MgO is a heterogeneous surface reaction, and therefore the role of the surfaces which ar e exposed to liquid water or water vapor is important. The reactivity of MgO depends on surface area and surface morphology, especially defect sites and density of defects. 146 Different fabrication process and firing temperature result in varied surface areas and distinct surface mo rphology which leads to different hydration resistance. Dissolution of MgO is a thermally ac tivated process, and the temperature dependence of the dissolution rate typically follo ws Arrhenius relation. The dissolution 108

PAGE 109

rate can be either diffusion controlled or reaction controlled. 138,147,148 The activation energy reported for the MgO hy dration process varies from 50 to ~100 kJ/mol at a temperature range typically from 20 C to 200 C, which attributes to different reactivity and physical properties of MgO as well as test conditions. 139,144,147,149-152 The activation energy for MgO hydration at temperature higher than 200 C has not been reported, yet. 5.1.2 Hydration of Polycrystalline MgO The hydration behavior of polycrystalline MgO is quite different from single crystal MgO because of the presenc e of grain boundaries. The dissolution of MgO usually starts from defect sites where water mo lecules are easily adsorbed and dissociated, and therefore MgO grain boundaries are more susc eptible to hydration. 153 Kitamura et al. 154 have proposed a hydration mechanism for MgO polycrystals based on these observations: hydration begins on t he grain boundaries near the surface of polycrystalline magnesia, which causes gr ain boundary separation as a consequence of lattice expansion; the bulk material is t herefore separated into finer particles or aggregates and then into single crystalline gr ains, with consequent hydration of the single crystals. This process is thus re ferred to dusting processing. Therefore, polycrystalline MgO hydrates much faster than single crystal MgO at the same test conditions. Although the hydration of MgO in varied fo rms has been well studied, the hydration behavior for MgO based composites is not well understood, and the relationship between hydration resistance and microstructu re has yet to be studied. Moreover, the stability of Mg 2 SnO 4 in hot water is unknown. Ther efore, the hydrot hermal corrosion 109

PAGE 110

behavior of the MgO-Nd 2 Zr 2 O 7 composites and Mg 2 SnO 4 were both studied experimentally. The results are discuss ed in the following part of this chapter. 5.2 Experimental Procedure 5.2.1 Hydrothermal Corrosion Testing Setup To simulate the condition in case of cladding breach accident s, an autoclave was used to provide a high temperature and high pressure environment. Since the coolant water circulates in the reactor, the ideal test should be a dynamic study involving water circulation in the setup. However, static test was first applied as a simple and low cost alternative to investigate the hydrothermal corrosion behav ior of the potential IM materials. The tests were performed in a 600 ml commercial stainless steel pressure vessel (Parr Instrument Company, model 4768). The pressure vessel was equipped with a pressure relief valve set, a rupture disk, a pressure gauge, a heating mantle, and a temperature controller with one thermocouple. As-sintered ceramic pellets were placed into the vessel which was filled with 200 ml of de-ionized water and sealed. The tem perature controller was set at 300 C and a heating time of approximate 45 minutes wa s required for water to reach the final temperature and saturation pressure (~9 MP a). The test cont inued until desired exposure time was attained. The autoclave was shut off periodically, and samples were taken out and rinsed to remove the hydration pr oduct. The pH values of the water in the autoclave were measured by pH meter (OAKT ON, pH 5 series) and recorded. Pellets were visually inspected, dried, and weighed. After the measurement s were finished, the pellets were placed back into the autoclave which was refilled with fresh DI water and the corrosion test was resumed. 110

PAGE 111

5.2.2 Sample Preparati on for Characterization In order to characterize the microstruc ture, chemical composition and crystal structure, samples were examined by SEM, EDS and XRD. A detailed sample preparation for characterization and instrum ent operating conditi ons were described previously in Chapter 3. The cross section of the hydrated pellet for SEM characterization was prepared by embedding t he pellet in an epoxy resin and polishing down to desired cross section. 5.2.3 Composite Microstr ucture Quantification Corrosion originates from the hydration of MgO; t herefore, the hydrothermal corrosion resistance of an MgO composite is natur ally related to the distribution of MgO. The contiguity of MgO can be quantified by calculating the fraction of the total interface area of the MgO phase that is shared by the MgO phase. 155 For a composite structure with MgO grains dispersed in Nd 2 Zr 2 O 7 matrix, the contiguity of MgO is given by the following equation: 156 MNV MMV MMV MMSS S C )()(2 )(2 (5-2) where (S V ) MM and (S V ) MN are the interface areas per unit volume between MgO-MgO and MgO-Nd 2 Zr 2 O 7 phases. Saltykov 157 derived a basic equation that correlates the area of surfaces and the intercepts with grain boundaries on a section plane: L VPS 2 (5-3) where S V is the interface areas and P L is the number of in tercepts. Combining Equations (2) and (3), the contigui ty of MgO can be expressed as: 156 MNL MML MML MMP P P C )(2)(4 )(4 (5-4) 111

PAGE 112

where (P L ) MM and (P L ) MN are the number of intersections per unit length of test line with MgO-MgO and MgO-Nd 2 Zr 2 O 7 grain boundaries. About 20 straight test lines were drawn randomly on one SEM im age at 2000X or 5000X per composition and at least 500 intersections were counted for each image. The grain size of MgO and Nd 2 Zr 2 O 7 was calculated using m odified lineal intercept technique for two-phase polycrystalline ceramics. 158 The formula can be expressed as: ) 2 1 ( )1( 56.1ab aaNNM C D (5-5) where D is the average grain size of phase a C is the test-line length, is the volume fraction of phase a M is the magnificati on of the SEM image, N aa is the number of intercepts with the boundaries of contiguous grains of phase a and N ab is the number of intercepts with the interfaces between the phase a and phase b The microstructure homogeneity was characterized by the spatial distribution of MgO. A dimensionl ess parameter (HP MgO ) was used to indicate the degree of homogeneity and can be expressed by t he normalized standard deviation ( MgO ) of the distribution of MgO in sm all blocks separated by a NN grid as shown below: 159,160 MgO MgO MgOHP (5-6) 1 )(2 1 N xMgO N j j MgO (5-7) N j j MgOx N11 (5-8) where x j is the MgO area fraction in the jth separated block, MgO is the total area fraction, and N is t he number of blocks. Maximum ho mogeneity was characterized by a minimum value of HP MgO To quantify the microstr ucture homogeneity of the MgO112

PAGE 113

Nd 2 Zr 2 O 7 composites, one microstructure image at 5000x per each composition was converted into a binary image and separat ed by a 4 grid into 16 equal size nonoverlapping squares. The MgO area frac tion was measured and calculated for each square using the softwar e Image J (V. 1.38x) 161 and the homogeneity parameter was calculated using Equations (6), (7), and (8). 5.3 Results and Discussion 5.3.1 Hydrothermal Corrosion of the MgO-Nd 2 Zr 2 O 7 Composites 5.3.1.1 Preliminary results The composites made from all three mixing methods were exposed to 300 C deionized water in the autoclave to assess t he hydrothermal corrosion resistance. Pure Nd 2 Zr 2 O 7 and pure MgO pellets were also test ed in order to compare with the composite samples. As expec ted, the pure MgO pellet completely dissolved in less than 1 hour. The fast hydrati on rate of the MgO pellet can be attributed to the presence of grain boundaries and the dusting process described in the previous section. 154 By contrast, the pure Nd 2 Zr 2 O 7 pellet showed no mass loss and no phase transformation for up to 30 days. The phase stability of Nd 2 Zr 2 O 7 is an advantage compared with yttria stabilized zirconia which transforms into the monoclinic phase in hydrothermal conditions. 162 It is consequently antici pated that if introducing Nd 2 Zr 2 O 7 could prevent the dusting process, the corrosion resist ance of an MgO composite can be improved. The pH of the DI water before corrosion test was measured to be ~5.5 at room temperature, and the pH of t he water after corrosion test wa s measured to be ~6.9. The increase of the pH value wa s probably due to dissolution of hydration product, i.e. Mg(OH) 2 113

PAGE 114

As described in Chapter 3, the compos ites mixed by ball milling had more homogeneous microstructures than the composit es mixed by mortar and pestle grinding and water magnetic bar stirring wh ich contained agglomerates of Nd 2 Zr 2 O 7 and MgO. As expected, the hydrother mal corrosion resistance was strongly microstructure dependent. The inhomogeneous composites made by the mortar and pestle mixing and water magnetic bar mixing comp letely dissolved in less t han 1 hour after exposure to 300 C deionized water regardless of MgO vo lume fraction from 40 vol% to 70 vol% showing poor corrosion re sistance as pure MgO. In order to characterize the microstructu re of the corroded composite, one test was conducted in a less aggressive condition at 150 C and saturation pressure. Figure 5-2 shows the surface of a com posite (40 vol% MgO) made by mortar and pestle mixing after 24 h exposu re. The surface was highly heterogeneous as shown in the figure. Minimum corrosion was observed for the Nd 2 Zr 2 O 7 agglomerate regions with few missing grains caused by dissolution of MgO grains. The regions that were destructed much more severely were covered by hydr ation product suggesting that they were the MgO agglomerate regions. The observation i ndicates that hydrat ion of isolated MgO grains or small MgO grain clusters did not cause extensive destruction; on the other hand, the hydration of MgO agglomerated regions caus ed destruction not only on the surface but deep into the interior of the co mposite. Although not shown here, it was observed that a large amount of hydration product built up on the fractured surfaces suggesting that the composit es fractured along the ext ensively destructed regions where MgO grains were agglomerated. The observed hydration product Mg(OH) 2 had a plate-like shape which is c onsistent with literature. 139,149 114

PAGE 115

A B Figure 5-2. Characterization of the hydrated composite with inhomogeneous microstructure. (A) Surface of compos ite pellet after exposure to 150 C DI water for 24 h, and (B) surface of t he composite at higher magnification, where morphology of hy dration product Mg(OH) 2 is shown. 115

PAGE 116

The homogeneous composite made by the bal l milling process exhibited much higher hydration resistance. Instead of being dissolved instantly in DI water at 300 C, the composites with up to 60 vol% of Mg O were able to withstand hundreds of hours in the autoclave. However, the composite with 70 vol% MgO did not survived but dissolved after 1 h exposure in 300 C wa ter. Due to the improved hydrothermal corrosion resistance, the composites mixed by ball milling process were selected for further study. 5.3.1.2 Microstructu re characterization Figure 5-3 (A)-(D) shows the microstructure evolut ion of the composites made by ball milling as the volume fraction of MgO incr eased from 40% to 70%. The composites were sintered at 1550 C for 4 h. The microstructure of the composites can be qualitatively characterized as follows: (A) interconnected Nd 2 Zr 2 O 7 grains with dispersed MgO grains; (B) and (C) in terpenetration of the two phas es; and (D) interconnected MgO phase with dispersed or clustered Nd 2 Zr 2 O 7 phase. The average grain size of MgO and Nd 2 Zr 2 O 7 and contiguity and homogeneity of MgO were calculated and given in Table 5 Table 5. Microstructure anal ysis results for the MgO-Nd 2 Zr 2 O 7 composites (C MM is contiguity of MgO, HP MgO is homogeneity parameter, and errors are calculated as standard dev iations of the mean). MgO (vol%) MgO ( m) Nd 2 Zr 2 O 7 ( m) C MM HP MgO 40 1.0 0.2 0.8 0.1 0.14 0.08 50 0.9 0.2 0.8 0.1 0.27 0.09 60 2.2 0.5 1.1 0.3 0.34 0.14 70 2.7 0.3 1.2 0.2 0.42 0.17 116

PAGE 117

It was found that the average grain size of MgO increased with the volume fraction of MgO while the aver age grain size of Nd 2 Zr 2 O 7 remained relatively constant. The results indicate that the MgO phase wa s the faster growing species and the Nd 2 Zr 2 O 7 phase was the slower growing species that limited the grai n growth of MgO. The observation was consistent with the pinning and constraint behavior, also known as coupled grain growth, which has been well studied in the zirconia-toughened alumina (ZTA) composite system. 163-166 A B C D Figure 5-3. Microstructure of MgO-Nd 2 Zr 2 O 7 composites made by ba ll milling process. The MgO volume fractions are A) 40%, A) 50%, C) 60%, and D) 70 %, respectively. The dark phase is MgO and the light phase is Nd 2 Zr 2 O 7 117

PAGE 118

The calculated contiguity of the MgO phase increased from 0.14 to 0.42 as the MgO volume fraction increased from 40 vol% to 70 vol% which was in agreement with visual inspection of SEM images. The homogeneity parameter also in creased from 0.08 to 0.17 as the MgO volume fraction increased from 40 vol% to 70 vol% indicating that the dispersion of MgO became less homogeneous which was probabl y due to coarsening of MgO grains. 5.3.1.3 Hydrothermal co rrosion of the composites made by ball milling The hydration product was found being atta ched to the composite surface. The thickness of the hydration product layer increased as corrosion time and could be as large as a few hundred microns after 5 days of corrosion testing. In order to analyze the microstructure of the hydr ated surface, a cross secti on of an un-rinsed hydrated composite containing 40 vol% of MgO wa s polished and then examined by SEM. Figure 5-4 (A) shows an area of the interface bet ween the hydration product and the unattacked composite at relatively low magni fication. The hydration product formed a porous layer where water can diffuse through readily. XRD was performed on the surface of the hydration produ ct layer to characterize t he crystal structures of the hydration products and the profile is shown in Figure 5-4 (B). The XRD profile for the unhydrated composite is also shown in the figure for a com parison. As expected, only the brucite phase Mg(OH) 2 and the pyrochlore phase Nd 2 Zr 2 O 7 were identified from the hydration product. The results confirm that MgO dissolved and precipitated out only as Mg(OH) 2 and that Nd 2 Zr 2 O 7 did not react with water at 300 C. The hydration product Mg(OH) 2 and pyrochlore Nd 2 Zr 2 O 7 can be simply distinguished in the SEM image by contrast due to different atomic masses (the light phase is Nd 2 Zr 2 O 7 and the dark phase 118

PAGE 119

is Mg(OH) 2 ). In the hydrati on product layer, the Nd 2 Zr 2 O 7 grains were randomly embedded in the agglomerated Mg(OH) 2 and the platelet shape of Mg(OH) 2 was consistent with previous observation. A 2025303540455055606570 Hydration Product Intensity (Arb. Units)2(Degrees) Pyrochlore Mg(OH)2MgOUnattacked Composite B Figure 5-4. Characterization of th e hydrated composit e with homogeneous microstructure. A) Cross-section of a hydrated composite pellet, and B) XRD profiles of hydration produc t and unattacked composite. 119

PAGE 120

Figure 5-5 shows the interface at a higher magnification where different microstructural features on both sides of t he interface can be seen. Hy dration induced destruction was only observed near the hydrated surface but not in the interior composite. The observation suggests that the hydration of MgO was primar ily limited to the surface region of composites. Cracks 10 m Detached grains and cracks Detached Nd2Zr2O7and MgO grain clusters Fractured Nd2Zr2O7 grain 1 m Figure 5-5. Microstructure of the interface between the composit e and the hydration product layer at high magnification. Several destructed areas near the interfac e are marked in the figure show cracked MgO grains, fractured Nd 2 Zr 2 O 7 grains, and detached grain clusters with unhydrated MgO grains. Most Nd 2 Zr 2 O 7 grains were connec ted with other Nd 2 Zr 2 O 7 grains in the unattacked composite forming Nd 2 Zr 2 O 7 grain clusters. Howe ver, the majority of Nd 2 Zr 2 O 7 grains in the hydration product laye r were single grains. The surface 120

PAGE 121

destruction and Nd 2 Zr 2 O 7 grain separation are probably c aused by the stresses induced by hydration of MgO grains and grain boundaries. Similar to the dusting process for polycrystalline MgO, water can migrate fa st along the connect ed MgO grain boundaries and cause grain boundary hydr ation and grain separation. However, due to the presence of Nd 2 Zr 2 O 7 grains, the contiguity of MgO phase was reduced significantly as shown in Table 5 As a result, the migration of water through MgO grain boundaries was localized and primarily limited to the surf ace region of the com posite. Therefore, the hydration of MgO grains and grai n boundaries caused destruction only on the surface but not in the interior. The hydration product could be easily remo ved after rinsing and wiping with a paper towel. The agglomeration of hydration product and its weak adhesion to the composite surface are probably due to the pr esence of attractive forces as a result of secondary bonds such as attractive Van der Waal s bonds and hydrogen bonds. It is worth noting that the hydrotherma l tests were performed under a st atic condition and the free convection in the autoclave seemed ineffect ive in removing the hy dration product from composite surface. However, in LWRs wate r is dynamically circulating and the forced convection may result in washing hydr ation product away instantly depending on the forced flow rate and the magnitude of the attractive forces. 5.3.1.4 Normalized mass loss rate and temperature dependence Given that the homogeneous composites ex hibited improved corrosion resistance, the performance of the composites was further evaluated by the mass loss rate. It was expected that the mass loss rate depends on t he exposure area of the sample, so the mass loss was normalized by the geometric ar ea and is referred to normalized mass 121

PAGE 122

loss (NML). 11,162 The hydration product was remov ed from the surface after every exposure. The pellet diameter thickness, and weight we re then recorded and the NML was calculated as follows: A mm NMLfi (5-9) where m i and m f are sample mass before and after each exposure and A is an average of the geometric area before and after eac h exposure. The NML was plotted against exposure time shown in Figure 5-6 (A) and the systematic errors were taken into consideration. A linear re lationship between the NML and the exposure time was found, so the data points were fitted by linear regression and the NML rate for each composition was empirically determined by the slope of the fit. The NML rate remained constant even at different time intervals suggesting that the corrosion process was not affected by the thickness of the built-up hydration layer and thus not limited by the diffusion of water through the built-up laye r. The observed linear relationship was consistent with liter ature for the MgO-ZrO 2 composites system, and the NML rate was on the same order of magnit ude compared to the composites with similar volume fraction of MgO. 11 It was also observed that as the volume fraction of MgO in creased, there was a corresponding increase in the NM L rate. Medvedev and coworkers 11 correlated the NML rate with the MgO weight fraction by using an exponential fit for MgO-ZrO 2 composites. A similar exponential rela tionship could be fou nd here for the MgONd 2 Zr 2 O 7 composites. Nevertheless, the ex ponential relationship was determined empirically and requires further investigatio n and physical explanation. The increased NML rate with MgO volume fr action is most likely due to the improved MgO contiguity 122

PAGE 123

because the connected MgO grains and grain boundar ies serve as fast hydration paths. Even though the corrosion was limited to the su rface region, water molecules were still able to migrate and initiated hydration through the connected MgO grain boundaries causing destruction and grain separation near the composite surface. The surface area consequently increased and the NML rate was enhanced. 0150300450600750 0.00 0.05 0.10 0.15 0.20 0.25 60 vol% MgO 50 vol% MgO 40 vol% MgO Exposure Time (h)Normalized Mass Loss (g/cm2) 18202224 -10 -8 -6 -4 300250200150 60 vol% MgO Ea=41 kJ/mol 50 vol% MgO Ea=41 kJ/mol 40 vol% MgO Ea=39 kJ/molT (oC) ln k10000/T (K-1) A B Figure 5-6. Quantitative analysis results. A) NML versus ex posure time in 300 C DI water, and B) extracted Arrhenius relationship. The temperature dependence on the NML rate was also investigated by conducting the corrosion tests at 150 C, 200 C, and 250 C, respectively. Similar to the tests conducted at 300 C, a linear kinetics wa s found between the NML and the corrosion time. The NML rates were det ermined for all the tests by the slope of the linear fit and the results are summarized in Table 5 The natural logarithm of NML rate versus the reciprocal value of absolute temperature at which the hydrotherma l corrosion tests were conducted is plotted in Figure 5-6 (B). The data points were fitted by linear regression and the apparent activation energy ( E a ) was extracted using the Arrhenius equation. In this case, the apparent E a was merely a measure of th e temperature dependence of the 123

PAGE 124

NML rate. The final results are listed in Table 5 showing similar apparent E a for the composites with 40, 50, and 60 vol% of Mg O and the calculated values were ~39-41 kJ/mol. The results suggest that same corrosion mechani sm(s) may be applied to these composites at the te mperature range from 150 C to 300 C. The SEM observation indicates that the corrosion process is ma inly governed by surfac e hydration of MgO, but the estimated E a for the composites is slightly lo wer than the typica l values reported for hydration of MgO in literature (50-65 kJ/mol). 144,147,154 The difference in activation energy is most likely due to different test conditions. The testing temperatures in the present work are in the range of 150 300 C, which is higher than the other experiments reported in literature. Moreov er, the physical values used to quantify the hydration process are differen t. In the present work the mass loss of the composites was quantified but in other st udies, the hydration of MgO was quantified by conversion rate. Even though the mass loss of the composit es is due to hydration of MgO, it is also related to other physical properties of the composites such as the MgO volume fraction and sample density. The discr epancy also suggests that t he hydrothermal corrosion of the composites may be governe d by multiple mechanisms which were competing with surface hydration such as water migration and hydration along MgO grain boundaries. Table 5. Normalized mass loss (NML ) rate and apparent activation energy (E a ) NML Rate (10 -4 gcm -2 h -1 ) MgO (vol%) 150 C 200 C 250 C 300 C Apparent E a (kJ/mol) 40 Not tested 0.36 0.01 1.27 0.05 2.01 0.03 39 8 50 Not tested 1.44 0.04 3.56 0.02 9.3 0.2 41 3 60 1.40 0.07 4.13 0.04 Not tested 29.5 0.3 41 2 124

PAGE 125

It is worth noting that th e hydration of MgO is a t hermodynamically favorable reaction. Introducing Nd 2 Zr 2 O 7 as a second phase cannot stop hydration of MgO but it can slow down the NML rate by reducing t he surface area and conti guity of MgO. The composite with a higher MgO volume fracti on is desirable due to its higher thermal conductivity. Nevertheless, t he current status shows that the composite with 70 vol% MgO failed because of high contiguity of MgO. In order to impr ove the hydrothermal corrosion resistance of 70 vol% MgO composite, the contiguity of MgO must be reduced. Reducing the contiguity of Mg O may lower the thermal conduc tivity of the composite, but the effect is not expe cted to be large because of the volume fraction of MgO remains the same. 93 The contiguity usually increases with volume fraction; however, some composites may keep phase contigui ty independent of volume concentrations such as the composite spheres assemblage (CSA), a theoretical model for spherical particles surrounded by a concentric matrix shell. 167 Therefore, the desired microstructure can be made fo r the composite with 70 vol% MgO by sintering a compact of Nd 2 Zr 2 O 7 -coated MgO particles with desired c oating thickness. The hydrothermal corrosion resistance may be further improved by enlarging grain si ze of MgO. The density of MgO grain boundaries c an be reduced if the grain size of MgO is enlarged. Since the dusting process and surface destr uction are caused primarily by MgO grain boundary hydration, minimizing MgO grain boundaries can reduce the hydration pathways thus lowering the NML rate. 5.3.1.5 Coprecipita tion and gel-casting In order to improve the hydration resistanc e of the composites with 70 vol% MgO, some other processing methods such as c oprecipitation and gel-casting were used as 125

PAGE 126

attempts to produce the desired composite microstructure. Even though these processes were not successful to produce the composites with improved hydration resistance by far, they are briefly reported here so that the same paths will not be repeated or new ways to succeed based on these experiments can be realized. The coprecipitation method is de scribed here in brief. An MgCl 2 solution was prepared by dissolving 0.0141 mol of MgCl 2 6H 2 O (Fisher, 99.0-101.0%) in DI water. The MgCl 2 solution was then added drop by drop into an NH 3 H 2 O solution agitated by a magnetic bar at the pH of 9.5-10 to form Mg(OH) 2 suspension. The pH of the solution was then increased to ~11 by adding NH 3 H 2 O. Another solution of NdCl 3 and ZrCl 4 in a stoichiometric ratio was prepared. To prepare this solution, 0.0015 mol of NdCl 3 6H 2 O (Acros Organics, 99.9% ) and 0.0015 mol of ZrCl 4 (Acros Organics, 98%) were dissolved in DI water. The soluti on was then added into the Mg(OH) 2 suspension drop by drop. Figure 5-7 shows the flowchart of the coprecipitation process. MgCl2H2O 28.9261 g 28 ml H2O NH3 H2O Solution pH = 9.5-10 T = 80 C Drop by Drop Nd(NO3)3 5.3255 g ZrCl4 3.5269 g Mixture solution Mg(OH)2 Suspension pH = 11, T = 80 C Drop by drop Mg(OH)2 Nd(OH)3 Zr(OH)4 H2O 29 ml Figure 5-7. Flowchart of th e coprecipitation process for fabrication of the MgO-Nd 2 Zr 2 O 7 composites. 126

PAGE 127

The precipitates were collected, dried and calcined at 1300 C for 6 h. The pyrochlore phase and MgO phase were confir med by XRD. The powder was pressed to form green pellets wh ich were sintered at 1550 C for 4 h. A detailed procedure on composite forming and sintering was de scribed previously in Chapter 3. Figure 5-8 shows the microstructure of the com posites made by copr ecipitation. The microstructure produced by the coprecip itation process was not homogeneous and can be characterized as Nd 2 Zr 2 O 7 agglomerates surrounded by a continuous MgO phase. Therefore, the desir ed composite microstructure was not successfully produced using coprecipitation in the present study. Figure 5-8. Microstructure of the compos ites made by the coprecipitation method. The hot water corrosion tests for the compos ites made by the c oprecipitation method were conducted at 300 C and 9 MPa. The co mposites dissolved quickly in one hour. It is not surprising that the composites exhibited poor hy drated resistance because the desired composite microstruc ture was not achieved. The other processing method is gel-c asting. In an attempt to produce homogeneously dispersed Nd 2 Zr 2 O 7 coated MgO particles in aqueous solutions, 127

PAGE 128

polyelectrolyte dispersants was used to incr ease the surface charge of particles and to stabilize the slurry. Figure 5-9 shows the conceptual design of the experiment to achieve the CAS microstructure. The Nd 2 Zr 2 O 7 and MgO particles were first coated with polyelectrolytes of opposite char ges. Under electrostatic forces, Nd 2 Zr 2 O 7 can be attracted onto the surface of Mg O to form the coated particles In this experiment, poly sodium 4-styrenesulf onate (PSS) (Aldrich, 20% solution) and poly diallyldimethylammonium chlo ride (PDAC) (Aldrich, 20 wt% in water) were chosen as the polyelectrolytes. + + + + + + +Nd2Zr2O7 -MgO PSS PDAC + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + 10 um 10 um MgO Nd2Zr2O7 MgO Nd2Zr2O7 Figure 5-9. Schematics of experiment design. The Nd 2 Zr 2 O 7 slurry was prepared by dispersing Nd 2 Zr 2 O 7 powder with 0.5 wt% PDAC in 70 ml DI water. The Nd 2 Zr 2 O 7 slurry was ball milled for 8 days. The MgO 128

PAGE 129

slurry was prepared by dispersing MgO powder with 0.5 wt% PSS in 70 ml DI water. The MgO slurry was ball milled for one day. The particle size and size distribution of the two slurries after ball milling are shown in Figure 5-10 0.1 1 10 100 0 2 4 6 8 10 12 14 77 nm 275 nm MgO dispersed with PSS Volume (%)Particle Size (m)Nd2Zr2O7 dispersed with PDAC Figure 5-10. Particle size distribution of the MgO and Nd 2 Zr 2 O 7 slurry. The mixed slurry was prepared by adding the PDAC coated Nd 2 Zr 2 O 7 slurry into the PSS coated MgO slurry drop by drop. The purpose of the mixing sequence and the slow mixing speed was to avoid rapid coagulation and sedimentation of the two slurries by electrostatic attraction. The slurry was vigorously agitated by magnetic bar stirring. Zeta potentials for the noncoated and PSS coated MgO sl urries, non-coated and PDAC coated Nd 2 Zr 2 O 7 slurries, and the mixed slurry were measured and the results are shown in Table 5 As the table shows, the sign of the zeta potentials of the MgO and Nd 2 Zr 2 O 7 slurries were reversed when polyelectrolytes were added. The values were 129

PAGE 130

large enough to stability the slurries. The zeta potential of the mixed slurry had the same sign and similar value to the PDAC coated Nd 2 Zr 2 O 7 slurry, suggesting the Nd 2 Zr 2 O 7 coated MgO particles may be formed in the solution. Table 5. Zeta potential and pH of the slurries. Slurry Zeta Potential pH Nd 2 Zr 2 O 7 -21.9 2.5 8.5 PDAC coated Nd 2 Zr 2 O 7 68.4 3.1 6.5 MgO 36.1 1.9 10.0 PSS coated MgO -78.4 2.2 10.3 Mixture (70 vol% MgO) 64.7 3.2 9.8 SEM and TEM were used to examine the mor phology of the particles in the mixed slurry. The images are shown in Figure 5-11 It can be seen that some of the Nd 2 Zr 2 O 7 particles (mainly below 1 m) attached to the surf ace of MgO (or Mg(OH) 2 ) but a completed coverage was not achieved. The partial coverage was e ffective to reverse the sign of the zeta potential of the PSS coated MgO slurry, but it may be too low to achieve the final desired CAS microstructure. Nd2Zr2O7MgO or Mg(OH)2 Nd2Zr2O7MgO or Mg(OH)2 A B Figure 5-11. Morphology of th e particles in the mixed sl urry. A) SEM image and B) TEM image. 130

PAGE 131

Figure 5-12 shows the pictures of the MgO and Nd 2 Zr 2 O 7 slurries with and without dispersants before and after 24 h. It can be see that the sl urries without dispersants precipitated out in less than 24 h, while the slurries with di spersants were still stable. The observation confirms that the slurries wi th dispersants were more stable than the slurries without the dispersants. Figure 5-12. Stability of slurries with and without polyelectrolytes. In order to maintain the characteristics of the mixed slurry, a di rect casting method should be applied to form green body so as to avoid further processes such as dry grinding and pressing. It wa s intended to use the gel-casti ng technique to form the green body. Due to the limited solid oxide powder, the slu rry was prepared starting with low solid loading (5-10 vol%) and agitated by a magnetic bar in a beaker to evaporate water to increase the solid loading. The hi ghest solid loading of the mixed slurry that could be handled in the experiment was onl y ~22 vol%. The solid loading need to be 131

PAGE 132

further increased to at least 45 vol% for green body forming and sint ering. Therefore, the process must be optimized to reduce the viscosity of the slurry and to increase the solid loading, such as optimiz ing the amount of dispersant. 5.3.2 Hydrothermal Corrosion Resistance of Mg 2 SnO 4 To evaluate the hydrothermal corrosion resistance of Mg 2 SnO 4 sintered Mg 2 SnO 4 pellets were exposed to water at 300 C and saturation pressure in the autoclave. The mass and geometry of the sample were measur ed and recorded at a time interval of 10 days. Figure 5-13 shows the mass and volume changes of Mg 2 SnO 4 as a function of time. As can be see from the figure, both the mass and the volume the sample increased at the first few days and then decr eased. However, the changes were small and less than 1% up to 30 days. The mass change was so small that the value was within the uncertainty of the measurement, and thus Mg 2 SnO 4 exhibited no measurable mass change after 30 days corrosion. The results indicate that sintered Mg 2 SnO 4 pellets were able to maintain the geomet ry integrity without a mass change up to 30 days. 0200400600800 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 0200400600800 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 Volume Change (%)Time (h) Original weight: 0.1353 g Diameter: 6.20 mm Thickness: 1.11 mm Mass Change (%) Figure 5-13. Mass and volume change as a function of corrosion time for Mg 2 SnO 4 132

PAGE 133

To evaluate the phase stability of Mg 2 SnO 4 under hydrothermal c onditions, XRD was performed on the surface of an Mg 2 SnO 4 pellet after 120 h exposure to water at 300 C and 9 MPa. Figure 5-14 shows the XRD pattern. Thre e phases can be identified from the XRD profile, which are Mg(OH) 2 SnO 2 and Mg 2 SnO 4 Some peaks which do not belong to either of them are marked as star s in the pattern, indicating other crystalline materials also existed. 10203040506070 (220)(310)(211)(531) (200) (101) (110)SnO2 PDF 41-1445 (440) (511) (331) (400) (222) (110) (102) (101) (001) (311) (220) (111)Intensity (Arb. Units) Mg2SnO4 PDF 24-0723 Mg(OH)2 PDF 44-1482 Figure 5-14. X-ray Diffracti on profile of the corroded Mg 2 SnO 4 surface after 120 h exposure to H 2 O at 300 C and saturation pressure. The 3 highest intensity d-spacing method was used in order to identify the peaks marked with stars. The dspacing values for three unidentified peaks with highest intensities were 1.7202, 2.5651 and 2.2931 The se arch range was set as 0.01 of the d-spacing values and there were 111 hits found in the database. All the patterns were compared to the obtained XRD pattern but a match could not be found. The unidentified phases could bel ong to hydration products formed by Mg and Sn. The XRD results indicate that Mg 2 SnO 4 was not stable in water at 300 C and saturation pressure, and the two main phases formed afte r hot water corrosion were Mg(OH) 2 and SnO 2 133

PAGE 134

The microstructure of the corroded surface of Mg 2 SnO 4 was also characterized using SEM. Figure 5-15 shows the SEM images recorded at different magnifications for the corroded surface of Mg 2 SnO 4 after exposure to water at 300C and saturation pressure for 120 h. The volumetric c hange was not noticeable by examining the morphology of the corroded surface shown in Figure 5-15 (A). Figure 5-15 (B) shows the surfaces of grains at a relatively high m agnification. It can be seen that the grain surface was attacked by water and covered wi th precipitates of nanometer size and some platelets. Figure 5-15 (C) shows the morphology of the precipitates an d a platelet formed near grain boundaries at a high magnification. 50 m 5 m A B 2 mMg(OH)2SnO2 C Figure 5-15. Mor phology of the Mg 2 SnO 4 pellet after exposu re to water at 300 C and saturation pressure for 120 h. Images were taken at a magnification of A) 1000x, B) 10, 000x, and C) 20,000x. T he precipitates are SnO 2 and the platelets are Mg(OH) 2 as labeled in the image. 134

PAGE 135

In order to identify the chemical composition of the nano precipitates, point EDS was performed and the EDS spectra we re collected. Since the in teraction volume of the electron beam with the samp le was on the order of 1 m 3 it was not feasible to get exact chemical composition of the precipit ates of nanometer size without collecting background information, so it wa s merely a qualitative analysis. Figure 5-16 shows two EDS spectra collected at different region s where the surface was covered with and without precipitates. It was found that the weight ratio of Sn to Mg for the surface covered with precipitates was higher than the surface wi thout precipitates, suggesting the areas covered with precipit ates were rich in Sn. T he EDS results imply that the precipitates may be crystalline SnO 2 identified from the XRD pattern, and the platelets could be Mg(OH) 2 based on the morphology. The fo rmation of nano precipitates suggests that Sn 4+ may also be leached out from the matrix but subsequently precipitated out as SnO 2 012345 MgKSnLSnL2SnLSnLSnL1 OKIntensity (Arb. Units)KeVCK 012345 MgKSnLSnL2SnLSnLSnL1 OKIntensity (Arb. Units)KeVCK A B Figure 5-16. Electron Dispersive Spectroscopy (EDS) spectra. A) The spectrum was collected on a region covered with prec ipitates and B) wit hout precipitates. 135

PAGE 136

Based on the XRD results and SEM observations, the corrosion process for Mg 2 SnO 4 is proposed as follows: the surface of Mg 2 SnO 4 is attacked by water; both Mg 2+ and Sn 4+ ions are leached out from the matrix; the Mg 2+ ions precipitate out as Mg(OH) 2 which may loosely attach to the corroded Mg 2 SnO 4 surface, and Sn 4+ forms SnO 2 Due to the low solubility of Sn 4+ in water, it is expected that precipitation of SnO 2 occurs instantly, and the size of precipitates is essentially small. The results shown in Figure 5-13 indicate that the dissolution of Mg 2 SnO 4 in water at 300 C is a slow process, and could be kinetically hindered by the a ttached hydration products on the sample surface. 5.4 Summary and Conclusions The corrosion behavior of the MgO-Nd 2 Zr 2 O 7 composites in hydrothermal conditions was studied. It was found that the corrosion resistanc e depends on microstructural homogeneity. The inhomogeneous composites made by mo rtar and pestle mixing and water magnetic bar stirring dissolved quickly in water at 300 C showing no improvement over pure MgO. The homogeneous composites made by ball milling exhibited improved hydration resistance and the corrosion was prim arily limited to the surface region. A linear relationship was found between the NML and the corrosion time. The NML rates for the composites with MgO volume frac tion of 40%, 50% and 60% were determined to be 0.0002, 0.0009, and 0.0028 g cm -2 h -1 at 300C. It was observed that the NML rates increased with MgO volume fraction, which can be attributed to the contiguity of MgO and hydration induced surface destruction. The NML rate also increased with corrosion temperature from 150 to 300 C and it followed an Arrhenius relationship. The apparent activation energy was calculated to be ~3941 kJ/mol. A desired microstructure with 136

PAGE 137

enhanced corrosion resistance should have minimum contiguity of MgO, good homogeneity, and large MgO grain size (minimizing grain boundaries). Mg 2 SnO 4 was not stable in water at 300 C and saturation pressure. Both Mg 2+ and Sn 4+ were leached out from the matrix and precipitated out as Mg(OH) 2 and SnO 2 The corrosion was limited to the sample su rface and probably hindered by the formed hydration products on t he corroded surface. Mg 2 SnO 4 maintained its geometric integrity with volume expansion less than 1% and no mass change after exposure for 720 h. Therefore, si ngle phase Mg 2 SnO 4 exhibited much higher hydration resistance than the MgO-Nd 2 Zr 2 O 7 composites. 137

PAGE 138

CHAPTER 6 EVALUATION OF AQEUOUS REPROCESSING FEASIBILITY 6.1 Introduction Reprocessing was originally used in weapo n programs to extract Pu for fabrication of atomic bombs. 168 The US was against reproce ssing in the commercial nuclear program since middle 70s mainly bec ause of proliferation concerns. 4 However, the spent nuclear fuels become nuclear waste instead of a valuable ener gy source without reprocessing. Recently the interests of reprocessing spent nuclear fuels have revived in the Advanced Fuel Cycle Initiative (AFCI) and the Global Nuclear Energy Partnership (GNEP). 4 Two aqueous based reprocessing techniques are briefly reviewed here. The first one is called PUREX process. PUREX is an acronym standing for plutonium and uranium recovery by ex traction. It is a liquid-liquid extraction method, and is the only large-scale system at present for reprocessing spent nuclear fuel to separate Pu, U, and fission products. 169 The important chemical used in this process is a complexing molecule called tributyl phosphate (TBP). U and Pu can be complexed by TBP and extracted from solution while other fissi on fragments and minor actinides cannot. 170 The separated Pu is then mixed with depleted U to form MOX fuel for LWRs. A major concern about PUREX is that Pu is able to be recovered seperately from commercial spent fuels, posing a big threat to proliferation. 171 The second technique is called UREX+1a, wh ich is a set of processes that has been designed recently to overcome the drawba cks of the PUREX proc ess, such as the proliferation problem. 45 The first step is the UREX pr ocess which separates only U and 138

PAGE 139

Tc, and the transuranium (TRU) isotopes stay in solution with the fission products. The following step removes Cs and Sr using chlorinated cobalt dicarbollide (CCD) and polyethylene glycol (PEG). T he third step removes all but the lanthanides and actinides from the waste, and the last step separates TRU from the lanthanides. Pu stays with the minor actinides a nd is no longer separated. 172 Mixing Pu with the minor actinides not only makes Pu less attractive for making nuclear weapons but also raises the power output by having more fissionable materi als such as the minor actinides. 4 Today, the main objective of reprocessing nuclear spent fuels is to better utilize U and Pu and to reduce demands on nat ural resources. Since the motivation for utilizing IMF is for transmutation of Pu and minor acti nides, which come from dismantled nuclear weapons and spent nuclear fuels, it is natural to design IMF as a multi-recycling type of fuel so that the unconsumed fissile materi als can be recycled more than once. Among those reprocessing techniques such as PUREX and UREX+1a, dissolution of spent nuclear fuels in aqueous soluti on is the head-end process. 168 The research motivation in this chapter is to study the dissoluti on behavior of the potent ial IM materials in aqueous solutions as an evaluation for the feasibility of aqueous reprocessing. The research objective is to understand the dissolution behavior and kinetics of the potential IM materials in acidic solutions. The rese arch goal is to find simple, effective, and economic methods and etchants that are capable of disso lving the potential IM materials. 139

PAGE 140

6.2 Experimental Procedure 6.2.1 Dissolution Test Setup Three dissolution methods were applied, including static dissolution, dynamic dissolution and ultrasonic dissolution. The st atic dissolution tests were carried out in a 250 ml beaker filled with 100 ml aqueous solu tion. Samples were immersed in the solution with no agitation. The dynamic dissolution tests were performed in 100 ml solution in a flask connected with a water cooled condenser. To provide uniform heat and constant mechanical agit ation, the flask and thermo meter were immersed in a water bath heated on a hot plat e and agitated by a magnetic bar in the solution. The magnetic bar was usually stirring at a const ant rate of 350 rpm. In the ultrasonic dissolution tests, an ultrasonic bath with a capacity of 2.8 L and 40 kHz transducer (Fisher Scientific Mechanical Ultrasonic Cleaners, Model FS20H) was used, and the tests were carried out by positioning a 250 ml flask in the central top zone of the sonication bath at 60 C. The ultrasound power was obt ained calorimetrically by the temperature increase over the first 5 min using the following equation: 173,174 sm dt dT WPower )( (6-1) where T is the temperature (K), t is time (s), m is mass of solution (g) and s is the specific heat of the solution (J/g K). Solution samples of 0.3 ml were drawn from the beaker at certain time intervals using a micron pipette with an error in the range of 10%, and these samples were diluted to 12 ml solution by adding distilled water and stored in a 15 ml low alkali glass vial. The ion concentrations in sample solutions were measured using inductively coupled plasma atomic emission spectrosc opy (ICP-AES, Perkin-Elmer Plasma 3200). 140

PAGE 141

Linear calibration was applied. The standard solutions with different ion concentrations varying from 1 to 100 mg/L were me asured to generate t he linear intensityconcentration relationship. The calibration curves with R value greater than 0.99 were used. The dissolved ion concentrations were calculated by multip lying the measured ion concentration, the dilution fa ctor and the volume of solvent. In an attempt to dissolve pyrochlore Nd 2 Zr 2 O 7 the dissolution tests were also performed using boil ing concentrated H 2 SO 4 in collaboration with Pr of. Czerwinski in the Department of Chemistry and Harr y Reid Center at the University of Nevada, Las Vegas. The tests were performed by his graduate student Ki el Holliday. Samples were placed in concentrated sulfuric acid (200 ml) filled in a round bottom flask equipped with reflux condensers and heated to boiling usi ng a heating mantle. Samp le solutions of 0.5 ml were taken periodically using gl ass Pasteur pipettes and diluted to 10 ml using DI water, and were then run through a Spectro Ciros ICP-AES to determine ion concentrations. 6.2.2 Structural Characterization The microstructure and crystal structure of the composites were characterized using scanning electron microscopy (SEM) and X-ray diffraction (XRD). The particle size and size distribution of dissolution residue were characterized using laser light scattering. The sample preparation for characterization and instrument specific ation are described previously in Chapter 3. 6.3 Results and Discussion 6.3.1 Dissolution of the MgO-Nd 2 Zr 2 O 7 Composites in HNO 3 Since HNO 3 is widely used in the current repr ocessing techniques, the dissolution behavior of MgO-Nd 2 Zr 2 O 7 composites in HNO 3 was the initial focus of this study. High 141

PAGE 142

molarity of HNO 3 (11 M) was selected for all tests to achieve high activity and dissolution rate. The fraction of consumed H + ions during dissolution was on the order of 10 -3 suggesting that the dissolution reaction has a minimum effect on the pH. The molarity of dissolved Mg 2+ ions was always less than 0.1 mol/L, which is at least one order of magnitude lower than the saturati on concentration (the solubility of Mg(NO 3 ) 2 is calculated to be 6.2 mol/L at 60 C 175 ), indicating that the dissolution reaction did not reached equilibrium. 6.3.1.1 Dynamic dissolution a nd dissolution rate of MgO Figure 6-1 shows the results of one test conducted at 60 C with magnetic bar stirring. The composite consisted of 70 vol% MgO with a density of 96 %. It was found that only Mg 2+ ions were leached out from the composite, and that no Nd 3+ or Zr 4+ ions were detected in solution. 0510152025 0 10 20 30 40 50 0.0 0.2 0.4 0.6 0.8 1.0 Mg2+ Nd3+ Zr4+ PenetrationDissolved Ions (mol/m2)Dissolution Time (h)Fraction of MgO DissolvedNDR = 3.61 0.04mol/m2 h p = 40.3 0.5 m/h Figure 6-1. Dynamic dissoluti on of the composite with 70 vol% of MgO in 11 M HNO 3 at 60C. The number of moles is normalized by the geometric area of MgO. For an explanation of the uni ts utilized, see text. 142

PAGE 143

As discussed previously in Chapter 2, t he dissolution of metal oxides in aqueous solutions is a heterogeneous reaction, and the rate of reaction depends on the intrinsic mass transfer coefficient k s1 the surface area S, and the driving force of concentration difference (C eq C). To compare the dissolution rates at different dissolution conditions, examining the initial rate where the solu tions are far from equilibrium is a common approach. 50,176 The dissolution rate is typically normalized by the surface area. Since only MgO can be dissolved, the dissoluti on rate of MgO was normalized by its geometric area. The geometric area of MgO was calc ulated by multiplying the MgO volume fraction and the total geometric area of the composite. The calculation is based on the conclusion that the area fraction of one phase in a random 2D section plane in a composite equals to its overall volume fraction. 156 In Figure 6-1 the y-axis on the left shows the geometric area normalized ions in so lution, which is converted into dissolved fraction shown on the right y-axis. As t he figure shows, the dissolution of MgO was completed after 25 h. A linear relationship was found bet ween the number of dissolved Mg 2+ ions and the dissolution time at the begi nning of dissolution, so the normalized dissolution rate (NDR) of MgO can be described by: dt dC A V NDRMgO (6-2) where V is equal to the volume of solvent, A MgO is the initial geomet ric area of MgO, and dC/dt is the change of the concen tration as a function of time, which can be obtained by fitting the data using linear regression. The calculated NDR for this test is 3.61 mol/m 2 h. The temperature effect on the NDR of MgO was ev aluated by conducting another dynamic test at room temperature around 20 C and comparing the results with the one obtained at 60 C. It was found that th e NDR of MgO increased by a factor of 36 as the 143

PAGE 144

temperature increases from 20 to 60 C, indicating the temperat ure has great effect on the NDR of MgO. The dissolution of spherical particles in aqueous solutions can be associated with the well-known shrinking core models, and the rate controlling mechanism can be classified into liquid film diffusion, surface reaction and product layer diffusion. 55 However, the composite has a finite cylindr ical shape and the dissolution is selective, which significantly complicates the model. Since the pellet has a relatively large aspect ratio greater than 8 with diameter to thick ness, the flat plate model can be used as an approximate for the initial dissolution of Mg O because the initial change on the radial direction is not significant. In the case of flat plate, the three mechanisms can be expressed as follows: 55 tkxF for film diffusion control (6-3) tkxD2 for product layer diffusion control (6-4) tkxS for surface reaction control (6-5) where x is dissolved fraction at time t and k F k D k S are apparent rate constants. For the dynamic dissolution, the film diffusi on can be eliminated due to strong stirring. 177 Therefore, the obtained li near relationship between t he dissolved fraction and dissolution time indicates t hat dissolution of MgO is a surface reaction controlled process at this reaction condition. To obtain a better model for describing the di ssolution behavior of Mg O, the flat plate model was modified here, and a new model was proposed by taking into account the shrinkage on the radial direction. By assuming that nitric ac id penetrates at a constant rate from the outer surface of the composite to the interi or and is able to dissolve MgO 144

PAGE 145

completely at the penetrati on depth, the penetration rate p (or the rate of displacement of the interface), can then be calculated in this formula: 50 R M p (6-6) where p is the penetration rate, M is the molecular mass of MgO, is the density of MgO and R is the NDR of MgO. For the same test conducted at 60 C, the penetration rate was calculated to be 40.3 m/h. The dissolution curve was then calculated based on this model and is plotted as a dashed line shown in Figure 6-1 6.3.1.2 Effect of poro sity and MgO content Figure 6-2 (A) compares the initial dissolution for the composites with densities varying from 88.2% to 95.5%. It was found that, as expected, the NDR of MgO increased with sample porosity. The surface area of MgO increases due to the presence of open porosity, and the closed por osity may become open porosity when the adjacent MgO grains are dissolved. The pores also provide additional tr ansport channels for the migration of protons and Mg 2+ ions, so the penetration rate is expected to be increased with porosity. Figure 6-2 (B) compares the initial dissolution curv es for the composites with the same density but different MgO volume fraction from 40% to 70%. It wa s found that the NDR of MgO increased with the volume fraction of MgO. For a surfac e reaction controlled process, the dissolution rate was expected to be proportional to the surface area. The geometric area was just the initial surface ar ea, but the true surfac e area of MgO during dissolution was unknown. It is anticipated t hat the dissolution reaction proceeds along connected MgO grains. So the c ontiguity of MgO, which is defined as the fr action of the interface area of MgO that is shared by MgO, is an import ant determining factor for the 145

PAGE 146

surface area of MgO during dissolution. 156 As discussed previously in Chapter 5, the contiguity of MgO in the composites c an be quantified using st ereology described in Underwoods book, 156 and was found to increase from 0.14 to 0.42 as the volume fraction of MgO increases from 40% to 70%. Therefore, higher c ontiguity of MgO may result in a larger surface area of MgO during dissolution and correspondingly a higher NDR of MgO, which is in agreement with the experiment. 01234 0 10 20 30 40 Dissolved Mg2+ (mol/m2) 88.2% 92.4% 95.5% Dissolution Time (h)NDR = 3.61 0.04 mol/m2 h NDR = 12.41 0.49 NDR = 7.03 0.18 A 012345 0 10 20 30 40 Dissolved Mg2+ (mol/m2) 70 vol% MgO 60 vol% MgO 40 vol% MgO Dissolution Time (h)NDR = 12.41 0.49 mol/m2 h NDR = 2.10 0.04 NDR = 4.45 0.05 B Figure 6-2. Dissolution of MgO-Nd 2 Zr 2 O 7 composites in HNO 3 A) Effect of sample porosity on NDR of MgO, and B) effect of MgO volume fraction on NDR of MgO. 146

PAGE 147

6.3.1.3 Characterization of the composites before and after dissolution The microstructural evolution of the compos ites tested with magnet ic bar stirring was recorded and shown in Figure 6-3 (A)-(C). Figure 6-3 (A) shows microstructure of a sintered composite with 60 vol% of MgO. The dark grains are MgO and the light grains are Nd 2 Zr 2 O 7 Figure 6-3 (B) shows a porous matrix whic h developed during dissolution. The developed porosity was a result of dissolution of MgO grains, and the interconnected pores served as transport channels for HNO 3 to continue digesting the MgO grains inside the matrix. Further ma gnetic bar stirring resu lted in a complete disintegration of th e porous matrix due to the cons tant mechanical agitation. Figure 6-3 (C) shows the morphology of the dissolution residue collected at the end of the test, which consists of single grains and grain cl usters. The large grain clusters are shown as an inset at the upper right corner. The particle size and size distribution of the residual powder are shown in Figure 6-3 (D). The dissolution residue had a broad size distribution primarily ranging from 1 to 100 m. The main peak between 1 and 10 m corresponds to the size of single grains and small grain clusters. The few minor peaks between 10 and 100 m indicate that large grain clusters with size above 10 m were also present in the residue. The results are consistent with the SEM observation in general. Even though HNO 3 alone cannot completely dissolve the composite, disintegration of the undissolved porous matrix is a benef it for dissolution because the surface area of dissolvable phases is expect ed to increase, which leads to a higher dissolution rate and dissolved fraction. 147

PAGE 148

10 m 20 m A B 30 nm 30 m 30 m 0.1110100 0 1 2 3 4 5 Volume %Particle Diameter (m) C D Figure 6-3. Structural characterization. A) the compos ite with 60 vol% MgO, B) the porous matrix developed during dissoluti on, C) the dissolution residue, and D) particle size and size distri bution of the dissolution residue. Figure 6-4 shows the XRD profiles fo r the surface of the compos ite, the surfaces of the porous matrix, and the dissolution residue. By comparing these X RD profiles, it was concluded that the pyrochlore Nd 2 Zr 2 O 7 remained intact during the whole dissolution, and all MgO has been dissolved. The results confirm a selective dissolution of MgO, and there was no phase transformation for pyrochlore Nd 2 Zr 2 O 7 after exposure to 11 M HNO 3 at 60 C. 148

PAGE 149

10203040506070 (220) (200) (110)Residual PowderNd2Zr2O7(311) (220) (400) (331) (511) (440) (531) (622) (444) (551) (711) Dissolved Surface MgO-Nd2Zr2O7Pellet Intensity (Arb. Units)2Degrees)MgO(111) Figure 6-4. X-ray Diffraction profile of the sint ered composite, surface of the porous matrix, and the residual powder collected fr om the flask after dissolution test. 6.3.1.4 Static and ul trasonic dissolution The composites tested here were made from the same batch as those tested in the dynamic dissolution. Figure 6-5 (A) shows the results of t he static dissolution test conducted at 60 C in 11 M HNO 3 It was found that only MgO can be dissolved, and the dissolution behavior was similar to the dy namic dissolution. Without agitation, the porous pyrochlore matrix remained intact des pite the dissolution of the MgO phase. The NDR of MgO was calc ulated to be 1.50 mol/m 2 h, which is less than half of the NDR obtained from the dynamic diss olution, suggesting the mass tr ansfer rate was lower in 149

PAGE 150

the static dissolution, and t he dissolution process was limit ed by film diffusion. The calculated penetration rate was 16.0 m/h. The dissolution of MgO was not completed at the end of test, and t he dissolved fraction reached about 90% and then leveled off. This could be a result of isolated MgO grai ns that have minimum connectivity to other MgO grains. Dissolution of these MgO grains was limited by diffus ion of reactants and products through Nd 2 Zr 2 O 7 grain boundaries, which was a much slower process compared with the dissolution reaction. By contrast, magnetic bar stirring provided additional mechanical force to disintegrate the porous matr ix and break the clusters apart, resulting in a completed dissolution of MgO. 0102030405060 0 10 20 30 40 50 0.0 0.2 0.4 0.6 0.8 1.0 010203040506070 0 10 20 30 40 50 0.0 0.2 0.4 0.6 0.8 1.0 Mg2+ Mg2+ Nd3+ Zr4+ PenetrationDissolved Mg2+ (mol/m2)Dissolution Time (h)Fraction of MgO DissolvedNDR=1.50 0.03mol/m2 h p = 16.0 0.3 m/h Static Nd3+ Zr4+ PenetrationDissolved Ions (mol/m2)Dissolution Time (h) Fraction of MgO DissolvedNDR=1.80 0.02mol/m2 h p = 20.1 0.2 m/h Ultrasound applied A B Figure 6-5. Dissoluti on of the MgO-Nd 2 Zr 2 O 7 composites 11 M HNO 3 at 60 C. A) Static dissolution and B) ultrasonic dissolution. It has been found that ultrasound can accele rate dissolution of metal oxides in aqueous solutions and trigger reactions t hat do not occur at normal conditions. 178,179 Studies show there are two main effects: one is mechanical effect that can increase mass transfer rate, surface area, and driving forc e; the other one is chemical effect that initiate reactions caused by formed reactive free radicals. 173,180,181 The preliminary 150

PAGE 151

ultrasonic dissolution test was an attempt to dissolve Nd 2 Zr 2 O 7 and further increase NDR of MgO in HNO 3 The ultrasonic power was measured and the value was 10.3 1.0 W/s. The ultrasound wa s applied for the first 8 h and then turned off, but the dissolution test continued at a static condition. Figure 6-5 (B) shows the results of the ultrasonic dissolution conducted at 60 C in 11 M HNO 3 The obtained NDR of MgO was 1.8 mol/m 2 h, which is comparable to the static dissolution but much lower than the dynamic dissolution. Nd 2 Zr 2 O 7 did not dissolve, indicating that ultrasound had no chemical effect on dissolving Nd 2 Zr 2 O 7 at this reaction condition. The inefficiency of ultrasonic dissolution of MgO was probably due to the large sample size compared with cavitation bubbles. The microstreaming effect is efficient when the size of reactant is comparable to the cavitation bubbles, which are typically in the micron range (at 20 kHz, the critical size is ~170 m in diameter). 182 Asymmetric implosion of cavitation bubbles on a large surface produces mi crojets that erode th e surface and create active sites, which can also enhance the dissolution rate. 173 Nevertheless, Mg O underwent dramatic surface reconstruction immediately when exposed to aqueous solutions, and the active sites might be already satura ted before ultrasound is applied. 183 It is worth noting that the ultrasonic effects depends on the ambient conditions of the reaction system. 173 More reaction conditions such as variant reac tion temperatures, diffe rent sample sizes, and whether agitation is applied, should be test ed to fully evaluate the ultrasonic effects on dissolution of the composites in HNO 3 6.3.2 Dissolution the MgO-Nd 2 Zr 2 O 7 Composites in H 2 SO 4 The dissolution of MgO-Nd 2 Zr 2 O 7 composite in H 2 SO 4 was first conducted in 7.9 M H 2 SO 4 at 60C with magnetic bar stirring in the el ectroceramics processing lab at the 151

PAGE 152

University of Florida, and the results are plotted in Figure 6-6 (A). The results show similar dissolution behavior and comparable NDR of MgO as to the dynamic dissolution in HNO 3 0102030405060 0 10 20 30 40 50 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Penetration Mg2+ Nd3+ Zr4+Dissolved Mg2+ (mol/m2)Dissolution Time (h)Fraction of MgO Dissolved 0246810 0 10 20 30 Mg2+ Nd3+ Zr4+ R = 4.22 0.04 A 04080120160200 0 5 10 15 First Order Kinetic Fit First Order Kinetic Fit Mg2+ Nd3+ Zr4+Ion Concentration (mmol/L) Dissolution Time (h)Ceq= 0.0133 0.0003 k = 0.0259 0.0014 Ceq= 0.0041 0.0003 k = 0.0220 0.0039 B Figure 6-6. Dissolution of the composites in H 2 SO 4 A) 70 vol% of MgO composite in 7.9 M H 2 SO 4 at 60 C, and B) 50 vol% of MgO composite in the boiling concentrated H 2 SO 4 152

PAGE 153

In an attempt to dissolve pyrochlore Nd 2 Zr 2 O 7 the dissolution tests were carried out in boiling concentrated H 2 SO 4 and the results are plotted in Figure 6-6 (B). The y-axis shows the concentration of the three ions in so lution. The obtained results indicate that both MgO and Nd 2 Zr 2 O 7 can be dissolved in boiling concentrated H 2 SO 4 The ability to dissolve Nd 2 Zr 2 O 7 can be attributed to the pres ence of strong nucleophile (SO 4 ) 2in concentrated H 2 SO 4 The results indicate that the nucleophilic attack by (SO 4 ) 2on the cation sites in pyrochlore Nd 2 Zr 2 O 7 is effective and also efficient at the bo iling point of H 2 SO 4 It is worth noting t hat dissolution of MgO-Nd 2 Zr 2 O 7 composites in concentrated H 2 SO 4 at room temperature didnt lead to a noticeable dissolution of Nd 2 Zr 2 O 7 which was probably hindered by kinetics factors. Therefore, the concentration of H 2 SO 4 and reaction temperature bot h play important roles in dissolving Nd 2 Zr 2 O 7 The dissolution of Nd 3+ and Mg 2+ reached equilibrium and neither of them was completely dissolved, indicating their solubility in concentrated H 2 SO 4 was limited. The fraction of dissolved Mg 2+ and Nd 3+ were calculated to be 72.9% and 91.8% respectively for a 0.4 g composite with 50 vol% of MgO in 200 ml concentrated H 2 SO 4 at its boiling temperature around 338C. The dissolution of Mg 2+ and Nd 3+ followed first order kinetics and the data points were fitted using first order equation as below: ))exp(1(kt CCeq (6-7) where C is the concentration of ions in solution, t is dissolution time, C eq is the equilibrium concentration, and k is the reaction constant indi cating how fast the reaction reaches equilibrium. The reaction constant k and equilibrium constant C eq were obtained from the fit at the boiling point of H 2 SO 4 and are shown in Figure 6-6 (B). The reaction constant for Nd 3+ and Mg 2+ were calculated to be 0.0220 0.0039 h -1 and 153

PAGE 154

0.0259 0.0014 h -1 and the equilibrium concentrations of Nd 3+ and Mg 2+ in solution were determined to be 0.0041 0.0003 mol/ L and 0.0133 0.0003 mol/L. It was observed that the diss olution rate of Nd 3+ was nearly twice as that of Zr 4+ at the first 20 h, suggesting that Nd 2 Zr 2 O 7 dissolved non-stoichiometrically in boiling concentrated H 2 SO 4 and it was incongruent dissolution. Moreover, dissolution of Zr did not follow the same kinetics. The dissolved Zr was firs t suspended as colloids in solution through the dissolution of Nd 3+ from the pyrochlore. As the co lloids polymerized they precipitated out of solution leaving an insoluble compound at the bottom of the flask. This was due to the low solubility of the Zr in concentrated H 2 SO 4 By constantly removing dissolution products from the solution, the dissolution reaction can be maintained until desired dissolution fraction is achieved. Therefor e, it is possible to dissolve the whole composite in boiling concentrated H 2 SO 4 thus eliminating the undesirable use of hydrofluoric acid (HF). Nevertheless, additi onal steps are required to transfer materials from H 2 SO 4 to HNO 3 which may increase the complexi ty and cost of the process. 6.3.3 Dissolution of Mg 2 SnO 4 in HNO 3 In order to study the aqueous dissolution behavior of Mg 2 SnO 4 dissolution tests were conducted for Mg 2 SnO 4 pellets in 11 M HNO 3 at 60 C. Figure 6-7 shows the results of one dissolution test. As shown in the figure, Mg 2+ could be leached out and completely dissolved in HNO 3 but the solubility of Sn 4+ ions was very limited. The molarity of Sn 4+ ions in solution at the end of test was less than 2 10 -3 mol/L. 154

PAGE 155

0102030405060 0 20 40 60 80 100 Sn Dissolution Fraction (%)Dissolution Time (h) Mg Figure 6-7. Dissolution of Mg 2 SnO 4 in 11 M HNO 3 at 60 C with magnetic bar stirring. The microstructure evolution of Mg 2 SnO 4 after exposure to HNO 3 was characterized by SEM and two SEM images are shown in Figure 6-8 (A)-(B). Figure 6-8 (A) shows the morphology of a partially dissolved surface wher e some of the grains were not attacked. EDS analysis was performed on the white prec ipitates, and the semi -quantitative results indicate that they were Mg-depl eted regions but rich in Sn. Figure 6-8 (B) shows an SEM image of a completely dissolved surface, where EDS was performed and the spectra is shown in Figure 6-8 (C). Except oxygen and carbon, which ca me from the conductive coating, only Sn was detected and no Mg peak showed in the spectrum. The results were in agreement with the ICP measurement, and the morphology was similar to the white precipitates s hown on the partially dissolved surface. 155

PAGE 156

A B 01234 5 0 1000 2000 3000 4000 5000 MgKSnLSnL2SnLSnLSnL1 OK IntensityKeVCK C Figure 6-8. Microstructure and c hemical analysis of dissolved Mg 2 SnO 4 A) Partial dissolved surface, B) completely di ssolved surface and C) EDS spectra collected from image B. The dissolution caused porosity to in crease, and consequently lowered the mechanical strength of the pellet. Because of the constantly me chanical agitation by magnetic bar stirring, the pellet was comple tely disintegrated and became powder after ~7 days agitation in HNO 3 The residual powder was colle cted after the dissolution test, washed with DI water and dried in an oven at 120 C for 10-15 min. Figure 6-9 (A) shows an SEM picture of the residual powder collected a fter the dissolution. The 156

PAGE 157

particle size of the powder was measured by laser light scattering and the results are plotted in Figure 6-9 (B). The multiple peaks indicate that the residual powder consisted of large number of submicron particles as well as large particles that were above 100 m. In order to characterize the crystal structure of the undissolved substance, XRD was performed on the residual powder and was compared with the X RD spectra for the sintered Mg 2 SnO 4 as shown in Figure 6-9 (C). The strong background of the XRD pattern suggests that there was a cons iderable amount of gl assy non-crystalline substance formed in the residual powder. None of the spinel peaks showed on the pattern for the dissolution residual, which indica tes that the spinel crystal structure was completely destroyed by nitric acid. The new peaks were identified as the SnO 2 phase by comparing with the PDF card #41-1445, implying that SnO 2 phase was formed during the dissolution. The crystallize size was determi ned by broadening of the XRD peaks using the Scherrers formul a that is expressed below, 184 cos 9.0 B t (6-8) where t is the crystallite size (), is the wave length of the X-rays (1.54056 ), is the Braggs angle, and B is the full-width at half-maxi mum measured in radians. The calculation was performed based on the peak (211), and the calculated crystallize size is ~3 nm. The instrument broadening and broadening caused by temperature or strain effects were not considered here, so the calcul ated crystalline size was merely an estimate. The crystallinity and crystallite size can be further investigated and confirmed using high resolution TEM. 157

PAGE 158

0.11101001000 0 1 2 3 4 Volume %Particle Diameter (m) A B 10203040506070 (112) (301) (211) (101) (110)Residual powder after dissolution SnO2(#41-1445) Intensity (Arb. Units) 2(o)Mg2SnO4(#24-0723)(111) (220) (311) (222) (400) (331) (422) (511) (440) (531) (620) C Figure 6-9. Characterization of dissolution residue. A) SEM image of the dissolution residue, B) particle size distribution of the dissolution residue, C) XRD profiles of the Mg 2 SnO 4 pellet and the dissolution residue. Theoretically, SnO 2 can be formed by either lattice reaction or nucleation after dissolution. The SnO 2 particles in the nanometer ra nge detected by laser scattering shown in Figure 6-9 (B) may be a result of dissolution fo llowed by precipitation, which is similar to the decomposition of Mg 2 SnO 4 in water at 300 C. Nevertheless, one should 158

PAGE 159

not expect substantial amorphous material formed in the dissolution residue if SnO 2 is formed by dissolution and precip itation. The formation of the amorphous material was probably due to an incongruent dissolution of Mg 2 SnO 4 in HNO 3 and that the dissolution of Mg 2+ was faster than dissolution of Sn 4+ When Mg 2+ ions were leached out from their lattice sites, the spinel structure was destroyed and the undissolved Sncontaining matrix tended to form crystalline SnO 2 which is thermodynamically more stable than amorphous SnO 2 The crystalline SnO 2 has a rutile structure and a same coordination number for Sn as Mg 2 SnO 4 (CN=6), which could reconcile such transformation. Since the reaction was conducted at 60 C, it was mostly likely that the transformation was hindered by kinetics factors, and thus lar ge amount of glassy substance was left behind. T he dissolution reaction of Mg 2 SnO 4 in HNO 3 was proposed as the equation shown below: )(2)()(2)(4)(2 2 2 42aqOHsSnOaqMgaqHsSnOMg (6-9) Based on the observations discussed above, it was concluded t hat dissolution of Mg 2 SnO 4 in 11 M HNO 3 at 60C with magnetic bar stirring resulted in a selective dissolution of Mg 2+ and the undissolved substance formed SnO 2 in either an amorphous phase or a nano crystalline phase. It is worth noting that SnO 2 is difficult to dissolve in aqueous solutions and would r equire using hydrobromic acid (HBr). 185 However, HBr is a toxic et chant and thus unlikely to be implemented in the current reprocessing scheme. Neverthel ess, it is unclear right now that whether a completed dissolution of Mg 2 SnO 4 is necessary in order to fully dissolve the materials of interest such U, Pu and minor actinides. Furthe r investigation on dissolution of Pu (or 159

PAGE 160

surrogate)-bearing inert matrix is needed to determine the effects of selective leaching of Mg 2+ on dissolution of Pu. 6.4 Summary and Conclusions In order to evaluate the feas ibility of aqueous reprocessing t he potential IM materials, the dissolution behavio r of the MgO-Nd 2 Zr 2 O 7 composites and Mg 2 SnO 4 in acidic solutions was investigated. The dissolution behavi or of the MgO-Nd 2 Zr 2 O 7 composites was studied in HNO 3 and H 2 SO 4 at different conditions. It was shown that MgO wa s able to be dissolved in 11 M HNO 3 and 7.9 M H 2 SO 4 at 60C, but Nd 2 Zr 2 O 7 was insoluble. The NDR of MgO depended on the MgO volume fraction, sample porosity, dissolution temperature, and agitation methods. Magnetic bar stirring was an efficient agitation me thod to accelerate dissolution process and di sintegrate the undissolved Nd 2 Zr 2 O 7 porous matrix into residual powder. The NDR value was determined to be 3.6 mol/m 2 h for the dynamic dissolution, which corresponded to a penetration rate of ~40 m/h. Both MgO and pyrochlore Nd 2 Zr 2 O 7 were able to be dissolved in boiling concentrated H 2 SO 4 but the solubility of Mg 2+ Nd 3+ and Zr 4+ were limited compared within aqueous solutions. The dissolution of Mg 2+ and Nd 3+ followed first order kinetics, but Zr 4+ precipitated out due to low solubility in concentrated H 2 SO 4 The reaction constant for Nd 3+ and Mg 2+ were calculated to be 0.0220 0.0039 h -1 and 0.0259 0.0014 h -1 and the equilibrium concentrations of Nd 3+ and Mg 2+ in solution were deter mined to be 0.0041 0.0003 mol/L and 0.0133 0.0003 mol/L. By constantly removing di ssolution product, it is possible to dissolve the whole co mposite in boiling concentrated H 2 SO 4 The dissolution behavior of the MgO-Nd 2 Zr 2 O 7 composites in acidic solutions suggests that 160

PAGE 161

aqueous reprocessing is possible for this iner t matrix material, but may require using boiling concentrated H 2 SO 4 and additional steps to transfer materials from H 2 SO 4 to HNO 3 The dissolution behavior of Mg 2 SnO 4 was studied in 11 M HNO 3 at 60C. It was an incongruent dissolution with selectively leaching Mg 2+ from the matrix. Sn was not able to dissolve but formed amorphous or nano crystalline SnO 2 Other acid su ch as HBr is needed to dissolve SnO 2 but unlikely to be utilized in repr ocessing due to its toxicity and corrosiveness. The dissolution behavi or of the MgO-Nd 2 Zr 2 O 7 composites and spinel Mg 2 SnO 4 in acidic solutions suggests that aqueous repr ocessing may be feasible for the potential inert matrix materials, but additional steps such as changing acids would be needed. 161

PAGE 162

CHAPTER 7 REACTOR TESTING 7.1 Introduction As discussed in Chapter 1, the performance of nuclear fuel materials can be better assessed by doing reactor irradiation tests. Compared with other irradiation techniques, neutron irradiation in thermal reactors offers several advantages as listed below: 1. Irradiation induced damages can be achieved homogeneously in bulk specimens, making it easier to measure physical property changes such as thermal diffusivity and macroscopic swelling. 2 2. Some transmutation damages can be char acterized such as the generation of particles by ( n ) reactions due to fast neutron irradiation. 186 3. If fission or fissile nuclides are incor porated, the fission fragments damage can be characterized. 4. Since radiation induced damage is usually sensitive to damage production rate, reactor tests can provide the most representative irradiation c ondition compared to other irradiation techniques such as accelerated charged-particle irradiation using accelerators. Due to these merits of neutron ir radiation, reactor testing of the potential IM materials is highly desirable. In April 2007, the U.S. Depar tment of Energy (DOE) designated the Advanced Test Reactor (ATR) at the I daho National Laboratory (INL) as a National Scientific User Facility (NSUF), which offers great opportunities to conduct reactor testing. In an effort to evaluate irradi ation behavior of the potential IM materials in a test reactor, a proposal for conducting irradiation tests in the ATR was drafted by t he author with guidance from his advisor who served as th e principle investigator of the project at UF. The proposal was submitted by his advisor and was awarded by the ATR-NSUF. Performing irradiation tests in test reactors is a great effort, and the preparation process itself can be a long term proj ect, which may last one or a few years depending on the type 162

PAGE 163

of experiment. Involvement in the whole preparation process is a rewarding and valuable research experience. Since it is an on-going projec t, experimental result s are not the focus in this chapter. Instead, the preparation process for the irradi ation test is presented in brief and the internship activities are summarized in the end. 7.2 Objectives The main objective is to study the irradiat ion behavior of the potential IM materials in a real reactor environment. Research is going to be focused on a systemic investigation of irradiation induced structur al evolution and thermophysi cal properties changes. The irradiation tests will be conducted at two tem peratures (nominally 200 C or the available lowest temperature and ~700 C or the highest temperature) and two dose accumulations (nominally 1 dpa and 2 dpa). The effects of temperature and radiation dose on materials properties will be determined. 7.3 Description of the Process Figure 7-1 shows the flowchart of the overall expe riment process. As mentioned earlier in this chapter, this is an on-going project. From the proposal subm ission to the final characterization and discharge of irradiated samp les, the whole process takes more than two and half years. Followi ng the flowchart shown in Figure 7-1 the completed work is summarized here. The process started with pr oposal preparation. After the proposed was awarded to conduct irradiation test in t he ATR, a project schedule was developed under a joint effort with INL staff, including the ATR-NSUF in terim director Mitche ll K. Meyer, the ATR irradiation testing manager Franc es M. Marshall, the project manager Julie A Foster, the 163

PAGE 164

ATR hardware specialist Gregg W. Wachs, and t he INL principle inve stigator Pavel G. Medvedev. Figure 7-1. Roadmap for experiment irradiation. 58 164

PAGE 165

Detailed experiment specific ations were determined, su ch as neutron energy, total fluence, temperature, and heati ng rate. Based on the specified irradiation conditions and position availability of the AT R, the position B-1 was assigned for the experiment. The next important step was develop ing the irradiation test plan. In the test plan, the experiment conditions were clearly defined, such as i rradiation position, te mperature, and target neutron dose. In order to maximum the usage of the space in the ATR, the existing capsule was modified to accommodate more test conditions such as radiation temperature and dose. The conceptual hardware design was documented in t he irradiation test plan, serving as the basis for technical drawings and fabrication. He re the hardware was referred to irradiation test vehicles which include holders, capsules, baskets, and all other components for holding and positioning target ma terials. In Appendix B, the conceptual hardware design is described in detail. Based on the experime nt configuration defined in the irradiation test plan, the safety analyse s such as neutron, thermal, and structural analysis were subsequently performed. T he neutron analysis was performed by neutron analyst James R. Parry and Joseph W. Nielsen. The thermal and safety analysis was performed by the author under guida nce of the thermo hydraulic analyst Paul E. Murray. Appendix C presents a det ailed thermal analysis and safety evaluation. In the meanwhile, the supporting hardware such as specimen hol ders and capsules was fabricated at INL, and the target materials were fabricated at t he University of Flor ida by the UF student Donald T. Moore in Nino Rese arch Group (NRG). The safety document, experiment safety assurance package (ESAP), was developed by the nuclear safety analyst Tamara. E. Shokes to address safety basis requirements. Specimen loading wa s performed at INL by the UF student Donald T. Moor e, INL principle investigator Pavel G. Medvedev, hardware 165

PAGE 166

specialist Gregg W. Wachs and ot her INL technical supporting st aff. The sealed capsules were tested to ensure no leaks or mechani cal flaws, and were then loaded into an aluminum sleeve and basket, packed, trans ported to the ATR for insertion. The present and future work of this project is briefly summarized here. As-run neutron analysis will be performed to provi de the information of the actu al irradiation conditions. After the desired radiation dose is achieved, the capsules will be removed from the ATR and stored in canal unt il they can be safely handled for trans portation. Once the capsules can be safely handled, they w ill be transport to the hot ce ll and dissembled. The post irradiation examination will be performed with INL experts and the PIE report will be compiled. The activated capsules, targets and all other materials will be properly disposed. 7.4 Summary of Internship Activities Some of the internship activities and involvement at INL are listed below: 1. developing and delivering t he irradiation test plan under Dr. Medvedevs guidance 2. designing hardware with Mr. Wachs assistance 3. performing thermal analysi s using Abaqus model under Dr. Murrays guidance 4. developing and deliv ering the thermal and safety analysis report. 166

PAGE 167

CHAPTER 8 SUMMARY AND FUTURE WORK 8.1 Summary There is an increasing inventory of r adioactive nuclear waste from both spent nuclear fuel and weapon programs, such as Pu and minor acti nides. In order to reduce the current stock of Pu and ot her transuranium elements and to provide electricity, IMF has been proposed for a uranium fr ee transmutation of fissile actinides which excludes continuous U-Pu conversion in nuclear reac tors. There are seve ral requirements for candidate IM materials that include high me lting point (> 2173 K) low thermal neutron adsorption cross section (< 2.7 barns), minimum solubilit y in coolant, adequate thermal conductivity ( UO 2 ) and good radiation stability. Bas ed on literature survey, the MgONd 2 Zr 2 O 7 composites and the spinel Mg 2 SnO 4 were selected as pot ential IM materials for the present study. The selected potential IM materials were successfully fabricated through conventional solid state processing. Pyrochlore Nd 2 Zr 2 O 7 was synthesized through the conventional solid state processing and the sol-gel processing. The calcination temperature for t he sol-gel derived Nd 2 Zr 2 O 7 powder was 1200 C, which was ~150 C lower than the so lid state derived Nd 2 Zr 2 O 7 powder. The sol-gel processing can significantly reduce the main particle size from a few microns down to a few hundred nanometers, however, some particles aggregated and grew as large as microns. To form composites, three mixing methods were used which were mortar and pestle mixing, water magnetic bar stirring, and ball milling. Distinct microstructure s of the composites 167

PAGE 168

were obtained from the three mixing methods The ball milling produced a relatively homogeneous microstructure compared with the other two processing methods. The spinel phase of Mg 2 SnO 4 was achieved by calcination of ball milled MgO and SnO 2 powder mixture in the stoichiometric ratio at 1200 C for 12 h. Rietveld refinement was performed for spinel Mg 2 SnO 4 and the obtained lattice parameter ( a ) and oxygen dilation parameter ( u ) were 8.6065(4) and 0.3834(4) The electrostatic potential calculation suggests that Mg 2 SnO 4 has an ordered inverse spin el structure; however, the thermodynamic calculati ons suggest that the B type ca tions are randomly mixed and disordered. The calculated GII for Mg 2 SnO 4 was determined to be 0.15, indicating that the structure is stable, even though combined with some lattice strain. The first evaluation of the potential IM materials was i rradiation tolerance. A preliminary study was conducted to evaluate the re sistance of Mg 2 SnO 4 to radiation induced amorphization. The in situ irradiation tests were performed using IVEMTandem facility at ANL. Samples were irradiated with 1.0 MeV Kr 2+ ions at 50 K and 150 K to a maximum fluence of 5 10 19 Kr ions/m 2 and 10 20 Kr ions/m 2 respectively. Microstructure and crystal structure evolut ions were monitored and recorded by BF images and SAED patterns. The amorphization doses for Mg 2 SnO 4 irradiated by 1.0 MeV Kr 2+ ions at 50 K and 150 K were determined to be 5 10 19 Kr ions/m 2 and 10 20 Kr ions/m 2 which corresponded to an atomic di splacement of 5.5 dpa and 11.0 dpa, respectively. Thermal annealing at room temperature was efficient for Mg 2 SnO 4 to restore its crystallinit y from the amorphous phase irradi ated at 50 K. The electronic stopping power exceeded the nuclear stopping pow er except the end of range of ions, but is only ~1.5 keV/nm. This suggests that the irradiation induced amorphization by 1 168

PAGE 169

MeV Kr 2+ in Mg 2 SnO 4 was mainly due to ballistic displacement damage and defect accumulation. Mg 2 SnO 4 shows less irradiation resistance compared with MgAl 2 O 4 against atomic displacement damage, which c an be attributed to its inverse structure, higher covalency of the bond, larg er ionic size and charge difference between Mg 2+ and Sn 4+ The second evaluation in this work was water corrosion resistance. The corrosion behavior of the MgO-Nd 2 Zr 2 O 7 composites in hydrothermal conditions was studied. It was found that the corrosion resistance depended on microstructural homogeneity. The inhomogeneous composites made by mortar and pestle mixing and water magnetic bar stirring dissolved quickly in water at 300C showing no improvement over pure MgO. The homogeneous composites made by ball milling exhibited improved hydration resistance and the corrosion was primarily limited to the surface region. Due to the improved hydrat ion resistance of the homo geneous composites, the ball milling produced composites were further in vestigated. The microstructure of the composites made by ball milling was analyzed in terms of MgO and Nd 2 Zr 2 O 7 grain sizes, MgO contiguity and homogeneity. It was found the contigui ty of the MgO phase in the composites increased from 0.14 to 0.42 with t he volume fraction of MgO increasing from 40% to 70%. The dispersi on of MgO became less homogeneous as the MgO volume fraction increased from 40 vol% to 70 vol% probably due to coarsening of MgO grains. The average grain size of MgO increased with the volume fraction of MgO while the average grain size of Nd 2 Zr 2 O 7 remained relati vely constant. The mass loss of MgO-Nd 2 Zr 2 O 7 composites was due to hydration of MgO. A linear relationship was found between the NML and the corrosion time. The NML rate 169

PAGE 170

increased with MgO volume fraction which c an be attributed to the contiguity of MgO and hydration induced surface destruction. The NML rate also increased with corrosion temperature from 150 to 300 C and it followed an Arrhenius re lationship. The activation energy was calculated to be ~39-41 kJ/mol. It can be inferred that a desired microstructure with enhanced corrosion resist ance should have minimum contiguity of MgO, good homogeneity, and large MgO grain size (minimizing grain boundaries). Mg 2 SnO 4 was not stable in water at 300 C and saturation pressure. Both Mg 2+ and Sn 4+ were leached out from the matr ix and precipitate out as Mg(OH) 2 and SnO 2 However, the dissolution of Mg 2 SnO 4 in water at 300 C was a slow process and probably hindered by the formed hydration products on the corroded surface. As a result, the corrosion was limited only to t he sample surface, and the mass and volume changes were less than 1% up to 30 days. The results indicate that single phase Mg 2 SnO 4 exhibited much higher hydration resistance than the MgO-Nd 2 Zr 2 O 7 composites. The third evaluation was aqueou s reprocessing feasibility. The dissolution behavior of the MgO-Nd 2 Zr 2 O 7 composites was studied in HNO 3 and H 2 SO 4 at different conditions. It was found that MgO was able to be dissolved in 11 M HNO 3 and 7.9 M H 2 SO 4 at 60C, but Nd 2 Zr 2 O 7 was insoluble. The NDR of MgO depended on the MgO volume fraction, sample porosity, dissolution temperature, and agitation methods. Magnetic bar stirring was effective to accele rate dissolution process and disintegrate the undissolved Nd 2 Zr 2 O 7 porous matrix into residual pow der. A completed dissolution of MgO was achieved with magnetic bar stirring. It could be inferred that soluble species in the composites such as U and Pu are able to be fully dissolved in agitated HNO 3 in a 170

PAGE 171

similar way as to MgO. Both MgO and pyrochlore Nd 2 Zr 2 O 7 were able to dissolve in boiling concentrated H 2 SO 4 but the solubility of Mg 2+ Nd 3+ and Zr 4+ were limited compared within aqueous solutions. The dissolution of Mg 2+ and Nd 3+ followed first order kinetics, but Zr 4+ precipitated out due to low solubility in concentrated H 2 SO 4 By constantly removing dissolution products, it is possible to dissolve the whole composite in boiling concentrated H 2 SO 4 The dissolution behavior of the MgO-Nd 2 Zr 2 O 7 composites in acidic solutions suggests t hat aqueous reprocessing is possible for this inert matrix material, but may re quire using boiling concentrated H 2 SO 4 and additional steps to transfer materials from H 2 SO 4 to HNO 3 The dissolution behavior of Mg 2 SnO 4 was studied in 11 M HNO 3 at 60C. It was found that it was an incongruent disso lution with selectively leaching Mg 2+ from the matrix. Sn was not able to dissolve in HNO 3 but formed amorphous or nano crystalline SnO 2 Dissolution of SnO 2 requires using HBr which is unlikely to be utilized in reprocessing due to its toxicity and co rrosiveness. Similar to the MgO-Nd 2 Zr 2 O 7 composites, the dissolution behavior of spinel Mg 2 SnO 4 in HNO 3 suggests that aqueous reprocessing may be feasible, but additional steps such as changing acids would be needed. Final evaluation will be performed after irra diating the potential IM materials in the ATR. Target materials were inserted into the reactor and will be irradiated to 1 and 2 dpa at ~350 and ~700 C, respectively. The PIE will start in 2010, and will be performed by a different NRG member focusing on radi ation induced modifications of the potential IM materials, such as thermophysical prope rties changes, structural and microstructural evolution, and volumetric swelling. It is envisioned that a be tter understanding of the 171

PAGE 172

irradiation behavior of the potent ial IM materials can be achieved by performing reactor irradiation tests. 8.2 Future work 8.2.1 Processing and Wa ter Corrosion Resistance It was observed that the hydrothermal corrosi on resistance of MgO-Nd 2 Zr 2 O 7 composites was microstructural dependent. The corrosion resistance can be further improved by minimizing MgO aggregates, r educing MgO contiguity, and enlarging MgO grain size. Ball milling is able to produce homogeneous microstructure but it limits the grain size of MgO due to particle size reduct ion by mechanical grinding. Sintering at a higher temperature (i.e. 1700 C) and extending sintering time may lead to grain growth and result in large grain size. One way to produce a homogeneous microstructure with minimum contiguity of MgO has been described previously in Chapter 5, which is to compact and sinter Nd 2 Zr 2 O 7 -coated MgO particles. The desired microstructure was not able to be produced through the coprecipit ation and gel-casting methods. Chemical vapor deposition (CVD) may be used to produce the Nd 2 Zr 2 O 7 -coated MgO particles and achieved completed coating, but the co st of production increases and the coating thickness may be limited due to the slow production rate. 8.2.2 Radiati on Tolerance Due to the high operating temperature of nuclear fuels (average ~700-900 C, but center peak value may reach 1200 C or even higher), effect of dynamic thermal annealing at this temperatur e range becomes important. Even though amorphization may not occur, materials could undergo significantly structur al evolution. They may contain large number of defects, form nano crystalline aggregates 187 or even 172

PAGE 173

decompose into other phases 13 The undesirable behavior coul d result in significant swelling and catastrophic material failure. Therefore, high temperat ure irradiation tests should be performed for the potential IM materi als. Moreover, the irradiation behavior of Mg 2 SnO 4 under electronic stopping regime is still unknown. To further evaluate the irradiation stability of Mg 2 SnO 4 irradiation tests should be conducted using swift heavy ions to simulate fission fragment effects. A side-by-side comparison should be made between the MgAl 2 O 4 and Mg 2 SnO 4 The transmutation damage caused by neutron irradiation should also be characterized. As a step forward, a systemat ic study on radiation tolerance of spinel materials should be conducted. Series of spi nel compounds should be investigated and contrasted in order to fully understand th e irradiation damage mechanisms in spinel compounds especially in the electronic stopping regime. The potential spinel compounds described in the materi al selection section in Chapt er 3 should be the focus. An in-pile irradiation te sting is much desirable. 8.2.3 Aqueous Reprocessing It was shown that dissolution in HNO 3 alone resulted in a se lective dissolution for both the MgO-Nd 2 Zr 2 O 7 composites and spinel Mg 2 SnO 4 Even though it is possible to completely dissolve the two potential IM mate rials in other acids such as concentrated H 2 SO 4 at boiling and HBr, changing acids r equire additional steps associated with increased processing cost, but may not be necessary for reprocessing. Providing adequate mechanical agitation, other soluble materials of interest may dissolve completely similar to Mg 2+ in a reasonable time frame. Ho wever, in order to validate it, Pu-bearing or U-bearing (or surrogate) IM materials should be fabricated and the 173

PAGE 174

dissolution behavior in HNO 3 should be characterized. To simulate fission fragments, the IM materials can be doped with Cs, Sr or other fission produc ts of interest and tested for dissolution. If breaking down t he insoluble materials into powder is not effective to fully dissolve fissile material s or interested fission products, then changing acids becomes a necessary step for aqueous reprocessing the IMFs. 8.2.4 Reactor Testing and Post Irradiation Examination (PIE) The PIE for the irradiated potential IM materi als taken out from th e ATR will start in 2010. The main focus in the PIE is to in vestigate radiation induced thermophysical properties changes and structural evolution. The macroscopic response of the materials to neutron irradiation such as volumetric changes will be quantified and correlated to microscopic features observed in electron microscopes such as defects and defect clusters. Effects of radiation dose and temperature on the radiation damage in the potential IM materials will be evaluated. 8.2.5 Other In-service Engineering Para meters for Potential IM Materials Irradiation tolerance, water corrosion resistance, and aqueous dissolution capability are the three main in-servi ce engineering paramet ers that have been characterized for the potential IM materials in the present study. As menti oned above, the thermal conductivity of these materials after irradiati on will be characterized in the PIE. It is worth noting that several other in-ser vice engineering param eters have not been considered here, but they are equally important, such as the chemical compatibility with cladding, in-pile irradiation behavior (fissile materials contained experiment), chemical stability with fission products, and fission gas so lubility and retention. The exact melting temperature for Mg 2 SnO 4 should be determined. The safe ty margin for the potential 174

PAGE 175

IMFs should be evaluated using finite element software such as Abaqus to predict the fuel temperatures at nor mal and transient conditions based on the determined thermophysical properties (melting point, specific heat, and thermal conductivity). 175

PAGE 176

APPENDIX A NEUTRONIC PROPERTIES OF POTENTIAL IM MATERIALS A.1 Introduction The effect of the additives absorption cro ss section on the reactivity of the fuel element needs to be examined for IM mate rials. These calcul ations establish the acceptability of a material fr om a nuclear reactivity standpo int. The calculation can be performed using the computer codes, such as CASMO-3 code, MCNP code, ORIGIN 2 code, and MONTEBURNS code. The CASMO-3 Code is a fuel assembly burnup program develop ed by Studsvik of America Inc. It is a mu lti-group, two-dimensional transport theory code for burnup calculations on BWR and PWR assemblies or simple pin cells. The code handles a geometry consisting of cylindric al fuel rods of varying composition in a square pitch array with allowance for fuel rods l oaded with gadolinium, bur nable absorber rods, cluster control rods, in-core instrument channels, water gaps, boron steel curtains and cruciform control rods in the regions separating fuel assemblies. The MCNP stands for M onte Carlo N-Particle Transport. It is a particle transport code for neutrons and photons created by Los Alamos National Laboratory and distributed by the Radiation Safety Informat ion Computational Cent er (RSICC) in Oak Ridge, Tennessee. It uses the Monte Carlo technique to generate a statistical history for a particle based on random samples from probability distributions It can be used as a stand alone code for single step modeling to get a point in time analysis of flux, energy deposition, or critica lity. It can also be used in conjunction with ORIGEN 2 inside MONTEBURNS in depletion calculations. 176

PAGE 177

The ORIGEN 2 code is another RSICC c ode package that computes decay and isotope depletion information for irradiated materials. The code considers timedependent formulation, destructi on, and decay concurrently. The input file for ORIGEN 2 should include the following information: (1) chemical composition and amount of material, (2) one-group microscopic cross-sections for each isotope, (3) material feed and removal rates, (4) length of irradiation period, and (5) the flux or power of the irradiation. The MONTEBURNS code is a RSICC code t hat couples MCNP and ORIGEN 2 to generate depletion and burnup calc ulations. Users have to generate three files to complete the depletion model, including a M CNP input file, a MONTEBURNS input file, and a feed rate input file. A.2 Neutronic Prope rties of MgO-Nd 2 Zr 2 O 7 Composite and Mg 2 SnO 4 Monteburns and MCNP were used for this calculation performed by Jiwei Wang. There were 50 burnup steps in Monteburns calculation for 2000 days total (About 60 MWD/KG equivalent burnup). A hundred active cycles were used in MCNP calculation, and one thousand histories per cycle. Same volume percent (8%) weapon grade PuO 2 ( 239 Pu: 240 Pu: 241 Pu = 93:6.5:0.5) was used, which means that the amount of PuO 2 was the same in all kinds of fuels. In this way, the amount of PuO 2 in 1 cm 3 fuel was equal to the volume fraction of PuO 2 multiplied by its density or (1cm 3 ) x (8 %) x (11.04 g/cm 3 ) = 0.8832 g. The neutronic calculation shows that Mg 2 SnO 4 has a similar behavior to MgAl 2 O 4 and the composite with 70 vol% of MgO c an provide enough reactivity to sustain the reaction at the end of life (EOL). 177

PAGE 178

0102030405060 0102030405060 0.8 1.0 1.2 1.4 1.6 0.8 1.0 1.2 1.4 1.6 keffEquivalent Burn Up (MWD/KG) MOX MgO-Nd2Zr2O7 (70/30 by vol.) MgAl2O4 Mg2SnO4 Figure A8-11. Neutron multiplication factor as a function of equivalent burn up. 178

PAGE 179

APPENDIX B CONCEPTUAL HARDWARE DESIGN FOR REACTOR TESTING B.1 Introduction This appendix is based on the INL internal report (INL-PLN -2874) titled Irradiation Test Plan for the ATR NSUF University of Fl orida Experiment. The existing hardware design for B-1 position consists of a single 18 in. long capsule. In order to maximize the usage of the space in the reactor and obtain more results for anal ysis, the existing design was modified so that samples of di fferent geometries can be loaded and different irradiation conditions such as neutron fluence and radiation te mperature can be achieved simultaneously. B.2 Conceptual Design The experiment capsule assembly designed for the B-1 position utilized fixtures and internal TEM sample holders. The original single capsule design was modified so that two combined capsules with equal length are stacked and the l ength of the overall capsule stack matches that of the original capsule. The fi xture assemblies were used to position the samples and prevent from contact with the capsule to stabilize the temperature. The design of the sample holder for the TEM samples is shown schematically in Figure B-1. TEM samples were loaded into the sample holder and the sample holder was inserted into the fixture loaded into the capsule. The experiment capsule was vacuumed and filled with He/Ar gas, and the final end plug was welded to assure a leak tight environment. The helium/argon mixture was adjusted during experiment fabrication and t he thermal insulating discs were used to achieve desired irradiation temperatures. 179

PAGE 180

Figure B8-21. Conceptual design of the Transmissi on Electron Microscopy (TEM) sample holder. A total of 3 capsule assemblies were us ed for the UF experiment, which included two low-dose capsules (capsule A and caps ule B) and one high-dose capsules (capsule C). The two low-dose capsules have the same outer dimensi ons and will be exchanged after approximately half of the irradiation durat ion so that they both attain approximately 1 dpa while the high-dose capsule attains 2 dpa. A schematic of the capsule configuration is s hown in Figure B-2. Peak axial fluence plane Capsule A or B Tube C2 (Samples) Tube C1 (Samples) Tube A1 or B1 (Dummy) Tube A2 or B2 (Samples) Capsule C 9.125 in. Al Spacer Stack of diffusivity samples TEM samples with holder End Plug Thermal insulator DWG. No. 426469 Capsule A or B Tube A1 or B1 (Dummy) Capsule connecting end plug 0.288 Tube A2 or B2 (Samples) 9.125 in. Tube C2 Tube C1 4.5 in. 4.5 in. 0.288Diffusivity samples TEM samplesFixture Capsule connecting end plug Capsule C Diffusivity samples TEM samples A B C Figure B8-32. Capsule assembly. A) configuration of stacked capsules, B) Capsule A or B with the dummy tube, and C) Capsule C. 180

PAGE 181

APPENDIX C THERMAL AND SAFETY ANALYSIS FOR REACTOR TESTING C.1 Introduction This appendix is based on the INL inter nal report (INL-ECAR455) titled Thermal Analysis of ATR-NSUF University of Flor ida Experiment. The analysis was performed by the author under guidance of the thermal-hydraulic anal yst Paul E. Murray. The purpose of the analysis was to demonstrat e compliance with the thermal safety requirements and to determine the temperat ure of the specimens during reactor operation at nominal cycle power. The evaluated thermal load cases include (1) steady-state reactor operation at nominal cycle power; (2) steady-state calcul ation for pump coast-down conditions at 25% over-power, outer shim control cylinders at 155 and 25% reduced coolant flow rate; (3) a condition 4 r eactivity insertion accident (RIA) tr ansient at initial conditions of 25% over-power and outer shim control cylinders at 155; (4) steady-state heat transfer in air after the experiment is removed from the reactor 5 h after shutdown. The pump coast-down transient is a reactor accident caused by loss of the commercial to the recirculation pumps in the r eactor. In the pump coast-dow n transient, the coolant flow rate decreases with time and the reactor is shut down in ~1 s after losing the commercial power. The transient conditi on was simplified into a steady-state calculation that bounds the safety limit. The power transient in condition 4 RIA is shown in Figure C-1. 181

PAGE 182

Figure C8-41. Reactor power transient in condition 4 Reactivity Insertion Accident (RIA). C.2 Experimental Procedure A finite element, steady-state and transient heat transfer analysi s of the capsule assembly, cooling water channels, and adjacent reflector was performed using ABAQUS version 6.7-3 on an SGI Altix ICE 8200 distributed memory cluster. The 8node linear brick element was used to model the capsules, TEM holders and specimens. The 4-node linear shell element was used to model the fixtures, sleeve, basket, and reflector. The 8-node forced convection br ick element was used to model the cooling water with a prescribed mass flow rate in t he axial direction. He at transfer across gas gaps was modeled using the gap conductance model s. Contact surfac es for He gas or He-Ar gas gap conduction included the inside surface of capsule, inside and outside surfaces of fixture and TEM holder, and outside surface of specimen. Thermal contact between specimens, inside and outside surfaces of fixture bottom, and inside surface of bottom end plug of capsule was modeled using a tie constraint. Heat transfer to the primary cooling water was modeled usin g the gap conductance model. Contact surfaces for forced convection include the outsi de surface of capsule, inside and outside 182

PAGE 183

surfaces of sleeve, inside and outside surfac es of basket, inside surface of reflector, and water. An example is given in Figure C-2, showing the finite element mesh of tube 2 in capsule C assembly. The parts are colo r coded for identification: capsule is yellow, fixture is blue, TEM sample holder is red, s pecimens are gray, thermal insulator is gray. Thermal insulator Diffusivity specimens TEM specimens Fixture TEM sample holder Capsule C-tube 2 Figure C8-52. Finite element mesh of Capsule C assembly. C.3 Results and Discussion C.3.1 Steady-State Operati on at Nominal Cycle Power Contour plots of the temper ature of capsule assembly A, B and C are shown in Figure C-3(A)-(C). The plots include fixtures, TEM sample holders and stainless steel spacers (only in capsule A and B) Temperature units are shown in Kelvin in the plots. Tube 1 (A1) and Tube 2 (A2) in capsule A, T ube 1 (B1) in Capsule B and Tube 2 (C2) in Capsule C contained 100% pure He gas. T he desired specimen tem perature (473 K or 183

PAGE 184

973 K) is indicated in the figures. Tube 2 (B 2) in Capsule B and Tube 1 (C1) in Capsule C contained a gas mixture of 15% He and 85% Ar. Capsule A, Tube 2 (473 K) Capsule A, Tube 1 (Dummy) TEM holder Fixture Stainless Steel Spacer Capsule A, Tube 2 (473 K) Capsule A, Tube 1 (Dummy) TEM holder Fixture Stainless Steel Spacer A Capsule B, Tube 2 (473 K) Capsule B, Tube 1 (Dummy) TEM holder Fixture Stainless Steel Spacer Capsule B, Tube 2 (473 K) Capsule B, Tube 1 (Dummy) TEM holder Fixture Stainless Steel Spacer B Figure C8-63. Temperature contour plots for A) C apsule A, B) Capsule B and C) Capsule C. Temperature is shown in Kelvin. 184

PAGE 185

Capsule C, Tube 2 (473 K) Capsule C, Tube 1 (973 K) TEM holder Fixture C Figure C-3. Continued. Peak capsule temperature wa s observed at the end plugs which contacted fixtures or thermal insulators, and is shown in Figur e C-4 for high temperature Tubes B2 and C1. A B Figure C8-74. Temperature of bottom end pl ug. A) Tube B2 and B) Tube C1. 185

PAGE 186

The minimum and maximum temperatures of the specimens are shown in Figure C5. The experiment aimed to contro l the specimen temperature at 700 and 200 C (or as low as possible) in different capsule tubes. The temperature of the specimens in the high temperature capsule tubes were wi thin an acceptable variation of C of nominal high temperature (~700 C), and the temperat ure of the specim ens in the low temperature capsule tubes can be as low as ~300 C with a variation of C. The obtained results indicate target temperatures were able to be achieved by the current design and gas mixture. TEM specimens had relatively uniform temperatures which were higher than the diffusivity disks fo r the same type of samples, because the stainless steel TEM holders had a great er mass and a higher heating load than specimens. 0 200 400 600 800 A2 Disk Minimum 0 200 400 600 800 A2 Disk Minimum Maximum Temperature (oC)Mg2SnO4A2 TEM B2 Disk B2 TEM C1 Disk C1 TEM C2 Disk C2 TEM Maximum Temperature (oC)MgO-Nd2Zr2O7 (70/30 in vol.)A2 TEM B2 Disk B2 TEM C1 Disk C1 TEM C2 Disk C2 TEM A B Figure C8-85. Temperature of sele cted specimens. A) MgO-Nd 2 Zr 2 O 7 composites (70 vol% MgO), and B) Mg 2 SnO 4 C.3.2 Pump Coast-down and Reactivity Insertion Accident (RIA) A steady state calculation was carried out assuming the water flow rate was reduced by 25% for the pump coast-down accide nt. The forced convection heat transfer 186

PAGE 187

coefficient was re-calculated at this condi tion. It was found that the maximum water channel temperature and the maximum heat flux were 67.9 C (341.1 K) and 140.1 W/in 2 respectively. Among all three water channels, the capsule-sleeve water channel had the maximum temperature because it directly contacted capsule. The temperature profile of the capsule-sleeve water channel and the heat flux at the capsule outer surface are shown in Figure C-6. The model was orient ed with the reactor cooling water flowing from left to right. The maxi mum temperature rise of the water was 16.3 C compared with inlet water te mperature (fro m 51.6 to 67.9 C). The critical heat flux for forced convection was obtained from the Bernath correlati on. For the hydrodynamic conditions occurring in the cooling channel, the critical heat flux was 4015 W/in 2 The ratio of critical heat flux to maximum heat flux (DNBR) was 28.7. The cooling water saturation temperature at t he outlet pressure was 208.9 C. The difference between the saturation temperature and inlet temperature was 157.2 C. Since the water temperature rise was 16.3C, the ratio of the critical temperature rise to the actual temperature rise (FIR) was 9.7. A B Figure C8-96. Pump coast-down analysis. A) Temper ature profile of the capsule-sleeve water channel, and B) heat flux at t he pressure tube outer surface. 187

PAGE 188

Another evaluation was performed using a transient calculation for condition 4 RIA (SIPT pump discharge breach), assuming the in itial conditions were 25% over-power and outer shim control cylinders at 155 It was found that the maximum temperature of the experiment components occurred at 0.225 s after the start of t he RIA transient. A contour plot of temper ature of the Caps ule C assembly is shown in Figure C-7(A). As shown in the figure, the maximum temperature was 940C (1213 K). The peak temperature of all experim ent components increased by 158 C from 782 to 940 C compared with the nominal operation condition. Figure C-7( B) shows the hot spots in the whole assembly. The peak temperat ure in RIA was less than the melting temperature of all component s in the capsule assembly. The peak water channel temperature was 68.4 C (341.4 K) and it occu rred at 0.225 s as well. As a result, the transient power increase dur ing RIA will not post danger to the experiment and reactor safety. A B Figure C8-107. The Reactivity Insertion Accident (R IA) transient. A) Te mperature of the experiment components at 0.225 s, and B) hot spots. 188

PAGE 189

C.3.3 Passive Cooling in Air An evaluation was performed to deter mine the maximum temperature of the experiment when it is stored in air after bei ng removed from the re actor. The heat load was due to decay heat of activation products evaluated at 5 h decay time following 6 irradiation cycles (~300 days) for Capsul e C, basket, and sleeve, and 3 irradiation cycles (~150 days) for Capsule A and B. A contour plot of tem perature of the Capsule C assembly is shown in Figure C-8. The maximum steady-state temperature is 39.5C (312.5 K), which is much less than the melting temperature of the experiment components. Figure C8-118. Temperature of Capsule B and C a ssembly with sleeve and basket in air after 5 h decay. 189

PAGE 190

C.4 Summary and Conclusions The calculated results for normal operati on indicate that the high specimen temperature (~700 C) can be achieved with a gas mixture of 15% He and 85% Ar, and the low specimen temperature (~200 C, or as low as possible) can be achieved with pure He gas. Peak structural material temperature occurred at t he bottom end plugs of capsules which contacted fixt ures or thermal insulators The temperature of the specimens in the high temperat ure capsule tubes were within an acceptable variation of C of nominal high temperature (~700 C), and the temperature of the specimens in the low temperature capsule tubes can be as low as 300 C with variation of C. Results of this evaluation indicate target temperature can be ac hieved by the current design and gas mixture. The analysis results for pump coast-down to emergency flow and RIA demonstrate compliance with safety requirements. The DNB R (departure fr om nucleate boiling ratio) was 28.7 and the FIR (f low instability ratio) was 9.7, which exceeded the minimum value of 2 with a substantial ma rgin of safety. The maxi mum temperatur e of all the experiment component s during RIA was 940 C (1213 K), which was less than the melting temperature. The results indicate that the transient pow er increase during RIA will not cause danger to experiment and reacto r safety. Assuming an activation product decay heat load after 5 h decay time followi ng six irradiation cycles for Capsule C, basket, and sleeve, and three irradiation cycl es for Capsule A and Capsule B, the maximum steady-state temperature of the assembly stored in air was 39.5 C, which was much less than the melting temperat ure of the experiment components. 190

PAGE 191

LIST OF REFERENCES 1. C. Degueldre and J. M. Paratte, "Concepts for an Iner t Matrix Fuel, an Overview," J Nucl Mater 274 [1-2] 1-6 (1999). 2. W. J. Weber, R. C. Ewi ng, C. R. A. Catlow, T. D. de la Rubia, L. W. Hobbs, C. Kinoshita, H. Matzke, A. T. Motta, M. Nastasi, E. K. H. Salje, E. R. Vance and S. J. Zinkle, "Radiation Effects in Crystalline Ceramics for the Immobilization of HighLevel Nuclear Waste and Plutonium," J Mater Res 13 [6] 1434-1484 (1998). 3. C. Degueldre, U. Ka semeyer, F. Botta and G. Ledergerber, "Plutonium Incineration in Lwrs by a Once th rough Cycle with a Rock-Like Kuel," Mater Res Soc Proc, 412 15-23 (1996). 4. D. Bodansky, "Reprocessi ng Spent Nuclear Fuel," Phys Today, 59 [12] 80-81 (2006). 5. T. Wiss and H. Matzke, "H eavy Ion Induced Damage in MgAl 2 O 4 an Inert Matrix Candidate for the Transmutat ion of Minor Actinides," Radiat Meas 31 [1-6] 507514 (1999). 6. T. A. G. Wiss, P. M. G. Damen, J. P. Hier naut and C. Ronchi, "Helium and Xenon Behaviour in Irradiated Am-Containing MgAl 2 O 4 (Reactor Experiment Efttra-T4)," J Nucl Mater, 334 [1] 47-57 (2004). 7. F. C. Klaassen, K. Bakke r, R. P. C. Schram, R. K. Meulekamp, R. Conrad, J. Somers and R. J. M. Konings, "Post I rradiation Examination of Irradiated Americium Oxide and Uranium Dioxide in Magnesium Aluminate Spinel," J Nucl Mater 319 108-117 (2003). 8. Butterma.Wc and W. R. Fost er, "Zircon Stability and ZrO 2 -SiO 2 Phase Diagram," Am Mineral 52 [5-6] 880 (1967). 9. L. M. Wang and R. C. Ewing, "Ion-Beam-Induced Am orphization of Complex Ceramic Materials Minerals," MRS Bull 17 [5] 38-44 (1992). 10. S. R. P. David R. Clarke, "T hermal Barrier Coating Materials," Mater Today 8 2229 (2005). 11. P. G. Medvedev, S. M. Frank, T. P. O'Holleran and M. K. Meyer, "Dual Phase MgO-ZrO 2 Ceramics for Use in LWR Inert Matrix Fuel," J Nucl Mater 342 [1-3] 4862 (2005). 12. K. R. Mikeska, S. J. Bennison and S. L. Grise, "Corrosion of Ceramics in Aqueous Hydrofluoric Acid," J Am Ceram Soc 83 [5] 1160-1164 (2000). 13. L. M. Wang, W. L. Gong, S. X. Wang and R. C. Ewing, "Com parison of Ion-Beam Irradiation Effects in X 2 YO 4 Compounds," J Am Ceram Soc 82 [12] 3321-3329 (1999). 14. World Nuclear Association, 25th Feb 2009, ( http://www.world-nuclear.org/) 15. Japan Nuclear Fuel Limited, 25th Feb 2009, ( http://www.jnfl.co.jp/english/) 191

PAGE 192

16. N. F. M. Henry and K. Lonsdale, Internat ional Tables for XRay Crystallography, Kynoch Press, Birmingham, 1952. 17. M. A. Subramanian, G. Aravamudan and G. V. S. Ra o, "Oxide Pyrochlores a Review," Prog Solid State Ch 15 [2] 55-143 (1983). 18. D. C. Rubie and A. J. Brearley, "Phase-Transiti ons between Beta(Mg,Fe) 2 SiO 4 and Gamma(Mg,Fe) 2 SiO 4 in the Earths Mantle Mechanisms and Rheological Implications," Science 264 [5164] 1445-1448 (1994). 19. N. W. Grimes, "The Spin els: Versatile Materials," Phys Technol, 6 22-27 (1975). 20. L. W. Finger, R. M. Hazen and A. M. Hofmeister, "High-Pressure CrystalChemistry of Spinel (MgAl 2 O 4 ) and Magnetite (Fe 3 O 4 ) Comparisons with Silicate Spinels," Phys Chem Miner 13 [4] 215-220 (1986). 21. K. E. Sickafus, J. M. Wills and N. W. Grimes, "Structure of Spinel," J Am Ceram Soc 82 [12] 3279-3292 (1999). 22. E. J. W. Verwey and E. L. Heilmann, "Physical Properties and Cation Arrangement of Oxides with Spinel Structures .1. Cation Arrangement in Spinels," J Chem Phys 15 [4] 174-180 (1947). 23. Schmocke.U, F. Waldner and H. R. Boesch, "Direc t Determination of Cation Disorder in MgAl 2 O 4 Spinel by Esr," Phys Lett A A 40 [3] 237-& (1972). 24. I. D. Brown, The Chemical Bond in Inorganic Chemistry: T he Bond Valence Mode, Oxford University Press, Oxford, 2002. 25. I. D. Brown and D. Altermatt, "BondValence Parameters Obtained from a Systematic Analysis of the Inorga nic Crystal-Structure Database," Acta Crystallogr B 41 [Aug] 244-247 (1985). 26. N. E. Brese and M. Okeeffe, "B ond-Valence Parameters for Solids," Acta Crystallogr B 47 192-197 (1991). 27. K. E. Sickafus, "Radiati on Damage Effects in Solids," Los Alamos Unclassified Report LA-UR-07-5573 (2007). 28. R. Devanathan, N. Yu, K. E. Sickafus and M. Nastas i, "Structure and Property Changes in Spinel Irradi ated with Heavy Ions," Nucl Instrum Meth B 127 608-611 (1997). 29. S. J. Zinkle and V. A. Skuratov, "Track Formation and Dislocation Loop Interaction in Spinel Irradiated with Swift Heavy Ions," Nucl Instrum Meth B 141 [1-4] 737-746 (1998). 30. Y. N. Yavlinskii, "Coulomb Repulsi on of Lattice Ions under Swift Heavy Ion Irradiation," Nucl Instrum Meth B 245 [1] 114-116 (2006). 31. Fleische.Rl, P. B. Pric e and R. M. Walker, "Ion Explosion Spike Mechanism for Formation of Charged-Parti cle Tracks in Solids," J Appl Phys 36 [11] 3645-& (1965). 32. R. L. Fleischer, "Fission Tracks in Solids Production Mechanisms and Natural Origins," J Mater Sci 39 [12] 3901-3911 (2004). 192

PAGE 193

33. G. Szenes, "Thermal Spike Model of Am orphous Track Formation in Insulators Irradiated by Swift Heavy Ions," Nucl Instrum Meth B 116 [1-4] 141-144 (1996). 34. G. Szenes, "Ion-I nduced Amorphization in Ceramic Materials," J Nucl Mater 336 [1] 81-89 (2005). 35. E. M. Bringa and R. E. Johnson, "C oulomb Explosion and Thermal Spikes," Phys Rev Lett 88 [16] (2002). 36. S. Lutique, D. Staicu, R. J. M. Konings, V. V. Rondine lla, J. Somers and T. Wiss, "Zirconate Pyrochlore as a Transmutation Target: Thermal Behaviour and Radiation Resistance against Fission Fragment Impact," J Nucl Mater 319 59-64 (2003). 37. J. F. Ziegler, J. P. Biersack and U. Littmark, The Stopping and Range of Ions in Solids, Pergamon Press, New York, 2008. 38. K. E. Sickafus, R. W. Grimes, J. A. Valdez, A. Cleave, M. Tang, M. Ishimaru, S. M. Corish, C. R. Stanek and B. P. Uber uaga, "Radiation-Induced Amorphization Resistance and Radiation Tolerance in Structurally Related Oxides," Nat Mater 6 [3] 217-223 (2007). 39. L. M. Wang, S. X. Wang, W. L. Gong, R. C. Ewing and W. J. Weber, "Amorphization of Ceramic Materi als by Ion Beam Irradiation," Mat Sci Eng aStruct 253 [1-2] 106-113 (1998). 40. Macdonal.Dd and D. Owen, "Dissolution of Magnesium Oxide in Dilute Sulfuric Acid," Can J Chemistry 49 [20] 3375-& (1971). 41. I. G. Gorichev, N. A. Kipriyanov, S. K. Vainman and N. P. Shevelev, "Analysis of the Processes of Dissolution of Metal-Ox ides in Acids on the Basis of Affine Transformations of Kinetic Curves," J Appl Chem-Ussr 54 [1] 43-47 (1981). 42. I. G. Gorichev and N. A. Ki priyanov, "Kinetics Regularity of the Process of Metallic Oxide Solutions in Acid-Media," Usp Khim+ 53 [11] 1790-1826 (1984). 43. V. V. Batrakov, I. G. Gorichev and N. A. Kipriyanov, "Effect of Electrical DoubleLayer on the Kinetics of Me tal-Oxide Dissolution," Russ J Electrochem 30 [4] 399412 (1994). 44. H. Mineo, H. Isogai, Y. Mo rita and G. Uchiya ma, "An Investigation into Dissolution Rate of Spent Nuclear Fuel in Aqueous Reprocessing," J Nucl Sci Technol 41 [2] 126-134 (2004). 45. A. J. Bakel, D. L. Bowers K. J. Quigley, M. C. Regal buto, J. A. Stillman and G. F. Vandegrift, "Dissolution of Irradiated Nuclear Fuel from the Big Rock Point Reactor," Acs Sym Ser 933 71-88 (2006). 46. P. G. Medvedev, M. J. Lambregts and M. K. Meyer, "Thermal Conductivity and Acid Dissolution Behavior of MgO-ZrO 2 Ceramics for Use in LWR Inert Matrix Fuel," J Nucl Mater 349 [1-2] 167-177 (2006). 193

PAGE 194

47. G. P. Nikitina, Y. E. Ivanov, A. A. Listopadov and L. B. Shpunt, "Existing Methods for Dissolution of Plutonium Dioxide .1 Dissolution in Mineral Acids and Their Mixtures," Radiochemistry 39 [1] 12-25 (1997). 48. J. M. Cleveland, "Dissolution of Refractory Plutonium Dioxide," J Inorg Nucl Chem 26 [8] 1470-1471 (1964). 49. B. Tuck, "Chemical Polishing of Semiconductors," J Mater Sci 10 [2] 321-339 (1975). 50. M. A. Blesa, P. J. Morando and A. E. Regazzoni, C hemical Dissolution of Metal Oxides, CRC Press, Inc., Boca aton, 1994. 51. I. G. Gorichev and N. A. Kipriyanov, "Regular Kinetic F eatures of the Dissolution of Metal Oxides in Acidic Media (Kinetics R egularity of t he Process of Metallic Oxide Solutions in Acid-Media)," Russian Chemical Reviews 53 [11] 1039-1061 (1984). 52. S. Myhra, R. S. Smar t and P. S. Turner, "The Surf aces of Titanate Minerals, Ceramics and Silicate-Glasses Surface Analytical and Electron-Microscope Studies," Scanning Microscopy 2 [2] 715-734 (1988). 53. S. Myhra, H. E. Bi shop, J. C. Riviere and M. Stephenson, "Hydrothermal Dissolution of Perovskite (CaTiO 3 )," J Mater Sci 22 [9] 3217-3226 (1987). 54. K. Sangwal and S. K. Ar ora, "Etching of MgO Cryst als in Acids Kinetics and Mechanism of Dissolution," J Mater Sci 13 [9] 1977-1985 (1978). 55. O. Levenspiel, Chemical Reaction Engi neering, John Wiley & Sons, Inc., New York, 1999. 56. E. Caldin, Fast Reactions in Solution, Wiley, New York, 1964. 57. N. Valverde and C. Wagner, "Consi derations on Kinetics and Mechanism of Dissolution of Metal-Oxides in Acidic Solutions," Ber Bunsen Phys Chem 80 [4] 330-333 (1976). 58. F. M. Marshall, S. B. Grover and D. J. Utterbeck, "Fy 2008 Advanced Test Reactor National Scientific User Facility Users Guide," Idaho National Lab Report INL/EXT-07-13577 (2008). 59. H. Kleykamp, "Selection of Materials as Diluents for Bu rning of Plutonium Fuels in Nuclear Reactors," J Nucl Mater, 275 [1] 1-11 (1999). 60. E. A. C. Neeft, K. Bakker, R. P. C. Schram, R. Conrad and R. J. M. Konings, "The Efttra-T3 Irradiation Experim ent on Inert Matrix Fuels," J Nucl Mater 320 [1-2] 106-116 (2003). 61. N. Chauvin, T. Albiol, R. Mazoyer, J. Noirot, D. Lespiaux, J. C. Dumas, C. Weinberg, J. C. Menar d and J. P. Ottaviani, "In-Pile St udies of Inert Matrices with Emphasis on Magnesia and Magnes ium Aluminate Spinel," J Nucl Mater 274 [1-2] 91-97 (1999). 62. J. Lian, J. Chen, L. M. Wang, R. C. Ewing, J. M. Fa rmer, L. A. Boatner and K. B. Helean, "Radiation-I nduced Amorphization of Rare-Ear th Titanate Pyrochlores," Phys Rev B 68 [13] (2003). 194

PAGE 195

63. J. Lian, X. T. Zu, K. V. G. Kutty, J. Chen, L. M. Wang and R. C. Ewing, "IonIrradiation-Induced Am orphization of La 2 Zr 2 O 7 Pyrochlore," Phys Rev B 66 [5] 054108 (2002). 64. J. Lian, K. B. Helean, B. J. Kennedy, L. M. Wang, A. Navrotsky and R. C. Ewing, "Effect of Structure and Thermodynamic Stability on the Response of Lanthanide Stannate-Pyrochlores to Ion Beam Irradiation," J Phys Chem B 110 [5] 2343-2350 (2006). 65. K. E. Sickafus, L. Minervini R. W. Grimes, J. A. Valdez M. Ishimaru, F. Li, K. J. McClellan and T. Hartmann, "Radiati on Tolerance of Co mplex Oxides," Science, 289 [5480] 748-751 (2000). 66. S. Lutique, R. J. M. K onings, W. Rondinella, J. Some rs and T. Wiss, "The Thermal Conductivity of Nd 2 Zr 2 O 7 Pyrochlore and the Thermal Behaviour of PyrochloreBased Inert Matrix Fuel," J Alloy Compd 352 [1-2] 1-5 (2003). 67. R. J. Hill, J. R. Craig and G. V. Gibbs, "Systematics of the Spin el Structure Type," Phys Chem Miner 4 [4] 317-339 (1979). 68. M. Burghartz, H. Matzke, C. Leger, G. Vambenepe and M. Rome, "Inert Matrices for the Transmutation of Actinides: Fabric ation, Thermal Properties and Radiation Stability of Ceramic Materials," J Alloy Compd 271 544-548 (1998). 69. V. F. Sears, "Neutron Scatte ring Lengths and Cross Sections," Neutron news 3 [3] 26-37 (1992). 70. T. D. Shen, S. Feng, M. Tang, J. A. Valdez, Y. Wang and K. E. Sickafus, "Enhanced Radiation Tolerance in Nanocrystalline MgGa 2 O 4 ," Appl Phys Lett, 90 [26] (2007). 71. D. Simeone, C. Dodane-Th iriet, D. Gosset, P. Daniel and M. Beauvy, "OrderDisorder Phase Transition Induced by Swift Ions in MgAl 2 O 4 and ZnAl 2 O 4 Spinels," J Nucl Mater 300 [2-3] 151-160 (2002). 72. G. Baldinozzi, D. Simeone, D. Gosset, S. Surble, L. Mazerolles and L. Thome, "Why Ion Irradiation Does Not Lead to t he Same Structural Changes in Normal Spinels ZnAl 2 O 4 MgAl 2 O 4 and MgCr 2 O 4 ?," Nucl Instrum Meth B 266 [12-13] 2848-2853 (2008). 73. D. Bacorisen, R. Smith, B. P. Uber uaga, K. E. Sickafus, J. A. Ball and R. W. Grimes, "Atomistic Simulations of Radiat ion-Induced Defect Formation in Spinels: MgAl 2 O 4 MgGa 2 O 4 and MgIn 2 O 4 ," Phys Rev B 74 [21] 214105 (2006). 74. B. P. Uberuaga, D. Bacoris en, R. Smith, J. A. Ball, R. W. Grim es, A. F. Voter and K. E. Sickafus, "Defect Kinetics in Spinels: Long-Time Simulations of MgAl 2 O 4 MgGa 2 O 4 and MgIn 2 O 4 ," Phys Rev B 75 [10] 104116 (2007). 75. F. Studer, M. Hervieu, J. M. Costantini and M. Toulemonde, "High Resolution Electron Microscopy of Tracks in Solids," Nucl Instrum Meth B 122 [3] 449-457 (1997). 195

PAGE 196

76. M. Treilleux, G. Fuchs, A. Perez, E. Balanzat and J. Dural, "High-Energy HeavyIon Irradiation Effects in MgO MgFe 2 O 4 Ceramics," Nucl Instrum Meth B 32 [1-4] 397-400 (1988). 77. C. Houpert, M. Hervieu, D. Groult, F. Studer and M. Toulemonde, "Hrem Investigation of Gev Heavy-I on Latent Tracks in Ferrites," Nucl Instrum Meth B 32 [1-4] 393-396 (1988). 78. E. Wenda, "High Temperat ure Reactions in the MoO 3 -Ag 2 O System," J Therm Anal Calorim 53 [3] 861-870 (1998). 79. Pistoriu.Cw, "Phase Di agrams of Sodium Tungstate and Sodium Molybdate to 45 Kbar," J Chem Phys 44 [12] 4532-& (1966). 80. W. J. Weber, R. C. Ewing and L. M. Wang, "The RadiationInduced Crystalline-toAmorphous Transition in Zircon," J Mater Res 9 [3] 688-698 (1994). 81. P. Asanti and E. J. Kohlmeyer, "U ber Die Thermischen Eigenschaften Der Verbindungen Von Kobalt Mit Sauerstoff Und Schwefel," Z Anorg Allg Chem 265 [1-3] 90-98 (1951). 82. E. N. Bunting, "Phase E quilibria in the System SiO 2 -ZnO," J Am Ceram Soc 13 [1] 5-10 (1930). 83. N. L. Bowen and J. F. Sc hairer, "The System FeO-SiO 2 ," Am J Sci, 24 [141] 177213 (1932). 84. F. P. Glasser, "The System MnO-SiO 2 ," Am J Sci 256 [6] 398-412 (1958). 85. A. A. Al-Shahrani, "Sintering Be havior and Thermal Property of Mg 2 SnO 4 ," J Mater Sci-Mater El 16 [4] 193-196 (2005). 86. A. M. Azad and L. J. Min, "Mg 2 SnO 4 Ceramics I. Synthesis-ProcessingMicrostructure Correlation," Ceramics International, 27 [3] 325-334 (2001). 87. S. Raghavan, "Gibbs Free-Energy of Formation of Magnesiu m Stannate from EmfMeasurement," Thermochim Acta 122 [2] 389-393 (1987). 88. G. Pfaff, "Synthesis of Magnesium Stannates by Thermal-Decomposition of Peroxo-Precursors," Thermochim Acta 237 [1] 83-90 (1994). 89. W. W. Coffeen, "Ceramic and Diel ectric Properties of the Stannates," J Am Ceram Soc 36 [7] 207-214 (1953). 90. I. N. S. Jackson, Lieberma.Rc and A. E. Ringwood, "Disproporti onation of Spinels to Mixed Oxides Significance of Ca tion Configuration and Implications for Mantle," Earth Planet Sc Lett 24 [2] 203-208 (1974). 91. R. D. Shannon, "Revised Effective I onic-Radii and Systematic Studies of Interatomic Distances in Halides and Chalcogenides," Acta Crystallogr A 32 [Sep1] 751-767 (1976). 92. H. P. Abicht, D. Voltzke and T. Mu ller, "Methods of Powder Preparation for Technical Ceramics," Z Chem 30 [11] 385-395 (1990). 196

PAGE 197

93. S. J. Yates, Processing-Thermal Conductivity Relationships in MgO-Pyrochlore Composite Inert Matrix Mate rials, Ph.D Thesis, University of Florida, 2009. 94. R. A. Young, The Rietveld Method, Ox ford University Press, New York, 1995. 95. L. Lutterotti, "Material Anal ysis Using Diffraction (Maud V2.058)," (2007). 96. S. J. Yates, P. Xu, J. Wang, J. S. Tulenko and J. C. Nino, "Processing of Magnesia-Pyrochlore Composites for Inert Matrix Materials," J Nucl Mater 362 [23] 336-342 (2007). 97. M. P. Vandijk, J. H. H. Termaat, G. Roelofs, H. Bosch, G. M. H. Vandevelde, P. J. Gellings and A. J. Burggraaf, "Electrical and Catalytic Properties of Some Oxides with the Fluorite or Pyrochlore Stru cture .1. Synthesis, Characterization and Conductivity," Mater Res Bull 19 [9] 1149-1156 (1984). 98. E. J. Harvey, K. R. Whittle, G. R. Lumpkin R. I. Smith and S. A. T. Redfern, "Solid Solubilities of (La,Nd) 2 (Zr,Ti) 2 O 7 Phases Deduced by Neutron Diffraction," J Solid State Chem 178 [3] 800-810 (2005). 99. D. G. Wickham, N. Menyuk and K. Dwight, "Evidence for Canted Magnetic Moments in Manganous Stannate (Mn 2 SnO 4 )," J Phys Chem Solids 20 [3-4] 316318 (1961). 100. F. Deboer, J. H. Vansanten and E. J. W. Verwey, "The El ectrostatic Contribution to the Lattice Energy of Some Ordered Spinels," J Chem Phys, 18 [8] 1032-1034 (1950). 101. K. T. Jacob and J. Val derraman, "Gibbs Free Energy of Formation of Magnesium Stannate," J Solid State Chem 22 [3] 291-295 (1977). 102. S. V. Yanina and C. B. Carter, "Te rraces and Ledges on (001) Spinel Surfaces," Surf Sci 513 [2] L402-L412 (2002). 103. F. W. Clinard, G. F. Hurley and L. W. Hobbs, "Neutron-Irradi ation Damage in MgO, Al 2 O 3 and MgAl 2 O 4 Ceramics," J Nucl Mater 108 [1-2] 655-670 (1982). 104. G. W. Groves and A. Kelly, "Neutron Damage in MgO," Philos Mag 8 [93] 1437-& (1963). 105. R. S. Wilks, "Neutr on-Induced Damage in BeO Al 2 O 3 and MgO a Review," J Nucl Mater 26 [2] 137 (1968). 106. B. D. Evans, H. D. Hendricks and J. M. Bunch, "FastNeutron and Ion-Beam Damage in Crystalline MgO and Al 2 O 3 ," Am Ceram Soc Bull 55 [4] 458-458 (1976). 107. R. W. Davidge, "Irradiat ion Damage and Irradiation Ha rdening in Single Crystal MgO after Low Neutron Doses," J Nucl Mater 25 [1] 75 (1968). 108. M. Stevanovic and J. Elst on, "Effect of Fast Neutr on Irradiation in Sintered Alumina and Magnesia," Proc Br Ceram Soc 7 423-437 (1967). 109. R. S. Wilks, "Gas Format ion by Transmutation of MgO and Al 2 O 3 ," J Nucl Mater 19 [3] 351-& (1966). 197

PAGE 198

110. G. J. Russell, E. A. E. Ammar and J. S. Thorp, "Cavit y Growth in NeutronIrradiated Magnesium Oxide," J Mater Sci, 11 [10] 1961-1966 (1976). 111. T. S. Elleman, R. B. Price and Sunderma.Dn, "Fission Fragment Induced Expansion in Ceramic Materials," J Nucl Mater, 15 [3] 164-& (1965). 112. M. Beauvy, T. Duvernei x, C. Berlanga, R. Mazoyer and C. Duriez, "Actinide Transmutation: New Investigati on on Some Actinide Compounds," J Alloy Compd 271 557-562 (1998). 113. L. M. Wang, S. X. Wang, S. Zhu and R. C. Ewing, "Effects of Fission Product Incorporation on the Microstr ucture of Cubic Zirconia," J Nucl Mater 289 [1-2] 122127 (2001). 114. K. Nakai, K. Fukumoto and C. Kinoshi ta, "Characteristics of the Loop Formation Process in the MgO-Al 2 O 3 System Irradiated with Fission Neutrons," J Nucl Mater 191 630-634 (1992). 115. L. W. Hobbs, F. W. Clinar d, S. J. Zinkle an d R. C. Ewing, "R adiation Effects in Ceramics," J Nucl Mater 216 291-321 (1994). 116. C. Kinoshita, K. Fukumoto, K. Fukuda, F. A. Garner and G. W. Hollenberg, "Why Is Magnesia Spinel a Radiation-Resistant Material," J Nucl Mater 219 143-151 (1995). 117. S. J. Zinkle, "Hardness and Depth-Depen dent Microstructure of Ion-Irradiated Magnesium Aluminate Spinel," J Am Ceram Soc 72 [8] 1343-1351 (1989). 118. K. E. Sickafus, N. Yu and M. Nastasi, "Amorphization of MgAl 2 O 4 Spinel Using 1.5 Mev Xe + Ions under Cryogenic Irradiation Conditions," J Nucl Mater 304 [2-3] 237241 (2002). 119. K. E. Sickafus, N. Yu R. Devanathan and M. Nastasi, "The Irradiation Damage Response of MgO3Al 2 O 3 Spinel Single Crys tal under High-Fluence IonIrradiation," Nucl Instrum Meth B 106 [1-4] 573-578 (1995). 120. G. P. Pells, "Radiation Effects and Da mage Mechanisms in Ceramic Insulators and Window Materials," J Nucl Mater 155 67-76 (1988). 121. J. Mayer, L. A. Giannu zzi, T. Kamino and J. Michael, "TEM Sample Preparation and Fib-Induced Damage," MRS Bull 32 [5] 400-407 (2007). 122. K. L. Smith, N. J. Za luzec and G. R. Lumpkin, "In Si tu Studies of Ion Irradiated Zirconolite, Pyrochlore and Perovskite," J Nucl Mater 250 [1] 36-52 (1997). 123. W. J. Weber, "Models and Mechanisms of Irradiation-Induced Amorphization in Ceramics," Nucl Instrum Meth B 166 98-106 (2000). 124. N. Bordes, L. M. Wang, R. C. Ew ing and K. E. Sickafus, "Ion-Beam-Induced Disordering and Onset of Amorphization in Spinel by Defect Accumulation," J Mater Res 10 [4] 981-985 (1995). 125. H. M. Naguib and R. Kelly, "Criteria for Bombardment-Induc ed Structural-Changes in Non-Metallic Solids," Radiat Eff Defect S 25 [1] 1-12 (1975). 198

PAGE 199

126. K. Trachenko, J. M. Pruneda, E. Artacho and M. T. Dove, "How the Nature of the Chemical Bond Governs Resistance to Amorphization by Radiation Damage," Phys Rev B 71 [18] 184104 (2005). 127. K. Trachenko, "Understanding Resist ance to Amorphization by Radiation Damage," J Phys-Condens Mat, 16 [49] R1491-R1515 (2004). 128. L. Pauling, The Nature of the Chemical Bond, Cornell Un iversity Press, Ithaca, NY, 1960. 129. S. S. Batsanov, "System of Electr onegativity and Effective Atomic Charges for Crystalline Compounds," Zh Neorg Khim 20 [10] 2595-2600 (1975). 130. Z. J. Chen, H. Y. Xiao, X. T. Zu, L. M. Wang, F. Gao, J. Lian and R. C. Ewing, "Structural and Bonding Proper ties of Stannate Pyrochlores: A Density Functional Theory Investigation," Comp Mater Sci 42 [4] 653-658 (2008). 131. J. Lian, R. C. Ewing, L. M. Wang and K. B. He lean, "Ion-Beam Irradiation of Gd 2 Sn 2 O 7 and Gd 2 Hf 2 O 7 Pyrochlore: Bond-Type Effect," J Mater Res 19 [5] 15751580 (2004). 132. B. J. Kennedy, B. A. Hunt er and C. J. Howard, "Str uctural and Bonding Trends in Tin Pyrochlore Oxides," J Solid State Chem 130 [1] 58-65 (1997). 133. K. E. Sickafus, A. C. Larson, N. Yu, M. Nastasi, G. W. Holle nberg, F. A. Garner and R. C. Bradt, "Cation Disorder in High-Dose, Neutron-Irradiated Spinel," J Nucl Mater 219 128-134 (1995). 134. O. Engkvist and A. J. Stone, "Adsorption of Water on the MgO(001) Surface," Surf Sci, 437 [1-2] 239-248 (1999). 135. M. B. Kruger, Q. Willi ams and R. Jeanloz, "Vibrational-Spectra of Mg(OH) 2 and Ca(OH) 2 under Pressure," J Chem Phys, 91 [10] 5910-5915 (1989). 136. Rothbaue.R, F. Zigan and H. Odaniel, "R efinement of Struct ure of Bayerite Al(OH) 3 Including a Suggestion for H Position," Z Kristallogr Krist 125 [1-6] 317-& (1967). 137. O. Fruhwirth, G. W. Herzog, I. Hollere r and A. Rachetti, "Dissolution and Hydration Kinetics of MgO," Surf Technol 24 [3] 301-317 (1985). 138. V. S. Birchal, S. D. F. Rocha, M. B. Mansur and V. S. T. Ciminelli, "A Simplified Mechanistic Analysis of t he Hydration of Magnesia," Can J Chem Eng 79 [4] 507511 (2001). 139. S. D. F. Rocha, M. B. Mansur and V. S. T. Ciminelli, "Kinetics and Mechanistic Analysis of Caustic Magnesia Hydration," J Chem Technol Biot 79 [8] 816-821 (2004). 140. E. M. van der Merwe and C. A. St rydom, "Hydration of Medium Reactive Magnesium Oxide Using Hydration Agents," J Therm Anal Calorim 84 [2] 467-471 (2006). 141. D. Abriou and J. Jupille, "Self-Inhibition of Water Dissociation on Magnesium Oxide Surfaces," Surf Sci 430 [1-3] L527-L532 (1999). 199

PAGE 200

142. G. K. Layden and G. W. Brindley, "K inetics of Vapor-Phase Hydration of Magnesium Oxide," J Am Ceram Soc 46 [11] 518-522 (1963). 143. W. Feitknecht and H. Braun, "Der Mechanismus Der Hydratation Von Magnesiumoxid Mit Wasserdampf," Helvetica Chimica Acta 50 [7] 2040 (1967). 144. M. Maryska and J. Blaha, "Hydration Kinetics of M agnesium Oxide Part 3 Hydration Rate of MgO in Terms of Temperature and Time of Its Firing," CeramSilikaty 41 [4] 121-123 (1997). 145. V. S. S. Birchal, S. D. F. Rocha and V. S. T. Cimine lli, "The Effect of Magnesite Calcination Conditions on Magnesia Hydration," Miner Eng 13 [14-15] 1629-1633 (2000). 146. M. Maryska and J. Blaha, "Kinetics of Hydration of Magnesium Oxide in Aqueous Suspension .2. The Effect of Conditions of Firing Basic Magnesium Carbonate on the Specific Surface Ar ea of Magnesium Oxide," Ceram-Silikaty 41 [1] 21-27 (1997). 147. G. L. Smithson and N. N. Bakhshi, "Kinetics and Me chanism of Hydration of Magnesium Oxide in a Batch Reactor," Can J Chem Eng 47 [5] 508-513 (1969). 148. R. Salomao, L. R. M. Bittencourt and V. C. Pandolfe lli, "A Novel Approach for Magnesia Hydration Assessment in Refractory Castables," Ceram Int 33 [5] 803810 (2007). 149. J. Blaha, "Kinetics of Hydration of Magnesium-Oxide in Aqueous Suspension .1. Method of Measurement and Eval uation of Exper imental-Data," Ceram-Silikaty 39 [2] 45-51 (1995). 150. H. Y. Qian, S. Y. Li, M. Deng and S. M. Zhang, "Hydration Dynamic of LightBurned Magnesia," Huagong Kuangwu Yu Jiagong 36 [12] 1-4 (2007). 151. S. Kitahara, "Rate of the Hydration of Magnesium Oxide," Fukuoka Kyoiku Daigaku Kiyo, Dai-3-bunsatsu: Rika Hen 26 69-75 (1976). 152. D. A. Chauhan, "Hydration Kinetics of the Electrolytic Magnesium Oxide," Salt Research and Industry 18 [2] 1-9 (1982). 153. C. A. Scamehorn, N. M. Harrison and M. I. Mccart hy, "Water Chemistry on Surface Defect Sites Chemidissociation Ve rsus Physisorption on MgO(001)," J Chem Phys, 101 [2] 1547-1554 (1994). 154. A. Kitamura, K. Onizuka and K. Tanaka, "Hydration Characteristics of Magnesia," Taikabutsu Overseas 16 [3] 112-122 (1996). 155. R. T. Dehoff, "Estimation of Particle -Size Distributions from Simple Counting Measurements Made on R andom Plane Sections," T Metall Soc Aime 233 [1] 2529 (1965). 156. E. E. Underwood, Quantitative Stereology, Addison-Wesley Publishing Company, Reading, 1970. 157. S. A. Saltykov, Stereometric Metallography, Me tallurgizdat, Moscow, 1958. 200

PAGE 201

158. J. C. Wurst and J. A. Nelson, "Linear Intercept Technique for Measuring GrainSize in 2-Phase Polycrystalline Ceramics," J Am Ceram Soc 55 [2] 109-111 (1972). 159. M. J. G. W. Heij man, N. E. Benes, J. E. ten Elshof and H. Verweij, "Quantitative Analysis of the Microstructural Homogeneity of Zirconia-Toughened Alumina Composites," Mater Res Bull 37 [1] 141-149 (2002). 160. M. G. H. M. Hendriks, M. J. G. W. Heijman, W. E. van Zyl, J. E. ten Elshof and H. Verweij, "Quantitative Analysis of Micr ostructural Homogeneity and Capacitance Correlations in Palladium/Yttria-Stabilized Zirconia Composites," J Am Ceram Soc 85 [8] 2097-2101 (2002). 161. R. C. Smart and J. Nowotny, Ceramic Interfaces Properties and Applications Woodhead Publishi ng London, 1998. 162. M. Herrmann, B. Seipel, J. Schilm, K. G. Nickel, G. Michael and A. Krell, "Hydrothermal Corrosion of Zirconia-T oughened Alumina (ZTA ) at 200 Degrees C," J Eur Ceram Soc 25 [10] 1805-1812 (2005). 163. K. B. Alexander, P. F. Becher, S. B. Waters and A. Bleier, "Grain-Growth Kinetics in Alumina-Zirconia (Cezta) Composites," J Am Ceram Soc 77 [4] 939-946 (1994). 164. I. W. Chen and L. A. Xue, "Development of Superplastic Structural Ceramics," J Am Ceram Soc 73 [9] 2585-2609 (1990). 165. J. D. French, M. P. Ha rmer, H. M. Chan and G. A. Miller, "Coarsening-Resistant Dual-Phase Interpenetrating Microstructures," J Am Ceram Soc 73 [8] 2508-2510 (1990). 166. F. F. Lange and M. M. Hirlinger, "Grain-Growth in 2-Phase Ceramics Al 2 O 3 Inclusions in ZrO 2 ," J Am Ceram Soc 70 [11] 827-830 (1987). 167. Z. Hashin, "The Elastic-Moduli of Heterogeneous Materials," J Appl Mech 29 [1962] 143-150 (1962). 168. J. T. Long, Engineer ing for Nuclear Fuel Reprocessi ng, American Nuclear Society, La Grange Park, IL, 1978. 169. O. D. Fox, C. J. Jones, J. E. Birkett, M. J. Carrott, G. Crooks, C. J. Maher, C. V. Roube and R. J. Taylor, "Advanced Purex Flowsheets for Future Np and Pu Fuel Cycle Demands," Acs Sym Ser 933 89-102 (2006). 170. A. Tahraoui and J. H. Mo rris, "Decomposition of Solv ent-Extraction Media During Nuclear Reprocessing Literature-Review," Separ Sci Technol 30 [13] 2603-2630 (1995). 171. T. Inoue and L. Koch, "Dev elopment of Pyroprocessing and Its Future Direction," Nucl Eng Technol 40 [3] 183-190 (2008). 172. D. Olander, "Nuclear Fuels Present and Future," J Nucl Mater, In Press (2009). 173. L. H. Thompson and L. K. Dora iswamy, "Sonochemistry: Science and Engineering," Ind Eng Chem Res 38 [4] 1215-1249 (1999). 201

PAGE 202

174. J. P. Lorimer, T. J. Mason and K. Fiddy "Enhancement of Chemical-Reactivity by Power Ultrasound an Alternative In terpretation of the Hot-Spot," Ultrasonics 29 [4] 338-343 (1991). 175. O. Sohnel and P. Novotny, Densit ies of Aqueous Solutions of Inorganic Substances, Elsevier, Amsterdam, 1985. 176. M. I. Martinez and W. B. White, "A Laboratory Inve stigation of the Relative Dissolution Rates of the Lirio Limest one and the Isla De Mona Dolomite and Implications for Cave and Karst Development on Isla De Mona," J Cave Karst St 61 [1] 7-12 (1999). 177. M. T. Larrea, I. GomezPinilla and J. C. Farinas "Microwave-Assisted Acid Dissolution of Sintered Advanced Ceramics for Inductively Coupled Plasma Atomic Emission Spectrometry," J Anal Atom Spectrom 12 [11] 1323-1332 (1997). 178. P. Moisy, S. I. Nikitenko, L. Venaul t and C. Madic, "Sonochemical Dissolution of Metallic Plutonium in a Mixture of Nitric and Formic Acid," Radiochimica Acta 75 219-225 (1996). 179. K. S. Suslick, Ultrasound:Its Chemical Physical, and Biological Effects, VCH Publishers, New York, N.Y., 1988. 180. K. S. Suslick, Y. Didenko, M. M. Fang, T. Hyeon, K. J. Kolbeck, W. B. McNamara, M. M. Mdleleni and M. Wong, "A coustic Cavitation and Its Chemical Consequences," Philos T Roy Soc A 357 [1751] 335-353 (1999). 181. K. S. Suslick and G. J. Price, "Applications of Ultrasound to Materials Chemistry," Annu Rev Mater Sci 29 295-326 (1999). 182. K. S. Suslick, "Sonochemistry," Science, 247 [4949] 1439-1445 (1990). 183. R. S. C. Smart and J. Nowotny, Cerami c Interfaces Properties and Applications, IOM Communications Ltd., London, 1998. 184. B. D. Cullity and S. R. St ock, Elements of X-Ray Diffr action, Prentice Hall, Upper Saddle River, 2001. 185. P. Walker and W. H. Tarn, Handbook of Metal Etc hants, CRC Press, Boca Raton, FL, 1990. 186. W. J. Weber, R. C. Ewi ng, C. A. Angell, G. W. Arnold, A. N. Cormack, J. M. Delaye, D. L. Griscom, L. W. Hobbs, A. Navrotsky, D. L. Pr ice, A. M. Stoneham and W. C. Weinberg, "Radiat ion Effects in Glasses Us ed for Immobilization of High-Level Waste and Plut onium Disposition," J Mater Res 12 [8] 1946-1978 (1997). 187. J. Lian, S. X. Wang, L. M. Wang and R. C. Ewing, "R adiation Damage and Nanocrystal Formation in Ur anium-Niobium Titanates," J Nucl Mater 297 [1] 89-96 (2001). 202

PAGE 203

BIOGRAPHICAL SKETCH Peng Xu was born in 1982 in Xingtai, Ch ina. He attended Sichuan University, obtaining his B.S. degree in materials sci ence and engineering. He spent one year studying in the Department of Materials Sc ience and Engineer ing at the University of Washington (UW) from 2003 to 2004 as an ex change student. While in UW, he joined Dr. Guozhong Caos research group and studied on synthesis of TiO 2 nanorods and high piezoelectric nano PLZT ceramics. In 2005, he began his PhD st udy in Dr. Juan C. Ninos research group. His Ph.D study fo cuses on synthesis and characterization of MgO-pyrochlore composite and spinels for light water reactors inert matrix fuels. In 2008, he spent five and half month in Idaho Na tional Lab (INL) as an intern student with guidance of Dr. Mitch K. Meyer and Dr. Pavel G. Medvedev. There, he assisted the irradiation test preparation fo r the ATR-NSUF University of Florida project. After spending about 1160 days at UF in Gainesvill e and 170 days at INL in Idaho Falls, he received his Ph.D from the University of Florida in the spring of 2009. 203