<%BANNER%>

High Resolution Transmission Electron Microscopy Analysis of the Influence of Grain Boundary and Triple Grain Junction C...


PAGE 1

HIGH RESOLUTION TRANSMISSION ELE CTRON MICROSCOPY ANALYSIS OF THE INFLUENCE OF GRAIN BOUND ARY AND TRIPLE GRAIN JUNCTION CRYSTALLINITY AND CHEMISTRY ON SILICON CARBIDE-BASED ARMOR WITH SMALL ADDITIONS OF AL UMINUM, BORON, AND CARBON By SAMANTHA CRANE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Samantha Crane

PAGE 3

This document is dedicated to my parents for never letting me give up.

PAGE 4

iv ACKNOWLEDGMENTS I would like to acknowledge Ce ramatec, Inc., for the funding of this project. I would also like to thank Dr. Darryl Butt fo r chairing my committee and for his continued guidance throughout my undergraduate and gr aduate studies. My committee members Drs. J. J. Mecholsky, Jr. and Amelia Dempere were invaluable resources on this project. In addition, I would like to thank Kerry Seibien of MAIC at the University of Florida and Dr. Helge Heinrich of MCF at the University of Central Florida for their help with performing the TEM and EFTEM studies presented in this thesis. Edgardo Pabit’s insight, help, and guidance were essential to th e completion of my research. I would like to thank Dr. Erik Kuryliw and Soroya Be netiz for teaching me how to perform TEM sample preparation and how to polish. Finally I would like to thank Dr. Butt’s research group for their help and support.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 2 EFFECT OF SINTERING PARAMETERS ON MICROSTRUCTURAL DEVELOPMENT AND MECHANICAL PROPERTIES...........................................5 Mechanical Behavior of Ceramic Armor Systems.......................................................5 Toughening Mechanisms.......................................................................................5 Crack bowing.................................................................................................6 Crack deflection.............................................................................................6 Crack bridging................................................................................................7 Microcracking................................................................................................8 Transformation toughening..........................................................................10 Toughening Mechanisms in SiC..........................................................................10 Structural Properties of Silicon Carbide.....................................................................11 Sintering of Silicon Carbide.......................................................................................12 Solid State Sintered SiC......................................................................................14 Liquid-Phase Sintered SiC..................................................................................15 Liquid-phase sintering aides.........................................................................15 Al-B-C additive system................................................................................17 Microstructural Features of Liquid-Phase Sintered SiC.............................................19 Phase Transformation..........................................................................................21 Grain Boundary Films in Li quid-phase Sintered SiC..........................................21 Triple Junction Phase Formation and Crystallization.........................................21 3 MATERIALS AND METHODS...............................................................................24 Materials Processing of A BC-SiC by Ceramatec Inc.................................................24 Characterization..........................................................................................................24 Toughness Testing by Ceramactec, Inc...............................................................24

PAGE 6

vi Microhardness Testing by Ceramatec, Inc..........................................................25 Reitveld Analysis by Ceramatec, Inc..................................................................25 Grain Morphology Characteriza tion by Ceramatec, Inc.....................................25 Transmission Electron Microscopy (TEM).........................................................26 Transmission electron micros copy sample preparation...............................27 High resolution lattice imaging....................................................................28 Energy dispersive spectroscopy (EDS)........................................................28 Energy-filtered transmission el ectron microscopy (EFTEM)......................29 4 RESULTS AND DISCUSSION.................................................................................31 Transmission Electron Micr oscopy Characterization.................................................31 Grain Boundary and Intergranular Film Characterization...................................31 Triple Grain Junctions.........................................................................................34 Compositional Studies................................................................................................36 0 wt. % Al............................................................................................................36 0.5 wt. % Al.........................................................................................................37 1 wt. % Al............................................................................................................42 1.5 wt. % Al.........................................................................................................44 4 wt. % Al............................................................................................................47 Materials Characterization Performed at Ceramatec, Inc...........................................48 Fracture Mode Determination by Ceramactec, Inc.............................................50 Relationship Between Hardness and Toughness by Ceramactec, Inc.................54 Grain Morphology Characteriza tion by Ceramatec, Inc.....................................54 XRD and Reitveld Analysis by Ceramatec, Inc..................................................54 Correlation Between Microstructure and Mechanical Properties...............................55 Toughness............................................................................................................58 Weak interfaces............................................................................................58 Residual stresses...........................................................................................60 High aspect ratio grains................................................................................68 Large grains..................................................................................................69 Hardness..............................................................................................................71 Effect of grain size on hardness...................................................................72 Effect of aspect ratio on hardness................................................................73 5 SUMMARY AND CONCLUSIONS.........................................................................74 Conclusions.........................................................................................................74 Suggested Future Work.......................................................................................75 LIST OF REFERENCES...................................................................................................76 BIOGRAPHICAL SKETCH.............................................................................................81

PAGE 7

vii LIST OF TABLES Table page 2-1. Ramsdell Notion of Common SiC Polytypes............................................................12 4-1. Grain Boundary Width and Interg ranular Film Determination by HRTEM.............34 4-2. Toughness and Hardness Measuremen ts of SiC with 0-6 wt. % Al..........................58 4-3. Characterization of SiC with 0-6 wt. % Al................................................................59 4-4. Reitveld Analysis for SiC Samples with 0-6 wt. % Al...............................................63

PAGE 8

viii LIST OF FIGURES Figure page 2-1. Fracture toughness increase with part icle size due to microcracking........................10 2-2. Si-C tetrahedra, which form the ba sic structural unit of SiC polytypes....................11 2-3. A close packed plane of spheres w ith the sphere centers denoted by A. Subsequent planes can be stacked in th e sphere valleys denoted by B or C to form the different SiC structures..............................................................................12 2-4. Schematic of the competing energy terms during densification................................13 3-1. The above diagram illustrates electron-matter interactions in transmission electron microscopy.................................................................................................26 4-1. The grain boundaries in a) 0 wt. % Al, b) 0.5 wt. % Al, and c) 1 wt. % Al were completely crystalline and contained no in tergranular film. In d) 1.5 wt. % Al and e) 4 wt. % Al the grain boundaries are amorphous and 1 nm wide. Please note that the scale marker in a), b), and c) is 2 nm. The scale marker in d) and e) is 5 nm......................................................................................................................33 4-2. Triple grain junctions fo r ABC-SiC. a) 0 wt. % Al, b) 0.5 wt. % Al, c) 1 wt. % Al, d) 1.5 wt. % Al, and e) 4 wt. % Al. Please note that the scale marker is 20 nm in a) and b), 10 nm in c), 200 nm in d), and 100 nm in e).................................35 4-3. This is a series of micrographs and EDS spectra from a grain boundary in the 0 wt. % Al sample. There is no varia tion observed in composition between the grains and grain boundary........................................................................................37 4-4. EFTEM elemental maps of a grain boundary in the 0 wt. % Al sample...................38 4-5. The above figure is a micrograph of a three grain junction in the 0 wt. % Al sample and the corresponding EDS spectra The EDS data shows no secondary phase present in the three grain j unction, and no variation in composition between the three grain junc tion and surrounding grains........................................38 4-6. EFTEM elemental maps of a triple junc tion in the 0 wt. % Al sample. There is no change in composition between the thre e grains and the three-grain junction...39 4-7. The above STEM image and correspond ing EDS spectra show the presence of carbon inclusions between SiC grains in the 0 wt. % Al samples...........................39

PAGE 9

ix 4-8. EFTEM elemental maps of a C incl usion surrounding a pore in the 0 wt. % Al sample.......................................................................................................................40 4-9. The above a) STEM image and b)-e ) corresponding EDS spectra are taken from b), e) the grain boundary and c), and d) the surrounding grains..............................41 4-10. EFTEM elemental maps of a grain boundary in the 0.5 wt. % Al sample..............41 4-11. The above STEM image and corres ponding EDS spectra ar e taken from the triple junction and the surrounding grains in the 0.5 wt. % Al samples. The secondary phase is confined to the trip le junction and does not extend into the grain boundaries. The scale marker is 100 nm.........................................................42 4-12. EFTEM elemental maps of an Al-O ri ch inclusion in the 0.5 wt. % Al sample.....42 4-13. EELS spectra taken from a triple j unction in the 0.5 wt. % Al sample shows does not show crystalline Al2O3...............................................................................43 4-14. A STEM image and corresponding ED S spectra showing metal and B-C rich inclusions. The scale bar is 200 nm.........................................................................43 4-15. A STEM image of two grains and a gr ain boundary from the 1 wt. % Al sample and the corresponding EDS spectra. The composition does not vary between the grains and grain boundary........................................................................................44 4-16. EFTEM elemental maps of a grain boundary and transgranular crack in the 1 wt. % Al sample. There is no vari ation in composition across the grain boundary or at the site of the crack..........................................................................45 4-17. A STEM image of a triple junction in 1 wt. % Al showing that aluminum and oxygen segregate on the triple point a nd that the composition changes as a function of depth into th e triple junction. The scale marker is 100 nm..................45 4-18. EFTEM elemental maps of a triple j unction in the 1 wt. % Al sample. Notice that the very center of the triple junction is Si and C free........................................46 4-19. A STEM image and corresponding EDS spectra of grains and contaminations observed in the 1.0 wt. % Al sample........................................................................46 4-20. A STEM image of two grains a nd a grain boundary from 1.5 wt. % Al ABC sample and the corresponding EDS spectra.............................................................47 4-21. EFTEM elemental maps of a grain bounda ry in the 1.5 wt. % Al sample. Note the formation of an Al-O rich intergra nular film. Scale marker is 20 nm...............47 4-22. A STEM image and EDS spectra of a trip le junction in the sample containing 1.5 wt. % Al....................................................................................................................48

PAGE 10

x 4-23. EFTEM elemental maps of a triple j unction in the 1.5 wt. % Al sample. Notice that the entire triple junc tion is void of Si and C.....................................................48 4-24. EELS spectra taken from the triple j unction in the 1.5 wt. % Al sample shows that the triple junction is filled with Al2O3...............................................................49 4-25. A STEM image and corresponding EDS spectra of two grains and a grain boundary in 1.5 wt. % Al sample. This s ite shows two different grains. The upper grain is a SiC grain and the lower grai n is that of an Al -O rich grain. The presence of aluminum-rich grains were only observed for 1.5 wt. % Al and 4.0 wt. % Al....................................................................................................................49 4-26. The above STEM image and correspond ing EDS spectra are of a SiC-SiC grain boundary, two SiC grains, and a B-C rich in clusion in the 4 wt. % Al sample.......50 4-27. The above STEM image and correspond ing EDS spectra are of a SiC-SiC grain boundary, a B-C rich inclusion, and an Al4C inclusion in the 4 wt. % Al sample...51 4-28. The above STEM image and correspond ing EDS spectra are of a SiC-SiC grain boundary, two SiC grains, and an Al-O-C rich inclusion in the 4 wt. % Al sample. Note that the area of the gr ain boundary that is farthest from the inclusion contains Al, while the area closest to the inclusion is depleted in Al.......51 4-29. The above STEM image and correspond ing EDS spectra are of a SiC-SiC grain boundary, a SiC grain, and a carbon inclus ion imbedded in an Al-O-C rich inclusion in the 4 wt. % Al sample...........................................................................52 4-30. The above STEM image and correspond ing EDS spectra are of a SiC-SiC grain boundary, and two SiC grains in the 4 wt. % Al sample. Note the presence of Al at the grain boundary................................................................................................52 4-31. EFTEM elemental maps of a grain boundary between two SiC grains in the 4 wt. % Al sample.......................................................................................................53 4-32. EFTEM elemental maps of a multiple gr ain junction in the 4 wt. % Al sample. The circled grain is Al-B -C rich and may be Al8B4C7. The area between the grains is Al-O rich....................................................................................................53 4-33. SEPB fracture surfaces for samples with 0-6 wt. % Al. There is a clear change in fracture mode between transgranular and intergranular be tween 1 and 15 wt. % Al. The scale marker is 10 m............................................................................55 4-34. Comparison of HV1 (left) and HK1 (right ) for various SiC materials. A change in fracture mode is eviden t between 1 and 1.5 wt. % Al..........................................56 4-34 continued. Comparison of HV1 (left) a nd HK1 (right) for various SiC materials. The onset of the transgranular fracture mode is accompanied by a high degree of

PAGE 11

xi crack branching and crushing. A change toward mixed mode fracture occurs at 4 wt. % Al at which point the cracks ar e straight and there is no crushing.............57 4-35. The above graphs plot Knoop hardness and SEPB toughness as a function of Al content. The change in hardness an d toughness with changing Al content is inversely related.......................................................................................................59 4-36. Change in aspect ratio a nd grain size with Al content............................................60 4-37. Polished and chemically or plasma-etched surfaces of SiC samples with 0wt. % Al..........................................................................................................................6 1 4-38. X-ray diffraction patterns SiC with 0-6 wt. % Al. Note that starting powder was beta-3C and that in addi tion to SiC polytypes, Al8B4C7 phase is noted at high Al contents.....................................................................................................................62 4-39. Polytypes from Rietveld analys is for SiC-0.6 wt. % B-0.2 wt. % C samples with Al contents ranging from 0 to 6 wt. %.............................................................62 4-40. Plot of Stress vs. Temperature for a particle in an infinite matrix...........................66 4-41. Effect of residual stre ss fields around triple juncti ons on crack propagation..........67 4-42. Intergranular cracking in the 1.5 wt. % sample. The figure on the left shows cracking through the intergra nular film at the grain boundaries. The figure on the right shows a missing triple juncti on due to the crack deflecting around the triple junction, but within the amorphous secondary phase. Note that the image on the right is slightly out of focus, wh ich makes the crack appear to be a broad region that is lighter in color....................................................................................68 4-43. Plot of toughness vs. aspect ratio shows a negative slop e and argues against crack bridging. The three data points at the bottom of the graph correspond to the samples that fail transgranularly.........................................................................70 4-44. Plot of toughness vs. grain size. T oughness generally increases with increased grain size..................................................................................................................71 4-45. Plot of the effect of grain size on hardness. The plot shows that hardness increases with grain size for the samples that fail via interg ranular fracture...........72 4-46. The plot of hardness vs. aspect rati on shows that the hardness is not dependant on aspect ratio...........................................................................................................73

PAGE 12

xii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science HIGH RESOLUTION TRANSMISSION ELE CTRON MICROSCOPY ANALYSIS OF THE INFLUENCE OF GRAIN BOUND ARY AND TRIPLE GRAIN JUNCTION CRYSTALLINITY AND CHEMISTRY ON SILICON CARBIDE-BASED ARMOR WITH SMALL ADDITIONS OF AL UMINUM, BORON, AND CARBON By Samantha Crane August 2005 Chair: Darryl Butt Major Department: Materials Science and Engineering Recent studies have shown that toughness, hardness, and fracture mode are a function of interface crystallinity and chemistr y in silicon carbide-based armor systems. Therefore, a thorough investigati on of the effects of varying a dditive concentration on the microstructure and chemistry of the inte rgranular phases including the grain-boundary films and triple-junction phases is essential in order to engineer a better ceramic-based armor. For this object four SiC-based materi als were prepared and studied. The silicon carbide materials were created with the boron and carbon content held constant while varying the amount of aluminum available.

PAGE 13

1 CHAPTER 1 INTRODUCTION Ceramic armor systems have been in use ag ainst small caliber projectiles since the 1960s, however, the way in which combat is c onducted has changed w ith the advance of defense related technology. For this reas on, there is now greater emphasis on creating lower weight combat vehicles that need minimal combat preparation time and can endure attack by long rod penetrators (Gooch 2002). The armor is designed to “defeat” the projec tile at the interface by increasing its “dwell” time. To obtain long dwell times in ba llistic tests, the armor material must have high hardness. In addition, an increase in duc tility will result in improved resistance to penetration of the projectile if the hardness is maintained (Flinders et al. submitted). Advanced silicon carbide-based armor is amongs t the United States Army’s top choices for ceramic armor systems. Difficulty in obtaining fully dense compacts in silicon carbide due to the low self diffusion coefficient and the highly covalent nature of the Si-C bond (Huang et al. 1986) hindered the use of silicon carbide until Prochazka (Prochazka 1975) developed a technique in 1975 using boron and carbon additives for pressurele ss solid-state sintering. While Prochazka’s technique produced fully dense silicon carbide its low fracture toughness, which is approximately 2.5 MPa-m1/2 as measured by the single-edge precracked beam (SEPB) technique, limited it s application (Flinders et al. submitted). Since Prochazka’s pioneering work, a variet y of additive systems including oxides, carbides, nitrides, and metals have been used with silicon carbide to promote liquid phase

PAGE 14

2 sintering, which aids in lo wering the full densification sintering temperature and obtaining compacts close to the theoretical de nsity (Biswas et al. 2001). Liquid phase sintering, however, introduces intergranular phases at the grain-boundaries and triplejunctions during sintering, which have a pr onounced influence on the final properties (Zhou et al. 2002). For example, strong inte rgranular bonding is desirable for creep resistance and high temperature strength, wh ile the activation of toughening mechanisms such as crack deflection, crack bridging, and micro-cracking depend on “sufficiently weak” intergranular bonding (Moberlychan a nd De Jonghe 1998). In addition, recent studies have shown that toughne ss, hardness, and fracture mode are a function of grain boundary and triple grain juncti on crystallinity and chemistr y in silicon carbide-based systems (Moberlychan and De Jonghe 1998). Fo r the specific application of ceramic armor, it is important to achieve both a high level of toughening and strength (Flinders et al. submitted). Therefore, control of the microstructure and chemistry of the intergranular phases is of particular interest when engineering ceramic armor systems. Of particular interest is the aluminum, boron, and carbon system, which exhibits insitu toughening. The additives in this system aid in the cubic to hexagonal phase transformation, which can be controlled to dictate the final micr ostructural morphology (Moberlychan et al. 1998). In addition, the si ntering aides diminish the number of flaws that limit the strength and t oughness of the final part (Zha ng et al. 1998). While the toughness of the final part usually increases due to a change in fracture mode from transgranular to intergranular cracking, the hardness diminish es with increased additive content. Thus, it is important to use a suffic iently small amount of additive in order to maintain the high hardness, while still aiding in the sintering process and microstructural development (Moberlychan and De Jonghe 1998).

PAGE 15

3 There are several in-situ toughening m echanisms that can be employed in aluminum-boron-carbon silic on carbide (ABC-SiC); however, many of these mechanisms diminish the material’s hardness. Therefor e, a thorough investigati on of the effects of varying additive concentration on the microstr ucture and chemistry of the intergranular phases including the grain-boundary films and tr iple-junction phases is essential in order to engineer a ceramic-based armor with both high hardness and toughness. For this objective, four SiC-based materials were pr epared and studied. The silicon carbide materials were created with the boron and car bon content held consta nt while the amount of aluminum was varied. To address the effect of varying ad ditive concentration on the intergranular microstructure and chemistry, the following tasks were performed: the measurement of the size of the grain-boundaries and triple-jun ctions, and the determination of the amount of crystallinity of the grain-boundaries and tr iple-junctions; and the quantification of the grain-boundary and triple-junction chemistries. To accomplish this task, high resolution transmission electron microscopy (HRTEM) in conjunction with the following techniques were performed: lattice imaging to determ ine grain boundary width and secondary phase crystallinity; and energy dispersive spectro scopy (EDS), and energy filtered transmission electron microscopy (EFTEM) to quantify chemical compositions. The work presented in this thesis was done in cooperation with Ceramatec, Inc. of Salt Lake City, Utah, as part of an SBIR for the United States Army. Ceramatec, Inc. processed the ABC-SiC, performed toughne ss testing, microhardness testing, Reitveld analysis, X-ray diffraction, and grain morphol ogy characterization. Wh en the results of work performed at Ceramatec, Inc. are presented in this thesis, they will be identified as such in the section headings.

PAGE 16

4 The thesis is organized as follows: a review of previous work on SiC-based ceramics with an emphasis on secondary phase formation in liquidphase sintered SiC and the effects of SiC microstructure on mechanical properties will be given in Chapter 2. The experimental methods, including the anal ytical electron microscopy methods used in this study are introduced in Chapter 3. The re sults of the detailed investigation of the intergranular phases of ABC-SiC are presente d in Chapter 4. Finally, in Chapter 5, conclusions from the results presented in Ch apter 4 are drawn and a correlation is made between the microstructure and chemistry of intergranular phases and the mechanical behavior of the final part.

PAGE 17

5 CHAPTER 2 EFFECT OF SINTERING PARAMETERS ON MICROSTRUCTURAL DEVELOPMENT AND MECHANICAL PROPERTIES Mechanical Behavior of Ceramic Armor Systems As discussed in the previous chapter, ceramic armor is designed to “defeat” the projectile at the inte rface by increasing its “dwell” time. To obtain long dwell times in ballistic tests, the armor material must have high hardness. In a ddition, an increase in ductility will result in improved resistance to pe netration of the projectile if the hardness is maintained (Flinders et al. 2003). In th is chapter, toughening mechanisms in ceramics and the necessary microstructura l features for their activati on are discussed. In addition, the effect of processing parameters such as type of sintering and sintering additives on the formation of specific microstruc tural features are discussed. Toughening Mechanisms Two classes of toughening mechanisms exis t in ceramics. Intrinsic mechanisms, which include crack bowing and crack deflecti on, operate ahead of the crack tip and act as the material’s inherent defense agai nst microstructural damage and cracking by increasing the materials resist ance to crack initiation. Ex trinsic toughening mechanisms operate in the wake of the crack and reduce the local stress intensity at the crack tip. Extrinsic mechanisms, which include cr ack-bridging, microcrack toughening, and transformation toughening, are the main sour ce of toughness in brittle materials. Extrinsic toughening mechanisms, which act in the crack wake, give rise to R-curve, behavior, (Nalla et al. 2004). In a material exhibiting R-cu rve behavior, a greater applied load must be applied to conti nue the advance of the crack.

PAGE 18

6 Crack bowing Crack bowing occurs when the crack front interacts with tough particles or inclusions. The crack is initially pinned by the tough obstacle, but can pass it by bowing on the same plane. While the stress fields are quite different, crack bowing is conceptually similar to the interaction of a dislocation line with a precipitate. Crack bowing assumes that the obstacle is impenetrabl e; however, if the obstacle were to break before the bowing process is complete a cr acked ligament would be left in the crack wake. If the crack passes without breaking the particle an uncracked ligament will be left in the crack wake and thus, crack bowing can be a precursor to crack-bridging (Green 1998). Crack deflection Crack deflection occurs when the crack is tilted or twisted out of the plane normal to the applied stress. The change in orient ation reduces the crack extension force, which means that a larger applied stress is require d for fracture and the toughness will therefore increase. This reduction is greater for twis ting than tilting and has been shown to be dependant on the volume fraction and shape of the particle providing the obstruction with rod shaped particles providing the greatest in crease in toughness. As the aspect ratio of the particles increases, the toughening increment increases due to the increase of twist angle with increased aspect ratio. In addition to the effect of geomet ry and the effect of the volume fraction of particles, crack deflecti on has been shown to also be the result of several factors: the local st ress field at the obstacle, the presence of a low toughness interface, or the pres ence of a cleavage pl ane (Green 1998). In multiple phase systems, such as one containing grains and intergranular films or triple junction phases, the difference in thermal expansion mismatch between the grains

PAGE 19

7 and secondary phases can give rise to residua l stresses. If the difference in thermal expansion between the grains and secondary phases is positive, a compressive radial stress is developed at the matrix-particle boundary and a te nsile tangential stress is developed in the matrix. This will attract the crack to the particle. When the difference in thermal expansion is positive, the crack wi ll be deflected between particles and a more tortuous crack path will be created, which will increase t oughness (Hertzberg 1996). If one considers a material with an intergranular film in which the intergranular film is the matrix and the grains are the particles, this second case can be used to explain the origin of intergranular fracture. With respect to intergranular films, th e local compressive residual stresses around the grains cause diminishing stress intensity di rectly at the tip of the deflected crack and are a result of a difference in thermo-elastic properties between the intergranular film and grains (Sternitzke 1997). Crys tallizing triple junctions can also cause residual stresses, the magnitude of which is dependent on the volume change upon crystallization (Pezzotti and Kleebe 1999). Crack bridging In the discussion of crack bowing it was mentioned that cracks can by-pass an obstacle and leave an intact ligament in th e crack wake (Green 1998). For crack bridging to occur, the uncracked ligament, usually an elongated grain, must remain intact as the crack approaches and it must be energetically favorable for the crack to deflect along the ligament surface, rather than cutting through it. This can occur if interfacial bonding occurs due to the presence of a weak interf ace, which would reduce the stress intensity at the crack tip and allow the ligament to survive and bridge the crack (N alla et al. 2004).

PAGE 20

8 The uncracked ligaments span the crack wake and sustain part of the applied load that would otherwise be used to advance th e crack. Thus, the crack bridges shield the crack tip from part of the applied load. This increases fracture toughness and gives rise to R-curve behavior (Green 1998 and Nalla et al. 2004). Several testing methods can be employed to determine if crack bridging is the dominant toughening mechanism. First, the crack tip and immediate crack tip wake can be viewed directly with either scanning electron microscopy or transmission electron microscopy (Yuan et al. 2003). Second, a co mparison of elastic compliance of the cracked specimen can be compared to the th eoretical compliance of an ideal bridgeand microcrack-free crack of the same length. Si nce crack-bridging increases the modulus, a reduction in compliance would suggest the pres ence of crack-bridging (N alla et al. 2004). Finally, when the ligaments have a plate-like geometry, the fracture toughness is given by d E K KL L o Ic Ic 67 1 2 2 281 0 where KIc is the fracture toughness measured by the SEBP technique, KIc o is the toughness in the absence of crack bridging, E is Young’s modulus, is the frictional stress between grains, L is the bridging rupture strain, L is the aspect ratio, and d is the mean grain size. A plot of KIc 2 as a function L 1.67 d should yield a straight line if L 2 is a constant (Flinders et al. submitted). Microcracking Stress-induced microcracking can give ri se to crack-tip shielding, which may increase a material’s toughness. During mi crocrack toughening a frontal process zone is developed in which the microcracks form. Microcrack formation, and subsequent opening, gives rise to a volumetric increase (Green 1998). The resulting dilation and reduction in modulus of the frontal process zone, if constrained by the surrounding rigid

PAGE 21

9 material, will shield the crack tip and cause an increase in toughne ss (Nalla et al. 2004). In addition, microcracking can lead to crack branching, When crack branching occurs, a greater applied stress is required to driv e the increased number of cracks and thus toughness will increase (Meyers and Kumarchawla 1999). Microcracking occurs when residual st resses build-up due to the following phenomena: phase transformations, thermal expa nsion anisotropy in one phase materials, or thermo-elastic property mismatch in multiphase materials (Green 1998). Phase transformation can cause volumetric expansi on and contraction. When this volumetric change occurs in a confined phase, residual stress can arise and lead to the formation of microcracks. Thermal expansion anisotropy arises when the crystal symmetry is less than cubic, such as hexagonal, rhombohedral, monoclin ic, tetragonal, and tr iclinic. Expansion is different in each crystal direction. Ther mal expansion anisotropy results in contraction of individual grains from neighboring grains at differing rates, depending on crystal orientation. This produces mechanical st resses at the interface, which can generate microcracks. Thermo-elastic property mismat ch produces residual stresses in a similar manner as the two aforementioned mechanis ms (Case et al. 2005). Areas of low toughness are particularly susceptible to microcracking: grain boundaries and intergranular films (Green 1998). The degree of toughness increase due to microcracking is dependent on the grain size. As the particle size increases, the t oughening increment will increase due to the creation of a tortuous crack path until the crit ical grain size is reached. At this point spontaneous microcracking will occur and the toughness will decrease since microcrack toughening is only effective in material s that experience microcracking upon crack

PAGE 22

10 propagation. It should also be noted that ther e is also a critical particle size below which cracking will not occur (Green 1998). No Discrete Microcracking MicrocrackingZone Spontaneous Microcracking Particle SizeFracture Toughness No Discrete Microcracking MicrocrackingZone Spontaneous Microcracking Particle SizeFracture Toughness Figure 2-1. Fracture toughness increase with particle size due to microcracking (Green 1998). Transformation toughening Transformation toughening occurs when a stress induced phase transformation of an unstable phase occurs in the process zo ne. The phase transformation is associated with a dissipation of energy and the development of compressive residual stresses that will oppose crack advance (Hertzberg 1996). Toughening Mechanisms in SiC Of the five toughening mechanisms discu ssed, crack deflection, crack bridging and microcracking are the most possible mechanisms active in SiC. The necessary conditions for these mechanisms are similar. Weak inte rfaces and a residual st ress field are needed for the activation of all three mechanisms. In addition, crack deflection and crack bridging also rely on the particles, in this case grains, having a high aspect ratio, while

PAGE 23

11 microcracking is dependant on grain size. Al l of these features can be manipulated by choosing the right processing conditions and additive systems. The remainder of this chapter will discuss the effect of sintering parameters on microstructural development. Structural Properties of Silicon Carbide Silicon carbide exists in several polytypes, of which the following are the most common: 3C, 2H, 4H, 6H, and 12R (Fischer et al. 1990). The polytypes consist of strong, primarily covalently bonded (88% co valent, 12% ionic) Si-C tetrahedra, which contain a Si-C bond length of 1.89 . The te trahedra join at the corners to form the different SiC polytypes (Ye 2002). Figure 2-2. Si-C tetrahedra, which form the basic structural unit of SiC polytypes. The structure of the polytypes can be simulated by stacking sheets of close packed spheres, each of which consists of both a la yer of Si atoms and a layer of C atoms. The sequence in which these sheets are stacked dis tinguishes one polytype from another. In Figure 2-3, site A represents the position of the spheres in the first sheet. Subsequent sheets of spheres can either be stacked in the B or C sites, which represent the valleys between the spheres. Thus, the sheets can be denoted as A, B, or C-sheets, depending on the position of the spheres (Ye 2002). The simplest stacking sequences are as follows: ...ABAB... and ...ABCABC..., which correspond to the hexagonal wurtzite an d cubic zinc-blend structures, respectively.

PAGE 24

12 The hexagonal and rhombohedral polyty pes are collectively referred to as -SiC, whereas the cubic structure of SiC an d is often referred to as -SiC (Ye 2002). Figure 2-3. A close packed plane of sphere s with the sphere centers denoted by A. Subsequent planes can be stacked in th e sphere valleys denoted by B or C to form the different SiC structures. In the Ramsdell notation, the number of sheets needed to define the stacking sequence is followed by a letter, which represents the crystal structure: C for cubic, H for hexagonal, and R for rhombohedral. Table 2-1 lists the common polytypes and their corresponding stacking sequences (Ramsdell 1947). Table 2-1. Ramsdell Notion of Common SiC Polytypes. Ramsdell Notation 3C 2H 4H 6H 15R Type -SiC -SiC -SiC -SiC -SiC Stacking Sequence (ABC Notation) ABC AB ABCB ABCACBABCBACABACBCACB The effects of different additive sy stems on polytype transformation and its relationship to mechanical properties will be discussed later in this chapter. Chapter 4 will discuss the specific effects of the Al, B, C additive systems used in this body of work to polytype transformation and its re lationship to mechanical properties. Sintering of Silicon Carbide Sintering is the most common processing route for polycrystalline SiC; however, due to the highly covalent nature of the Si -C bond, it is difficult to produce final parts

PAGE 25

13 with densities approaching the material’s theo retic density. Several factors contribute to the limited thermodynamic driving force and sluggish kinetics: high grain boundary energy, low self-diffusion coefficients, and coarsening (Prochazka 1987, Tanaka 1991). The thermodynamic driving force for dens ification and mass transport is the reduction of free energy in the system du e to the formation of interfaces from free surfaces. In the specific cas e of grain boundary formation, the grain boundary energy, GB, must be less than twice the solid vapor energy, SV. For pore shrinkage, where a three grain junction is formed, the ratio of the grain boundary energy to the solid-vapor interfacial energy, GB/ SV, should be less than 3 (Greskovich and Rosolowski 1976). In a highly covalent material such as SiC, the grain boundary energy is so great that the reduction of excess free energy is extremely small and densification is practically inhibited (Prochazka 1987, Tanaka 1991). Figure 2-4. Schematic of the compet ing energy terms during densification. In addition to having an extremely low thermodynamic driving force for densification, the kinetics of SiC densification are sluggish (Ye 2002). As previously mentioned, the high degree of covalency in the Si-C bond limits densification. This arises from the extremely low self diffusio n coefficients of Si and C in SiC (~ 10-13 cm2/s for Si and ~ 10-11 cm2/s for C), which limits mass transport and makes diffusion-

PAGE 26

14 controlled solid-state sintering infeasible (Hong 1979). In addition, coarsening, not densification, dominates during the sintering process due to silicon carbide’s relatively high vapor pressure, which causes larger grains with larger radii of curvature to grow at the expense of smaller grains with smaller ra dii of curvature. This leads to coarsening without the elimination of pores (Prochazka 87). Therefore, to produce final parts with densities approaching the material’s theoretical density, the cond itions contributing to the poor thermodynamics and kinetics must be overcome. This has led to the use of sintering aides, wh ich can lower the grain boundary energy, increase the solid vapor in terfacial energy, and increase the mass transport. Depending on the additive system, the aforementioned behavior can be achieved in the absence or presence of a liquid phase, which are referred to as solid-state sintering and liquid-phase sintering, respectivel y (Ye 2002). It is also important to note that the mechanical properties in ceramics ar e controlled by the flaw size and flaw size distribution. Sintering ai des are commonly used to achieve high density and thus decrease the flaw size due to porosity (Zhang et al. 1998). Solid State Sintered SiC Prochazka was the first to develop pressure less sintering of Si C with small amounts of B and C. He gave a thermodynamic expl anation of the effect of additives on the sintering of SiC. The Carbon removes the surface oxide by carbothermal reduction, which increases the solid/vapor interfacial energy. The B segregates to the grain boundaries and reduces the grain boundary interfacial energy (Prochazka 1975). As discussed in the previous section, an increas e in the solid-vapor interfacial energy and a decrease in the grain boundary energy w ill favor pore shrinkage and grain boundary formation.

PAGE 27

15 Tanaka argued that the C also acted to decrease the grain boundary energy, since a certain amount of carbon was necessary, rega rdless of the oxide c ontent of the starting powder (Tanaka 1984). In addition, B and C have the added effect of increasing selfdiffusion in SiC (Birnie 1986). While the combined effects of B and C allow for solid state sintering of SiC, the low fracture toughness, which is 2.5 MPa-m1/2 as measured by the single-edge precracked beam (SEPB) technique, and exaggerated grai n preclude the use of sold-state sintered SiC as ceramic armor (Flinders et al. submitted). Liquid-Phase Sintered SiC Liquid-phase sintering, which can be acco mplished via pressureless sintering, gaspressure sintering, hot-pressing, and hot-iso static-pressing, is achieved when the melting temperatures of the additives are lower than the sintering temperat ure (Ye 2002). This condition allows for the formation of a liquid phase that will spread among the grains during sintering. Upon cooling, the liquid phase may persist as a secondary glassy or crystalline phase at the grain boundaries and triple grain junc tions. Liquid-phase sintering results in more uniform densificat ion and suppression of the exaggerated grain growth that may occur in solid state sintering. In addition, the sec ondary phases can alter the properties of the sintered ceramic (Kaneko et al. 2000). Liquid-phase sintering aides Several additive systems have been used in liquid-phase sintering of SiC. Aluminum additives have been used in several forms, such as Al-metal, Al2O3, Al4C3, and AlN (Kaneko et al. 2000), and in combination with several other additives, such as Y2O3, B, C, B4C, and CaO (Kim et al. 1995, Gu et al. 1996, Zhou et al. 1999). Each additive

PAGE 28

16 system has different effects on the microstruc ture and mechanical properties of the final sintered SiC (Ye 2002). Al2O3 is frequently used in combination with Y2O3 to promote high aspect ratio SiC grains and a YAG (Y3Al5O12) intergranular phase, which improve the fracture toughness and creep resistance. Chia an d Lau suggested that the main toughening mechanism was microcracking between the YAG secondary phase and the SiC grains (Chia and Lau 1991). Nitrogen in the form of AlN is used either alone or in combination with Al2O3 and Y2O3 to retard the transformation, suppress grain growth, and promote the formation of oxynitride glass, which have also been found to increase the fracture toughness (Kim and Mitomo1999). One example of AlN-containing SiC armor is SiC-N, which is manufactured by Cercom. Inomata et al. (Inomata et al. 1980) showed that in the Al-B-C additive system a liquid phase was present at 1800C near the composition of Al8B4C7. This non-oxide system has recently received attention for its ability to produce a three-fold increase in the fracture toughness of SiC (Chen and Lau 200 0). With respect to increased toughness, the Al-B-C system can achieve high toughness at lower processing temperature than the YAG system when a comparable amount of li quid phase is used (Flinders et al. 2003). However, for a given toughness SiC with YAG as a secondary phase is harder than SiC sintered with a comparable amount of Al-B-C additives (Flinders et al. 2003). The effects of B and C additives were discussed in the previous section. However, the combined effect of Al, B, and C differ greatly from those of B and C alone. This additive system promotes the phase transformation, the 6H-SiC to 4H-SiC polytype transformation, the formation of elongated grains, the formati on of intergranular films, and the formation of secondary phases at the triple-gra in junctions (Cao et al. 1996).

PAGE 29

17 Al-B-C additive system The effects of B and C additive were discussed in the previous section, however the combined effect of Al, B, and C differ greatly from those of B and C alone. Researchers at the Lawrence Berkeley National Laborato ry performed experiments on ABC-SiC in which they varied the amount of alum inum from 1-3wt%. The boron and carbon concentrations were held constant at 0.6wt% B and 2 wt% C. Similar to the work of Chia and Lau, the ABC-SiC exhibited R-curve behavior. They observed the formation of either crystalline or amorphous secondary phases in the SiC structure as large regions that wet amongst many matrix grains, triple point ju nctions, and grain boundaries. The Berkeley group observed the formation of an amorpho us secondary phase at the grain boundaries, the crystallinity of which is dependent on the aluminum concentration with increasing crystallinity at higher wt% Al. According to the Berkeley group, the lowered work-ofadhesion afforded by the amorphous second ary phase that promotes toughening and intergranular cracking can be controlled by the second phase chemistry (Moberlychan and De Jonghe 1998). In a different study at Lawrence Berkel ey National Laboratory, two different experiments on ABC-SiC were performed. In the first set of experiments, aluminum foil was imbedded in ABC-SiC powder and hot presse d. This allowed for the formation of an aluminum gradient and the measurement of Si C properties as a function of wt% Al. In the second set of experiments, the wt% Al was varied from 3-7%. The boron and carbon concentrations were held constant at 0.6wt% B and 2 wt% C. In both sets of experiments the researchers observed a lowered densifica tion temperature and an increased number of triple junctions at higher wt% Al (Zhang et al. 2003).

PAGE 30

18 In the first set of experiments, they observed that the area density, aspect ratio, and size uniformity of the elongate d grains varied as a function of aluminum content. The aspect ratio is particularly de pendent on aluminum concentration. At low concentrations of approximately 1wt% Al, the aspect ratio is reduced to nearly unity, thus grains are equiaxed. In addition, th e researchers observed a decrea se in indentation toughness and a decreased tendency toward intergranular cracking as the wt% Al was decreased (Zhang et al. 2003). The second set of experiments was run to better understand the effect of aluminum concentration on the properties of ABC-SiC. The variation in aluminum content from 37wt% Al altered the microstructure. As with the first set of experiments, the area density, aspect ratio, and size uniformity of the elongat ed grains varied as a function of aluminum content. A nearly linear increase in the aspe ct ratio is observed up to 6wt% Al while the length reaches a maximum at 5wt% Al. A bimodal grain distribution was observed at aluminum contents greater than 3wt%. In addition, the toughness degraded at concentrations greater than 6w t% Al (Zhang et al. 2003). The increased aluminum content also had a profound effect on th e crystallinity of the grain boundaries. As prev iously stated, the Berkeley gr oup found that at aluminum concentrations lower than 3wt% the second phase was amorphous. Increasing the wt% Al lead to increased cr ystallinity of the grain boundaries. At 3wt% Al, 85% of the second phase films examined were amorphous and at 6wt% Al and above all of the second phase films examined were crystalline. It is also important to note that increasing the overall aluminum content raised the concentration of aluminum at the grain boundaries and in the grain. At aluminum concentrations great er than 5wt% the grains were saturated and metallic aluminum precipitated at the grai n boundaries (Zhang et al. 2003).

PAGE 31

19 According to the Berkeley experiments, increasing the boron content increases the number density of elongated grains, coarse ns the grains, reduces the aspect ratio, promotes phase transformations, enhances grain boundary diffusion, and produces a driving force for mass transport. In additi on, boron is more effective in promoting the transformation (Zhang et al. 2003). Increased carbon content promotes elongated grains, enhances phase transformation, enhances grain boundary diffus ion, and produces a driving force for mass transport. It also increases the self-diffusi on rate, as well as the grain boundary diffusion rate of SiC. In addition, the carbon incr eases apparent density up to 4wt% because of pore elimination and decreases apparent dens ity after 4wt% because of appearance of carbon phase in materials. Furthermore, a decrease in grain growth was observed with increased carbon content. It is import ant to note that boron and carbon do not form intergranular films by themselves and in th e absence of aluminum will incorporate into the SiC lattice (Zhang et al. 2003). The grain morphology is controlled by the B/C ratio and when the B/C ratio favors elongated grain growth, the aluminum accelerat es it and produces higher aspect ratios. When the Al/B and Al/C ratios are reduced, there are less liquid phases and the aluminum effects are diminished. When the Al/B and Al/C ratios are increased, boron and carbon are depleted from the lattice and dissolve into the liquid phases. In addition, at constant Al/B/C ratios, changing the to tal amount of additives will still change grain morphology and phase composition (Zhang et al. 2003). Microstructural Features of Liquid-Phase Sintered SiC As mentioned in the previous two sec tions, the sintering additives used and processing route can greatly affect the final mi crostructure of SiC materials. However,

PAGE 32

20 there are several common microstructural feat ures of liquid-phase sintered SiC: SiC grains, intergranular films at the grain bound aries, and secondary phases at the triple grain junctions (Moberlychan et al. 1996). However, cases have been reported where direct grain-to-grain interf aces are present (Kaneko et al. 1999). Figure 2-5 is a schematic of the aforementione d microstructural features. Both -SiC and -SiC starting powders may be used in liquid-phases sintering of SiC. Typically -SiC powders are used when a fine equiaxed microstructure is desired (Sigl and Kleebe 1993). However, wh en increased toughness is desired, -SiC is generally used as a starting powder since the -tophase transformation upon sintering will result in an interlocking microstructure of elongated plate-like grain, which has been shown to increase fracture toughness via crack -bridging and crack deflection (Gilbert et al. 1996). The phase transformation can be accelerated by seeding the -SiC starting powder with -SiC (Baud and Theveno 2001), an d retarded by incorporating Ncontaining sintering aides, such as AlN (Jun et al. 1997). Crystalline Secondary Phase Amorphous Triple Junction Intergranular Film Large Glass Pocket Direct Grain-toGrain Boundary Crystalline Secondary Phase Amorphous Triple Junction Intergranular Film Large Glass Pocket Direct Grain-toGrain Boundary Figure 2-5. Schematic of the microstructura l features of liquid-phase sintered SiC (Ye 2002).

PAGE 33

21 Phase Transformation As mentioned previously, the Al-B-C additive system promotes the phase transformation, the 6H-SiC to 4H-SiC poly type transformation, and the formation of elongated grains. The high aspect ratio grains offer reinforcement and increased toughness by promoting several toughening mechanisms. Grain Boundary Films in Li quid-phase Sintered SiC While the presence of intergranular films may appear to be a kinetic phenomenon, the existence of intergranular films at variou s experimental conditions suggests that these films may be in a state of thermodynamic equilibrium. Clarke devised a model to explain the existence of intergranular films based on a force balance betw een attractive Van der Waals forces across the grain boundaries and re pulsive steric forces, capillary forces, and electric double layer forces in the film (Clarke 1987, Chen et al. 1993). The model predicted the presence of intergranular films on the order of 1-2 nm in several materials, which agrees with experimental data presen ted to date. While developing his model, Clarke considered the case of SiC without additives and concluded that SiC is the only material where the attractive forces are greater than the repulsive forces (Clarke 1987). Thus SiC without additives would not have a stable intergranular film. Varying the width and composition of the intergranular film has a large effect on the mechanical properties of SiC. In Al-B -C-containing SiC, the intergranular film has been shown to increase the fracture toughness by a factor of three. Therefore, tailoring of the intergranular film is an important task. Triple Junction Phase Formation and Crystallization The Berkeley group observed that the reduc tion of large secondary phases resulted in the appearance of triple junctions. Thes e regions are several nanometers wide and up

PAGE 34

22 to 10 m long. They form by heterogeneous nuc leation on the basal plane of the SiC. Typically, only partial crysta llization can occur and an amorphous region is observed between the crystallin e secondary phase and the SiC gr ain (Moberlychan and De Jonghe 1998). Full crystallization is hindered by several factors: the composition of the intergranular phase may shift into regions of the phase diagram in which crystallization capability is low; the volume change due to crystallization of the intergranular phase is constrained by the surrounding SiC grains, wh ich causes internal stresses that create a thermodynamic barrier to complete crystallizat ion; and there may be kinetic hindrance to crystallization (Bonnell et al. 1987 Raj 1981, Kessler et al. 1992). The crystallization process greatly affects the composition distribution in the triple junction. The amorphous region between the crystalline secondary phase and SiC grains has a different composition from the crysta lline secondary region. This is due to undesirable solute rejection to the side and ahead of the so lidification front, which allows the secondary phases to grow free of SiC. Typically, the secondary phases at the triple junctions are a crystalline ternary or binary containing Al-O-B, Al-B-C, Al-C, or Al-O (Moberlychan and De Jonghe 1998). In addition, the extent of triple junction crystallization can affect the mechanical properties of the final part. Kessler et al (Kessler et al. 1992 ) found that the volume change due to crystallization is strongly de pendant on the degree of crystallization with increased crystallization resulting in incr eased volume change. This volume change results in residual stresses whose magnitude is proportional to the volume change of the constrained phase, or triple junction. Pezzotti and Kleebe (Pezotti and Kleebe1999) found that a nearly 100% fraction of crystall ized secondary phase at the triple junction

PAGE 35

23 will create substantial tensile stresses due to volume contraction, which drives cracks toward the triple junction and causes su bsequent interface delamination. This phenomenon can lead to crack brnching and intergranular fracture. Crack deflection will manifest itself as roughness in the final fracture surface. When SiC fractures in an intergranular fashion, the fracture surface has a high degree of surface roughness, which is associated with twist and tilt crack deflection (Moberlychan et al. 1998).

PAGE 36

24 CHAPTER 3 MATERIALS AND METHODS Material processing, characterization, a nd electron microscopy methods used to investigate the microstructure and chemistry of the SiC-based materials are presented in this chapter. Materials Processing of ABC-SiC by Ceramatec Inc. -SiC (Superior Graphite grade HSC-05 9) with surface area in the 15-17 m2/g range was mixed with Al, B, and C additiv es dispersed with a polyamine polyester polymer (Avecia Chemical grade Solsperse 2400 0) by adding one wt. % of the polymer, based on solids, to 400 grams of reagent grade toluene. The Al powder (Valimet grade H3), which has an average size of 3 m, was added in amounts of 0, 0.5, 1, 1.5, and 4 wt. %, whereas the boron (H. C. Starck, amorphous B grade S-432B) was ke pt constant at 0.6 wt. %. The carbon was introduced as 4 wt. % Apiezon grade W wax, assuming a 50 % yield after pyrolysis to give ~2 wt. % C. The slurries were deagglomerated for two hours with a paint shaker and then rolled over night before drying. Powders were passed through a 44 m screen before hot pressing at 28 MPa in stagnant Ar inside graphite dies at 2100 C for 1 h. Characterization Toughness Testing by Ceramactec, Inc. The hot pressed billets were ground in order to form 3 mm x 4 mm x 45 mm bars, which were subsequently indented with a 98 N Knoop indenter. The samples were then pre-cracked to initiate crack growth. The single edge pr e-cracked beam (SEPB) tests

PAGE 37

25 were performed with loads rangi ng from 4 kN to 18 kN using spans of 4, 5, or 6 mm. The original crack was marked by a dye (Magnaflux Zyglo ZL-60D), using vacuum infiltration and oven dried at 110C overnight and cooled before testing. All crack planes were parallel to the hot pressing direction (Flinders et al. submitted). Microhardness Testing by Ceramatec, Inc. A Leco microhardness machine was used to perform Vickers and Knoop hardness measurements on polished SEPB bars under a 9.8 N load. The load was applied at 50 m/s, with a 15 second dwell time (Flinders et al. submitted). Reitveld Analysis by Ceramatec, Inc. Rietveld analysis was used to determine SiC polytypes present in the densified samples with X-ray diffraction patterns collected from 30-80o 2-theta, with a step size of 0.02o/step and a counting time of 4 sec/step (Flinders et al. submitted). Grain Morphology Characterization by Ceramatec, Inc. Polished samples of liquid phase sintered Si C with Al content ranging from 0.5 to 6 t. % Al were plasma-etched by evacuatin g and back-filling with 400 millitorr of CF410% O2 and etching for 20-40 minutes. The SiC sample with 0 wt. % Al was etched in molten KOH at 550C for 10-15 seconds. Grain size was determined by the lineintercept method, where the multiplication cons tant ranged between 1.5 (equiaxed grains) and 2.0 (elongated, plate-shaped grains). Typically, 200-300 grains were measured for each composition in order to get a mean grain size. The aspect ratio of the five most acicular grains in each of three micrographs was used to estimate a comparative aspect ratio (Flinders et al. submitted).

PAGE 38

26 Transmission Electron Microscopy (TEM) Transmission electron microscopy, which uses a finely focused beam of electrons, is an important part of the experimental pro cedure used to investigate the microstructure and chemistry of the SiC-based materials. A wide variety of secondary signals from the specimen are produced when the electron beam interacts with the matter in the specimen. In addition to producing signals used for imag ing, these secondary signals are also used to obtain chemical information (Williams and Carter 1996). Figure 3-1 illustrates some of the interactions between the electron beam and the sample. Figure 3-1. The above diagram illustrates el ectron-matter interactions in transmission electron microscopy (Williams and Carter 1996). Electron interactions can be divided into tw o classes: elastic and inelastic-scattering events. While elastic and inelastic scattering give rise to many us eful signals, only the

PAGE 39

27 signals relevant to TEM and the associat ed characterization techniques will now be discussed. Elastic scattering events affect th e trajectories of the beam electrons inside a specimen without altering the kinetic energy of the electrons. In TEM, elastically scattered electrons are the major source of cont rast in images and also create the intensity distributions in diffraction patterns. Inelastic scattering events result in a transfer of energy from the beam electrons to the atoms of the specimen. Inelastic scattering events lead to the generation of characteristic X -rays, which are used in energy dispersive spectroscopy (EDS), and inelastically scatte red electrons, which are used in electron energy loss spectrometry (EELS) (Williams and Carter 1996). Transmission electron microscopy sample preparation In order to perform the electron characteriz ation techniques discussed later in the chapter, the samples must first be prepared for use in the TEM. Transmission electron microscopy specimens were prepared by the standard mechanical thinning method, which creates a self-supporting disk. A 3 x 4 mm rectangle with a thickness of 1 mm was cut from each bulk specimen (3 x 4 x 45 mm ba rs made for SEPB testing) with a low speed diamond saw. The samples were then moun ted on an aluminum polishing stub with crystal-bond adhesive and wet polished to a 100 m thickness and 3 m finish. The samples were then cut into 3 mm diameter disc s using an ultrasonic dr ill and then further thinned on a precision dimpling machine to 50 m using a 3 m diamond solution and a flattening tool. The center of the sample was subsequently thinned to 20 m with a dimpling tool. Then four to five hours of Argon ion milling at 4 keV, 1and a beam angle of 12 was carried out un til the sample was perforated.

PAGE 40

28 High resolution lattice imaging The high resolution lattice images presented in this thesis were generated on a JEOL JEM-2010F FEG with a point-to-point resolution of 1.9 . The images were obtained with the Help of Kerry Seibien at the Major Analytical and Instrumentation Center at the University of Florida. High resolution lattice imaging, or phase-contrast imaging, requires the selection of several diff racted beams, in addition to the transmitted beam, to create the image. The collected b eams interfere and cause the intensity of the beam to vary sinusoidally with different periodicities for different values of the diffraction vector. This in turn causes lattice fringes in an area of crystallinity. At high magnifications these fringes can be imaged and information on the spacing of the planes normal to the diffraction vector of the beams can be obtained (Williams and Carter 1996). Lattice imaging can be used to investigat e interfaces such as grain boundaries and three grain junctions by orienting the interface parallel to the electron beam and tilting the surrounding grains on-axis. The aforemen tioned experimental conditions will create lattice fringes in the surrounding grains. In the absence of a secondary phase, an abrupt interface of near atomic dimensions can be observed. If an amorphous region exists at the grain boundary, it will not exhibit lattice fringes and can be imaged directly. Thus, lattice imaging may be used to determine th e existence of amorphous secondary phases at interfaces (Williams and Carter 1996). Energy dispersive spectroscopy (EDS) The EDS data presented in this thesis was generated on the JEOL 2010F equipped with an Oxford INCA 200 EDS system, whic h has a probe size of approximately 5 nm. Energy dispersive spectroscopy produces plots of the X-ray counts, or intensity, versus the X-ray energy. The plot consists of tw o types of signals: Bremsstrahlung X-rays, and

PAGE 41

29 characteristic X-rays. Bremsstrahlung X-rays are the radiation which is emitted when electrons are decelerated due to interaction with the sample. It is characterized by a continuous distribution of radiation a nd gives no information on the chemical composition of the sample. On an EDS plot, Bremsstahlung X-rays make up the background peaks. The larger Gaussian peaks are the characteristic X-ray peaks. Characteristic X-rays occur when an incident electron ionizes an inner-shell electron, and an electron drops down from a higher energy level to fill the vacancy. The radiation energy produced is equal to the difference between the atomic energy levels. Since the difference in energy levels is unique to each element, characteristic X-rays give compositional data (Williams and Carter 1996). Energy-filtered transmission electron microscopy (EFTEM) The EFTEM performed in this thesis was done at the Materials Characterization Facility at the University of Central Florida with the help of Dr. Helge Heinrich on a Technai F30 equipped with a FEG and a GATAN GIF. Elemental maps can be formed using EFTEM by imaging with electrons that ha ve lost energy corresponding to the innershell ionization edge of a particular elemen t of interest (Hofer et al. 1997). For elemental mapping, the intensity of any part of the electron energy loss spectrum can be selected to form an electron spectroscop ic image (ESI) (Williams and Carter 1996); however, it is necessary to remove the bac kground contribution to the image intensity (Hofer et al. 1995). The three window tec hnique was employed to create ESIs with the background contribution removed. This te chnique involves recording images at two energy-loss values before and one energy lo ss value after the ionization energy of the element under investigation, as seen in Fi gure 3-2 (Schaffer et al. 2003). Then an

PAGE 42

30 extrapolated background image is calculate d and subtracted from the ionization edge image, which gives an elemental map (Hofer et al. 1995). Figure 3-2. Schematic of the three-window t echnique. Two pre-edge images are used to estimate the background and calculate an elemental map.

PAGE 43

31 CHAPTER 4 RESULTS AND DISCUSSION The discussion in chapter 2 focused on toughening mechanisms in ceramic materials and the microtructural features necessary for their activation. Per that discussion it was concluded th at the most likely tougheni ng mechanisms active in SiC depend on weak interfaces and the presence of a residual stress field. Tests performed at Ceramatec, Inc. showed a change in fractur e mode from transgranular to intergranular between 1 and 1.5 wt. % Al. Therefore, this chapter will focus on characterizing the changes in microstructure of each of the five SiC materials described in chapter 3 and the effect of these changes on mechanical behavior. Transmission Electron Microscopy Characterization High resolution lattice imaging, energy dispersive spectroscopy, and energy filtered transmission electron microscopy data were taken for samples of SiC with Al wt. % content that varied from 0 wt. % Al to 4 wt % Al, and a constant 0.6 wt. % B and 2 t.% C. The following sections include a detailed analysis of these samples based upon the crystallinity, thickness, and composition of the grain boundaries and triple grain junctions. Grain Boundary and Intergranular Film Characterization Based on the high resolution TEM data ta ken from the samples, shown in Figure 41, the thickness of the grain boundary and the intergranular fi lm (IGF) varies between the samples with 0-1 wt. % Al, and the samples with 1.5 wt. % Al or greater. In the samples with 0-1 wt. % Al added, there is a direct tran sition from grain to grain, which results in a

PAGE 44

32 negligible grain boundary thickness. The sa mples with 1.5 wt. % Al or greater contain grain boundaries on the order of 1 nm. The crystallinity of the grain boundary, based on the lattice image of the samples also varies. While the samples with 0-1 wt. % Al show a completely crystalline grain boundary, the samples with greater than 1.5 and 4 wt. % Al show an amorphous grain boundary. These observ ations are summarized in Table 4-1. Konishita et al. found that the solubili ty limit of Al in SiC is 0.2 wt. % Al (Konishita 1997). Hence, it would be expected that at Al contents greater than 0.2 w.% Al, the Al would be expelled from the grain and form an intergranular film. Since this is not the observed phenomenon, the presence, or lack, of the amorphous film can not be explained based on the solubility limit alone As Clarke explai ned in reference to equilibrium intergranular film thickness, SiC in the absence of additives is one of the only materials in which the attractive Van der Waals forces across the grain boundaries are greater than the repulsive steric for ces, capillary forces, and electric double layer forces in the film (Clarke 1987). Accordin g to the equilibrium se gregation theory of McLean, the amount of solute present at th e grain boundary is directly related to the solute content in the grains and inversely related to temperature. / QRT DCACe CD is the solute atomic fraction at the gr ain boundary. A is a constant related to vibrational entropy. Q is the free energy of segregation at the grain boundary, which is related to the energy difference between an at om in the bulk and an atom at the grain boundary. R is the gas constant and T is the absolute temperat ure (Konishita 1997). Therefore, it is possible that there exists a critical level of additives necessary to promote the formation of an intergranular f ilm and it was not reached at the prescribed

PAGE 45

33 processing temperature until 1.5 wt. % Al. This topic will be revisited in the discussion of the compositional studies performed on the five SiC materials. a) b) c) d) e) Figure 4-1. The grain bounda ries in a) 0 wt. % Al, b) 0. 5 wt. % Al, and c) 1 wt. % Al were completely crystalline and containe d no intergranular film. In d) 1.5 wt. % Al and e) 4 wt. % Al the grain boun daries are amorphous and 1 nm wide. Please note that the scale marker in a), b) and c) is 2 nm. The scale marker in d) and e) is 5 nm.

PAGE 46

34 Table 4-1. Grain Boundary Width and In tergranular Film Determination by HRTEM Amorphous 1 nm 4.0 Amorphous 1 nm 1.5 Crystalline ~0 nm 1.0 Crystalline ~0 nm 0.5 Crystalline ~0 nm 0.0 Grain Boundary Crystallinity Grain Boundary Width Al Content (wt.%) Amorphous 1 nm 4.0 Amorphous 1 nm 1.5 Crystalline ~0 nm 1.0 Crystalline ~0 nm 0.5 Crystalline ~0 nm 0.0 Grain Boundary Crystallinity Grain Boundary Width Al Content (wt.%) Triple Grain Junctions In general the sintering aides will attract impurities in the starting powder, react with the native oxide on th e particle surface, and form a mass transport medium during densification (Falk 1997). In the specific case of Al-B-C additives, boron and carbon will not form a secondary phase without alum inum. This assertion has been proven experimentally by many researchers (Prochazka 1975 and Moberlychan and De Jonghe 1998). Per the discussion in chapter 2, in SiC si ntered without Al the carbon reacts with the native oxide to form SiO and CO. This reaction depletes the oxygen from the system (Stobierski and Guberat 2002). However in th e presence of Al, O will be trapped in the system because of the highly negative free energy of formation and comparatively low vapor pressure of aluminum oxides. Thus, th e O content will increase with increasing Al. The evidence of this behavior will be presented in later sections of this chapter. This phenomenon produces a greater volume and cr ystallinity of secondary phase for higher wt. % Al compositions, when the B and C co mpositions are held constant. Figure 4-2 illustrates the formation of triple junction phases once the solubility limit of Al in SiC has been reached (0.2 wt. % Al). The figure also illustrates that the volume of the triple junction phase increases with in creasing aluminum content.

PAGE 47

35 20 nm 0 wt% Al a) b) c) d) e) Figure 4-2. Triple grain juncti ons for ABC-SiC. a) 0 wt. % Al, b) 0.5 wt. % Al, c) 1 wt. % Al, d) 1.5 wt. % Al, and e) 4 wt. % Al. Please note that the scale marker is 20 nm in a) and b), 10 nm in c) 200 nm in d), and 100 nm in e). 4 wt% Al 0.5 wt% Al

PAGE 48

36 Compositional Studies Energy dispersive spectroscopy (EDS) was performed on the samples using the HRTEM in scanning transmission electron microscope (STEM) mode. EDS is a semiquantitative chemical analysis method. There are several factors to take into consideration when performing EDS on TEM samp les. First, the quality of the data is dependent upon the electron beam, or probe size with relation to the area of sample being studied. While the manufacturer lists the minimum probe size as 0.8 nm, the minimum attainable probe size in practi ce is closer to 3-5 nm. The grain boundaries in this set of samples are typically 1 nm wide. Thus, ther e may be signal coming from the surrounding grains as well. It should also be noted that the data is aver aged through the bulk. Therefore, if there are overla pping grains, the grain boundary is not oriented parallel to the beam direction, or the grain boundary do es not go all the wa y through the sample thickness characteristics X-rays will be genera ted from the grain as well and affect the chemical analysis. Therefore, it is more prac tical to use the differen ce in characteristic peaks present and in peak heights between the secondary phases and th e grains to identify variations in composition than to try to use EDS as a quantitative chemical analysis technique. In addition to EDS, which uses characte ristic X-rays as the signal, EFTEM was also used to create elemental maps of th e samples. Since the signal for EFTEM comes from the electron energy loss spectrum, EFTEM provides a method of verifying the presence of species in the samples. 0 wt. % Al According to the literature, boron and car bon do not form intergranular films by themselves and in the absence of aluminum w ill incorporate into the SiC lattice (Zhang et

PAGE 49

37 al. 2003). This can be seen in the EDS data in Figures 4-3 through Figure 4-6. However, several carbon inclusions were observed in this sample due to excess free carbon. This phenomenon can be seen in Figures 4-7 and 4-8. As observed in Figures 4-3 through 4-6, there is very little variation in composition between the grain, the grain boundary, and the three grain junctions. The presence of a clean grain boundary combined with the gr ain boundary width and crystallinity data suggests the absence of an intergranular film. In addition, the triple grain junctions in this sample were usually clean except for the occasional carbon inclusion. An EDS spectra and EFTEM analysis of a triple junc tion in this sample shows no variation in composition between the grains and three grain junctions, which suggests the absence of a secondary phase at the triple junctions. The presence of secondar y phases at the grain boundaries and triple grain j unctions generate residual st resses that promote several toughening mechanisms and intergranular fr acture. The absence of these secondary phases may explain the low toughness of solid state SiC. 0.5 wt. % Al While this sample is above the solubility limit of Al in SiC, the grain boundaries are free of secondary phase and the composition does not change between the grains and 400 nm Grain Grain Boundary Figure 4-3. This is a series of micrograph s and EDS spectra from a grain boundary in the 0 wt. % Al sample. There is no variation observed in composition between the grains and grain boundary.

PAGE 50

38 Si C O B Grain Boundary Si Si C C O O B B Grain Boundary Grain Boundary Figure 4-4. EFTEM elemental maps of a grain boundary in the 0 wt. % Al sample. S2 S3 S4 S5 C O Si C O Si C O Si Triple Junction Grain Grain Boundary Figure 4-5. The above figure is a micrograph of a three grain juncti on in the 0 wt. % Al sample and the corresponding EDS spectra. The EDS data shows no secondary phase present in the three grain junction, and no variation in composition between the three grai n junction and surrounding grains.

PAGE 51

39 Si C B O TJ Si Si C C B B O O TJ TJ Figure 4-6. EFTEM elemental maps of a trip le junction in the 0 wt. % Al sample. There is no change in composition between the three grains and the three-grain junction. S2 S5 S7 Carbon Inclusion SiC Grain SiC Grain SiC Grain Figure 4-7. The above STEM image and corresponding EDS spectra show the presence of carbon inclusions between SiC grai ns in the 0 wt. % Al samples. grain boundaries. This can be seen in Figures 4-9 and 4-10. This data in combination with the clean, narrow, crystalline grai n boundary observed in the HRTEM images

PAGE 52

40 Si C O B Si C O B Figure 4-8. EFTEM elemental maps of a C inclusion surrounding a pore in the 0 wt. % Al sample. testifies to the assertion that no intergranu lar film exist at the grain boundaries. Unlike the 0 wt. % Al sample, the 0.5 wt. % Al sample shows a small O peak at the grain boundary. This may be due to the diminished effectiveness of the C additive at removing the native oxide in the presence of Al. Since the solubility limit of Al in SiC has been reached in this sample, rejection of Al from the grains and the formation of a re gion of secondary phase are expected. While this was not observed at the grain boundaries, it was observed in the triple junctions as an Al-O rich phase. EELS spectrum taken from these regions did not show the characteristic peaks for Al or O in Al2O3 (see Figure 4-13); however, the spectrum did show the peaks of elemental Al and O. This would suggest that th e region is amorphous. The triple junctions may retain the second ph ase since the energy to create a film free grain boundary is less than that to create a clean triple junc tion. In addition, since the

PAGE 53

41 S2 S3 S4 C O Al Si C Al Si Grain Grain Boundary Figure 4-9. The above a) STEM image a nd b)-e) corresponding ED S spectra are taken from b), e) the grain boundary and c), and d) the surrounding grains. 100 nm Si 100 nm C 100 nm Al 100 nm B 100 nm O GB 100 nm Si 100 nm Si 100 nm C 100 nm C 100 nm Al 100 nm Al 100 nm B 100 nm B 100 nm O 100 nm O GB Figure 4-10. EFTEM elemental maps of a grain boundary in the 0.5 wt. % Al sample. triple junctions start out as pores, which ar e larger in volume, the liquid may be pushed into the pores as the SiC-SiC grain boundaries form first. In addition to the triple junction phase, transition metal inclusions and B-C rich inclusions, which are larger than the triple junctions, were observed. These can be seen in Figure 4-14.

PAGE 54

42 C O Si S2 S3 C O Si Al Triple Grain Figure 4-11. The above STEM image and co rresponding EDS spectra are taken from the triple junction and the surrounding grains in the 0.5 wt. % Al samples. The secondary phase is confined to the trip le junction and does not extend into the grain boundaries. The scale marker is 100 nm. C 0.1 m Si 0.1 m Al 0.1 m B 0.1 m O 0.1 m 0.1 m C 0.1 m C C 0.1 m 0.1 m Si 0.1 m Si Si 0.1 m 0.1 m Al 0.1 m Al Al 0.1 m 0.1 m B 0.1 m B 0.1 m 0.1 m O 0.1 m O O 0.1 m 0.1 m 0.1 m 0.1 m 0.1 m Figure 4-12. EFTEM elemental maps of an Al-O rich inclusion in the 0.5 wt. % Al sample. 1 wt. % Al Although the change from transgranular to intergranular fraction occurs between 1 and 1.5 wt. % Al, there is still very little va riation in composition between typical grains and grain boundaries in this sample as s hown in Figures 4-15 and 4-16. However, the

PAGE 55

43 Elemental O Peak Elemental Al Peak Elemental O Peak Elemental Al Peak Al -k O -k Elemental O Peak Elemental Al Peak Elemental O Peak Elemental Al Peak Al -k O -k Figure 4-13. EELS spectra taken from a triple junction in the 0.5 wt. % Al sample shows does not show crystalline Al2O3. S2 S3 S4 S5 Transition Metal Inclusion B-C Rich Grain SiC S2 S3 S4 S5 S2 S3 S4 S5 Transition Metal Inclusion B-C Rich Grain SiC Figure 4-14. A STEM image and corres ponding EDS spectra showing metal and B-C rich inclusions. The scale bar is 200 nm. overall counts of O in the sample are greater than those of the 0.5 wt. % sample with the majority of the increase in the triple junctions This lends credence to the assertion that the Al is retaining the O in the system. The nature of the triple grain junction of this sample differs from that of the 0.5 wt. % Al sample as shown in Figures 4-17 an d 4-18. While the increase in volume of secondary phase is insignificant, the cha nge in composition is great. The secondary

PAGE 56

44 phase in the 0.5 wt. % Al sample was found to be amorphous; however, the 1 wt. % Al sample contains crystalline Al2O3 in the center of the triple junction and an amorphous phase separating the triple junction phase from the grains. In addi tion, the concentration of aluminum and oxygen increases with the distance into the tr iple junction. This is in agreement with the Berkeley group’s assertio n that the crystallization process greatly affects the composition distribution in the triple junction and that the amorphous region between the crystalline secondary phase and SiC grains has a different composition from the crystalline secondary region (M oberlychan and De Jonghe 1998). In addition to the secondary phases observ ed at the triple j unctions, grains of transition metals were also observed. Thes e grains, which are shown in Figure 4-19, are the result of transition metals present in the starting powder. S2 S3 S4 C O Al Si Grain C O Al Si Grain Boundary Figure 4-15. A STEM image of two grains and a grain boundary from the 1 wt. % Al sample and the corresponding EDS spectra. The composition does not vary between the grains and grain boundary. 1.5 wt. % Al The change in fracture mode from transgranu lar to intergranular fracture is evident at 1.5 wt. % Al. This sample differs fr om those at lower wt. % Al, which fracture transgranularly, in that there is now an interg ranular film that is rich in Al and O present

PAGE 57

45 GB Si C Al B O 100 nmGB Crack GB Si C Al B O 100 nm Si Si C C Al Al B B O O 100 nm 100 nmGB Crack Figure 4-16. EFTEM elemental maps of a gr ain boundary and transgranular crack in the 1 wt. % Al sample. There is no variation in composition across the grain boundary or at the site of the crack. Figure 4-17. A STEM image of a triple junc tion in 1 wt. % Al showing that aluminum and oxygen segregate on the triple poin t and that the composition changes as a function of depth into th e triple junction. The scale marker is 100 nm. at the grain boundaries (Figures 4-20 and 421), and the triple junction is filled with Al2O3 (Figures 4-22 through 4-24). In addition there is an increase in O at the triple S2 S3 S4 S5 S6 C O Al Si CO Al Si C O Al Si Grain Triple Junction Center Triple Junction Edges

PAGE 58

46 Al O B C Si TJ Al Al O O B B C C Si Si TJ Figure 4-18. EFTEM elemental maps of a tr iple junction in the 1 wt. % Al sample. Notice that the very center of the triple junction is Si and C free. Grain Grain Boundary Secondary Phase Figure 4-19. A STEM image and corre sponding EDS spectra of grains and contaminations observed in the 1.0 wt. % Al sample. junctions with respect to the lower wt. % Al samples. Moreover, the appearance of Al-O rich grains is observed in this sample (Figure 4-25).

PAGE 59

47 S3 S4 S5 C O Al Si C O Al Si Grain Boundary Grain Figure 4-20. A STEM image of two grains and a grain boundary from 1.5 wt. % Al ABC sample and the corresponding EDS spectra. Si C Al O B GB Si Si C C Al Al O O B B GB Figure 4-21. EFTEM elemental maps of a gr ain boundary in the 1.5 wt. % Al sample. Note the formation of an Al-O rich inte rgranular film. Scale marker is 20 nm. 4 wt. % Al The 4 wt. % Al containing sample also exhi bits high concentrations of Al and O at the grain boundaries (see Figures 4-26 and 4-27). The triple junctions are high in aluminum, and oxygen. In addition, this sa mple had aluminum-rich grains that may be Al4C, carbon precipitates, and a B-C rich grains that may be B4C. In addition, the XRD

PAGE 60

48 performed by Ceramatec showed that a crystalline phase of Al8B4C7 at 3 wt. % Al. This phase was observed as an Al-B-C rich grain in the EFTEM images. Materials Characterization Performed at Ceramatec, Inc. This section will detail the material characterization performed at Ceramatec, Inc. Determination of fracture mode, grain morphology, and phase assemblage will be discussed. S2 S3 S4 S5 C O Al Si C O Al Si Triple Junction Grain Figure 4-22. A STEM image and EDS spectra of a triple junction in the sample containing 1.5 wt. % Al. Si C Al B O TJ Si Si C C Al Al B B O O TJ Figure 4-23. EFTEM elemental maps of a tr iple junction in the 1.5 wt. % Al sample. Notice that the entire triple junction is void of Si and C.

PAGE 61

49 Characteristic Al-k Peak in Al2O3Characteristic O-k Peak in Al2O3 Characteristic Al-k Peak in Al2O3Characteristic O-k Peak in Al2O3 Characteristic Al-k Peak in Al2O3Characteristic O-k Peak in Al2O3 Al -k O -k Characteristic Al-k Peak in Al2O3Characteristic O-k Peak in Al2O3 Characteristic Al-k Peak in Al2O3Characteristic O-k Peak in Al2O3 Characteristic Al-k Peak in Al2O3Characteristic O-k Peak in Al2O3 Al -k O -k Figure 4-24. EELS spectra taken from the tr iple junction in the 1.5 wt. % Al sample shows that the triple ju nction is filled with Al2O3. Figure 4-25. A STEM image and correspondi ng EDS spectra of two grains and a grain boundary in 1.5 wt. % Al sample. This site shows two different grains. The upper grain is a SiC grain and the lower gr ain is that of an Al-O rich grain. The presence of aluminum-rich grains were only observed for 1.5 wt. % Al and 4.0 wt. % Al. SiC Grain Al-O-rich Grain Al-O-rich Grain Grain Boundary

PAGE 62

50 S2 S3 S4 B-C Rich Grain SiC Grain Grain Boundary S2 S3 S4 S2 S3 S4 B-C Rich Grain SiC Grain Grain Boundary Figure 4-26. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain boundary, two SiC grains, and a BC rich inclusion in the 4 wt. % Al sample. Fracture Mode Determination by Ceramactec, Inc. The fracture mode was determined by insp ecting the fracture surface of samples after SEPB testing. Figure 4-33 shows the fract ure surfaces. From this data it is apparent that there is a change in fracture mode from transgranular to intergranular between 1 and 1.5 wt. % Al. The samples develop some de gree of mixed mode fracture above 3 wt. % Al. While SEPB measurements give more accurate values of toughness, often indent surfaces are more instructive as to the fract ure mode. Thus, the change to mixed mode can be seen more clearly in the Vickers hardness indents in Figure 4-34.

PAGE 63

51 S2 S3 S4 B-C Rich Grain Al4C Grain SiC Grain Boundary Figure 4-27. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain boundary, a B-C rich inclusion, and an Al4C inclusion in the 4 wt. % Al sample. S2 S3 S4 S5 S6 SiC Grain Grain Boundary Al-O-C Rich Grain Grain Boundary Figure 4-28. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain boundary, two SiC grains, and an Al-O-C rich inclusion in the 4 wt. % Al sample. Note that the area of the grain boundary that is farthest from the inclusion contains Al, while the area clos est to the inclusion is depleted in Al.

PAGE 64

52 S2 S3 S4 S5 S6 Carbon Inclusion Al-O-C Rich Grain Al-O-C Rich Grain SiC-SiC Grain Boundary SiC Grain S2 S3 S4 S5 S6 S2 S3 S4 S5 S6 Carbon Inclusion Al-O-C Rich Grain Al-O-C Rich Grain SiC-SiC Grain Boundary SiC Grain Figure 4-29. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain boundary, a SiC grain, and a carbon inclusion imbedded in an Al-O-C rich inclusion in the 4 wt. % Al sample. Grain Boundary Grain Grain Boundary Grain Figure 4-30. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain boundary, and two Si C grains in the 4 wt. % Al sample. Note the presence of Al at the grain boundary.

PAGE 65

53 Figure 4-31. EFTEM elemental maps of a gr ain boundary between tw o SiC grains in the 4 wt. % Al sample. C Si Al O B Figure 4-32. EFTEM elemental maps of a multiple grain junction in the 4 wt. % Al sample. The circled grain is Al-B-C rich and may be Al8B4C7. The area between the grains is Al-O rich. Si C Al B O GB

PAGE 66

54 Relationship Between Hardness and Toughness by Ceramactec, Inc. The hardness of the SiC samples initially di minishes as Al is added; however, once the fracture mode changes to intergranular the hardness increases again and reaches a maximum at 4 wt. % Al. The toughness remains constant until the fracture mode changes to intergranular. From this point, the toughness increases dramatically and reaches a maximum at 2 wt. % Al. This relationship can be seen in Table 4-2 and Figure 4-35. Grain Morphology Characterization by Ceramatec, Inc. Several possible toughening mechanisms in SiC are dependant on the size and aspect ratio of the grains. The data on the si ze and aspect ratio of the grains is presented in Table 4-3. Figure 4-36 illustrates how aspect ratio and grain size change as a function of Al content. In general, the grain size incr eases as the aspect ratio increases. Figure 437 presents the SEM micrographs upon whic h the grain morphology calculations were performed. XRD and Reitveld Analysis by Ceramatec, Inc. Per the discussion in chapter 2, the transformation from -SiC to -SiC and the formation of secondary phases can affect th e toughness. When SiC transforms from a cubic to hexagonal structure, the grains gr ow anisotropically into a three dimensional structure of interlocking elonga ted grains. The formation of this network of interlocking grains has been shown to offer some degree of hardness in other studie s. In addition, the formation of secondary phases has been known to produce residual stresses, which are needed to initiate several toughening mechanisms in ceramics. It is therefore important to understand the phase assemblage in the SiC samples under study. Figure 4-38 is the XRD data collected from the samples containing 0-6wt. % Al. The formation of the secondary phase Al8B4C7 begins at 3 wt. % Al. It is important to

PAGE 67

55 SiC-0Al-0.6B-2C Transgranular SiC-0.5Al-0.6B-2C Transgranular SiC-1Al-0.6B-2C Transgranular SiC-1.5Al-0.6B-2C Intergranular SiC-0Al-0.6B-2C Transgranular SiC-0.5Al-0.6B-2C Transgranular SiC-1Al-0.6B-2C Transgranular SiC-1.5Al-0.6B-2C Intergranular SiC-6Al-0.6B-2C Intergranular SiC-2Al-0.6B-2C Intergranular SiC-3Al-0.6B-2C Intergranular SiC-4Al-0.6B-2C Intergranular SiC-6Al-0.6B-2C Intergranular SiC-2Al-0.6B-2C Intergranular SiC-3Al-0.6B-2C Intergranular SiC-4Al-0.6B-2C Intergranular Figure 4-33. SEPB fracture surfaces for samp les with 0-6 wt. % Al. There is a clear change in fracture mode between tran sgranular and intergranular between 1 and 15 wt. % Al. The scale marker is 10 m (Flinders et al. submitted). note that Al2O3, Al4C, and B4C do not appear in the XRD spectra. This would suggest that the volume fraction of these sp ecies is much less than that of Al8B4C7. Figure 4-39 and Table 4-4 show the results of the Reitveld analysis. Unlike the Berkeley group who processed at 1900C, all of the -SiC has transformed to -SiC.

PAGE 68

56 Figure 4-34. Comparison of HV1 (left) and HK1 (right) for various SiC materials. A change in fracture mode is evident betw een 1 and 1.5 wt. % Al (Flinders et al. submitted).

PAGE 69

57 Figure 4-34 continued. Comparison of HV1 (left) and HK1 (right) for various SiC materials. The onset of the transgranu lar fracture mode is accompanied by a high degree of crack branching and cr ushing. A change toward mixed mode fracture occurs at 4 wt. % Al at which poi nt the cracks are straight and there is no crushing (Flinders et al. submitted).

PAGE 70

58 However, this data does agree with the Be rkeley group assertion that the 4H content increases up to 3 wt. % Al and then decreases thereafter. Correlation Between Microstructure and Mechanical Properties Toughness Per the discussion in chapter 2, the viab le toughening mechanisms in SiC crack deflection, crack bridging, and microcracking depend on at least one of the following: weak interfaces, residual stresses, high aspect ratio grains, and large grains. This section will analyze the observed effect that these pr operties have on SiC sintered with Al-B-C additives and suggest a possi ble toughening mechanism. Weak interfaces The high resolution lattice imaging and co mpositional analysis presented earlier in this chapter support the formation of a weak intergranular film between 1 and 1.5 wt. % Al. This film was primarily composed of an amorphous film that contained Al and O. Konoshita et al. concluded th at the segregation of Al and O atoms to grain boundaries will weaken the grain boundary strength and therefore provide an energetically favorable crack path (Konoshita et al. 1997). Table 4-2. Toughness and Hardness Meas urements of SiC with 0-6 wt. % Al. Al Density Fracture Toughness HK1 Hardness (g/cc) Mode (MPa-m1/2) (GPa) 0.0 3.170.01 T 2.60.2 20.40.4 0.5 3.170.01 T 2.60.1 19.40.2 1.0 3.170.01 T 2.70.1 19.20.3 1.5 3.130.0 1 I 6.10.3 13.70.2 2.0 3.130.01 I 6.70.4 14.30.3 3.0 3.140.01 I 6.20.4 16.60.7 4.0 3.120.01 I 6.10.2 16.90.8 6.0 3.070.01 I 5.60.3 15.00.6 T = Transgranular Fracture I = Intergranular Fracture

PAGE 71

59 1 2 3 4 5 6 Al Content (wt. %)SEPB Toughness (MPa-m1/2)KnoopHardness, HK1 (GPa) 1 2 3 4 5 6 Al Content (wt. %)SEPB Toughness (MPa-m1/2)KnoopHardness, HK1 (GPa) Figure 4-35. The above gr aphs plot Knoop hardness and SEPB toughness as a function of Al content. The change in ha rdness and toughness with changing Al content is inversely related (Flinders et al. submitted). Table 4-3. Characterization of SiC with 0-6 wt. % Al (Flinders et al. submitted) x Density Grain Size Asp ect Fracture Toughness HK1 Hardness (g/cc) (m) Ratio Mode (MPa-m1/2) (GPa) 0.0 3.170.01 4.70.4 5.41.1 T 2.60.2 20.40.4 0.5 3.170.01 4.60.2 5.91.3 T 2.60.1 19.40.2 1.0 3.170.01 4.10.5 5.40.9 T 2.70.1 19.20.3 1.5 3.130.01 5.20.4 3.90.6 I 6.10.3 13.70.2 2.0 3.130.01 5.80.5 5.01.3 I 6.70.4 14.30.3 3.0 3.140.01 11.91.2 4.51.0 I 6.20.4 16.60.7 4.0 3.120.01 5.30.5 5.11.0 I 6.10.2 16.90.8 6.0 3.070.01 7.10.6 8.72.7 I 5.60.3 15.00.6

PAGE 72

60 Grain Size and Aspect Ratio vs. Aluminum Content0 2 4 6 8 10 12 14 01234567 Aluminum Content (wt.%) Grain Size Aspect RatioAspect RatioGrain Size (m) Grain Size and Aspect Ratio vs. Aluminum Content0 2 4 6 8 10 12 14 01234567 Aluminum Content (wt.%) Grain Size Aspect Ratio Grain Size Aspect RatioAspect RatioGrain Size (m) Figure 4-36. Change in aspect rati o and grain size with Al content. The effect of the weak intergranular film on toughness is apparent. Upon formation of the intergranular film at 1.5 wt. % Al, the toughness more than doubled. In addition, the formation of the intergranular film wa s accompanied by a change in fracture mode from transgranular to intergranular. It is therefore apparent that the intergranular film plays a large role in increasing the toughness due to a change in fracture mode. However, since all three of the previously mentioned to ughening mechanisms could be affected by this intergranular film, or at least weak interfaces, no conclusion can be drawn with respect to the dominant toughening mechanism. Residual stresses The analysis of residual stresses due to thermal expansion anisotropy in single phase systems and thermo-elastic property mi smatch in multiphase systems is important

PAGE 73

61 Figure 4-37. Polished and chemically or plasma-etched surfaces of SiC samples with 0wt. % Al (Flinders et al. submitted).

PAGE 74

62 0 200 400 600 800 1000 1200 1400 1600 1800 2000 3032343638404244462-Theta (degrees)Intensity (counts/sec)6.0% Al 4.0% Al 3.0% Al 2.0% Al 1.5% Al 1.0% Al 0.5% Al 0.0% Al A l8B4C74H4H 6H 6H Figure 4-38. X-ray diffraction patterns SiC w ith 0-6 wt. % Al. Note that starting powder was beta-3C and that in a ddition to SiC polytypes, Al8B4C7 phase is noted at high Al contents (Flinders et al. submitted). SiC Al 0.6 wt% B 2 wt% C, 2100oC / 1hr0 10 20 30 40 50 60 70 80 90 100 01234567Al Content (wt. %)SiC Polytype (wt. %) 4H 6H 15R Figure 4-39. Polytypes from Rietveld anal ysis for SiC-0.6 wt. % B-0.2 wt. % C samples with Al contents ranging from 0 to 6 wt. %.

PAGE 75

63 Table 4-4. Reitveld Analysis for SiC Samp les with 0-6 wt. % Al (Flinders et al. submitted). Al Content (wt. %)Phase Assemblage (wt. %) 3C 4H 6H 15R 0 0 0 93.1 6.9 0.5 0 0 87.5 12.5 1 0 22.4 68.2 9.4 1.5 0 34 58.9 7.1 2 0 40.7 51.2 8.1 3 0 89 6.4 4.6 4 0 81.4 14.5 4.1 6 0 52.3 37.6 10 in determining crack behavior. As mentione d in chapter 2, in multiple phase systems such as one containing grains and intergra nular films or triple junction phases, the difference in thermal expansion mismatch be tween the grains and secondary phases can give rise to residual stresses. If the diff erence in thermal expansion between the grains and secondary phases is negative ( m < p ), a compressive radial stress is developed at the matrix-particle boundary and a tensile ci rcumferential stress is developed in the matrix. The particle will be in tension and the crack will be attracted to the particle. When the difference in thermal expansion is positive ( m > p ), the crack will be deflected between particles and a more tortuous crack path will be created, which will increase toughness (Hertzberg 1996). If one considers a material with an intergranular film in which the intergranular film is the ma trix and the grains are the particles, this second case can be used to explain the origin of intergranular fracture. With respect to intergranular films a local compressive tangential residual stress around the grains can cause diminishing stress in tensity directly at the tip of the deflected

PAGE 76

64 crack and are a result of a difference in thermo-elastic properties between the intergranular film and grains (Sternitzke 1997). Crystallizing triple junctions can also cause residual stresses, the magnitude of which is dependent on the volume change upon crystallization (Pezzotti and Kleebe 1999). The extent of triple junction crystalliz ation can affect the mechanical properties of the final part. Kessler et al. found that th e volume change due to crystallization is strongly dependant on the degree of crystalliz ation with increased crystallization resulting in increased volume change. This volume change results in residual stresses whose magnitude is proportional to the volume change of the constrained phase, or triple junction (Kessler et al. 1992). Pezzotti and Kleebe found that a nearly 100% fraction of crystallized secondary phase at the triple ju nction will create substantial tangential tensile stresses at the interface between the grains a nd triple junction due to volume contraction, which drives cracks toward the triple ju nction and causes subsequent interface delamination. This phenomenon can lead to crack splitting and intergranular fracture (Pezzotti and Kleebe 1999). It is therefore necessary to determine the local residual stresses between the grains and grain boundaries, and between the grain and triple junctions. For the sake of simplicity, the residual stresses with respec t to the grain boundaries will be calculated with the assumption that the grains are fibers in an infinite matrix and are bonded to the grain boundaries. This is obviously a simplifi cation considering the fact that the matrix, or intergranular film, is narrower than the particle, but serves as an approximation and means of relative comparison. Superposition of stress fields would be expected, however, they are ignored in this calculation. The grain boundary film is an amorphous glass containing Al and O, and is most lik ely an aluminosilicate glass with a low silica

PAGE 77

65 volume fraction. The properties of an aluminosilicate with 64.4 % Al2O3 and 36.6 % SiO2 were used for the amorphous intergranula r film. Thus, the residual stress due to a fiber in an infinite matrix can be calculated as follows (Green 1998): Fiber: rr P Matrix: 222 222 222 222() () () ()rrParb rba Parb rba Where a =fiber radius, b= the radius of the fiber plus the radius of the matrix 0()()(1) (1)(12)(1) 43070(9.7 m/mC-4.5 m/mC)(1373K-298K)(1+0.17) (10.17)(70)(1.34)430(10.22) 340fmmfAf fmffmEETT P EE GPaGPa GPaGPa MPa Thus, the stresses in the fiber are compressi ve and the stresses in the matrix are tensile in the radial direction. This stress st ate will cause matrix cracking. This residual stress state can therefore explain the tran sition to intergranular fracture when an intergranular film is present. It is also important to understand the effect of the triple junction phases on residual stress. For the sake of simplicity, the effect of superposition of stresses will be ignored, and the triple junction phase is assumed to be a spherical particle in an infinite matrix. The properties of alumina will be used for the triple junction and those of SiC will be used for the matrix. The stress state will be calculated as a function of temperature. The residual stress due to a particle in an infinite matrix can be calculated as follows (Green 1998):

PAGE 78

66 Particle: rr P Matrix: 3 32p rrrP r 0 0 6 02()() 2(12)(1) 2(430)(370)(4.5 m/mC-8.2 m/mC)(-298K) 2(430GPa)(1-2*0.22)+370GPa(1+.17) 2.7210(-298K)mpmpA mppmEETT P EE GPaGPaT xPaT T0 may be one of several temperatures: the processing temperature, the melting point of alumina, the meting point of mulite or 2/3 of the homologous temperature. Therefore, the stress, P, is plotted vs. temper ature to determine the sign of P, which allow for the determination of the residual stress fi elds in the secondary phases at the triple junctions and in the surrounding SiC grains. Load vs. Temperature-7 -6 -5 -4 -3 -2 -1 0 050010001500200025003000 TemperatureP (GPa) Stress vs. Temperature Load vs. Temperature-7 -6 -5 -4 -3 -2 -1 0 050010001500200025003000 TemperatureP (GPa) Stress vs. Temperature Figure 4-40. Plot of Stress vs. Temperature for a particle in an infinite matrix.

PAGE 79

67 From the graph, the calculated load ca n be between -3 GPa using 2/3 of the homologous temperature and -6 GPa using the processing temperature. The entirety of this range is negative, thus the stresses in the particle are tensile and the stresses in the matrix are tensile in the radial direction and compressive in the circumferential direction. As the crack is attracted to regions of tension, this stress state will cause the crack to be driven toward the triple junction and cause delamination. The combined effect of tensile stresses in the grain boundary ph ase and the triple junction phase will cause the crack to propaga te within the intergranular film and around the triple junctions. This behavior was obs erved in the 1.5 wt. % Al sample amd is shown in Figure 4-42. Triple Junction in Tension Triple Junction in Compression Crack Tip Crack Tip Crack Attracted to Triple Junction Crack Deflected from Triple Junction Triple Junction in Tension Triple Junction in Compression Crack Tip Crack Tip Crack Attracted to Triple Junction Crack Deflected from Triple Junction Figure 4-41. Effect of resi dual stress fields around triple junctions on crack propagation (Pezzotti and Kleebe 1999).

PAGE 80

68 Figure 4-42. Intergranular cracking in the 1.5 wt. % sample. The figure on the left shows cracking through the intergranular film at the grain boundaries. The figure on the right shows a missing trip le junction due to the crack deflecting around the triple junction, but within the amorphous secondary phase. Note that the image on the right is slightly out of focus, which makes the crack appear to be a broad region that is lighter in color. Residual stresses contribute to all three toughening mechanisms discussed earlier in the section. Therefore, it would be difficu lt to conclude a specific toughening mechanism at this point. However, a probable origin of the intergranular cracking has now been proposed. High aspect ratio grains Both crack deflection and crack bridging de pend on the aspect ra tio of the grains (Green 1998). If either of these mechanisms were active, an increase in toughness with aspect ratio would be expected. In the case of crack deflection, th e SiC grains are taken to be the particles and thus the volume fraction of high aspect ratio grains is close to unity. Faber and Evans found that after the volume fraction reaches 0.2 no additional increase in toughening will be observed w ith increased volume fraction of elongated grains (Green 1998). In addition, the crack wake on the indented samples does not show

PAGE 81

69 the deflection of the crack and th erefore, it is unlikely that crack deflection is the main toughening mechanism in the samples. If crack bridging were to occur, the toughness would increase as a function of aspect ratio according to the following equation (Flinders et al. submitted): d E K KL L o Ic Ic 67 1 2 2 281 0 However, when the square route of toughne ss is plotted against aspect ratio raised to the 1.67, Figure 4-43, the slope is nega tive. In addition, no crack ligaments are observed in the crack wake in the hardness inde nts. Thus, the evidence points away from crack bridging. Large grains Microcracking occurs when residual stre sses build-up due to the following: phase transformations, thermal expansion anisotropy in one phase materials, or thermo-elastic property mismatch in multiphase materials (Green 1998). The previous section on residual stresses showed that residual st resses existed at the interfaces between the intergranular film and SiC grains, and the trip le junctions and SiC gr ains. These stresses were also shown to cause cracking at the inte rface and debonding of the triple junctions, thus satisfying one of the condit ions necessary for microcracking. The degree of toughness increase due to microcracking is dependent on the grain size. As the particle size increases, the to ughening increment should increase due to the creation of a tortuous crack path until the critical grain size is reached. Baud and Theveno found that the assertion that the t oughness increases with grain size, or more precisely the square route of grain size, hol ds true for liquid-phase sintered SiC (Baud and Theveno 2001). However, the compositi on of the starting powers was all the same and only grain size varied in the materials that Baud and Theveno. investigated. For the

PAGE 82

70 case of the SiC sintered with Al-B-C additi ves under investigati on, not only does the grain size change, but the types and volume fr action of secondary phases change. This would in turn affect the magnitude of the residual stresses develo ped and therefore the degree of toughening gained by microcracking. Therefore, there may be a synergistic effect between the grain size and the volume fraction of s econdary phases, which would result in the plot of toughn ess vs. grain size showing no co rrelation between the two. Therefore, the possibility that microcrack ing is the dominant toughening mechanism should not be dismissed. Indeed, an inspection of the Vickers inde nts reveals a high degree of crushing and crack branching in the 1.5 wt. % sample. The 4 wt. % Al sample experiences a much lower degree of crack branching and crushing than the 1.5 wt. % sample. This may be due to the formation of additional secondary phases in this sample such as Al8B4C7, Al4C, 0 5 10 15 20 25 30 35 40 45 50 051015202530354045Toughness2(MPa2m)L 1.67 Toughness2vs. Aspect Ratio1.67 0 5 10 15 20 25 30 35 40 45 50 051015202530354045Toughness2(MPa2m)L 1.67 Toughness2vs. Aspect Ratio1.67 Figure 4-43. Plot of toughness vs. aspect ratio shows a negative slope and argues against crack bridging. The three data points at the bottom of the graph correspond to the samples that fail transgranularly.

PAGE 83

71 Figure 4-44. Plot of toughness vs. grai n size. Toughness generally increases with increased grain size. B4C, and an Al-O-C rich phase. Not only ar e more species present, but they are present in much greater volume fraction than the sec ondary phases in the 1.5 wt. % Al sample. As shown in the calculation performed in the residual stresses se ction, the secondary phases surrounded by the intergranular film will create compressive stresses in the secondary phase, which will deflect the crack. If these phases are present in high enough concentration, it could explain the change in fracture mode from intergranular to mixed mode at this composition, and the drop in toughness. Hardness The hardness initially drops when small amounts of aluminum are added with the largest drop in toughness occurring when th e fracture mode changes to intergranular fracture. From this point, the hardness begins to increase again, reaching a maximum at 4 wt. % Al, and then decreases again. The la rge drop in hardness upon the formation of an intergranular film is due to the increased ability of the SiC grains to move under the

PAGE 84

72 indentation load, which results in lower hard ness (Flinders et al. 2003). The increase in hardness up to 4 wt. % may be explained in a similar fashion to the drop in hardness. The increased number of secondary phases that exist in a state of compression would resist grain movement under th e indentation load. By building-up compressive stresses in the material, a greater applied load is needed to deform the material. Effect of grain size on hardness While the grain size had little effect on toughness, it does affect the hardness. Often a Hall-Petch type relationship is appl ied to correlate the hardness with the grain size; however, this model is not as useful for ceramic as it is for metals. The relationship was originally designed to explain an increase in strength due to di slocation pile-up at Hardness vs Grain Size13 14 15 16 17 18 19 20 21 35791113 Grain Size (microns)Hardness (GPa) Transgranular Fracture Intergranular Fracture Figure 4-45. Plot of the eff ect of grain size on hardness. The plot shows that hardness increases with grain size for the samples that fail via intergranular fracture.

PAGE 85

73 grain boundaries, since the number of grai n boundaries increases with decreasing grain size. Yet, the hardness does scale inversely with grain size. This effect is more likely due to the reduction of the inherent size of the flaw s when the grain size is reduced than to an increase in dislocation piles. Effect of aspect ratio on hardness By plotting hardness vs. the aspect ratio, it observed that the aspect ratio has no effect on the hardness. This behavior can be seen in Figure 4-46. Hardness vs. Aspect Ratio13 14 15 16 17 18 19 20 21 3456789101112 Aspect RatioHardness (GPa) Intergranular Fracture Transgranular Fracture Figure 4-46. The plot of hardness vs. asp ect ration shows that the hardness is not dependant on aspect ratio.

PAGE 86

74 CHAPTER 5 SUMMARY AND CONCLUSIONS Conclusions With the use of high resolution lattice im aging, energy dispersive spectroscopy, and energy filtered transmission electron microscopy, it was determined that an amorphous intergranular film forms between 1 and 1.5 wt. % Al. The film is composed primarily of an Al-O rich phase and contains an equilibrium film thickness of approximately 1 nm. The residual tensile stresses in this film allow for cracking through the film, promote the change in fracture mode from transgranular to intergranular fracture, and increase the toughness values by a factor of two. The addition of aluminum to the sample s promoted the formation of secondary phases at the triple junctions. These sec ondary phases were Al-O rich and the oxygen content and crystallinity increased as the Al content was increased, with the formation of alumina occurring at 1.5 wt% Al. Additional secondary phases were formed as discrete grains and inclusions at Al contents greater than 1.5 wt. % Al. These phases may include: Al4C, an Al-O-C phase, Al8B4C7, and B4C. The crystallization of the secondary phases generated a compressive stress field in the particles that caused the crack to deflect through the amorphous phase around the crystal line triple junctions in the 1.5 wt. % Al sample, which has a lower volume fraction of secondary phase. However, as the volume fraction of secondary phases increased, as in the 4 wt. % Al sample, the amount of residual compressive stress in the samples in creased and caused a simultaneous increase in hardness and decrease in toughness.

PAGE 87

75 Suggested Future Work While some conclusions were made to possible toughening mechanisms in SiC sintered with Al-B-C additives, additional in vestigation into the toughening mechanisms is needed. In-situ cracking studies perf ormed in a scanning electron microscope in secondary electron mode could be used to monitor the advance of the crack and image the crack tip and crack wake. This study would produce potential identification of the toughening mechanism. In addition, a comparison of the elastic co mpliance of the cracked specimen can be compared to the theoretical compliance of an ideal bridgeand micr ocrack-free crack of the same length. Since crack-bridging incr eases the modulus, a reduction in compliance would suggest the presence of crack-bridging. Conversely, a decrease in the compliance would suggest microcracking (Nalla et al. 2004).

PAGE 88

76 LIST OF REFERENCES Baud, S., and Theveno, F., “Microstructural and Mechanical Properties of Liquid-phase Sintered Seeded Silicon Carbide,” Mater. Sci. Phys., 67, 165-174, 2001. Birnie, D. P., “A Model for Silicon Self-Di ffusion in Silicon Carbide: Anti-Site Defect Motion,” J. Am. Ceram. Soc., 69 [1], C33-C35, 1986. Biswas, K. Rixecker, G., Wiedmann, I ., Schweizer, M., Upadhyaya, G. S., and Aldinger, F., “Liquid Phase Sintering and Micros tructure-Property Relationships of Silicon Carbide Ceramics with Oxynitride Additives,” Mater. Chem. Phys., 67, 180-191, 2001. Bonnell, D. A., Tien, T. Y., and Rhle, M., “Controlled Crystallization of the Amorphous Phase in Silicon Nitride Ceramics,” J. Am. Ceram. Soc., 70, 460-465, 1987. Cao, J. J., MoberlyChan, W. J., DeJonghe, L. C., Gilbert, C. J., and Ritchie, R. O., “In Situ Toughened Silicon Carbide with Al-B-C Additions,” J. Am. Ceram. Soc., 79 [2], 461-69, 1996. Case, E. D., Smith, I. O., and Baumann, M. J., “Microcracking and Porosity in Calcium Phosphates and Implications for Bone Tissue Engineering,” Mater. Sci. Eng. A, 390, 246-254, 2005. Chen, D., Sixta, M. E., Zhang, X. F., De Jonghe, L. C., and Ritchie, R. O., “Role of the Grain Boundary Phase on the Elevated-Temperature Strength, Toughness, Fatigue an Creep Resistance of Silicon Carb ide Sintered with Al, B, and C,” Acta Mater., 48, 4599-4608, 2000. Chia, K. Y., and Lau, S. K., “High Toughness Silicon Carbide,” Ceram. Eng. Sci.Proc., 12[9-10], 1845-1861, 1991Clarke, D. R., “On the Equilibrium Thickness of Intergranular Glass Phases in Ceramic Materials,” J. Am. Ceram. Soc., 70 [1], 1522, 1987. Clarke, D. R., Shaw, T. M., Philipse, A. P., and Horn, R. G., “Possible Electrical Double-Layer Contribution to the Equilibrium Thickness of Intergranular Glass Films in Polycrystalline Ceramics,” J. Am. Ceram. Soc., 76 [5], 1201-1204, 1993. Falk, L. K. L., “Microstructral Development During Liquid Phase Sintering of Silicon Carbide Ceramics,” J. Eur. Ceram. Soc., 17, 983-994, 1997. Fischer, G. R., and Barnes, P., “Toward a Unified View of Polytypism in Silicon Carbide,” Phil. Mag. B., 61, 217-236, 1990.

PAGE 89

77 Flinders, M., Ray, D., and Cutler, R. A., “Toughness-Hardness Trade-Off in Advanced SiC Armor,” Ceram. Trans., 151, 37-48, 2003. Flinders, M., Ray, D., Anderson, A., and Cutle r, R. A., “Liquid Phase Sintered Silicon Carbide as Protective Armor,” J. Am. Ceram. Soc ., submitted. Gilbert, C. J., Cao, J. J., Moberlychan, W. J., De Jonghe, L. C., and Ritchie, R. O., “Cyclic Fatigue and Resistance-curve Be havior of an In-situ Toughened Silicon Carbide with Al-B-C Additions,” Acta Metal. Mater., 44 [8], 3199-3214, 1996. Gooch, W. A. Jr., “An Overview of Ceramic Armor Applications,” Ceram. Trans., 134, 3-21, 2002. Green, D. J., An Introduction to the Mechani cal Properties of Ceramics, Cambridge University Press, Cambridge, UK, 1998. Greskovich, C., and J. H. Rosolowski “Sintering of Colavent Solids,” J. Am.Ceram. Soc., 59, 336-343, 1976. Gu, H., Pan, X., Tanaka, I., Cannon, R. M., Hoffmann, M. J., Mllejans, H., and Rhle, R., “Structure and Chemistry of Intergranular Films in Ca-Doped Si3N4,” Materials Science Forum, 207-209, Hertzberg, R. W., Deformation and Fracture Mechani cs of Engineering Materials, John Wiley & Sons, Inc., New York, NY, 1996. Hofer, F., Warbichler, P., and Grogger, W., “I maging of Nanometer-sized Precipitates in Solids by Electron Spectroscopic Imaging,” Ultramicroscopy, 59, 15-31, 1995. Hofer, F., Grogger, W., Kothleitner, G., an d Warbichler, P., “Quantitative Analysis of EFTEM Elemental Distribution Images,” Ultramicroscopy, 67, 83-103, 1997. Hong, J. D., “Self-Diffusion in Alpha and Beta Silicon Carbide”, Proc. Of the 4th Int. Met. Mod. Ceram., Mat Sci. Monographs, 6, En ergy and Ceramics, P. Vincenzini (Ed.) Elsevier, 1979. Huang, J. L., Hurford, A.C., Bruner, S. L., Cutler, R. A., and Virkar, A. V., "Sintering Behavior and Properties of SiCAlON Ceramics," J. Mater. Sci., 21, 1448-1456, 1986. Inomata, Y., Tanaka, H., Inoue, Z., and Ka wabata, H., “Phase Relation in SiC-Al4C3-B4C System at 1800C,” J. Ceram. Soc. Jpn. (Yogyo Kyokaishi), 88 [6], 353-355, 1980. Jun, H-W., Lee, H-W., Kim, G-H., and Song, H. S., “Effect of Sintering Atmosphere on the Microstructure Evolution and Mechanical Properties of Silicon Carbide Ceramics,” Ceram. Sci. Eng. Proc. 18 [4], 487-504, 1997.

PAGE 90

78 Kaneko, K., Yoshiya, M., Tanaka, I., and Ts urekawa, S., “Chemical Bonding of Oxygen in Intergranular Amorphous Layers on High-purity -SiC Ceramics,” Acta Mater., 47, 1281-1287, 1999. Kaneko, K., Honda, S., Nagano, T., and Saitoh, T., “Analytical Inve stigation of Grain Boundaries of Compressive Deformed Al-doped Sintered -SiC” Mater. Sci. Eng. A, 285, 136-143, 2000. Kessler, H.,, Kleebe, H-J., Cannon, R. W ., and Pompe, W., “Influence of Internal Stresses on Crystallization of Intergranular Phases in Ceramics,” Acta Metall. Mater., 40 [9], 2233-2245, 1992. Kim, Y. W., Mitomo, M., and Hirotsuru, H ., “Grain Growth and Fracture Toughness of Fine-Grained Silicon Carbide Ceramics,” J. Am. Ceram. Soc., 78 [11], 3145-3148, 1995. Kim, Y-W., and Mitomo, M., “Fine-Grained Silicon Carbide Ceramics with Oxynitride Glass,” J. Am. Ceram. Soc., 82 [10], 2731-2736, 1999. Konoshita, T., Munekawa, S., and Tanaka, S -I., “Effect of Grain Boundary Segregation on High-temperature Strength of Hot-pressed Silicon Carbide,” Acta Mater., 45 [2], 801-809, 1997. Meyers, M. A., and Kumarchawla, K., Mechanical Behavior of Materials, Prentice-Hall, Inc., Upper Saddle River, NJ, 1999. MoberlyChan, W. J., Cannon, R. M., Chan, L. H., Cao, J. J., Gilbert, C. J., Ritchie, R. O., and De Jonghe, L. C., “Microstructural Development to Toughen SiC,” Mat. Res. Soc. Symp. Proc., 410, 257-262, 1996. Moberlychan, W. J., Cao, J. J., and De Jo nghe, L. C., “The Roles of Amorphous Grain Boundaries and the ransformation in Toughening SiC,” Acta Mater., 45 [5], 1625-1635, 1998. Moberlychan, W. J., and De Jonghe, L. C., “Controlling Interface Chemistry and Structure to Process and Toughen Silicon Carbide,” Acta Mater. 45 [7], 2471-2477, 1998. Nalla, R. K., Krunzic, J. J., and Ritchie, R. O., “On the Origin of the Toughness of Mineralized Tissue: Microcr acking or Crack Bridging?” Bone, 34, 790-798, 2004 Pezzotti, G., and Kleebe, H. J., “Effect of Residual Microstresses at Crystalline Multigrain Junctions on the T oughness if Silic on Nitride,” J. Eur. Ceram. Soc., 19, 451-455, 1999. Prochazka, S., “The Role of Boron and Carb on in the Sintering of Silicon Carbide,” pp. 171-181 in Special Ceramics 6. Edited by P. Popper. British Ceramic Research Association, Stoke-on-Trent, U. K., 1975.

PAGE 91

79 Prochazka, S., “Why is It Difficult to Sint er Covalent Substances?” General Electric R&D Center, Schenectady, USA, Techni cal Information Series, Class 1 GECI Program 89CRD025, 1987. Raj, R.“ Fundamental Research in Stru ctural Ceramics for Service Near 2000C,” J. Am. Ceram. Soc., 76 [9], 2147-2174, 1993. Ramsdell, R. I., “Studies on Silicon Carbide,” Amer. Mineral., 32, 64, 1947. Schaffer, B., Grogger, W., and Hofer, F., “Width Determination of SiO2-films in Si-based Devices Using Low-loss EFTE: Image C ontrast as a Function of Sample Thickness,” Micron, 34, 1-7, 2003. Sigl, L. S., and Kleebe, H. J., “Core/Rim Structure of Liquid-Phase-Sintered Silicon Carbide,” J. Am. Ceram. Soc.,76, 773-776, 1993. Sternitzke, M., “Review: Structural Ceramic Nanocomposites,” J. Eur. Ceram. Soc., 7, 1061-1082, 1997. Stobierski, L., and Guberat, A., “Sintering of Silicon Carbide I. Effect of Carbon,” Ceramics International, 29, 287-292, 2003. Tanaka, H., “Sintering of Silicon Carbide,” in Silicon Carbide Ceramics-1, Ceramics Research and Development in Japan Seri es, Eds. S. Somiya und Y. Inomata, London, Elsevier Applied Science, 213-236, 1991. Williams, D. B., and Carter, C. B., Tansmission Electron Microscopy, Plenum Press, New York, NY, 1996. Ye, Haihui, “Microstructure and Chemistry of Grain-Boundary Films and Triple Junction Phases in Liquid-Phase Sintered SiC Cera mics,” Ph.D. Dissertation, Max-PlanckInstitut fr Metallforschung, Stuttgart, Germany, 2002. Yuan, R., Kruzic,. J. J., Zhang, X. F., De Jo nghe, L. C., and Ritchie, R. O., “Ambient to High-temperature Fracture Toughness an d Cyclic Fatigue Behavior in Alcontaining Silicon Carbide Ceramics” Acta Mater., 51, 6477-49, 2003. Zhang, Y. H., Edwards, L., and Plumbridge, W. J., “Effect of Crystallization on Fatigue Crack Growth in a Sic-Reinforced Silicon Nitride Composite at 1200C,” Acta Mater., 46 [4],1327-132, 1998. Zhang, X. F., Yang, Q., and De Jonghe, L. C., “Microstructure Development in Hotpressed Silicon Carbide: Effects of Aluminum, Born, and Carbon Additives,” Acta Mater., 51, 3849-3860, 2003. Zhou, Y., Tanaka, H., Otani, S., and Ban do, Y., “Low-Temperature Pressureless Sintering of -SiC with Al4C3-B4C-C Additions,” J. Am. Ceram. Soc., 82 [8], 19591964, 1999.

PAGE 92

80 Zhou, Y., Hiroa, K., Yamauchi, Y., and Kanzaki, S., “Tailoring the Mechanical Properties of Silicon Carbide Ceramics by Modification of the Intergranular Phase Chemistry and Microstructure,” J. Eur. Ceram. Soc., 22, 2689-2696, 2002.

PAGE 93

81 BIOGRAPHICAL SKETCH Samantha Crane moved from her childh ood home in Roslyn Heights, NY, to Weston, FL, at the age of 15. She graduated high school from The University School of Nova Southeastern University in 1999. In the summer of 1999 she enrolled in the Honors College at the University of Florida. She graduated cum laude from the University of Florida with her Bachelor of Science in ma terials science and engi neering in May 2003 and will receive her Master of Science in materials scienc e and engineering in August 2005. Samantha has a particular interest in nuclear materials, which began with her senior research project on bur nable poison rod assemblies. Sh e has pursued this interest by participating in two summer internships with the United States Nuclear Regulatory Commission (NRC) in Rockville, MD. Samant ha will be returning to the NRC for fulltime employment in the fall of 2005.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101130_AAAAAM INGEST_TIME 2010-11-30T07:49:03Z PACKAGE UFE0010845_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 650 DFID F20101130_AAALTR ORIGIN DEPOSITOR PATH crane_s_Page_07.txt GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
40e610872f6fb4ae60aca7190c27ffce
SHA-1
b710537176d9273aeb0091b5aa5994e71c028b30
5796 F20101130_AAAMAA crane_s_Page_74thm.jpg
dea4744effff10f2b1f5cc6ee98bbcb4
a773fb9af62feac84bc3fc95d63abcc0aa1f4cdf
1850 F20101130_AAALUG crane_s_Page_27.txt
cf7a80cc2aeff3a2fc217b4ff87fdc55
37f62251bed32d54d7c2b92a6291d94f1efc4767
2766 F20101130_AAALTS crane_s_Page_08.txt
8f1731afd92b0e881b1afd184ef574f1
545432ab473f610a729d21d51929fcb51ac1c33a
6682 F20101130_AAAMAB crane_s_Page_76thm.jpg
e5ca03f616c135b32ed88e4d64a4083a
ffe4881007374c8952af026a1bf00c241b49dde5
1885 F20101130_AAALUH crane_s_Page_29.txt
e55ddd7cf090ff588314bac5a0bb5ef2
28fb094dde15a9847a5985b069f90d29f3538100
3167 F20101130_AAALTT crane_s_Page_09.txt
2a653a9ea0d27acbcc2349d1f1211f70
b4429bb8360d4b54c024b9e1fc55d646a316262f
17778 F20101130_AAAMAC crane_s_Page_77.QC.jpg
e88565c87ec59042faf5908c1adce865
9565f57e028026d7dbbcdf0b722c3bae96fa3318
2057 F20101130_AAALUI crane_s_Page_30.txt
aca301f74da78ef87af42156948e96c7
40b901b9fba43bab7dce41774fca63ef5ef89b8f
3338 F20101130_AAALTU crane_s_Page_10.txt
d8f59a5394e204ae583881983470adb9
a2b98adfb2b114ac443f74339e4b973f6c445ed8
5553 F20101130_AAAMAD crane_s_Page_77thm.jpg
018e3f651e697a6ae8fc9da967de29bd
efe2520ba054c853828bfec4360183927c2c10c9
1951 F20101130_AAALUJ crane_s_Page_33.txt
2e2c60c19fb21a02c65d4aa05dd9bd22
e800c20eebf74773a38152010797de76357983a6
3045 F20101130_AAALTV crane_s_Page_11.txt
c50c5aace99e30504656af9b4d2d4dd7
3ffc2d053d63092e1e2253c2c5e10a60b23f5e0d
18737 F20101130_AAAMAE crane_s_Page_79.QC.jpg
73b71457aa8918f5c70e1cdb85a31176
8bbe11080654407cf976d7a6254ef7cc926b8d4f
1825 F20101130_AAALTW crane_s_Page_13.txt
a3bc5ad1b1c5c0686d3feac66dff80c5
fc0078e885f7060c2c4acb97609b2e2931673312
5584 F20101130_AAAMAF crane_s_Page_79thm.jpg
797facf26e454a2eac25dea5cd7759ea
061db323da2a014e73d7aafb58b120b366a7d77f
2034 F20101130_AAALUK crane_s_Page_34.txt
431a45d287938d8ff3c2280010d10674
356458289fb3480d7ffc88d026cc8eefb10c85e5
755 F20101130_AAALTX crane_s_Page_16.txt
0621f4c1159649a162845fef2899179a
c8c7218e4c0459a6541651ce1320870ecdea9abb
21446 F20101130_AAAMAG crane_s_Page_80.QC.jpg
26c54325f975e04ad81f6086e67c7cc8
fbc8ab543e4e22012c6299fdd198c7635254c1d2
1149 F20101130_AAALVA crane_s_Page_57.txt
f21d0541554480edd91221499c1f747b
d6cab239ed3c0c511c7131ed351370b5b6da6a00
1643 F20101130_AAALUL crane_s_Page_36.txt
01951277f5f91530a223d4b7a2973fb1
2ca023a68927c1ccf879c8c88564057ce28bbf3e
1899 F20101130_AAALTY crane_s_Page_17.txt
0f829ca55ff82d952f6cea22959e51fc
f4c41ea82d11f4d9594c9c030ba2e829e8fd7191
6371 F20101130_AAAMAH crane_s_Page_81thm.jpg
4b29fefdda38d52d69841b0160fa223c
beca02423248ac6c896d950bf76ff03fec256d53
1167 F20101130_AAALVB crane_s_Page_58.txt
e102dade04f793ffbea656bca7beaad5
bc41efffd29906830b4d0882faa69b9e1330c724
1732 F20101130_AAALUM crane_s_Page_37.txt
2ac5b88b7727f858d567caf9efdb9396
f3d818c9ba3ce2d26796a93d4f16b51888ebe9c8
1927 F20101130_AAALTZ crane_s_Page_19.txt
d6c75276209b8a9f62c943d325be2944
ead756b6ccdc884073ed77efc026c7b3f6e89fc5
5265 F20101130_AAAMAI crane_s_Page_82thm.jpg
604b4a95050a34c3379e4fdc4dee9168
8289063170775bec6289d6e8b4b79bae9284311e
953 F20101130_AAALVC crane_s_Page_59.txt
81015c2866d95ee72cd3612de4fa3bcb
8bd356b7052a8b24b641c0f449e656f347fef8fd
1754 F20101130_AAALUN crane_s_Page_38.txt
4981fb2c94295efc63458dd9c1894edf
9813a9aadece96c2ee3b24843e2a0ffe986bc9fa
4832 F20101130_AAAMAJ crane_s_Page_84thm.jpg
df6bd807025442a0448b525ccf320a06
e4997c46245fcd798b18d8c2486fa06eeb3d2c8d
1241 F20101130_AAALVD crane_s_Page_62.txt
3c895ce10717f758c4b93e8321fd9625
8c5963695741842023a18171747a88c31f5df424
F20101130_AAALUO crane_s_Page_41.txt
96ada9f453d4d80750f9ef270b27701d
3e371424ba73de6bddc34807e9a85cda8177518e
4102 F20101130_AAAMAK crane_s_Page_85thm.jpg
4b191ee2709363be55ed292f05f53a49
70d5a0abc4403fb8d7de2d67588027af2b133046
1061 F20101130_AAALVE crane_s_Page_63.txt
4d7b2f4f449a180e0bca9e91171c911c
ea2cba392be919cf4d69b4aa24a7f173432f878c
432 F20101130_AAALUP crane_s_Page_42.txt
09155c5378776999af85a5537b0a42a4
a8d6ac9158c00d04b21f695c242123a6629fcffa
6323 F20101130_AAAMAL crane_s_Page_88thm.jpg
06555d857a755dc16d19c9d9f85cded0
f269127d4d7d73971a91b0809136599d12cf12ce
984 F20101130_AAALVF crane_s_Page_64.txt
dc5fa7e76a9d6c04532f693e625b74cb
8c577f0274fe4b5b7c8fe8d1bab18b86fe2c8ff9
1751 F20101130_AAALUQ crane_s_Page_43.txt
e0fd543b2aeea0e1655689d0dbc6c17d
05a373ec37957a569badd2ba5b86b0edd882c185
6848 F20101130_AAAMAM crane_s_Page_89thm.jpg
83aa591ebe98dd83ddc2f411d31e9874
3be037ec2c214967e4c153554b2760eac40bd425
381 F20101130_AAALVG crane_s_Page_65.txt
1810fe07e6d53ecf536069ed97e36e18
0383dac7e84efad6c1b2a6eb48afb001fb33f7ac
2026 F20101130_AAALUR crane_s_Page_44.txt
9ee6c6815ea4a0d7c55b0fd05831f876
ef27d2387607e9c04c4435c1da284e3448cbb1bc
7115 F20101130_AAAMAN crane_s_Page_90thm.jpg
b1a96103b016ed3dc0f177622cebd77c
7cefab19f3b9787499fa9678c96b831963b17c7e
1974 F20101130_AAALVH crane_s_Page_66.txt
9087833cb9d2b2d4bd15d8ae8a2c4d03
c8aa4f8c96a487167e55f79f91eeb65eb22b0d4f
636 F20101130_AAALUS crane_s_Page_45.txt
edf8c657094998e012c0b2f2f6e47397
6c3656c901b5eb265efbdd65a9f7538777e5f3ad
24932 F20101130_AAAMAO crane_s_Page_91.QC.jpg
62982b0cb1d4798fb8b983011bfaf590
81b55abbd212c2ed244fce2226b756652d7e8dde
689 F20101130_AAALVI crane_s_Page_67.txt
f49e9ab3deb632a03c78fe84bebb30cf
dd50c1e1c64cd24c8ace75677ac11267db5dec2f
1999 F20101130_AAALUT crane_s_Page_46.txt
ce8236cb74d40c2b4e006ffe93a5f329
37b46c4ef99f7624ada0581af990306ddb364671
5226 F20101130_AAAMAP crane_s_Page_92.QC.jpg
39dcb0a59256e4e885c80e82e1b0252c
ecd6ddb087d56af7ec9e945c8babec1ad8833238
1537 F20101130_AAALVJ crane_s_Page_68.txt
207251d9b720a2fcf3af31bac62d3d2a
f7c5ee288b5d0304daffd7adb286c5cbf28bdd25
405 F20101130_AAALUU crane_s_Page_47.txt
0433d4821ca6a81ff135575ac06b62b1
9df8f9765139b22d91d2ad4fb5641510d5edf29c
1884 F20101130_AAALVK crane_s_Page_70.txt
b43d7b1587d76e0e7c69082d79e6d6bc
f9462c9d4392001d51249f4bcb751b408c813ef4
1912 F20101130_AAALUV crane_s_Page_48.txt
79dc325dc1100062d6300ffd4dfa5cc4
2199424f8cafa2f155e6b7bff871cae4d6f9e4d1
748 F20101130_AAALUW crane_s_Page_53.txt
7d35c64ddf14f45b3e3ac239df223d4a
bf3dfba1519a88aa7dab10891eb92f35c89caec6
302 F20101130_AAALWA crane_s_Page_92.txt
ec54a20d9bf618514ee03bbdd4c036ec
71c1b668c36afc91e79e965f33643288cde801b1
1233 F20101130_AAALVL crane_s_Page_71.txt
6f1ba5fd1da24e6073ff42af99048179
973ffd0791bb3c446c69536197fe35a1413a7e80
1033 F20101130_AAALUX crane_s_Page_54.txt
a1195a6ba786fd3f6d029bc79dcf5565
ab7ea0c1a7be6ef4dbca8a9f29de0ffbc9d67d4c
9031254 F20101130_AAALWB crane_s.pdf
6696d1bb9b8e07fb5234e35381f14cad
59130d63e99d47f1380c7ada69b9fa00fe75f35c
1628 F20101130_AAALVM crane_s_Page_72.txt
9b8277475941e7485e33a6a59622b7f7
f57dd1b696f9789451b9a8acac7cbeeaedfe6425
F20101130_AAALUY crane_s_Page_55.txt
90bc5b4e9f1f57457513a6ba23c10fad
c4ba1e48108b217b781d34d5726ee1cd257fa105
5757 F20101130_AAALWC crane_s_Page_75thm.jpg
904cc32116cf54c998375eb572c624a3
d77b0f6ec935ba7612a172097844ac5bf3039862
249 F20101130_AAALVN crane_s_Page_73.txt
93bafd00684afdddf56edd751af5f033
35f0ef28aacca19fe481c35fc15edb7267313d40
1864 F20101130_AAALUZ crane_s_Page_56.txt
cb451fd8eb8824818b3772a449d59496
28dac9d66f62c8fe683a9f7f2ecea8a4dc226202
2249 F20101130_AAALWD crane_s_Page_42thm.jpg
44765eb1f23ac2598a973a0e199d601c
2c3c024485cbf89e4a625367875e469f2ace8863
746 F20101130_AAALVO crane_s_Page_74.txt
125eef8469fee029643d040126c3f066
a5d9e5d6b68261c56200f722696f6a6524652496
24820 F20101130_AAALWE crane_s_Page_08.QC.jpg
eb568d50ee5ba345b37b092a0ff868dd
2ecec33bd08eda4f16b63993ebe35e76a4a23a35
2060 F20101130_AAALVP crane_s_Page_75.txt
3b4c4a901e4a9fd670bb174d35f2e849
8656837e1b02130a9938dab27cef9be79fb2b248
18920 F20101130_AAALWF crane_s_Page_05.QC.jpg
dace86f1001f4a8c00c063fe026daaad
fb88a6f177d631301253b5928c2f51992772b5bb
2080 F20101130_AAALVQ crane_s_Page_76.txt
f5c4af46ac27ae52ce8344e472e5324c
79538f85041bd2ca5f748e3541d5ced875ff685f
5525 F20101130_AAALWG crane_s_Page_62thm.jpg
910851ed63c927688fac1091874f018b
6cdb6a82b82b8c39e4172e813043407746a6200b
1607 F20101130_AAALVR crane_s_Page_77.txt
7aec3a4dbe4375b346fffed140431087
4bfad83f28e26b79e45582464b0320eb884ce188
22136 F20101130_AAALWH crane_s_Page_56.QC.jpg
4fcf93c5aca5328bd0d7c7cbc3052518
0c8b49905a5b12b6799255ac7c247ad3cd8ad9ab
1197 F20101130_AAALVS crane_s_Page_78.txt
f26cee9b5e452658946cac9b01fa1749
30cbecc99f54793e9cf9233d8ffd1358971c71dd
6562 F20101130_AAALWI crane_s_Page_66thm.jpg
2b0e8138d47a4669c5a569f12f000100
ac67d8ba8aab4d824ced7c7610ce2d7e95cce498
1996 F20101130_AAALVT crane_s_Page_81.txt
0f6366e92669a261cb4d4b8da7f33019
f8b61e8a9cc98ed35605c73c331f11f82a0e6050
16725 F20101130_AAALWJ crane_s_Page_68.QC.jpg
bee0f5d1a9d9075729c311f34ab34a0b
0a2cda43c93a45428ff2a56264ed5d6edf6bcffe
1351 F20101130_AAALVU crane_s_Page_84.txt
a407e71dc75e8a1bec5e68c3d6b911fe
195feac6ad9f36abf2bf7b484f926a84fd980811
6234 F20101130_AAALWK crane_s_Page_80thm.jpg
fc40f82f403a018f4230381583ae2234
71e0ddcd73bb44840c696d7812a085bccd136617
1746 F20101130_AAALVV crane_s_Page_86.txt
a4d366abf2befb8928fa5be56844114a
89edc11c9151216d2594d741d60eb9d4b9e90034
5772 F20101130_AAALWL crane_s_Page_32thm.jpg
f867676ae3a90970c614f0f67264da68
dcad1277c9f16a8518643ca01525c334c96df9f3
915 F20101130_AAALVW crane_s_Page_87.txt
4a5cc2ec7a3f92166a72e97332a95619
b0d6f4b50e56da0abbb33a7363b257df8736033a
2317 F20101130_AAALVX crane_s_Page_88.txt
f72e11ba6675e4d596e4c3f282caabce
dc34fe35834392c83d4c05c39170c9d3a41b1eba
5737 F20101130_AAALXA crane_s_Page_53thm.jpg
c6835edc1e2fc035cb658d1aa642914d
4017115846b69bd0d5ebb5e35b2bf9c94e372ae4
23146 F20101130_AAALWM crane_s_Page_26.QC.jpg
53851001d905b104f7bd846b022b6d8d
36d3dd33c1c00c1fcb1ada7e42ff8abd4e783824
2409 F20101130_AAALVY crane_s_Page_89.txt
82d7d718531a2f65e603c75e26d9adf0
3c8fb21300bf14a57e6fb4f63fae5c09464c31d4
23234 F20101130_AAALXB crane_s_Page_52.QC.jpg
e5319bf1449a9e0ad3de66595a4ec6a3
8ae487856fa1967868869d2f4415fc87b1c37021
5551 F20101130_AAALWN crane_s_Page_38thm.jpg
23df973c46bf9f8f67156d96ce20e79c
2c85a8bdbb0d4a8820e62d2605148d6f9ed3316f
2467 F20101130_AAALVZ crane_s_Page_90.txt
18555f522cd5aed96240d676ba95ad37
2b6604fb84dda8b4360eef9370b06d4c861e7d75
21262 F20101130_AAALXC crane_s_Page_37.QC.jpg
225e42c95574692b70040e009731d0f2
334b01e42cacd5327f4a7b1191156899ca547a3c
6532 F20101130_AAALWO crane_s_Page_26thm.jpg
6db1669f66f2db9c909aa4c8dea56ed5
6b05d42e67fe62109974967559b3223df8a1e286
5304 F20101130_AAALXD crane_s_Page_22thm.jpg
297b0882715d111df595a30c8a7b522a
3409c5cae82c7f1a8d537f4640411727f38c5601
3790 F20101130_AAALWP crane_s_Page_78thm.jpg
228bd28f59231002d84c572234602183
81f7f6a6357cb37c6772324ebded140b77bae26f
6393 F20101130_AAALXE crane_s_Page_44thm.jpg
cd6753276f87e4a91812b3855170b143
c538fabfcb2f58cd524f7f096ab56adbfaf3bd58
16241 F20101130_AAALWQ crane_s_Page_84.QC.jpg
a47b2c6ae5e069f54f775f17c9fda4d4
3d6c6649f844658c63d686e7afebfffabfb4cb52
18200 F20101130_AAALXF crane_s_Page_83.QC.jpg
ce51e2f0e349f32b2bfdd310dac4d1b8
0192a03d0e4d57739ec5a500fae85801113e8f77
20008 F20101130_AAALXG crane_s_Page_55.QC.jpg
4850320010317eb48ee2e3811f60953c
97c2aa65f39ddad9f3402e30cf4dbf7b17ed75fa
3633 F20101130_AAALWR crane_s_Page_87thm.jpg
5007b1fc7fc007bb2d48b6a1c1dfd469
34c6eab7c010e29e218f6b9291088c44a8429e39
22373 F20101130_AAALXH crane_s_Page_06.QC.jpg
ca085c875665884997aa4871ba153e93
a06b775c9d268cea40f854775b3e77a87b1c6885
5463 F20101130_AAALWS crane_s_Page_06thm.jpg
5274c35471f604e43ac2b2e4a76052fa
eb86058b2179b639b57c8e6dbe3434e3792c1580
17686 F20101130_AAALXI crane_s_Page_63.QC.jpg
6fc0c10c686ff25dc0ed88684f3a4607
73c6b9fdadde42d57ba81011649abb7bdaf7ddc3
17374 F20101130_AAALWT crane_s_Page_71.QC.jpg
3b99248e6d9c3389307083bbcb13e3b9
369d56f56e20135d10607a2593fb6075567c7b3f
6135 F20101130_AAALXJ crane_s_Page_55thm.jpg
a2e4520d6b3260a15fa43ae4d19b7944
a2a3174aabb21f0bfb7d84ed221eb5e56b36ee50
21383 F20101130_AAALWU crane_s_Page_75.QC.jpg
79b6fa1da64d58d618373d97a790aa18
b7c6debf2dc6bf025457f0fffd292253823beb60
18721 F20101130_AAALXK crane_s_Page_53.QC.jpg
ce3f8b28c6f622acbd8982d2d6192089
bfd8643768c51ee6f1158099da4a3fd83b2ccfb6
3858 F20101130_AAALWV crane_s_Page_93thm.jpg
4ad4bd71a1d92f1db09e50572e642cd5
30e1b70ce671f4379a5745c9391ea19bb9d5a007
18576 F20101130_AAALXL crane_s_Page_82.QC.jpg
14009e1b106f09ccb388f0adfe07ef03
8d64e49b88a481781676b2060349697588551c2f
6129 F20101130_AAALWW crane_s_Page_23thm.jpg
092cf7c770962e495ba2c6221666cee8
11a394e23d61d9736281f7b780c24f42bd61105a
26242 F20101130_AAALYA crane_s_Page_11.QC.jpg
244c794b400248edfc0f5880a8fc9471
b8bee884f66de663f4abb3d5ea32d1193f0fc2a8
24810 F20101130_AAALXM crane_s_Page_14.QC.jpg
fa815d326ecb68ac86bbe0945c66211f
2c8b6d7b49f028f786bc090920d50bfaf29e7b99
7021 F20101130_AAALWX crane_s_Page_49thm.jpg
20d093fe5f1ab71dfc91c2047f8be421
a90f985d6f688de998353acaedac9ae646573d69
20908 F20101130_AAALYB crane_s_Page_13.QC.jpg
128ad2d25dd88b40498d888da0a853bc
b748efbc57a1910f0d3a04f8bf15e35cb3d6a5a6
12763 F20101130_AAALWY crane_s_Page_85.QC.jpg
d9ceb2ede68b8a5d7769835883430ffc
acb16d8b6d0859d6bcbd5cb4a78f3f5b8f10ca7a
5906 F20101130_AAALYC crane_s_Page_13thm.jpg
5099f0bd4bdcbfb912b03d44a5a232c3
283dde8f6881c4a3d6271932e59c8e2dbce90322
20651 F20101130_AAALXN crane_s_Page_60.QC.jpg
05f104b521e2cdf602114ed00e561f72
c56882c99b3f41b18ef4c042336cf8d23938e410
5420 F20101130_AAALWZ crane_s_Page_71thm.jpg
0aaf22af58e29010b7fcf42367c2fcc3
e2c1d0a3a86f63bbc7748c3d1157cb34964d77ec
6702 F20101130_AAALYD crane_s_Page_14thm.jpg
e7425032a37072ae16cabf529dd06c25
a0803d7a0b30a2759a9d0230ed362b1d09aefd2d
20850 F20101130_AAALXO crane_s_Page_32.QC.jpg
2635144361dac4a28c160d0651320225
e718b8e04a9e93ca196daf318143e111fd4baf6a
10338 F20101130_AAALYE crane_s_Page_16.QC.jpg
32cf4a13268db1868fd839f4669578ce
985469be2ad21232d9a087c4359747f2c968e9d9
25013 F20101130_AAALXP crane_s_Page_41.QC.jpg
f93d6ea2aaec0241f4a79d5bcd643668
0023c9d759200ba9d930c8d99dcfc09fb64ea9f1
21124 F20101130_AAALYF crane_s_Page_17.QC.jpg
9b5b38ca12a6d6955cfe5a0107fa12bb
57aa4d7e677b756c34a83550cf9495dc9b3aa0ce
7987 F20101130_AAALXQ crane_s_Page_07.QC.jpg
9f4180441b76c24f57066c17a6339f85
5d795c067dbbf9e4076edf561d0605cce9895c4f
22242 F20101130_AAALYG crane_s_Page_18.QC.jpg
f16598710b3fd12aba40e7315b7a7ed0
b04d11a4ea06aea185e814759c93bb11e35a335e
139701 F20101130_AAALXR UFE0010845_00001.xml FULL
b4c18abae5fcc1eb3395b657e4b997aa
298a026f3f72d50cae15fce8d53401cb27bef45b
6306 F20101130_AAALYH crane_s_Page_18thm.jpg
fd3d884c441ffe29941b507d097a3e70
f24c1090fd8e67cb6d4224c9774f96a45c01e576
3282 F20101130_AAALXS crane_s_Page_02.QC.jpg
4de8d5ce1e49847bfbc8cf5c16557543
f9edfdea6c2c3e9e690814caf7017e6aa86ef378
22648 F20101130_AAALYI crane_s_Page_19.QC.jpg
ac53b4278067860da3d2779dbf7191a1
00340eac1654abb70b59c4087f5af2ac23369e14
1370 F20101130_AAALXT crane_s_Page_02thm.jpg
2be84e67306d8523cfeb88458945ecc3
fbb6bbdf4ca758d1112971beb48d1b5e2148e8be
6417 F20101130_AAALYJ crane_s_Page_19thm.jpg
19d56152a2f841614323ad68a1e9099c
dbe36b7acccbc3a869a7fa53ecf72412cac6fef4
3528 F20101130_AAALXU crane_s_Page_03.QC.jpg
6c4b4478f594adbd31b2ba1215bca4be
cc15893c455edeae4e172774af93e963f8702d77
6841 F20101130_AAALYK crane_s_Page_20thm.jpg
82dc6f2f0829338e18ffc42727b4e849
c3da20b66f6a23a2ecc0e0b1acf6551e7f13c349
5113 F20101130_AAALXV crane_s_Page_05thm.jpg
2284c62a80a7dbd4ee266746f7917030
4af7357ab2b725636745110fc1af47732fe51a8f
24801 F20101130_AAALYL crane_s_Page_21.QC.jpg
43ba159535aa04a501659e8a9abbe07f
baa5ac5dde9c2858811693b75dd30e891c208733
2403 F20101130_AAALXW crane_s_Page_07thm.jpg
87b64efd325de9c92cbecc4970466c2a
beaca85ecce81bb575fdba25adfcdbe83cc8ced2
17435 F20101130_AAALYM crane_s_Page_22.QC.jpg
053f33ca4827b91b49e6f1aaee8f9eb8
9da67b632d87665e1c0844f86d29f81b42783921
28355 F20101130_AAALXX crane_s_Page_09.QC.jpg
5d7e004bd837f9702e4d7b8730fa2dcb
275714cde04139c78a2e620bba6945d852e32a86
6796 F20101130_AAALZA crane_s_Page_34thm.jpg
f521851e1c4ca6c1e7701dfeeb0745b5
ea55d4b7e294521f52fa75664729b2e7e473fd32
20701 F20101130_AAALYN crane_s_Page_23.QC.jpg
e260a2266362dbe7d8bac7ebed512e13
bf68fb7eb34926a2aa5816c0b2c34dd081cc331e
7357 F20101130_AAALXY crane_s_Page_09thm.jpg
1cda96c3a9d1b5a48f9357f8788927ba
7b895cb3d17769305149e3a68300f87f8d40f8b6
2670 F20101130_AAALZB crane_s_Page_35thm.jpg
616204a5263b3a3ab34af2627a46cd49
23848e3de594552fb0fda75f18e07e4239237b1f
7668 F20101130_AAALXZ crane_s_Page_10thm.jpg
0aa5ecf4f4245784c82ad8c84174f59c
27aee6f4a20daa2fa47ab7f6b13e42919d56407f
5610 F20101130_AAALZC crane_s_Page_36thm.jpg
2e05756cfd20f5ac6762099ea160aa00
ef4381e931b5601500b26c08c1a72da01be67597
21909 F20101130_AAALYO crane_s_Page_25.QC.jpg
aa8320bfc5f62b3925f3501dede1c70d
628056ca6953f3378b6344afcd5b3cdaeab61246
6419 F20101130_AAALZD crane_s_Page_39thm.jpg
a2a1169aa2a3ab5a65971eca0b397363
9aa62304943057d00351ee069bc440e83149f0ab
22444 F20101130_AAALYP crane_s_Page_27.QC.jpg
d5d6132ec79b1572670c1cf3e7423c71
5ddd551ecc182f8a4044fe384506e0709364c160
23958 F20101130_AAALZE crane_s_Page_40.QC.jpg
d2c8d925d1c3aeb122713c9270f47555
57da5750a9ba15cc52e36cdde2a9e3f6ed6170d0
6264 F20101130_AAALYQ crane_s_Page_27thm.jpg
d11c0e4bc3a52daf18eb856b08700485
7f1b4fb0549f70c07c556280a22df5fe092c3eb7
6887 F20101130_AAALZF crane_s_Page_41thm.jpg
5ea00a6009b212b367c0fde81bf159b2
0dfd9d055aee33a1ff623e70cd68267451bd1cca
24013 F20101130_AAALYR crane_s_Page_28.QC.jpg
04760236ef623085309910ab5527da9a
5cf476c2c9d736cf5278770633e28f5799a3fc4c
80424 F20101130_AAALBK crane_s_Page_41.jpg
96e92e59b5cb8c4b0ed0455dbb6a47cb
b64c9e59ab0884ae46bb3a9761e662026e52d71c
6878 F20101130_AAALZG crane_s_Page_42.QC.jpg
e78b00f82773eaa7dfb319f1d36d6fab
e7039834a39628aaa50f43ccb24b6345a39a57ba
22811 F20101130_AAALYS crane_s_Page_29.QC.jpg
685fcdfcb588616fc4bf46cbcbf6f8c4
ab82e6ae0f68180f23144cccdb04d39df0452d07
5407 F20101130_AAALBL crane_s_Page_58thm.jpg
745c2a78b555ce32fea495635786a75a
2ba946a1ad38105223d6cfabc988460a30c130b1
21881 F20101130_AAALZH crane_s_Page_43.QC.jpg
284d13b80aa1b96e08cf35e6f9d80230
ffda903733b094d10ec38a451c43149041809dc4
24421 F20101130_AAALYT crane_s_Page_30.QC.jpg
479146e219a0c05c8ef273917cda58ce
4c0e7c9ffab69cefab68da1739a8cee6121c46c6
25271604 F20101130_AAALCA crane_s_Page_10.tif
456c5e28e4b35835abf4b2d6f0b6c3a7
2c503a64b830e4da7b7b9668ccb3a008a5287bfb
60847 F20101130_AAALBM crane_s_Page_90.pro
16493094e89260aa8fac8375dd1baa73
0c3f8966e7d4125fd7008a9d024b2f824c6e6ba6
6108 F20101130_AAALZI crane_s_Page_43thm.jpg
89128d232fc001c929497518d7248c9d
a726a9efe14e1aa62175d250181656ae05d80d10
6602 F20101130_AAALYU crane_s_Page_30thm.jpg
66488b1662bda3e769d4aacac30e9df1
8322cce1ae587103e0ca9ad97d009759ae3532b8
60189 F20101130_AAALCB crane_s_Page_91.pro
57180ed7931635d0591d75d7de88f657
2bdb20e237066a8672efc2faf1f862a9c430a402
56173 F20101130_AAALBN crane_s_Page_71.jpg
ccd72917940f109e424c9c3f43b2c61f
e340b5b061c642f9ba7d75b93946a89c8591a9b0
19502 F20101130_AAALZJ crane_s_Page_45.QC.jpg
0c4f0d0789a798cebce5e9ce5097a677
c69b66eb08d840a7fafccfda2baacfca5cfc5753
23751 F20101130_AAALYV crane_s_Page_31.QC.jpg
459864d4eedf4b8b2865adc49da7667a
2812fd2bfa2bf22c4b2f1a79d3eb9a64bbd87096
1053954 F20101130_AAALCC crane_s_Page_18.tif
1dfcb465a33ffdc16a202fd9e1e635fb
048e06320bfb273f0950b692229b1b44f77667b1
5232 F20101130_AAALBO crane_s_Page_83thm.jpg
30caacc210f79d366cdbb2a455f193ae
0111cbec45663447444e30898a2d6ede7437838d
5864 F20101130_AAALZK crane_s_Page_46thm.jpg
31dc730b96a0d08c250cb9b61a431f2a
75d70f88c86b03f7fad05a56abe229a79b972dba
F20101130_AAALYW crane_s_Page_31thm.jpg
5db6016dae4d08419e11aa5d6003c6ee
20f4880d8cb3847f30b952e0b361da4f67d5b236
1950 F20101130_AAALCD crane_s_Page_18.txt
5ad98f7081e49420bc5af478bc87d738
a86467be0dacfe4754c7276a576c5ce78ff4b959
22770 F20101130_AAALBP crane_s_Page_44.QC.jpg
b33f5a0843a87b9ed98957932997f571
634954b67372007e7066151e4166589b2cab787c
22653 F20101130_AAALZL crane_s_Page_48.QC.jpg
4aedb4d1af54518f6194dede1cccba9e
ffddb101d18f3b2795c36fe179cc8416da10b679
22761 F20101130_AAALYX crane_s_Page_33.QC.jpg
caaaac671d62c9a9393812538eff106f
2d2034cea099ff8e26539d5901d3f105437a01ba
21079 F20101130_AAALCE crane_s_Page_46.QC.jpg
e94342bcf2bdfeb904a0dee04222bf7d
beb7b06045c97cd7998aef23064bad351e3046a7
23058 F20101130_AAALBQ crane_s_Page_88.QC.jpg
79884f86c7717357da0aa99e7484ca83
000a5ec6a533598e1c12247ba2487e88c6e6321b
16944 F20101130_AAALZM crane_s_Page_50.QC.jpg
e57e24886a5e2492cc507c357b0c8190
0349809eeb3b8942182834753a977ff7db55c3b9
6533 F20101130_AAALYY crane_s_Page_33thm.jpg
a439da765a2983376d5c4a504b7ea4f0
d0f2db301a64abbfd9bd3f3cbe9689ec41b9cc6b
95913 F20101130_AAALCF crane_s_Page_13.jp2
4bd23a3ea2b118e265b355b38f4c3c0d
835dc3d16fc85ac02e9464622f66842ca568b9cb
F20101130_AAALBR crane_s_Page_58.tif
d18f2a6cfd4df328687694901d37271c
9bf6335746e0c84ceff7ed7802c4fd3ac07cb24a
6440 F20101130_AAALZN crane_s_Page_56thm.jpg
611f390122cfa36ae8826d106c624687
91f5d031b9458a13b670e6835130d32d553bba15
24697 F20101130_AAALYZ crane_s_Page_34.QC.jpg
bbeb5e15b00484bff6f9b4ccea535517
5455634997d9046928a1d1ba139226ccaeff601f
741955 F20101130_AAALCG crane_s_Page_83.jp2
4e1ecb63ddd94b61bf501323dbffb90e
e932bae8d6368612700262e5800fa63ca1bb003f
2438 F20101130_AAALBS crane_s_Page_91.txt
ee97514946d145789a5e719d9622eea9
be3e419ed2909f2f6d4d18596f98c772e5afab93
18852 F20101130_AAALZO crane_s_Page_59.QC.jpg
a8fab7d02a3085ebb5b2b3980f75b4da
e5165bfcdbf14ea4c173286dd9da2a700b647da2
994 F20101130_AAALCH crane_s_Page_93.txt
afc331d92f79b4e61609756fb4af63bc
64c8d03fd83675af8990f0c6f42807079587620b
30044 F20101130_AAALBT crane_s_Page_10.QC.jpg
4c85fd365957cfe7301e6bc1586d559d
9a2d6321589c226ba15103f920b266ca98541a89
5753 F20101130_AAALCI crane_s_Page_57thm.jpg
5cb87c065a01c6cf8717b7f5db26e8db
451877318de10edbd6a8ded3e72aa65e2b78545f
71427 F20101130_AAALBU crane_s_Page_12.jp2
4a2332d7ef391fba4491aabcde3e235e
0242f5e64a0f0f6867f3c0e2592a2f76961f3b83
5850 F20101130_AAALZP crane_s_Page_59thm.jpg
d419967767b430a2dca6c449c097e381
b91871fad7bbbc520d165667dddd2bc36f37ce37
42874 F20101130_AAALCJ crane_s_Page_75.pro
b404a4f021e6ff792c6e7015edaf092b
442fb080b854961ecdcfc1bf358032969922be30
5343 F20101130_AAALBV crane_s_Page_69thm.jpg
a7cf18bf11f65fad9b43d3232ce20d93
b2aa2103d5d12593f4aeca598e1f5d049ce96bc1
16580 F20101130_AAALZQ crane_s_Page_61.QC.jpg
3362c69589bc3ebf78736f62c102348e
b74b0912369ad62b0780b31b64b1d1a69e02dfad
56463 F20101130_AAALCK crane_s_Page_64.jpg
df04ed4dadf9eb46dcbc15f088839025
7213a5fdbaf17c4c83959d569cfbb573fc68f035
1690 F20101130_AAALBW crane_s_Page_82.txt
83c76d1d03b5e45ce0417482ad56b347
bfcd21603a80282c607c1bb7d046ef896b3a196a
17426 F20101130_AAALZR crane_s_Page_64.QC.jpg
155752ea325be5f29732b574262e80f7
2b2db9fa659fa2c10797cb205e02ec28b823a25a
6411 F20101130_AAALCL crane_s_Page_48thm.jpg
1a481bb01ec07d02639638f811177e33
66c0249c893f21dcd4d04b5c8cdaf473562c2c75
5371 F20101130_AAALZS crane_s_Page_64thm.jpg
d56bc165d0504c92ff4ee66dda3fe199
34c53c648b6f9c7fdce0c8448fb7dff78d80bcdc
F20101130_AAALCM crane_s_Page_77.tif
7b0f0de28067c6d1f16f3e1f22be201e
499804fdc1a7c6e8a95bce7d08cb927e50e3eede
F20101130_AAALBX crane_s_Page_44.tif
06ecb477b25b82b892792c39353f17f4
f3e6d3722372a500d528874aeddf7ace6829a16b
23506 F20101130_AAALZT crane_s_Page_66.QC.jpg
ae72b2ded09db55bf22f98897b010759
800cad24ca38a3879a56c778ad2ac2a1e94e9760
90411 F20101130_AAALDA crane_s_Page_70.jp2
a0bc8c5829a96ccb25f4aa5f90d57035
aa0c6929b4fc03fb02a644483bae7deba3206167
73414 F20101130_AAALCN crane_s_Page_40.jpg
215f6db63738b61ec9ece5558f653dea
1dd40c02cb294112047f2ffaf1c8f5bd0e437deb
15378 F20101130_AAALBY crane_s_Page_47.QC.jpg
08cf5227a878c14b0f417712e36c6795
8be864f58e6c4869f5944636825291cb1260e189
6283 F20101130_AAALZU crane_s_Page_67thm.jpg
bbd9a4e0efe847b64870621c1e05f3c7
6b32cc2c66f602777338e272a4dc3204fa0955b5
81153 F20101130_AAALDB crane_s_Page_05.pro
323f8bd54bdd6c333c698aaec0413a90
72d14ecd717a104274c9811cee42b62be48fc4f2
1051946 F20101130_AAALCO crane_s_Page_65.jp2
06889d38396cd3287f8eb8d88a938bc1
2bec49aea3baf1ec2d101d1da2b827e252208e8f
12889 F20101130_AAALBZ crane_s_Page_04.QC.jpg
5feef0ce870563078458e75f315c93d9
db56b01b94cb3d5a4fe144e280139d967003bf56
4821 F20101130_AAALZV crane_s_Page_68thm.jpg
db2a6b1a5c7a00ab79e8bd43eb2d1486
4099bde0c2eef01e895352fd550d053c0eceaecf
10305 F20101130_AAALDC crane_s_Page_02.jpg
396d6c2a7049014eb0fcb663fe2260b9
b64aeff6d0d85757ea99c0a8fde77c7ef254d2f5
983 F20101130_AAALCP crane_s_Page_51.txt
0e653f47179d09397d4ce76af7e5469e
9bd57a1a02a10934f41dd0ae847a58b61add9b75
5742 F20101130_AAALZW crane_s_Page_70thm.jpg
8ebe78a00054c32a3c488199c9613caf
293c17e8c454fba91bedaa3b7ddb4310aba0ecae
6786 F20101130_AAALDD crane_s_Page_91thm.jpg
b257629f25a2af04033bd9b98986591d
cb996cde2833f51bc83651915c14c59a4063a2db
12890 F20101130_AAALCQ crane_s_Page_78.QC.jpg
7eabbec54a3b810303dd0501b85c3461
3081e383c90f2a1c2fb8fb0a4e2b346af415e4d9
16705 F20101130_AAALZX crane_s_Page_72.QC.jpg
30f9ddbb97b66db563f66c7bb3f2a16c
4816c1080d0cb723a18db8f5d601304d1a459e34
15154 F20101130_AAALDE crane_s_Page_74.pro
d8ae65c52498a1c2372ef5df46239e39
c2065643654373a425972f32e99862e9a7c61124
F20101130_AAALCR crane_s_Page_11.tif
707f17ba1eeb0315054a89b342b320ef
70e5f67715ede7f09f766e8cb4ea3c8a18bd9958
25064 F20101130_AAALZY crane_s_Page_73.QC.jpg
ec9882acf4ca3ef319a6e91d396cedf0
60444b8e466c11ad05b845709fa1e884d60399f9
126736 F20101130_AAALDF crane_s_Page_90.jp2
46b25d4d132b041e25d9669ccd972219
1fe575e53a3dbd1e83ca90fbc6f4e1cd6038dd67
39896 F20101130_AAALCS crane_s_Page_56.pro
96bbafa9646b01ce73107a74e0149213
b36b2a7aee89290fbd65e9f5c3867c3e251c2ee2
19732 F20101130_AAALZZ crane_s_Page_74.QC.jpg
3293bbdacf887b94b38ba3d93bede224
f19f9e02d2b65b1a968c38c51b820af2781c4be2
F20101130_AAALDG crane_s_Page_31.tif
c9e74f8ea0c2070b1ed1e20ac8f0d58e
f64a832200066751bdeebcea86e6282d8e487355
2121 F20101130_AAALCT crane_s_Page_14.txt
9f7766f597ba934a26230c98a8044a11
bb18615d41e49f3b0939e715ceac30cb93ac86b1
55380 F20101130_AAALDH crane_s_Page_61.jpg
3f14170aba1bcb3c1e1e1da2304e62f0
3a220f8289b124452fad7916a828a9c249aee00b
20194 F20101130_AAALCU crane_s_Page_69.QC.jpg
8fe5d7cb5c07ebc2781799865aa60c73
cf8fdeae15f16a5c725d7a6a86be32bbf82c4ece
16946 F20101130_AAALDI crane_s_Page_62.QC.jpg
d6a856f2dba2649e8b830f863f7b2919
0f5dfd7ed450d055e18d723330032a7702d7db7f
69397 F20101130_AAALCV crane_s_Page_18.jpg
3b2fc8c398daf2b567ea9ec7ad9301ab
37c8307603c1b5c3a10540d15e7fc9fa5d2a8b0d
1205 F20101130_AAALDJ crane_s_Page_52.txt
f8b5ddd1eac8fe5680c4e293c8ffa757
ce1401b884aa8e000c978a4f5af13cb4e9f10160
1376 F20101130_AAALCW crane_s_Page_79.txt
a5520ae72a8c0bdff8b7c7dc3873b1ae
681a907aff731ba8f783015b9eb6744f94bcf363
F20101130_AAALDK crane_s_Page_49.txt
b1f3d64a6cca824df624e43b03ab863c
3924413dc29381850723a6c423597d565d847b3b
22514 F20101130_AAALCX crane_s_Page_63.pro
259473cd865ee8b7fc5faf468d986795
6a2076d972f3af32f8f2338f0d7187995d7bc81d
F20101130_AAALDL crane_s_Page_12.tif
7e56c87b3ad7569525a5ad7eaa7bc467
a30d8bb5e787d978ff3d947dc5e6615f73618489
F20101130_AAALEA crane_s_Page_52.tif
ffd1a1ea4a9f982bbe1a41864b593b55
04546f8275df4ee66ae1efd342fe9f8c890b8833
2024 F20101130_AAALDM crane_s_Page_15.txt
edbd009d64eca30a22e14182f4af7226
181643ebc39584df5f75828beef1865f04484769
6101 F20101130_AAALCY crane_s_Page_37thm.jpg
b08b4de7f505247bc718000306e4e917
9d74a808683d302a0cad3bc69dd20fe73091a699
1493 F20101130_AAALEB crane_s_Page_03thm.jpg
52a9afa0f517b94f29f17551da6ecf5e
251188cc6def5eaf25ef6a64b7c9bfc0ce5d4bdd
17188 F20101130_AAALDN crane_s_Page_92.jpg
89bfcde00f31cb7e44149ea9c945f43d
eed8f67f0969c23625fc88d2903e7f3aaaaff909
2785 F20101130_AAALCZ crane_s_Page_01thm.jpg
a3dd6019a35d64ddca0e311866d0290d
689759d46590becd9780840bef1aec2b6c48fbb5
F20101130_AAALEC crane_s_Page_22.tif
3898949515381e0367e8fa71c76a914d
61c33d3151cbb07dc7343c2e7b7bbd21910a4368
49508 F20101130_AAALDO crane_s_Page_26.pro
90b0a9c3d1223e4fe65212734061bd36
c1b5877418db743d866a92303799084f85582018
2027 F20101130_AAALED crane_s_Page_40.txt
bf561b9c98164e9d331207131c09ed1b
2341b37e153d78a02e2e7f8600354862de7ef84a
68733 F20101130_AAALDP crane_s_Page_48.jpg
e4c84117ddb8af74dfd36116caaa70a4
b77a062c76938a17300dffce37ceae0eea6a3d9f
8687 F20101130_AAALEE crane_s_Page_47.pro
34353521c9a19e154afed845e81ee73c
f33bba6d91488d85f8142e5f917a8b188fe6ce63
76520 F20101130_AAALDQ crane_s_Page_51.jpg
ea8a29fa4b3df6f8a31539f7eb98016a
9c62bdf1ef36800e2a2d717ef1ddb9b90ac663f1
F20101130_AAALEF crane_s_Page_89.tif
45223fc5c3342027fa764febe57bf6a3
13ce95b3bb322151072f75715ce916feaba68066
F20101130_AAALDR crane_s_Page_15.tif
9d70313edad0b04937d10a1ad8166858
359c0553623d258b0564be9aebd530f3280602bd
6597 F20101130_AAALEG crane_s_Page_52thm.jpg
4cb3bacc57d603d005e2cc5c2fd7d8d4
f046f5ac981f155a7e14050b4be657b3fb656b39
911120 F20101130_AAALDS crane_s_Page_23.jp2
a62504f079f45e8766e744ab0b769a61
8f7644755acf707b259b03ef1e97243e1f084c83
72762 F20101130_AAALEH crane_s_Page_45.jpg
70ac72dd972831fe2dd4b6d855b50cb2
7741183b750848d664dd3da7d3cf90cdcdc53fd2
F20101130_AAALDT crane_s_Page_51.tif
f6eafa7b7866d188b7fc8b0a04ba86b9
f76e89f9023d6c9bb6376528613b4f1d0bb24657
110883 F20101130_AAALEI crane_s_Page_40.jp2
3c16926c6350162e1bf6645746b5e460
61af36c8240a74ef99b420f5009650bc63d97397
6467 F20101130_AAALDU crane_s_Page_29thm.jpg
98d9b294296e7dd8318c5cb54fe62f0c
d3f373eeed6db1ce0ac0e7c370c5fe6bd171ee28
87961 F20101130_AAALEJ crane_s_Page_67.jpg
2a30e9c4faacc96905624ca868d33717
b34ef643ff8b237cfca581acb6f7334f19af5345
F20101130_AAALDV crane_s_Page_57.tif
ba5e2a213de663e586e466187641e9c3
9466333c16d2756bc8e4843faaec1671c81f9bef
F20101130_AAALEK crane_s_Page_72.tif
6434ae6236b3bdb348e78ec07146fcfd
8a23b829f5d1190405675118ff244a1db30935c6
48905 F20101130_AAALDW crane_s_Page_81.pro
ebd36e2543b2007c2c8772d70bff574a
6e3894ffc1a0fcfb494af325ba8c5341ead71209
1500 F20101130_AAALEL crane_s_Page_80.txt
902931398a0eee27d6966e8292d203ee
4e0be64aaa10add9bca83ed464831e8c55e67f0e
68294 F20101130_AAALDX crane_s_Page_43.jpg
3ad1328791a1525423b39afae83f5859
4402031063a292a29def22dee48e3d96ee6fe8b4
6148 F20101130_AAALFA crane_s_Page_17thm.jpg
3fbfb399b6e382bfdcfa6ebb53bc27fa
a82589a34d24faa9c7296a488de84deb88d4dbc8
62548 F20101130_AAALEM crane_s_Page_58.jpg
88c44c2f63d30cc1033deceae2893e47
f32c7cd834a03f42adfb9cbb3b51296bf1907ecd
2017 F20101130_AAALDY crane_s_Page_28.txt
15ad38571e50ba20a0859f5d98108859
85873a77c9ee64aa03df58a01dc5708601ce36a3
852657 F20101130_AAALEN crane_s_Page_79.jp2
50adbf21053cb67b150a1059a7594421
7c5d32247fa32cf22a99763b7246f416231dfd0b
22646 F20101130_AAALFB crane_s_Page_81.QC.jpg
d3b3d69989b483d31cb6089050706f82
d8f1fee800b21ea7c557be7331dc06dc4956da39
20697 F20101130_AAALEO crane_s_Page_86.QC.jpg
1807fc87f9c93e57c5494595c1fd128d
e50b0e6b044d46c3123eb9babb393f48256a5566
1051934 F20101130_AAALDZ crane_s_Page_80.jp2
60c5d7a23775390858c081f21ca2c195
eed1cf819d9a86debc9c6f95bf9478231e2a57cd
92451 F20101130_AAALFC crane_s_Page_11.jpg
f86e0046a3a125a076d858a87940fe87
d2603166ed74b5869456a8227a7fb755f6a6b747
68282 F20101130_AAALEP crane_s_Page_65.jpg
eae74253e718bdd60c5e83f59e458554
9787372a31b4a1682fde3a1741e9227f84bde6b3
1822 F20101130_AAALFD crane_s_Page_92thm.jpg
924eb6fa1beeb7c0af73e2d71f38d398
023c4da470b4820b5c2fe2096e0ca7c8fc8ed85f
F20101130_AAALEQ crane_s_Page_40.tif
f8b499c221252fb8fa3904fcd39eb4fe
cefd44c6cfadcc861c8e2b1f07b9d78a0fcf0cfe
73302 F20101130_AAALFE crane_s_Page_15.jpg
40c83d9e0005b5617a59ca0b6606cc84
964721176c0960fa346b23bcc1ebc65c54686f84
70283 F20101130_AAALER crane_s_Page_26.jpg
0e8e11f4df80a74edeb96427fe10593b
a5fb60da0b91a7756a675e757784dfa655b7b333
49129 F20101130_AAALFF crane_s_Page_33.pro
e950f7b58ace8776925365c5493517e0
127ee9dfb4103c0af1e086a26491d08bfcdec347
463818 F20101130_AAALES crane_s_Page_85.jp2
bae2680e1cd2941badf8fc7074d364bb
4164492fbb574dd23f70c9921a17f525bc6bc42f
13617 F20101130_AAALFG crane_s_Page_35.pro
b9abd96637f721fd7903ec581409ba11
056e28f1feeb9387a696ad63661b8445a7d67d9d
5549 F20101130_AAALET crane_s_Page_65thm.jpg
47d6f00ef7cf267ab011efd681047d07
576e965528e98ed388bbc8ccd4f2b5d34a2332fd
1968 F20101130_AAALFH crane_s_Page_21.txt
43ec22572d7d40b74a6efec6d35c6711
3240941540006bbfd0ad31c761347550d9f87d33
5782 F20101130_AAALEU crane_s_Page_86thm.jpg
b1e511449d65b072409c384a14277b78
34cb7aef8062e6a7f1383cf17f2975e7372c66a4
23773 F20101130_AAALFI crane_s_Page_04.pro
7173da9d040480369fd3b9a6fda07d9e
40951bca2ae359532aa391d7c760f9735ef3cfd8
F20101130_AAALEV crane_s_Page_88.tif
4661f2ea40b5f4d905ed2a29ae5e081b
b47075c475b1eb4b1009c71341920eafe5085ed2
18011 F20101130_AAALFJ crane_s_Page_38.QC.jpg
762cdeedf57e194303bf52738b3339f1
97c6d4a8fa7173161775a4a447d349c797ce09e1
4467 F20101130_AAALEW crane_s_Page_12thm.jpg
fd35ceac67ed93bb216407311b2c6894
aebf8f76fdd7c8260182a14e75373060fa9d2d32
73756 F20101130_AAALFK crane_s_Page_30.jpg
23aa5f04e6e0cbab10335be6c2aba406
056dc5996cbeb397c16d9a10d452512b8972dace
591 F20101130_AAALEX crane_s_Page_35.txt
6022cd8f3941142804a12838dd03164a
71d78c964e845789b8220851df951c47e126ab98
1039 F20101130_AAALFL crane_s_Page_60.txt
fc841ac15573017c807caa17d02a4530
801fe451eded12610a64cd4b29f5d2c1d2a80edc
1038727 F20101130_AAALEY crane_s_Page_56.jp2
ae3c5f33a08e2fa7b08601da6195bab0
346b1c792f57dda5a2788a101f4b640b23d3e99e
78823 F20101130_AAALGA crane_s_Page_49.jpg
e4578040ef3462decee220fb32b742f1
68f237512f24eb8886624435896f88f604c5fa7f
777721 F20101130_AAALFM crane_s_Page_71.jp2
8a3a131a3e06886951685923a6ca315e
02c6755a304552c2ee510c3a0b3fff0e42bd1c4f
24038 F20101130_AAALEZ crane_s_Page_76.QC.jpg
10ee7888501d56ff55d35838407548ff
98277092fda6bb07f4c7dc56e56deb57ff89947f
21396 F20101130_AAALGB crane_s_Page_78.pro
9aee4a7ce7e23496b86fc4e6b31e48e0
a484fca612d8ac232deb2a15ca6a87101e0f1bc4
71184 F20101130_AAALFN crane_s_Page_80.jpg
75159aefcdb4f2d3e0086aba779311f1
72cf7f6daff597b124e7bacfd2b41fef0eea2a4e
8705 F20101130_AAALGC crane_s_Page_01.QC.jpg
4ffddb1ea65182182ecd184575d32c26
f09ba79f76924f6a65747bed508dd2ee7acaf54e
20033 F20101130_AAALFO crane_s_Page_51.QC.jpg
a1914fedb8f5293d20971f601694090a
4a0be0e9f8ad1aef5148591ab2e55c9d0ead6d5b
3872 F20101130_AAALGD crane_s_Page_04thm.jpg
f628fd33ee6a7007635822b16b25a7c6
e70a3f6ad4495d9256a46553b014f4325dddc3aa
25018 F20101130_AAALFP crane_s_Page_67.QC.jpg
6a2a4433c6c914e0c3120b0ab891270c
98867804f712bdacc30a88f07127fc4358524ce4
24475 F20101130_AAALGE crane_s_Page_55.pro
d35300815181f01e3eadfa9abcd17e3f
5e6c44a345ebde38f5cac70ed5edfa518da6d0de
6007 F20101130_AAALFQ crane_s_Page_25thm.jpg
de23e350dc4576d4159becc55f5f37dd
2e1449d2d6d05869eaa738ac300059435f69c27e
F20101130_AAALGF crane_s_Page_31.txt
cb917e29ee3199058ff6d8f7e88d002f
accfd6b98b0b6ca7ba6ab9a5f191da1090db67ee
106073 F20101130_AAALFR crane_s_Page_33.jp2
95f872c077e99be98217f98ea47c7eba
ddd97685e4ddeffe617b445b89a1b87c5ca588bb
F20101130_AAALGG crane_s_Page_36.tif
679ca137249f15a40f5a47fbeb152d76
2b6984e076df24512b95455cec0c9854fbada489
26324 F20101130_AAALFS crane_s_Page_62.pro
01f787d4dcccc4eccb0e906ec8ac8848
8beecd3d4a693e228f8572f5b9684e6e6cf9f13b
40670 F20101130_AAALGH crane_s_Page_04.jpg
02d50caf783537f2400d3e06cef0fc3f
f96a03dc82b1c238b1edc551524c45e7b4af0677
22834 F20101130_AAALFT crane_s_Page_64.pro
c8a672565020b792feae5ee1e5d771da
1bce9fcf1c704f56ebd01b33347c2efdd77b517c
5125 F20101130_AAALGI crane_s_Page_72thm.jpg
a5dde8cc350bf4b24c5beef16377600c
8ca884afbebdc01f6faa57e52ddc81d80508db48
F20101130_AAALFU crane_s_Page_05.tif
676dab38e94835fce31295a2b42a0526
855caf9ce7db8d9dd5ea0b51bb2956da17aeffb0
F20101130_AAALGJ crane_s_Page_55.tif
8bb1b28cbebfe98c778a332226b4fa27
0bea7795cd2f5acbac3a961b7c9824eacc0fa75d
1051970 F20101130_AAALFV crane_s_Page_58.jp2
2cea691ea5252d54d5c0633fccb2c587
e9231192d54c7a293cf67c19cbc5745267e259ec
15601 F20101130_AAALGK crane_s_Page_67.pro
14af69f19bbc2c79f220b1af0b3b6996
9ee6d3cbb8ba0a7deb4f357ae77f04f0b5ef3712
1381 F20101130_AAALFW crane_s_Page_12.txt
95727d6f1ad8160701dd52be5f68ced7
69b6d4cd8430ec19536a95576b6fd191f7934dfe
1665 F20101130_AAALGL crane_s_Page_32.txt
b55227f0273f0267c26fb3120cdb3c83
1d9718524c68038d0fdc354a287d4e3f23319ba4
6622 F20101130_AAALFX crane_s_Page_11thm.jpg
8d5a7eb9b70fa1e7cd54eca54556197e
e912fc3c89bec16e8036dad7b4d1d353a14638f7
F20101130_AAALHA crane_s_Page_46.tif
759ed3f79110512dcac41e9678a83bb9
13b1b38bfb89a60c9273bcbdb59a427d36f48831
F20101130_AAALGM crane_s_Page_02.tif
78f6f25bb391893b7cc83401f7c20fea
56abed97356c6a08fa40235d849a94d1939c7dec
18694 F20101130_AAALFY crane_s_Page_58.pro
1963c2050f387ac96335189065c8afbb
77027583c3949417034dd4c75af03fd1ad9c78ba
6612 F20101130_AAALHB crane_s_Page_15thm.jpg
0fa616c26f078e6db49d473c8dcdb6e6
1a2e981daf1786a702b1d6af44c92bdc1dd92004
17547 F20101130_AAALGN crane_s_Page_58.QC.jpg
f46a332f9cd8940d391c74782dcc9a34
dc8420c41f32963e94c755ca7239649b2b359e78
24914 F20101130_AAALFZ crane_s_Page_90.QC.jpg
e3ecec7b85eaa6f297b15b8e3d003424
42386a4f0265b26a920017dc81265f247b31a1f2
8495 F20101130_AAALHC crane_s_Page_35.QC.jpg
c325a1e7a242255eaec3b0983d3eff8a
364ce18da5188487fbfb0056816c36657d9fbb1b
5907 F20101130_AAALGO crane_s_Page_60thm.jpg
0195dacc7f3023ed075747c05376cf34
db53d47a0ae06fbc1d47a9e64362328595bfe8fd
102020 F20101130_AAALHD crane_s_Page_27.jp2
af08e88f32c8660dc5b8e1e8e072e175
9f6a5240baa9b90c6021e9315efafa53cd82c792
1894 F20101130_AAALGP crane_s_Page_39.txt
c57a4ad8854eac32000145db856140b9
1462523c3531bb67d7dde124c691b5af1b9b91ba
2105 F20101130_AAALHE crane_s_Page_03.pro
58a1a7e5b504597bd3aec5cb8f1b2c32
5580a088368d9182775e80763278a97b12388d94
48073 F20101130_AAALGQ crane_s_Page_39.pro
e10e2dc89e646e6e4db36f4855923322
decb5031d603c9be6d489c48385f9afcfacaafa6
6207 F20101130_AAALHF crane_s_Page_08thm.jpg
16a2c277341e7d938d60715332fe8aac
081a493aee30d0ce9b33e3aae0aef3d82884d0ba
F20101130_AAALGR crane_s_Page_19.tif
3d0be2541f856e78618269426b3f2531
71bf1ae02aaa5fc17a18415cd368325b2d30dba3
116719 F20101130_AAALHG crane_s_Page_88.jp2
5a3d8dba4a821d3265ce0d63c74231d8
3fb208093f0c21feb68fb18da6f7599b4f81f327
38439 F20101130_AAALGS crane_s_Page_77.pro
6e8fa280e5248cadae73fed2d546df84
80c41e9fa43774f86b3dbd7572b202cb3b722337
6446 F20101130_AAALHH crane_s_Page_73thm.jpg
bebbdc0a578964480a6c4f3f04492506
3dfe06f73745c0796313f0bd5543b6f451de9e99
78969 F20101130_AAALGT crane_s_Page_21.jpg
3708a01004e50e383f4c7fbcd083cde1
e97eec7202c60ede2eab97c86a7a9a1a2db8c995
96765 F20101130_AAALHI crane_s_Page_37.jp2
fbb1491006e6c7601886664a34782a32
773722c8e995d4efc1881829e57a2208e7e0d6e7
22559 F20101130_AAALGU crane_s_Page_39.QC.jpg
47e03e465faeef18d422a077d78cea8b
542abbb7377db723ef7894688d0af4ad3776b869
4886 F20101130_AAALHJ crane_s_Page_61thm.jpg
e022e8dda5d0f2cba86325e146ffdce7
439c4f3efea136e6b05b2725acd35976f2861e49
36666 F20101130_AAALGV crane_s_Page_32.pro
3516dad93e1103b02f638b69cec002d6
a57d0a6a93d7236b553f6a16a5c982730c5d6223
51653 F20101130_AAALHK crane_s_Page_12.jpg
e47b2b1039ec5efdcda7395841514361
fa0de6755935293dd2866e61186a660534edd200
19431 F20101130_AAALGW crane_s_Page_70.QC.jpg
f9114e55a8f29186960bf88870129fae
ade3fbf7e115605eaef18d52920d051245fd4b61
18317 F20101130_AAALHL crane_s_Page_65.QC.jpg
66800b7ab7f4157fbb8cf4451ff52344
6d7bee1406392ea4f7c6257ad2a33f6d65f76d8d
1153 F20101130_AAALGX crane_s_Page_85.txt
2e5f68aa1db7047f1b7166e25533cd4e
5f2831dc1c2da7e889b2967e4d01e6597310b647
1051965 F20101130_AAALHM crane_s_Page_64.jp2
c241de890ec40029e8b9326ae47508e2
e98a97dadd636920560bdb51b3e27381da4d8c13
111739 F20101130_AAALGY crane_s_Page_76.jp2
1ee551d4fcc8577fd7694d5c333cbbfb
287b8f26dc5bbb979b1733071a90b1cb09ae7fe6
1051906 F20101130_AAALIA crane_s_Page_60.jp2
67721d662aeff9ceb580c1dccc7e4b55
d854cf18e58546b778c740ba53c98d9370d2cd02
1525 F20101130_AAALHN crane_s_Page_83.txt
bf618421e9535c6525316b9ff2ae0c45
1e64e51a0e54ed57133b3b5e0baea8fa328f2716
23756 F20101130_AAALGZ crane_s_Page_49.QC.jpg
259c437288796918d61e80ce26b47b5d
6eec33597ad54c87359b1eeea6d5b225d0ecb7c9
913 F20101130_AAALIB crane_s_Page_61.txt
79d86b4cd20ee3b236de5c1a5ccca2b7
0fcf32ed5d03fee38043b9dcf474abd04ae7f558
110801 F20101130_AAALHO crane_s_Page_15.jp2
dfeab96470c21e0bd739f98d9369ec87
b43b253e96d3bb3593f2053496dbafcf01df0e17
107632 F20101130_AAALIC crane_s_Page_26.jp2
7b9db48e5dcfec8f9da8b1eea6431749
1378bb39ab08adadaf4c2ce8537075045837882f
68275 F20101130_AAALHP crane_s_Page_08.pro
55b6565e07f26e6a26ff7097600a7b71
f2440c8b8d519d3e23c81bf556d763952fd4c228
15795 F20101130_AAALID crane_s_Page_12.QC.jpg
37e18a6a105f18b31ed904f98530415a
2f4395abfcb07dfe047aa1ad46fd1c012385697a
5587 F20101130_AAALHQ crane_s_Page_63thm.jpg
47c4a2e84a97a1c2bb45621ec417ff21
3ce6e00bc8afbed73000b903dbae9f6ae4069c22
72694 F20101130_AAALIE crane_s_Page_66.jpg
d58b8d04381641f31a4db0c3924a06b4
783b8099d091dff9819d92fcb314139886b62ced
6871 F20101130_AAALHR crane_s_Page_21thm.jpg
07caff52f0c31fa72a1f2ce042b459cc
4cafde88a572701267f44195615005345837ee4b
112642 F20101130_AAALIF crane_s_Page_30.jp2
928e0434a7c6439136e31e88328457e5
e51833dc5c7583cd8a6cebf949972873a5c81d58
1051967 F20101130_AAALHS crane_s_Page_69.jp2
ffe5d17802cf9135f3727c8481764d8c
c69e471e9fe30a0718b9755dbd3b30d3fef24723
24491 F20101130_AAALIG crane_s_Page_89.QC.jpg
e324b34cbc7ecb33d5db75d7de3dac8a
5aae16fb696749a98be98a9eea4c093a17fb7953
1147 F20101130_AAALHT crane_s_Page_02.pro
20c6b1c0712a31cf1e3276de728baee6
11d66d1d773168e6b322d887619fb26439b0a09b
F20101130_AAALIH crane_s_Page_30.tif
231536013fed254109cf4de9755ffd15
89ccc6b2e5af61b201d4a20f769a9b6d73c8c264
29996 F20101130_AAALHU crane_s_Page_79.pro
842ffb97f7c66593232169890f1cc8fd
8febe38360d584e6bba749a6a727bc9e416c8ce7
1042 F20101130_AAALII crane_s_Page_50.txt
044ae710dcc5933fb52e66b72612a943
1b02a7145d58e7ce2a5cf3dccb5e22e7a35ecd57
F20101130_AAALHV crane_s_Page_13.tif
2891686f08cd6e0e0089ca92ef7a3967
f87acdeadb099d084f96e0d7559bbebf6959ae46
19362 F20101130_AAALIJ crane_s_Page_36.QC.jpg
6945297b877622ffd590f3147fc772ec
92233826e5f861a964ac0a7c653d84a80da26e26
20392 F20101130_AAALHW crane_s_Page_24.QC.jpg
7f61da9b24c84a59f013e98a0542db70
72da7f66019502cbd119b295ba754fc9de74acbf
78588 F20101130_AAALIK crane_s_Page_89.jpg
e1d94ee1b0c0c9c6f66902fe60476fd7
a030552dd5175bf3c0ce69e13fe689b2c20628a8
72286 F20101130_AAALHX crane_s_Page_60.jpg
e4ae303e7cce13c8bb1488fb3c43d1ce
c07706cb5990f6236edd3e07536e9e310e444a57
56082 F20101130_AAALIL crane_s_Page_77.jpg
c84a34d248d43f997cec9f43e774b27b
a3059cfa1e71f8f3935bd1ff9d618dc8cad72661
22045 F20101130_AAALHY crane_s_Page_54.QC.jpg
b28f8eda6d0abf9ffc9f28c55920fd0a
9e6535421cc66f22b8fef28f556fa7659c31b9b3
4718 F20101130_AAALJA crane_s_Page_47thm.jpg
24c74e99193fabdf4f55d31ebfc7cae7
16d6e1d57595c51408637f585ea8ca522dcabc37
105925 F20101130_AAALIM crane_s_Page_18.jp2
1ae88111d0aee47e2644ec539eecfcd6
8b090d63fb65f5d94f577b33a462edbb96822abe
19456 F20101130_AAALHZ crane_s_Page_57.QC.jpg
940a6a4a03f38e5d87e9f1b03f408317
d609e75ab41cdc34ce558cbf2e6c47afb7ccea0e
F20101130_AAALJB crane_s_Page_81.tif
bff72428ced951d10d7640e94c6f5d37
9ffb5d9b3c1ed11f670bc9c94c8e98c55aa63d64
5878 F20101130_AAALIN crane_s_Page_24thm.jpg
4c628bd3eb7b09f6018ff7caee3619e3
937546018f635b277b48b5859c5d76d4588e614c
17588 F20101130_AAALJC crane_s_Page_53.pro
f5fed1bbf674b7ee3be0e4060bda8841
e7005c5584545528a323e1429beda40e946bb764
1051983 F20101130_AAALIO crane_s_Page_09.jp2
8aaafbf724371917f8daa51a6c9b3ecf
5ad5797f0f8e3e46b353dd90d3abc97c168412bc
70316 F20101130_AAALJD crane_s_Page_29.jpg
93838a0c8128d18837503dc9d0aea4e0
06953dc6f709da9ec705dce3ef76cc40a27c6fec
12536 F20101130_AAALIP crane_s_Page_87.QC.jpg
16e1edb6f6cf4a3c51231a338b57711c
50797d62a23ef4498207e9396a5df48bba11731b
616 F20101130_AAALJE crane_s_Page_69.txt
ad4301b8b5474279a04a2dcb8875f394
15843339d3c4c337948484a943f8eb32975c157d
82313 F20101130_AAALIQ crane_s_Page_91.jpg
28e05c36a63ab307b3f7acb964a41bf0
efb1e9da72b3915061b591ab3d09252dfaa6916d
F20101130_AAALJF crane_s_Page_47.tif
1c687f5f8394d6a708046923b77ee5db
d9d13392db1e968234674b3bd97a6922a31501e3
3206 F20101130_AAALIR crane_s_Page_16thm.jpg
8b2df7429e6f9903e838e70aae9a2202
123aa81b61ceeae2177324c3caac8cc81ab6c128
F20101130_AAALJG crane_s_Page_49.tif
b5c41e8b33e4b70a3a9b89e6ade49bfa
a50d53abe46e64a354c6b94ba986e87c1562f920
6541 F20101130_AAALIS crane_s_Page_40thm.jpg
ba36c37e8edd1d967cbaae62b96d4cdc
3637fcee86fadfc160eb399c3f8d83aaa8abccbc
49366 F20101130_AAALJH crane_s_Page_18.pro
71abaa9e5277194da7d5ce9e66dfb975
703ccb9b140226442ad4647ce55e317271e566f4
66578 F20101130_AAALIT crane_s_Page_17.jpg
f7ef9c20839d6ff7affb8eb4912387a3
bc5a1dd361c977ac4d0c960ad030fdaa9bb7a062
F20101130_AAALJI crane_s_Page_52.jp2
d69c24bbe362fe21bc31a479009780d0
e2d4cc000e7eeffffb61075323be14e27bc72dc2
31976 F20101130_AAALIU crane_s_Page_16.jpg
0a1ed9adc979b1ea0ecc4aa9fa3ed955
e746212cd8045a72ba38a593643efef28d57d83c
51473 F20101130_AAALJJ crane_s_Page_84.jpg
9bc120cee45c300f8f3350c4e299ad96
7f4651019fcda632642b5ea8df5df8f8f9825dfe
F20101130_AAALIV crane_s_Page_80.tif
e7d3b2f0fd1292ce00a8f03ab65056d8
6f0257df066e8d537f6c5ed49be1a16416307c7a
64314 F20101130_AAALJK crane_s_Page_55.jpg
a9f0503d8922a909940d5ff105cd553e
4f2e9597e97f0421901e1669bbca715a2f1c1b11
82523 F20101130_AAALIW crane_s_Page_54.jpg
517134c907a80d759d8065b4c75e9378
f540ab17d1d62a6c88994ef419f7aeff2d903722
13186 F20101130_AAALJL crane_s_Page_93.QC.jpg
1396704d787cdc6d1bbd58eb595e2986
6293ef737d810c1f5a496ef14067658acfb25c2e
4982 F20101130_AAALIX crane_s_Page_50thm.jpg
91da8c6288e0ffe14596cf26c9f76a2d
8ed9f6c8ac11394861151c717b2c2a828faaba12
106779 F20101130_AAALJM crane_s_Page_29.jp2
9df785cdefc0966e76e92daeaf4b90fd
b1345ab181920827dc4a611e5e00d779187a939f
54061 F20101130_AAALIY crane_s_Page_87.jp2
85f79c1915ee49fc0fc88f15bd1056ef
e8c17778256206bd600fd7d1d6f62bec83c3bd5d
F20101130_AAALJN crane_s_Page_14.tif
b7229230b08dd75d7642298953fc3f7d
e99943d488286acfea0e453479d70d7b75f06d81
6629 F20101130_AAALIZ crane_s_Page_28thm.jpg
cf3cacf30a41cf02cad569fdbcd6ecc0
f64e4e7d9c2c3b18ca792f75dacd42d3223e0380
6388 F20101130_AAALJO crane_s_Page_54thm.jpg
dee864823b8fe60b5d87daa906b2c758
80cf5bde8b590f47f83285ac97b252860e354c1e
30061 F20101130_AAALKC crane_s_Page_01.jpg
0d747d5eafb3bfa069662144eeaa358a
7a174926d506ffc5c4657f800eedc87845d54067
49991 F20101130_AAALJP crane_s_Page_21.pro
275b9d414da350e45476ad0b2f98a072
5e7e95f9bf9ba4226e0c5bc3a60fc09fb6aa143a
10828 F20101130_AAALKD crane_s_Page_03.jpg
4c01247040b62bb1019bf644809883fe
0f254c449023ae985a5d72c06e3199240df26dc1
5556 F20101130_AAALJQ crane_s_Page_45thm.jpg
a440eaff16c7660fb4817a3eae6bb9bc
c55d5bf0d74c65d84dba91d005e3bcdd31a4a713
73360 F20101130_AAALKE crane_s_Page_05.jpg
27b4babf943cbd068caf7bfe3b58621f
5940e32e0253ade25147b12055060802f5c00558
45011 F20101130_AAALJR crane_s_Page_46.pro
85995aa2efa7175acfaf789da5f26f11
4d71c325e7c449982a78475a4e842ff98b62828b
90230 F20101130_AAALKF crane_s_Page_06.jpg
8aad2fb8c9dd42954969dcd894c60295
464394036924031082445f4e6b4ba4df39dbb34c
5984 F20101130_AAALJS crane_s_Page_51thm.jpg
9754e288553f0ddc76c0de12ea0914a2
641afd108aa400a852facc81e28ecf3515b3f569
25989 F20101130_AAALKG crane_s_Page_07.jpg
62ec242a5a182bf555e873822ce53b93
0b492f1e25ca287f7dc350cbd806736653b84bfb
67581 F20101130_AAALJT crane_s_Page_27.jpg
22854d77eac4ef64571d91783fe73607
4437899b9fa264d7b7246e0ca0892f5b3e7657f3
91592 F20101130_AAALKH crane_s_Page_08.jpg
e5189df80b59b3ae2d0de01cb1fb49d6
6e4a5248a5c7b3ee9de25be4ced0155c5e5d6869
F20101130_AAALJU crane_s_Page_34.tif
d6238faa861630a088fc07c5a7340944
7f35aa7980aaf0772481aee4d521024c1e410387
101572 F20101130_AAALKI crane_s_Page_09.jpg
3cf4136d15f5d5a0057f481cae5aebe9
030b4893a377323f7bbda01c3a4c9a427137023f
102917 F20101130_AAALJV crane_s_Page_06.pro
106234097246873f57e0aa452eae9331
da66f794038df8c9142c4f225df2d2fbf07d4dce
112920 F20101130_AAALKJ crane_s_Page_10.jpg
a32da8ba8995051e817bd3775d100c45
6fe5811aa79e0bcd889d964bc4fc2646f2d8ab99
63185 F20101130_AAALJW crane_s_Page_86.jpg
101228f0231d9d12832542510c940f0c
ac6203e06521fc18e911cac03c4502670d20b9d5
64094 F20101130_AAALKK crane_s_Page_13.jpg
1070a7a76e6536621e316dd77d09d2fa
fb68a57bf21054be4ba9e1e3d8099cc267c5f727
24057 F20101130_AAALJX crane_s_Page_15.QC.jpg
7c5867f48032c269771b127c0c3c2429
6dea8b1b21fc9ca80ab7a873ca2d46d285e7bb34
57811 F20101130_AAALLA crane_s_Page_38.jpg
d70408d1b49c4c620323687598b73815
8229e9e95dc850cf8785ac7f3baf7f38481511ae
76551 F20101130_AAALKL crane_s_Page_14.jpg
79ebb6e5e5408317561a0c1c5ad3ef68
67fa83c44c465a0dfd208a844cfe5e71675faf37
24596 F20101130_AAALJY crane_s_Page_20.QC.jpg
2c55f410c40279e3349b220f84e62c02
4441e1f4942c6d159804e0c511d19a04f865916a
67436 F20101130_AAALKM crane_s_Page_19.jpg
23c178a690a0d5b3cc22291e32a39b29
281f8b5196a89f91d92d1c60941208eae8730665
108038 F20101130_AAALJZ UFE0010845_00001.mets
39c5abcd93b0bc6ce181902eb42b6912
821f0499c9e27f08a517f4a8ab34ed3a35a1cbd1
69396 F20101130_AAALLB crane_s_Page_39.jpg
ea351d20bae73ad2546539b7cc67e32d
6a7152ca80105aa4af8e6f96be9812faabc5833c
78902 F20101130_AAALKN crane_s_Page_20.jpg
051bb7ac2325c9206bdc75076ea25cf8
e0d327adbdfa52e5f73d37d55a8d52c5c4e9c1f3
22999 F20101130_AAALLC crane_s_Page_42.jpg
e9d284690bd0dea8a9577e7827e91e19
e31217912d6cd675408b65b3843fa4c765647c17
55746 F20101130_AAALKO crane_s_Page_22.jpg
f2b8ff44a682fd21cfc219a9fae77383
7bbf769b479d9248b031e22be294f43e840ddf94
70044 F20101130_AAALLD crane_s_Page_44.jpg
e63f5e03a7a9f23e71e693a37584e2e5
69a71bcf6e29f7779af0706986225ad358335b0c
65342 F20101130_AAALKP crane_s_Page_23.jpg
c0fb4e4947eaa126d316aa243064e9ae
989cbdc033632d4fb3a48c030ef65dd0f585357f
66445 F20101130_AAALLE crane_s_Page_46.jpg
559350fc2b4f5a48aefcf1d8ee6b901b
b077787972bff7be708c20e739f076d91501d437
65557 F20101130_AAALKQ crane_s_Page_24.jpg
41dbea9ec0cbd333bfa1ee838d260e46
fe07843dbb9311eeaf3938bf9a5a09e2c05f15c1
57534 F20101130_AAALLF crane_s_Page_47.jpg
ba2ada9f63683af310bf629a363c625b
94cbd5c183990dc568d668d41f3f1bff3f3ddb06
68774 F20101130_AAALKR crane_s_Page_25.jpg
e8ad6dffcf501f9aec08bdd02f48308d
f82b5dd1975e467f9ff7a418f51d0ed3f84e2598
60602 F20101130_AAALLG crane_s_Page_50.jpg
cca79ad835ea1e1e670cc9ce995f1ccd
ebdbdad676fbb7f2c6040329e4351c845969f814
73719 F20101130_AAALKS crane_s_Page_28.jpg
fe26753aac2bc6352f65e7002f2366aa
c3aeef33adccb474d0a081867025defccf96499b
81626 F20101130_AAALLH crane_s_Page_52.jpg
29cd2e3e2359e08e1b55be5a39978c9a
9f4886a4ca904afdf38ee3adf9f34d21c98fef77
71076 F20101130_AAALKT crane_s_Page_31.jpg
65053762a443e70f8e57830c4c14c521
2194cc06e29e171560523a868aa1f3efe906819c
62770 F20101130_AAALLI crane_s_Page_53.jpg
6ed33f603b23c63bf287bb6427062f9c
f19441242520ebc7d2ea7e7d4b2adef70910809e
66073 F20101130_AAALKU crane_s_Page_32.jpg
01d76a3e102a083ca55628e772f298e5
1bbf5054430c5833844afa915716b5cc06cc26e8
70929 F20101130_AAALLJ crane_s_Page_56.jpg
dc559b7dad047c6d1124e4cab1b3d139
85dc2249df73216e8beffa6f4cdfb32f5789146e
71886 F20101130_AAALKV crane_s_Page_33.jpg
f72e9b7f7f432c25bf1b031a0219326a
e19e1a19540d608599f05ccb4025915ac86d4bae
70987 F20101130_AAALLK crane_s_Page_57.jpg
4957cdaeca9c77bd26c39aa67578fe13
d97925191256ac2b500ca456257c39644af143b0
73032 F20101130_AAALKW crane_s_Page_34.jpg
02ff5fa0d753123ff6fda6bbb4c0839f
a688b4e37899a1cc9a393e929c7013e430dd3507
56722 F20101130_AAALMA crane_s_Page_83.jpg
5f5dcc3a5ed7395d2b416c598223d02e
e26ca28ecba603ad79df9219cafd63d58bb1dac8
67289 F20101130_AAALLL crane_s_Page_59.jpg
223122e25e139093739deb49189f5441
d2103a291fa723e9fc6b01087607084ba209bdd2
25048 F20101130_AAALKX crane_s_Page_35.jpg
2fcfa1a793f6a7c1f57b7488384dda53
876e882f893d1194be9ae30cd188ce0e99384adf
37458 F20101130_AAALMB crane_s_Page_85.jpg
207452963a34ea819635ef5be4f58935
688eb66e4087e4913772048ad06e72d655bc8b5e
56781 F20101130_AAALLM crane_s_Page_62.jpg
9500c26ac13ae4ae1ffb2602c92e88e8
7d6c4bfbd7f6e96ff34f0eac484c623743a9728a
59005 F20101130_AAALKY crane_s_Page_36.jpg
98813d23867e0c3900148f4b1ca362d7
25a7efa48f36c576bae17a5ce8547e97bbbcc77b
59979 F20101130_AAALLN crane_s_Page_63.jpg
a1c641e2e2393be4947a20d5c42129a7
822efe9b377c9aec052a3ae1ca7e5530d48cee76
64937 F20101130_AAALKZ crane_s_Page_37.jpg
7af4b172d5cb230ee79300ead6239575
9a5e8bf7155803d5b2e13045a83eaf0e09375988
38514 F20101130_AAALMC crane_s_Page_87.jpg
b853cac841902c60fe32ff5cf995c215
75ad14cec3847f00d1dc94669b418f20d0c566af
55337 F20101130_AAALLO crane_s_Page_68.jpg
3da2847668ca4f4c539396e6e6a28eaa
1f3c5a4b857cfbfa992e2772daa510cd1b0fe7c5
82483 F20101130_AAALMD crane_s_Page_88.jpg
7064cb3f196758f0dddd8a7ae7f6c2de
9c9a696f07d24cf963251e91a05ca52174e76082
67335 F20101130_AAALLP crane_s_Page_69.jpg
284bda34e0858fb59c3fa762b516ee75
0cf55035b34faad0ba76951f440949d72d167a4c
84735 F20101130_AAALME crane_s_Page_90.jpg
0642f230a80bd6fb911911cd4fc4d31a
01afe897beb8438a24728bd9ed8038edcd94f27d
62583 F20101130_AAALLQ crane_s_Page_70.jpg
8bcae35626399aea368b3af92b896298
dc3edea9df8f227107532677ca3b7fb4b7de4bd9
40846 F20101130_AAALMF crane_s_Page_93.jpg
6e2c6a044d7c5c321ead73dc84a93b6e
cbeccc8d887a04abb561b4b1c86b32403035c2ec
52072 F20101130_AAALLR crane_s_Page_72.jpg
b9d46b5cd66028b3a9754b29464bf75a
243ebe5e9311d2e93d311ea11710cf7377d76ef1
34157 F20101130_AAALMG crane_s_Page_01.jp2
4d300d66a0e7603bb0212f2c394b058b
ad3553b812f83edcfa4dd2e7847f6d48c5ce6b7c
92929 F20101130_AAALLS crane_s_Page_73.jpg
9d74ff31e7052718839557cbc8b5ffb0
457fd52e432db2d5f1678c9dc297e65c0293c712
5615 F20101130_AAALMH crane_s_Page_02.jp2
d349af3083dd85267336797878abea38
4030885943d833484be21a5294f7e0a1ef997330
59635 F20101130_AAALLT crane_s_Page_74.jpg
b781c82cfadb29ecdd6d00829d908873
b63aa4b8e5a2fea2da5106b6b23fda15cb574d54
7680 F20101130_AAALMI crane_s_Page_03.jp2
11759a37ce95da773980c33c5db179ab
39a9ad5f33e94209c80f62e8b02bfbe572f67bd0
64939 F20101130_AAALLU crane_s_Page_75.jpg
f482f4983306fe56974f902a75d0ccee
f2346d3b6e8025492313fe4d88c9e5eb26e96ac8
53832 F20101130_AAALMJ crane_s_Page_04.jp2
946d689b6691fc800d674114d4f419c1
7b283aef7ef3d10562295005bdfabbb56c7d9645
73400 F20101130_AAALLV crane_s_Page_76.jpg
1f3164b7410000aed3cf22f909c74c8b
e4f67c9885bd19aa480df448e98509c01712a465
1051966 F20101130_AAALMK crane_s_Page_05.jp2
14d8be907ab9ce3064a312c94d8f2134
fee7ce434c0e2443317ba0cb211673a3af357bd1
39731 F20101130_AAALLW crane_s_Page_78.jpg
07cbf181cbe07af5d23c87b3e7da51b3
c9be25f5a8724d023612b97bc5acc40a79c3d35c
F20101130_AAALML crane_s_Page_06.jp2
82eb0895876d717b8daa54784cc994cb
64e1d1c03bf22cc21bc4a1fc69ea6de09cfa3de5
59981 F20101130_AAALLX crane_s_Page_79.jpg
68a38f3d39bd0c834359030e933a6102
60d6dcf94aa33485711541a0ba0b0ed8b5843155
109095 F20101130_AAALNA crane_s_Page_31.jp2
e10e565d693155f154265fffafb19274
f8762e76e5411cc68626dd53cdde9f1decfd09df
526316 F20101130_AAALMM crane_s_Page_07.jp2
30907b82d176b33a355cbb969d0cbdbc
58e7d5322d7e6f6cca8c14c0a29dd2f58ae7bb0d
68645 F20101130_AAALLY crane_s_Page_81.jpg
91ac8e0b8a35780eeef02ac242d725c2
4ec021f214abd3deb48b7bd53d9b4775de4ad055
868923 F20101130_AAALNB crane_s_Page_32.jp2
36b86e829e0f04074ed5e837b6a01c45
371b0dd628a05d28ea4753d5c12b36bb09c822b1
1051925 F20101130_AAALMN crane_s_Page_08.jp2
55c69da168849d71a2ab892ed4d136a1
2f8058aa4d35bd7eeb7d904529cd9c58b78634d9
57755 F20101130_AAALLZ crane_s_Page_82.jpg
74061aef5cbda1be915c1827535c5c7e
e3a1db1669516145daa79130dc41e96ae08fb08c
110599 F20101130_AAALNC crane_s_Page_34.jp2
995f687749220beaa1716ce5b87632bb
0e786f11cd422a095c3b5c9b5bfb00e492764615
1051982 F20101130_AAALMO crane_s_Page_10.jp2
3c5e8943f478ac29206f72215b9cfaec
33202ca3dbcc3a3015f451e953bb69d1d0912590
F20101130_AAALMP crane_s_Page_11.jp2
ce772809a46146ed981d717286b0c869
c33ca729311cd364dc98f827d70bf5e8bd45ca47
32284 F20101130_AAALND crane_s_Page_35.jp2
9315e61b8a1302e5a71093a01529ebde
5c3fef444a5d69140ce682341552948676a79310
115567 F20101130_AAALMQ crane_s_Page_14.jp2
10a69d62a6b115ac83bc769950122303
4e99b5c29b57519287388c474adf820fd0459bdb
86973 F20101130_AAALNE crane_s_Page_36.jp2
049f5c92adf92e40e93b9ce75ddb2e86
d66e7868d3f499130535f528ad15929b3bc11c4c
43527 F20101130_AAALMR crane_s_Page_16.jp2
553635d72c1699b204e979dea37ed50f
609586d6e19d959554b2db5805aca9cab4bed963
932613 F20101130_AAALNF crane_s_Page_38.jp2
0fa715ead79f3b772f25e2e2c5300dc2
134b0a697271621343d461d2d0f6f2e3b2f2ef76
97606 F20101130_AAALMS crane_s_Page_17.jp2
74eb77ea49b075c0412dd3df99f3f995
6bfe1d236261084fa74b591ac64cbee7cbbe45ae
103604 F20101130_AAALNG crane_s_Page_39.jp2
6ab4cb7c269d91b14cdf0b2b535541f5
3a86735424e0a03520eb68b9f3f472e2268a758d
104892 F20101130_AAALMT crane_s_Page_19.jp2
6a2e205b597dd7544925f7e2b2ca45e5
2a40b30e45cd4a3b2c0716926b2acc016beec261
1051978 F20101130_AAALNH crane_s_Page_41.jp2
97eb241fdf2c18c7ae3426725dbddf78
83525e60657982ba177c71dd57ddffa4afcabea1
F20101130_AAALMU crane_s_Page_20.jp2
4474580dcab2bcb130d494be98809ad6
af643a308225ea0682ddf5612eeabf09a205e7b1
288503 F20101130_AAALNI crane_s_Page_42.jp2
a9d673a5bec7c93ab48bb0c95986c9c2
14088bb614d455543411d770574c67775f9ea41c
1051852 F20101130_AAALMV crane_s_Page_21.jp2
bf6302f7642d3b2d2bc49102dc1aebb1
8ac84ebdaec52fd256f0ff68a24a0d46f041ea15
935130 F20101130_AAALNJ crane_s_Page_43.jp2
091df3b8a0fdbe88a5b9b7231f99a8c3
0b7677bea56543aa869bb43799b23919f718eb93
746844 F20101130_AAALMW crane_s_Page_22.jp2
e425de7a51901dc8bfc28de72b71a749
c91eb42558135356385ae08ed0b8a91e6daa1994
106532 F20101130_AAALNK crane_s_Page_44.jp2
735a7d75f81e283fcd222f6f2950233b
aa45020431eb767792e6da7af4f840354a783cf2
108233 F20101130_AAALOA crane_s_Page_66.jp2
aebd708301f1eef16e2f1a9163b0d223
ea02f38e71ff82381bf31b6e64c21a55e7a72471
1051953 F20101130_AAALNL crane_s_Page_45.jp2
1ca154d8debe6d3c3b508f3d01f98eea
b300429e3dbb299a95edfae0225e259e5eea9083
882106 F20101130_AAALMX crane_s_Page_24.jp2
1dc9004d62de397f022801c9bd0497f9
3e618e7953ea348c0aff18fd610a0e3eb5c03b15
1051968 F20101130_AAALOB crane_s_Page_67.jp2
6b2d40d0d5695ac72342bed8ea7dc541
edd22e77907efd5541fec2872424b097387239ab
93865 F20101130_AAALNM crane_s_Page_46.jp2
b787a4268966af852cf72d4b584ae01c
dfc773b3a1385d97f0e4296a9b69e2718e1c8fc9
945103 F20101130_AAALMY crane_s_Page_25.jp2
eedb26afdf0254863c9ba1d477fba2a4
286c45fb7576ecd24cac6b9f83ca7a43ea482335
1051913 F20101130_AAALOC crane_s_Page_68.jp2
31959c7029ddb0451aeb4a702216b872
663729d6155aed74c4542443e7354de579a24e77
1051813 F20101130_AAALNN crane_s_Page_47.jp2
0734023b660d99346cfaa3d23ccd0233
d4ed3676a00471e335583bfe6942d2bbcf056460
112278 F20101130_AAALMZ crane_s_Page_28.jp2
8e3fa63af7e4490df048a60294371a6c
91b46f20bebc6cff716c2756876ae0ea0cf2bf23
681105 F20101130_AAALOD crane_s_Page_72.jp2
53a91f5a91b21cac2134d94ab7cec37f
dea51e16d08792818c78e87fb2933d38ed43c351
102888 F20101130_AAALNO crane_s_Page_48.jp2
2ae6524a541f84d84ce0b5a09b5c95f1
881abf089b9d343e88ad92386c5d05fa5c6b9aa3
1051975 F20101130_AAALNP crane_s_Page_49.jp2
cf788718495833d7c933e7b9997f3b2c
63d2261fa9bafa61658522b4b8f55a42bb145807
1051785 F20101130_AAALOE crane_s_Page_73.jp2
3deee5dc4b77467fbc87cf6b1cb23e45
995dcf6ee79b95fa046e950d009d16aceb1e77a9
1051743 F20101130_AAALNQ crane_s_Page_50.jp2
09c91ed5393b9ee5c928199eb231d4f5
b88c757f557431a223ec5cdf5d194b6210597174
839285 F20101130_AAALOF crane_s_Page_74.jp2
2b0551a089a8444cece3a7c1ac4521fe
bb8c895ee2b117fe02a84e34c0455514db441dc9
1051712 F20101130_AAALNR crane_s_Page_51.jp2
37cb0bf3ac77748a8eb95e69011818a6
45f5396f61414f5339e75836c6515de31239d133
87552 F20101130_AAALOG crane_s_Page_75.jp2
83de8dd8de9b30e4adfa1df7ebad6c6d
e4d87b352047e692c142abc79d47ed1dcc20a793
F20101130_AAALNS crane_s_Page_53.jp2
11d3985d1e016ce0673f53e469900a85
7b079d7a53c952fc9f31e1081af73b52551e541f
82425 F20101130_AAALOH crane_s_Page_77.jp2
f9d187c485ca1e9fc05b304a60a2b9eb
720f297f8248ff33b715b85827e552229daa074e
F20101130_AAALNT crane_s_Page_54.jp2
28e4fcadcd83b8066ffaaaa14e07de08
39d4431afb32a5e0637d370199f77e8aacea39f2
491746 F20101130_AAALOI crane_s_Page_78.jp2
2a939905995223ad39ab8e5b8595e2b5
a8ecb8c768b9d9614f6b9885aaf2552a98353fbc
1051985 F20101130_AAALNU crane_s_Page_55.jp2
f2735e6ff92f8b96f2f3e9d7b09dc62e
051f0cd64ca1e4d645534858e7e1079c530b5abc
104499 F20101130_AAALOJ crane_s_Page_81.jp2
74439686c564e72552da6049b18d5684
5ebfe376394561360dc322196323e61ad890e7e8
1051890 F20101130_AAALNV crane_s_Page_57.jp2
a20eb8d23c793c41cfb56109de7b82c8
21366f451befeb8c797226a034ad44ac99aefabb
789255 F20101130_AAALOK crane_s_Page_82.jp2
fd479df57efc5d3a68d055a5ef0930b7
bffae215993cd9ccebd6c1ad77a08b56f7431698
1051947 F20101130_AAALNW crane_s_Page_59.jp2
2c008ad7a167d8c6c80590692fb38e44
3cc939c573637c7677a462786c5ef57676694e16
673959 F20101130_AAALOL crane_s_Page_84.jp2
dc74c772cdd25bc1f3c506d7bafb1dcc
1c4c2c325865cc16f652f9d18d2d20e1429e45d3
891262 F20101130_AAALNX crane_s_Page_61.jp2
1a5c4e3dca2a9092e941f8326bdd2c86
ef1f48f785fd91bdcc972272fcbaa719181fa7dd
F20101130_AAALPA crane_s_Page_20.tif
26ed0e0e8edacf5b28fa5ddefc4dd0c4
ff83f38a32f93c7b9675aa69fb3a4107538f54a1
92870 F20101130_AAALOM crane_s_Page_86.jp2
bb7e382c68c064d3c8ed44848796cf64
14f98810b6df01c10791690b5334d4b7a0cebc76
1051958 F20101130_AAALNY crane_s_Page_62.jp2
17c3314ba6dd92b5a26dd065faa2ab28
53374d9c9beb48de36c79103c5605feb24001909
F20101130_AAALPB crane_s_Page_21.tif
1bc8581a6c935b6fc55805bb89980138
1776bf36adf4cd4c72c5334aec8a98f85fd0d29c
124780 F20101130_AAALON crane_s_Page_89.jp2
8c177c5f77a92bc576fd7b737eb635e4
1bd52339221e0e0b91be24776ae4591984777fd2
1051981 F20101130_AAALNZ crane_s_Page_63.jp2
556238a268eb551d87e3519418dbef82
740b9efa3f7c624f246a2e28c954c40255239757
F20101130_AAALPC crane_s_Page_23.tif
c0770c00123d4739feec08284ccc2ac3
ef377d9cd2041f829d52164738091a93bf234df7
126764 F20101130_AAALOO crane_s_Page_91.jp2
6f8e0b1c89e1ecd4b6a46897d9d172da
4d8a6db91123fdcdf25bf6bc490ddabe7cb8c261
F20101130_AAALPD crane_s_Page_24.tif
9f649e66cf3c6d1b1e71368c2f789d0e
ffee6324a284ebc3ddfb5b75ae36e605a1e754eb
16423 F20101130_AAALOP crane_s_Page_92.jp2
63aa38b4649a36a0f2af47842996cacf
73bb9085dc57e7e2bf7ff21125e116fa14571719
F20101130_AAALPE crane_s_Page_25.tif
31822d0f0b371ca9511ff48aa0f1d5fb
e456dbfdb3ad3b3658d61a2f8d170bfd4313cace
55090 F20101130_AAALOQ crane_s_Page_93.jp2
5af917c2cc5f9c88551fd3bd90d95963
85946824214d41af48a0e133542fe024ebef33e3
F20101130_AAALOR crane_s_Page_01.tif
ed2bd6a37b37feaa18f41eea9cbf71e1
0426ef667cd15cf7d891e2cd51bb58f287ac536c
F20101130_AAALPF crane_s_Page_26.tif
b4dd914199be535d619af35d55a13969
9924d772131a202cfd7533a58c2ea6afb29adea4
F20101130_AAALOS crane_s_Page_03.tif
a8430aa275dbce74a00973288aed56bb
30578ac55002c8a1e14d5cf04525469c253e4ecc
F20101130_AAALPG crane_s_Page_27.tif
25158f4e4704179be05a09ba06b877e3
70fa1afd95a8171f6a9d68b8cc77b2a6382e8b68
F20101130_AAALOT crane_s_Page_04.tif
77a860c70d0a4c10dd1d45a452e4e35b
392b8a06b75b876bd34dd9e12416ed67fb56678b
F20101130_AAALPH crane_s_Page_28.tif
3d0f486afac16c7edf6e0f0ac4304bb0
1022dcdbfe2bcdbd46ac6e541cc4c8bcfeca3e53
F20101130_AAALOU crane_s_Page_06.tif
c5731e37b49f5ed09f243cdd249cfba2
0503c9f9b366045029347777c40fb01f559d4ec6
F20101130_AAALPI crane_s_Page_29.tif
26c337ae1817d98ef8435bb472926414
3e4926527c820d6174c999310ae1ef00f27f2709
F20101130_AAALOV crane_s_Page_07.tif
eb7a8aaa7307de698178f51823e49950
f8fe5537b561d740d676a2d08b146056f8c371bc
F20101130_AAALPJ crane_s_Page_32.tif
5608e682fb7173e01d822de0458d676b
a6790d27ea2d2a9afea2af693fdd2fc0ab0f9aab
F20101130_AAALOW crane_s_Page_08.tif
505cfc3c770c7d955b93779f2b177597
3ab116385c2cc86ce891057edf70a6f518ede4a1
F20101130_AAALPK crane_s_Page_33.tif
a549b4e00d0d800c455fa3482f8e08fd
9c599340be4d7c1bff49f143b576f8a08542d5d3
F20101130_AAALOX crane_s_Page_09.tif
436becb3dcfef2802150815b4a8f316f
715b988a5af1e151296c74db8df1a28e54a7926a
F20101130_AAALQA crane_s_Page_61.tif
8a77593a9bb7cd47232202fce5efd63e
4831fd550d4fa875c6b6dc51facbad9c2c623868
F20101130_AAALPL crane_s_Page_35.tif
c0f4bc1fcce7a153a233ade89b607974
63630da24f29102a1cc4de8ea3ca888b236c0529
F20101130_AAALOY crane_s_Page_16.tif
003fe70a3945f8900373113df72bfd2f
c207839da1a02bd245113aec986e67dfac5339c8
F20101130_AAALQB crane_s_Page_62.tif
4c88491a13cd54c1b84bff034c8680b0
1c9573bd3f64c44ba46115aec562ff246d219e58
F20101130_AAALPM crane_s_Page_37.tif
6427a05218c8866bd0287cefcdfdb28b
9ae61f882a32ea684ed44e6a69cae1fa0bb5ca09
F20101130_AAALOZ crane_s_Page_17.tif
633fafbeb9e243ca932115e3c0d66209
d5f69d945d8c1b0a7e1b95cd99d3b0ef09ff085e
F20101130_AAALQC crane_s_Page_63.tif
bfe63f8c9658df23349795afabd9187e
598bac5a0ec57a086427af9c2dc700c26f93e977
F20101130_AAALPN crane_s_Page_38.tif
26390a3ccd1baaf09532ec483d91ec58
3bc75b57ce383c1b657c6d00e43ab47789c8d2fa
F20101130_AAALQD crane_s_Page_64.tif
7e6623920df4ed47568c96781a4485ba
edc3e62223c1c4d9c35bc33beb992a9065a652f6
F20101130_AAALPO crane_s_Page_39.tif
4973c533b2364128999918e12580d54b
6beba4887d9e2fae8c24b3b2b00cefe9a22fd5c0
F20101130_AAALQE crane_s_Page_65.tif
bcf8999c7162c4ce4504c61c7c87b610
c7553c03a99bdad612e9243e63f560e5e3d617f2
F20101130_AAALPP crane_s_Page_41.tif
481be506b9809d9fcf2851b3dd18cf4c
2a33a2ba896508100d74c91b56c4c043b6a395c1
F20101130_AAALQF crane_s_Page_66.tif
327246c20c152f7081e217aa626a050f
7a7d7265458eb0505d8d5e6de5f57db87b407fdb
F20101130_AAALPQ crane_s_Page_42.tif
b5c2fa6add175343daaad788278d9670
f68d588eb824befd1f6027198676b5e27e945702
F20101130_AAALPR crane_s_Page_43.tif
6127fb8959791ed00d0077abd1c4d639
3405860ca8e1568293024bd46c7ce3ddf3a2db75
F20101130_AAALQG crane_s_Page_67.tif
9a4f1f0bdca32367d8877d0d676709ea
59d51ebf7930b5286ea3d8776871045dcbd5572e
F20101130_AAALPS crane_s_Page_45.tif
cb8099d36ed88563f5756c28c9e77ad3
12b07ce5802521ef5c2fe289e923ea6a89668852
F20101130_AAALQH crane_s_Page_68.tif
dc6deb8f625d275f7c04f44e91900bc4
c77a773ab15798709cb1977aa7be287254659541
F20101130_AAALPT crane_s_Page_48.tif
b891a2343bee348eb2b3e8a026e620d2
b03e9035428a49a86a32bc2b7d8101d0050ef579
F20101130_AAALQI crane_s_Page_69.tif
bac1349e73c653c699fa8de84f3781ce
a69b27e9319a17b554f8ad38b8b83ec9e8b304a8
F20101130_AAALPU crane_s_Page_50.tif
d1f500a913840e51d96be533f4c07b12
68413e549fe3eca6f74bb6706c58ff0ea8013495
F20101130_AAALQJ crane_s_Page_70.tif
1f918bd5a737aa108f8bf7a4237170ef
eb6f34b26fbcaf2aad54c3513c978b1dcae1afa7
F20101130_AAALPV crane_s_Page_53.tif
8e03a3ae7294b4f9dfe1ac2cc41ae26c
b2987cb94653eb703c8de42371a7fe2256ed5795
F20101130_AAALQK crane_s_Page_71.tif
21e9c871c8b3112dcc95e719e0e8444d
8c85953b14c90d6916d27751c7901fda5596d918
F20101130_AAALPW crane_s_Page_54.tif
92e0f4ab283ba1254ce0e9fe7b375cf9
61692e3f24354e7496efe675d7cd8a70206a7242
F20101130_AAALRA crane_s_Page_93.tif
ee5378c30e4329241baee55131ab4c57
38a83ebc7c34f3e908ee566172b366a701f4a340
F20101130_AAALQL crane_s_Page_73.tif
eccfd5b4730cb47db98d3146b21e01dd
0db6ff7af9ef8760572fbc3845e269398fef9a57
F20101130_AAALPX crane_s_Page_56.tif
a1b114c1870b9f4c1ccba2b70639315d
696602c55822e0d36138635c6aa0fc3efb238af2
11881 F20101130_AAALRB crane_s_Page_01.pro
ffca0ac9663f144502598cb892bfbf39
430a7f510c4e2b0e0c0496b120d7e6f4a5887fd3
F20101130_AAALQM crane_s_Page_74.tif
e1ee5e1279c3be60a31df2db2e5068dc
9f803e33623efe611ad2a9478ce220d99980b19e
F20101130_AAALPY crane_s_Page_59.tif
4a8db4b17d477e54070d8e79b46b5288
6c53ec8b53ea7bdf52baafed897b740868d6257b
15374 F20101130_AAALRC crane_s_Page_07.pro
b6df7ce98a41eece1ce8770a0dc650ba
f9dff8f0e905b1195a7c95d73136821fdd71b24c
F20101130_AAALQN crane_s_Page_75.tif
0468b8f0347f758ec55e78b94c872bfe
8700590002fa6338c24e2ae7f414d08ff036b34c
F20101130_AAALPZ crane_s_Page_60.tif
dcdec98ab930cc8bfb9fe793e00433d2
2a9da0d4da9d7f064188ae8d59487b3be539a7e9
79822 F20101130_AAALRD crane_s_Page_09.pro
50745a30adbf68fd01527bb82ad4dd9d
928a1265cb128b983f355281e415cb4a8be893dd
F20101130_AAALQO crane_s_Page_76.tif
09961a2ddb43dd54fd658720e09b28bd
d38928189f5f72ebf5f30ea92874ae4210806125
83602 F20101130_AAALRE crane_s_Page_10.pro
c6ecddda65396a5dc1770b87f13aeade
2b6ddb77196ce8f887e7729810fe01adcef9919d
F20101130_AAALQP crane_s_Page_78.tif
19cedf37bf38f857a3ca1925221466a0
ceb1a3c47fd0025b331f0e9652e4a7b2239402b1
75576 F20101130_AAALRF crane_s_Page_11.pro
d683d16e1565ad2d95cb341454ac5db1
f8023169f70dc385b0c30cf02bf9672e493b26ae
F20101130_AAALQQ crane_s_Page_79.tif
c54ca2a128f4a5cdb4b9d1441ac33b43
5eb8cb9087bad3ab27d18a643f9c0582dc42d32b
30163 F20101130_AAALRG crane_s_Page_12.pro
73e2e119a182800757255a97d249ed86
689f2b3636c24f16d4c38d744b4217f8871a0105
F20101130_AAALQR crane_s_Page_82.tif
75291c3ff50d9f73e2dd22021d8182f4
1b5be922ceaef8a3284b98d76f70d761c5f5c887
F20101130_AAALQS crane_s_Page_83.tif
a2aa5ebda9b2ca64ed84e07db998b082
3b387f7a43de78fc7ccd574ff54bd107bf349495
44138 F20101130_AAALRH crane_s_Page_13.pro
9b63635468ca65da92b62d4d2a776409
0ee11f88886d046fe17cf22d76e93b782f68134d
F20101130_AAALQT crane_s_Page_84.tif
90f210ff94e99827ff63e0656f7c1db6
a35fc2fe5d93ee3d7aa1c3613983da3f2aebcc5d
54171 F20101130_AAALRI crane_s_Page_14.pro
1606c9426c4d8ab48f8d7ce21b6306a5
25389504525cce3294dd46b7e59da57c8a546bcc
F20101130_AAALQU crane_s_Page_85.tif
fb8859b6ed57559f2d8847e7669e1781
93c3e26a75abe57bd8af38278f702da824746507
51235 F20101130_AAALRJ crane_s_Page_15.pro
b564e3605147341ac41f44a8e928552d
c43835a447b053781bb7a73bf7b9215c241edae0
F20101130_AAALQV crane_s_Page_86.tif
13ba38003dbd0f974034e165f79b4b00
3d64cb77465280ea4f959b76fc731b73969d811e
18813 F20101130_AAALRK crane_s_Page_16.pro
920ad51861d3b9a6a208871067e55c70
cd5f743cc981155d7a0b201770abb3c83ee0a50c
F20101130_AAALQW crane_s_Page_87.tif
bcbaa2593f2ed0010850198eadeda454
a14c074080c0d0af8235b96814862cc7daf6c935
45030 F20101130_AAALRL crane_s_Page_17.pro
f041925ddc22b2fc505f908d75af2932
97bde1fd19feed589b481e5667f80552ed956850
F20101130_AAALQX crane_s_Page_90.tif
610eb076be07817e052b6e7945131a29
5960b63515d37fc7a7142ed1d8927f10100b1a4e
34513 F20101130_AAALSA crane_s_Page_38.pro
9d5832efc38ddb9f27858e5c359d1d62
57404132cefcceb3f0b8ecc41f829f9eb0b1c090
48929 F20101130_AAALRM crane_s_Page_19.pro
5252715206140d7d016c08d82816bfc2
d7925fe96816e4d3d8f6cf9a96f273696b29468b
F20101130_AAALQY crane_s_Page_91.tif
02e769af56840fb83e5d5eb17918ce1d
e3555fb345ffc21472bff5189078dbb388b979fa
51368 F20101130_AAALSB crane_s_Page_40.pro
45c7c042a0e5a468885c64525e8949ff
881cf2ab7d0a8dd92c5425b3aeb1e3654ab4e561
49512 F20101130_AAALRN crane_s_Page_20.pro
f8358a41246f15b50e9352d074d16cbe
a9fdc4ff4d8cfdf08dfeacd207a814ef1e141411
F20101130_AAALQZ crane_s_Page_92.tif
36bd7c9e62343b9966343b194f9f3aaf
0953b0351e1af6fc9c2c44918b8c4712aec97038
49667 F20101130_AAALSC crane_s_Page_41.pro
8428daf4f2fab14d2b3307308fa52df8
4d2c64f76ad98bf48a937fdca8cb39725774c245
30560 F20101130_AAALRO crane_s_Page_22.pro
e4dcb5a26c69ba7d7bf1d6e18d779869
027d62133b3b69acf2c583373c85c99e1027c783
8155 F20101130_AAALSD crane_s_Page_42.pro
c2141f323212eed3423027cbab3d0064
16cc7e42f78587c96dd0532e603717a1c938c562
40032 F20101130_AAALRP crane_s_Page_23.pro
09703232fce9aa4db14c1c94ce4bba37
04503c437de3d357c1de87ede542ea14b1ea9f99
42092 F20101130_AAALSE crane_s_Page_43.pro
380c30c3ce6682770c7f50cb5c233c5e
6ff4bfbf7ed04cdfba4a9087d8b3ca070dbb882c
38280 F20101130_AAALRQ crane_s_Page_24.pro
d687cf050008f49f2f3b5c290f647466
f3f0216520beee1318bf34f7a87e82f0667fef37
50276 F20101130_AAALSF crane_s_Page_44.pro
c77df30a526d8160ee7a5800a59bbd76
56aa57fbe3b6e8f677f5d0c983e2f79e32a8496e
41403 F20101130_AAALRR crane_s_Page_25.pro
39d8c0fd7fd95cabc4142684f487bc9c
61c6c5e2612abc2391e98eb8e3b3cb71c69a470a
14432 F20101130_AAALSG crane_s_Page_45.pro
9db61303d213276f27ca37cc25dcb107
79773697064a5ec4f9b2338727b1c8a210cf18e4
46537 F20101130_AAALRS crane_s_Page_27.pro
1b12b6c2a8bb6c6542496954af71494d
b68b6700ea046df326da77569ce6163a88adf309
47399 F20101130_AAALSH crane_s_Page_48.pro
97955d685c4cdefa9a9962bba7c25cf0
4332c46a1d54df54eba0546ff45805b1a3a58eb9
51359 F20101130_AAALRT crane_s_Page_28.pro
b984f60d120c56b7f9337d3e5c337f2b
8b2b56a0cc512e0fb841e1910e6b1ed1a5288b31
44280 F20101130_AAALSI crane_s_Page_49.pro
640c388c56cef599c2d3b9bafedf63f4
e91a68128dea2e68ee452c0ec2c269d7b06e7f13
47756 F20101130_AAALRU crane_s_Page_29.pro
1e7fd92f51a93cd163ce604f7f08887d
7337a40f2ac3598a2ea9781edd4f70a69b971e05
18740 F20101130_AAALSJ crane_s_Page_50.pro
2751b5f3b90cf551910e55cdb6205e9b
2143ee0b5d2a7395e352f05d2612c09becb59cc8
52227 F20101130_AAALRV crane_s_Page_30.pro
ab55aa51eb840f79e3df57366e109aa4
e095734502b7d6d74f9bd4cbd92caf64be7836b1
21387 F20101130_AAALSK crane_s_Page_51.pro
88cc31612baa0c3566c81a27c7924a5f
4da463e87be087fd347e4939a8ed0364d957accb
50056 F20101130_AAALRW crane_s_Page_31.pro
0de7149e835362856c34c33e4336dce6
20cafd540577b3c405654aa0364d558ac581f441
36064 F20101130_AAALTA crane_s_Page_80.pro
0e27560632705c7d16304aad32887e3b
98681ad7f110cdc09fe5db9b77fa49d5d799dcfc
29597 F20101130_AAALSL crane_s_Page_52.pro
629324d9115719ab3232c620cdd56b5e
318a3177fad3e75f2b10e01c0c26ee28adda635f
51682 F20101130_AAALRX crane_s_Page_34.pro
53a560ccd8ca3b7212ae40fdb0027703
9bedee0c42b60a45673ab263ce7af3fd177654f6
35785 F20101130_AAALTB crane_s_Page_82.pro
f4fdf9ff1b409ff3d3beb18b0a741f41
9467668804b451c28cb364994aba7704da0bc6ef
21008 F20101130_AAALSM crane_s_Page_54.pro
2729b9a1f25ba595288c7c36e51e79e4
e3fb4571829b13de08ddd79f98335e4a6a74f7cc
38380 F20101130_AAALRY crane_s_Page_36.pro
c912dee13a8dadd36e4bb29e131ce38b
666fb86cdeda3910f135a633b02db97c89e6ed42
33246 F20101130_AAALTC crane_s_Page_83.pro
0d854d1cefe722f4598f9b4fef8eaebc
ab9a53171d9fb2c5ec0b70a55faf6b6d55bc594d
23479 F20101130_AAALSN crane_s_Page_57.pro
8e3545a4d445aba9dcbe243dcd24a0a9
20ef68474181e0f7c6a30cfa11a8d131c4dff08f
43470 F20101130_AAALRZ crane_s_Page_37.pro
0e8c667773c2f4bde0dd0a20bb0c9e8a
f98e12a61b148d77236ec7fc0200c6d0277e39e9
29727 F20101130_AAALTD crane_s_Page_84.pro
91851fb35f8a57ffbec454c5c33711a5
bef22aacd4b61dc1120fa6c839fffcdc726b7d75
22814 F20101130_AAALSO crane_s_Page_59.pro
e21f09a9497130d3b7a6e5ad1071a269
de490b39ad7982fa8fd44765964590d48f47e7e0
22836 F20101130_AAALTE crane_s_Page_85.pro
30ed7b0e168d94738ccfa28e09f05edf
d40c0dbd299db40d25597c966b4f57923881c8d6
24580 F20101130_AAALSP crane_s_Page_60.pro
b95515aeb38ea92b746f469061066d27
204763c45694040a27f23452167b643b984315c5
42570 F20101130_AAALTF crane_s_Page_86.pro
8aeb4d17df2f13b8dc4afc50eb00bc9d
c7f901f814adc4dac377279d30b8ae91c1c9621d
22026 F20101130_AAALSQ crane_s_Page_61.pro
c2ea7cca1a70cb162468abbdc3389732
7b7c7b9788068d6d80d55af9606a685c5b439dcf
22629 F20101130_AAALTG crane_s_Page_87.pro
c350e206f5b1c3045e602e3dddcb93a5
e512bb2dc3ceb68692d1f4082966a34d95f05abe
8214 F20101130_AAALSR crane_s_Page_65.pro
fd97f4412206a9f5f4522ca7f2fca590
7c06fd11759d8c1b36778c149dd85959538f9b71
57293 F20101130_AAALTH crane_s_Page_88.pro
d70241d37e3665f26219b1137fb7d62a
27d4e55948761d18060b98d98a7febc44a1dd2cf
49876 F20101130_AAALSS crane_s_Page_66.pro
552364946bd1c4226aa3809249247df4
efe29a2b2fefaa9cd4c08b9487c87f303924b46a
59721 F20101130_AAALTI crane_s_Page_89.pro
c2ee72f60d0cea7ff281b5e1b6b7fb78
aa9a2bb62984b82bdc2759300e32f47b8dd2f587
21608 F20101130_AAALST crane_s_Page_68.pro
d1ed27b32233581506f6a5c54abbe5f7
0f0d43d655c48a955e3a12315af2e3930da1ee99
8712 F20101130_AAALSU crane_s_Page_69.pro
f99b4a78ecf8395d716c0ae188072fc3
0dbc1e7c4957cd76479017360811edd487a0e090
6433 F20101130_AAALTJ crane_s_Page_92.pro
5649a3de4bda40fee2d28dbffc69eede
315987f9e0ceb5085a9a487b1efe7ce3ac1d3dad
43665 F20101130_AAALSV crane_s_Page_70.pro
e5ca80f8d70cdcce328e853444e2ba88
bcff0bb29e0549c7db12e41404ae5abac8592e6a
23799 F20101130_AAALTK crane_s_Page_93.pro
d576dca46283cc3f7b630fa836d61f3d
31b8ade9e920f4655dbe7922718e37396f59864c
24399 F20101130_AAALSW crane_s_Page_71.pro
dc9587191207ed7b6295a60e4a64cf5a
2c3b53c96627fae25d8078351b408356c5cbf295
599 F20101130_AAALTL crane_s_Page_01.txt
a56b053f2b96ba453f9867895ec3682b
6dbdde84be9f4da4e16546722fd00ae464759557
30712 F20101130_AAALSX crane_s_Page_72.pro
f4e799d34cd609a16a0f4808ed5373c3
d5ef3add7c41a7bc16fce3fdbbabc90d211e504b
1991 F20101130_AAALUA crane_s_Page_20.txt
4e7e96171913e7e9ff07a0fbfa293c66
f8125553215f4c530e8e50b302c1fd2ac7bca1cb
109 F20101130_AAALTM crane_s_Page_02.txt
398ab8e126d1927b83c739e89b5bc21d
52517e83557a2405c99d42b7ce28b194c39af395
3748 F20101130_AAALSY crane_s_Page_73.pro
c71ef53679252bd191e360c72fdc2e3f
a70a5d684608afa4cd533937e0c511ee6e57ee74
1440 F20101130_AAALUB crane_s_Page_22.txt
8af00d31afa583ecbeb5ce216706780e
14ee8562ea8ea4861123360d805e01f52d067f7b
125 F20101130_AAALTN crane_s_Page_03.txt
713d75d1922060456437636936ca582b
56b71188da62cf7f5e6ab9abf97039e61ae6a296
53011 F20101130_AAALSZ crane_s_Page_76.pro
3991c51136b8f472abb20956d007c07a
e31807321ee37fa3e3b3c8f842a7ecec2bc7d988
1662 F20101130_AAALUC crane_s_Page_23.txt
532f1b1cb1c0f14f6ea323f661177a22
52d73f6dd1c967d83a05e74e42be6d166562e0ce
992 F20101130_AAALTO crane_s_Page_04.txt
f76255bc23a17af5cf4e84f80144bc54
06b326156235b02fce3015d2f1fc919da3fcee6a
1634 F20101130_AAALUD crane_s_Page_24.txt
984eef469407e3996b942454ab9814d5
b11af5f6e49f4034656697f595a28c34ef19f46b
3345 F20101130_AAALTP crane_s_Page_05.txt
aa3394ffaa0f9fab6d6715f31742b529
5140b083b9850eb989f328933d4d201cc0b16dcf
1780 F20101130_AAALUE crane_s_Page_25.txt
4662d7db776a786f210907fa7eb46e55
ade5866a3668fd70620bb54f366b41335cacb222
4275 F20101130_AAALTQ crane_s_Page_06.txt
eb5d8ff10b10a588c4b048505897d551
482a84028d41a2b5f15ac930a314749985210522
1952 F20101130_AAALUF crane_s_Page_26.txt
8ea841bf0bfb4cf5abe45c09f1c7c5f8
1518e440c641312ecd189c1e3bd3e36d57b984fe


Permanent Link: http://ufdc.ufl.edu/UFE0010845/00001

Material Information

Title: High Resolution Transmission Electron Microscopy Analysis of the Influence of Grain Boundary and Triple Grain Junction Crystallinity and Chemistry on Silicon Carbide-Based Armor with Small Additions of Aluminum, Boron, and Carbon
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010845:00001

Permanent Link: http://ufdc.ufl.edu/UFE0010845/00001

Material Information

Title: High Resolution Transmission Electron Microscopy Analysis of the Influence of Grain Boundary and Triple Grain Junction Crystallinity and Chemistry on Silicon Carbide-Based Armor with Small Additions of Aluminum, Boron, and Carbon
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010845:00001


This item has the following downloads:


Full Text











HIGH RESOLUTION TRANSMISSION ELECTRON MICROSCOPY ANALYSIS OF
THE INFLUENCE OF GRAIN BOUNDARY AND TRIPLE GRAIN JUNCTION
CRYSTALLINITY AND CHEMISTRY ON SILICON CARBIDE-BASED ARMOR
WITH SMALL ADDITIONS OF ALUMINUM, BORON, AND CARBON














By

SAMANTHA CRANE


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Samantha Crane




























This document is dedicated to my parents for never letting me give up.















ACKNOWLEDGMENTS

I would like to acknowledge Ceramatec, Inc., for the funding of this project. I

would also like to thank Dr. Darryl Butt for chairing my committee and for his continued

guidance throughout my undergraduate and graduate studies. My committee members

Drs. J. J. Mecholsky, Jr. and Amelia Dempere were invaluable resources on this project.

In addition, I would like to thank Kerry Seibien of MAIC at the University of Florida and

Dr. Helge Heinrich of MCF at the University of Central Florida for their help with

performing the TEM and EFTEM studies presented in this thesis. Edgardo Pabit's

insight, help, and guidance were essential to the completion of my research. I would like

to thank Dr. Erik Kuryliw and Soroya Benetiz for teaching me how to perform TEM

sample preparation and how to polish. Finally, I would like to thank Dr. Butt's research

group for their help and support.















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ................................................... vii

LIST OF FIGURES ................................................ viii

ABSTRACT ................................................... ................. xii

CHAPTER

1 IN T R O D U C T IO N .................................................. .. ....................................1.. .. ... 1

2 EFFECT OF SINTERING PARAMETERS ON MICROSTRUCTURAL
DEVELOPMENT AND MECHANICAL PROPERTIES ................ ..................... 5

Mechanical Behavior of Ceramic Armor Systems.................................................5...
T oughening M echanism s........................................ ....................... ...............5...
C rack bow ing ........................................................................................ .. .6
C rack deflection ................ .............. .............................................. 6
C rack bridging .................................................................... ............. .. .7
M icrocracking .................................................................... ............. .. .8
Transform ation toughening ..................................................... ............... 10
Toughening M echanism s in SiC ..................................................... ................ 10
Structural Properties of Silicon Carbide.................................................................. 11
Sintering of Silicon C arbide ....................................... ........................ ............... 12
Solid State Sintered SiC ..................................... .. ......... .......... .. .............. ... 14
L iquid-P hase Sintered SiC ............................................................. ............... 15
L iquid-phase sintering aides.................................................... ............... 15
A l-B -C additive sy stem ................................................... ........ ............... 17
Microstructural Features of Liquid-Phase Sintered SiC........................................19
P hase T ransform ation ..................................................................... ................ 2 1
Grain Boundary Films in Liquid-phase Sintered SiC.....................................21
Triple Junction Phase Formation and Crystallization ....................................21

3 M A TERIALS AND M ETH OD S .......................................................... ................ 24

Materials Processing of ABC-SiC by Ceramatec Inc............................................24
C characterization ............. .. ................ .................. .................... ......... .. ................24
Toughness Testing by Ceram actec, Inc ................ ................................... 24


v









Microhardness Testing by Ceramatec, Inc. ....................................................25
R eitveld A analysis by Ceram atec, Inc. .......................................... .................. 25
Grain Morphology Characterization by Ceramatec, Inc. ...............................25
Transmission Electron Microscopy (TEM)................................26
Transmission electron microscopy sample preparation ..............................27
H igh resolution lattice im aging ............................................... ................ 28
Energy dispersive spectroscopy (ED S) ............................... ..................... 28
Energy-filtered transmission electron microscopy (EFTEM)...................29

4 RE SU L TS A N D D ISCU SSION ............................................................ .................. 31

Transmission Electron Microscopy Characterization............................................31
Grain Boundary and Intergranular Film Characterization...............................31
T riple G rain Junction s ......................................... ........................ ................ 34
C om positional Studies .............. .................. ................................................ 36
0 w t. % A l ............................................................................ .. . .......... .. 36
0 .5 w t. % A l .......................................................................... ....... ................. .... 3 7
1 w t. % A l ............................................................................ ........ ................. .... 4 2
1 .5 w t. % A l .......................................................................... ....... ................. .... 4 4
4 wt. % Al ............... ... ............ ... ........................ 47
Materials Characterization Performed at Ceramatec, Inc......................................48
Fracture Mode Determination by Ceramactec, Inc ........................................50
Relationship Between Hardness and Toughness by Ceramactec, Inc..............54
Grain Morphology Characterization by Ceramatec, Inc. ...............................54
XRD and Reitveld Analysis by Ceramatec, Inc ............................................54
Correlation Between Microstructure and Mechanical Properties...............................55
T o u g h n e ss ............................................................................................................ 5 8
W eak interfaces .............. ...... ............ ............................... ...............58
R esidual stresses ..................................................................................... 60
H igh aspect ratio grains........................................................... ................ 68
L arge grains ......................................................................................... 69
H a rd n e ss ...................... ... ................................................................................... 7 1
Effect of grain size on hardness .................................................. 72
Effect of aspect ratio on hardness ........................................... ................ 73

5 SUMMARY AND CONCLUSIONS....................................................................74

C o n c lu sio n s ......................................................................................................... 7 4
Suggested F uture W ork ........................................ ....................... ................ 75

L IST O F R E FE R E N C E S .................................................. ........................................... 76

B IO G R A PH ICAL SK ETCH .................................................................... .................. 81















LIST OF TABLES


Table page

2-1. Ram sdell Notion of Comm on SiC Polytypes ....................................... ................ 12

4-1. Grain Boundary Width and Intergranular Film Determination by HRTEM.............34

4-2. Toughness and Hardness Measurements of SiC with 0-6 wt. % Al....................... 58

4-3. Characterization of SiC w ith 0-6 wt. % Al........................................... ................ 59

4-4. Reitveld Analysis for SiC Samples with 0-6 wt. % Al..........................................63















LIST OF FIGURES


Figure page

2-1. Fracture toughness increase with particle size due to microcracking. ..................... 10

2-2. Si-C tetrahedra, which form the basic structural unit of SiC polytypes.................11

2-3. A close packed plane of spheres with the sphere centers denoted by A.
Subsequent planes can be stacked in the sphere valleys denoted by B or C to
form the different SiC structures ......................................................... ............... 12

2-4. Schematic of the competing energy terms during densification.............................. 13

3-1. The above diagram illustrates electron-matter interactions in transmission
electron m icroscopy ............. ................ ............................... ...............26

4-1. The grain boundaries in a) 0 wt. % Al, b) 0.5 wt. % Al, and c) 1 wt. % Al were
completely crystalline and contained no intergranular film. In d) 1.5 wt. % Al
and e) 4 wt. % Al the grain boundaries are amorphous and 1 nm wide. Please
note that the scale marker in a), b), and c) is 2 nm. The scale marker in d) and e)
is 5 n m .................................................................. ............................................... ... 3 3

4-2. Triple grain junctions for ABC-SiC. a) 0 wt. % Al, b) 0.5 wt. % Al, c) 1 wt. %
Al, d) 1.5 wt. % Al, and e) 4 wt. % Al. Please note that the scale marker is 20
nm in a) and b), 10 nm in c), 200 nm in d), and 100 nm in e). ..............................35

4-3. This is a series of micrographs and EDS spectra from a grain boundary in the 0
wt. % Al sample. There is no variation observed in composition between the
grains and grain boundary ....................................... ........................ ................ 37

4-4. EFTEM elemental maps of a grain boundary in the 0 wt. % Al sample.............. 38

4-5. The above figure is a micrograph of a three grain junction in the 0 wt. % Al
sample and the corresponding EDS spectra. The EDS data shows no secondary
phase present in the three grain junction, and no variation in composition
between the three grain junction and surrounding grains. ..................................38

4-6. EFTEM elemental maps of a triple junction in the 0 wt. % Al sample. There is
no change in composition between the three grains and the three-grain junction. ..39

4-7. The above STEM image and corresponding EDS spectra show the presence of
carbon inclusions between SiC grains in the 0 wt. % Al samples. ........................39









4-8. EFTEM elemental maps of a C inclusion surrounding a pore in the 0 wt. % Al
sam p le .................................................................................................... ....... .. 4 0

4-9. The above a) STEM image and b)-e) corresponding EDS spectra are taken from
b), e) the grain boundary and c), and d) the surrounding grains. .............................41

4-10. EFTEM elemental maps of a grain boundary in the 0.5 wt. % Al sample..............41

4-11. The above STEM image and corresponding EDS spectra are taken from the
triple junction and the surrounding grains in the 0.5 wt. % Al samples. The
secondary phase is confined to the triple junction and does not extend into the
grain boundaries. The scale m arker is 100 nm .................................... ................ 42

4-12. EFTEM elemental maps of an Al-O rich inclusion in the 0.5 wt. % Al sample. ....42

4-13. EELS spectra taken from a triple junction in the 0.5 wt. % Al sample shows
does not show crystalline A 120 3. ......................................................... ................ 43

4-14. A STEM image and corresponding EDS spectra showing metal and B-C rich
inclusions. The scale bar is 200 nm .................................................... ................ 43

4-15. A STEM image of two grains and a grain boundary from the 1 wt. % Al sample
and the corresponding EDS spectra. The composition does not vary between the
grains and grain boundary ............ .. ........................................................... 44

4-16. EFTEM elemental maps of a grain boundary and transgranular crack in the 1
wt. % Al sample. There is no variation in composition across the grain
boundary or at the site of the crack ..................................................... ................ 45

4-17. A STEM image of a triple junction in 1 wt. % Al showing that aluminum and
oxygen segregate on the triple point and that the composition changes as a
function of depth into the triple junction. The scale marker is 100 nm. ...............45

4-18. EFTEM elemental maps of a triple junction in the 1 wt. % Al sample. Notice
that the very center of the triple junction is Si and C free...................................46

4-19. A STEM image and corresponding EDS spectra of grains and contaminations
observed in the 1.0 w t. % A l sam ple .................................................. ................ 46

4-20. A STEM image of two grains and a grain boundary from 1.5 wt. % Al ABC
sample and the corresponding ED S spectra. ....................................... ................ 47

4-21. EFTEM elemental maps of a grain boundary in the 1.5 wt. % Al sample. Note
the formation of an Al-O rich intergranular film. Scale marker is 20 nm ............47

4-22. A STEM image and EDS spectra of a triple junction in the sample containing 1.5
w t. % A l ................................................................................................ ........ .. 4 8









4-23. EFTEM elemental maps of a triple junction in the 1.5 wt. % Al sample. Notice
that the entire triple junction is void of Si and C. ............................................... 48

4-24. EELS spectra taken from the triple junction in the 1.5 wt. % Al sample shows
that the triple junction is filled w ith A1203.......................................... ................ 49

4-25. A STEM image and corresponding EDS spectra of two grains and a grain
boundary in 1.5 wt. % Al sample. This site shows two different grains. The
upper grain is a SiC grain and the lower grain is that of an Al-O rich grain. The
presence of aluminum-rich grains were only observed for 1.5 wt. % Al and 4.0
w t. % A l ................................................................................................................ ... 4 9

4-26. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain
boundary, two SiC grains, and a B-C rich inclusion in the 4 wt. % Al sample .......50

4-27. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain
boundary, a B-C rich inclusion, and an A14C inclusion in the 4 wt. % Al sample...51

4-28. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain
boundary, two SiC grains, and an Al-O-C rich inclusion in the 4 wt. % Al
sample. Note that the area of the grain boundary that is farthest from the
inclusion contains Al, while the area closest to the inclusion is depleted in Al.......51

4-29. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain
boundary, a SiC grain, and a carbon inclusion imbedded in an Al-O-C rich
inclusion in the 4 w t. % A l sam ple...................................................... ............... 52

4-30. The above STEM image and corresponding EDS spectra are of a SiC-SiC grain
boundary, and two SiC grains in the 4 wt. % Al sample. Note the presence of Al
at th e g rain b ou n d ary ........................................................................... ............... 52

4-31. EFTEM elemental maps of a grain boundary between two SiC grains in the 4
w t. % A l sam p le ...................................................................................................... 5 3

4-32. EFTEM elemental maps of a multiple grain junction in the 4 wt. % Al sample.
The circled grain is Al-B-C rich and may be Al8B4C7. The area between the
g rain s is A l-O rich ........................................................ ..........................................5 3

4-33. SEPB fracture surfaces for samples with 0-6 wt. % Al. There is a clear change
in fracture mode between transgranular and intergranular between 1 and 15 wt.
% A l. The scale m arker is 10 [tm ...................................................... ................ 55

4-34. Comparison of HV1 (left) and HKI (right) for various SiC materials. A change
in fracture mode is evident between 1 and 1.5 wt. % Al....................................56

4-34 continued. Comparison of HV1 (left) and HKI (right) for various SiC materials.
The onset of the transgranular fracture mode is accompanied by a high degree of









crack branching and crushing. A change toward mixed mode fracture occurs at
4 wt. % Al at which point the cracks are straight and there is no crushing. ............57

4-35. The above graphs plot Knoop hardness and SEPB toughness as a function of Al
content. The change in hardness and toughness with changing Al content is
in v ersely related ...................................................................................................... 5 9

4-36. Change in aspect ratio and grain size with Al content. ......................................60

4-37. Polished and chemically or plasma-etched surfaces of SiC samples with 0- wt.
% A 1 ....................................................................................................... ........ .. 6 1

4-38. X-ray diffraction patterns SiC with 0-6 wt. % Al. Note that starting powder was
beta-3C and that in addition to SiC polytypes, A18B4C7 phase is noted at high Al
c o n te n ts .................................................................................................................. ... 6 2

4-39. Polytypes from Rietveld analysis for SiC-0.6 wt. % B-0.2 wt. % C samples
with Al contents ranging from 0 to 6 wt. %. ..................... ............................62

4-40. Plot of Stress vs. Temperature for a particle in an infinite matrix........................66

4-41. Effect of residual stress fields around triple junctions on crack propagation..........67

4-42. Intergranular cracking in the 1.5 wt. % sample. The figure on the left shows
cracking through the intergranular film at the grain boundaries. The figure on
the right shows a missing triple junction due to the crack deflecting around the
triple junction, but within the amorphous secondary phase. Note that the image
on the right is slightly out of focus, which makes the crack appear to be a broad
region that is lighter in color ...................................... ...................... ................ 68

4-43. Plot of toughness vs. aspect ratio shows a negative slope and argues against
crack bridging. The three data points at the bottom of the graph correspond to
the sam ples that fail transgranularly.................................................... ................ 70

4-44. Plot of toughness vs. grain size. Toughness generally increases with increased
g ra in siz e ............................................................................................................. . 7 1

4-45. Plot of the effect of grain size on hardness. The plot shows that hardness
increases with grain size for the samples that fail via intergranular fracture...........72

4-46. The plot of hardness vs. aspect ration shows that the hardness is not dependant
o n a sp ect ratio ........................................................................................................... 7 3















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

HIGH RESOLUTION TRANSMISSION ELECTRON MICROSCOPY ANALYSIS OF
THE INFLUENCE OF GRAIN BOUNDARY AND TRIPLE GRAIN JUNCTION
CRYSTALLINITY AND CHEMISTRY ON SILICON CARBIDE-BASED ARMOR
WITH SMALL ADDITIONS OF ALUMINUM, BORON, AND CARBON


By

Samantha Crane

August 2005

Chair: Darryl Butt
Major Department: Materials Science and Engineering

Recent studies have shown that toughness, hardness, and fracture mode are a

function of interface crystallinity and chemistry in silicon carbide-based armor systems.

Therefore, a thorough investigation of the effects of varying additive concentration on the

microstructure and chemistry of the intergranular phases including the grain-boundary

films and triple-junction phases is essential in order to engineer a better ceramic-based

armor. For this object four SiC-based materials were prepared and studied. The silicon

carbide materials were created with the boron and carbon content held constant while

varying the amount of aluminum available.















CHAPTER 1
INTRODUCTION

Ceramic armor systems have been in use against small caliber projectiles since the

1960s, however, the way in which combat is conducted has changed with the advance of

defense related technology. For this reason, there is now greater emphasis on creating

lower weight combat vehicles that need minimal combat preparation time and can endure

attack by long rod penetrators (Gooch 2002).

The armor is designed to "defeat" the projectile at the interface by increasing its

"dwell" time. To obtain long dwell times in ballistic tests, the armor material must have

high hardness. In addition, an increase in ductility will result in improved resistance to

penetration of the projectile if the hardness is maintained (Flinders et al. submitted).

Advanced silicon carbide-based armor is amongst the United States Army's top choices

for ceramic armor systems.

Difficulty in obtaining fully dense compacts in silicon carbide due to the low self

diffusion coefficient and the highly covalent nature of the Si-C bond (Huang et al. 1986)

hindered the use of silicon carbide until Prochazka (Prochazka 1975) developed a

technique in 1975 using boron and carbon additives for pressureless solid-state sintering.

While Prochazka's technique produced fully dense silicon carbide, its low fracture

toughness, which is approximately 2.5 MPa-m1/2 as measured by the single-edge

precracked beam (SEPB) technique, limited its application (Flinders et al. submitted).

Since Prochazka's pioneering work, a variety of additive systems including oxides,

carbides, nitrides, and metals have been used with silicon carbide to promote liquid phase









sintering, which aids in lowering the full densification sintering temperature and

obtaining compacts close to the theoretical density (Biswas et al. 2001). Liquid phase

sintering, however, introduces intergranular phases at the grain-boundaries and triple-

junctions during sintering, which have a pronounced influence on the final properties

(Zhou et al. 2002). For example, strong intergranular bonding is desirable for creep

resistance and high temperature strength, while the activation of toughening mechanisms

such as crack deflection, crack bridging, and micro-cracking depend on "sufficiently

weak" intergranular bonding (Moberlychan and De Jonghe 1998). In addition, recent

studies have shown that toughness, hardness, and fracture mode are a function of grain

boundary and triple grain junction crystallinity and chemistry in silicon carbide-based

systems (Moberlychan and De Jonghe 1998). For the specific application of ceramic

armor, it is important to achieve both a high level of toughening and strength (Flinders et

al. submitted). Therefore, control of the microstructure and chemistry of the intergranular

phases is of particular interest when engineering ceramic armor systems.

Of particular interest is the aluminum, boron, and carbon system, which exhibits in-

situ toughening. The additives in this system aid in the cubic to hexagonal phase

transformation, which can be controlled to dictate the final microstructural morphology

(Moberlychan et al. 1998). In addition, the sintering aides diminish the number of flaws

that limit the strength and toughness of the final part (Zhang et al. 1998). While the

toughness of the final part usually increases due to a change in fracture mode from

transgranular to intergranular cracking, the hardness diminishes with increased additive

content. Thus, it is important to use a sufficiently small amount of additive in order to

maintain the high hardness, while still aiding in the sintering process and microstructural

development (Moberlychan and De Jonghe 1998).









There are several in-situ toughening mechanisms that can be employed in

aluminum-boron-carbon silicon carbide (ABC-SiC); however, many of these mechanisms

diminish the material's hardness. Therefore, a thorough investigation of the effects of

varying additive concentration on the microstructure and chemistry of the intergranular

phases including the grain-boundary films and triple-junction phases is essential in order

to engineer a ceramic-based armor with both high hardness and toughness. For this

objective, four SiC-based materials were prepared and studied. The silicon carbide

materials were created with the boron and carbon content held constant while the amount

of aluminum was varied.

To address the effect of varying additive concentration on the intergranular

microstructure and chemistry, the following tasks were performed: the measurement of

the size of the grain-boundaries and triple-junctions, and the determination of the amount

of crystallinity of the grain-boundaries and triple-junctions; and the quantification of the

grain-boundary and triple-junction chemistries. To accomplish this task, high resolution

transmission electron microscopy (HRTEM) in conjunction with the following techniques

were performed: lattice imaging to determine grain boundary width and secondary phase

crystallinity; and energy dispersive spectroscopy (EDS), and energy filtered transmission

electron microscopy (EFTEM) to quantify chemical compositions.

The work presented in this thesis was done in cooperation with Ceramatec, Inc.

of Salt Lake City, Utah, as part of an SBIR for the United States Army. Ceramatec, Inc.

processed the ABC-SiC, performed toughness testing, microhardness testing, Reitveld

analysis, X-ray diffraction, and grain morphology characterization. When the results of

work performed at Ceramatec, Inc. are presented in this thesis, they will be identified as

such in the section headings.









The thesis is organized as follows: a review of previous work on SiC-based

ceramics with an emphasis on secondary phase formation in liquid-phase sintered SiC

and the effects of SiC microstructure on mechanical properties will be given in Chapter 2.

The experimental methods, including the analytical electron microscopy methods used in

this study are introduced in Chapter 3. The results of the detailed investigation of the

intergranular phases of ABC-SiC are presented in Chapter 4. Finally, in Chapter 5,

conclusions from the results presented in Chapter 4 are drawn and a correlation is made

between the microstructure and chemistry of intergranular phases and the mechanical

behavior of the final part.















CHAPTER 2
EFFECT OF SINTERING PARAMETERS ON MICROSTRUCTURAL
DEVELOPMENT AND MECHANICAL PROPERTIES

Mechanical Behavior of Ceramic Armor Systems

As discussed in the previous chapter, ceramic armor is designed to "defeat" the

projectile at the interface by increasing its "dwell" time. To obtain long dwell times in

ballistic tests, the armor material must have high hardness. In addition, an increase in

ductility will result in improved resistance to penetration of the projectile if the hardness

is maintained (Flinders et al. 2003). In this chapter, toughening mechanisms in ceramics

and the necessary microstructural features for their activation are discussed. In addition,

the effect of processing parameters such as type of sintering and sintering additives on the

formation of specific microstructural features are discussed.

Toughening Mechanisms

Two classes of toughening mechanisms exist in ceramics. Intrinsic mechanisms,

which include crack bowing and crack deflection, operate ahead of the crack tip and act

as the material's inherent defense against microstructural damage and cracking by

increasing the materials resistance to crack initiation. Extrinsic toughening mechanisms

operate in the wake of the crack and reduce the local stress intensity at the crack tip.

Extrinsic mechanisms, which include crack-bridging, microcrack toughening, and

transformation toughening, are the main source of toughness in brittle materials.

Extrinsic toughening mechanisms, which act in the crack wake, give rise to R-curve,

behavior, (Nalla et al. 2004). In a material exhibiting R-curve behavior, a greater applied

load must be applied to continue the advance of the crack.

5









Crack bowing

Crack bowing occurs when the crack front interacts with tough particles or

inclusions. The crack is initially pinned by the tough obstacle, but can pass it by bowing

on the same plane. While the stress fields are quite different, crack bowing is

conceptually similar to the interaction of a dislocation line with a precipitate. Crack

bowing assumes that the obstacle is impenetrable; however, if the obstacle were to break

before the bowing process is complete a cracked ligament would be left in the crack

wake. If the crack passes without breaking the particle an uncracked ligament will be left

in the crack wake and thus, crack bowing can be a precursor to crack-bridging (Green

1998).

Crack deflection

Crack deflection occurs when the crack is tilted or twisted out of the plane normal

to the applied stress. The change in orientation reduces the crack extension force, which

means that a larger applied stress is required for fracture and the toughness will therefore

increase. This reduction is greater for twisting than tilting and has been shown to be

dependant on the volume fraction and shape of the particle providing the obstruction with

rod shaped particles providing the greatest increase in toughness. As the aspect ratio of

the particles increases, the toughening increment increases due to the increase of twist

angle with increased aspect ratio. In addition to the effect of geometry and the effect of

the volume fraction of particles, crack deflection has been shown to also be the result of

several factors: the local stress field at the obstacle, the presence of a low toughness

interface, or the presence of a cleavage plane (Green 1998).

In multiple phase systems, such as one containing grains and intergranular films or

triple junction phases, the difference in thermal expansion mismatch between the grains









and secondary phases can give rise to residual stresses. If the difference in thermal

expansion between the grains and secondary phases is positive, a compressive radial

stress is developed at the matrix-particle boundary and a tensile tangential stress is

developed in the matrix. This will attract the crack to the particle. When the difference

in thermal expansion is positive, the crack will be deflected between particles and a more

tortuous crack path will be created, which will increase toughness (Hertzberg 1996). If

one considers a material with an intergranular film in which the intergranular film is the

matrix and the grains are the particles, this second case can be used to explain the origin

of intergranular fracture.

With respect to intergranular films, the local compressive residual stresses around

the grains cause diminishing stress intensity directly at the tip of the deflected crack and

are a result of a difference in thermo-elastic properties between the intergranular film and

grains (Sternitzke 1997). Crystallizing triple junctions can also cause residual stresses,

the magnitude of which is dependent on the volume change upon crystallization (Pezzotti

and Kleebe 1999).

Crack bridging

In the discussion of crack bowing it was mentioned that cracks can by-pass an

obstacle and leave an intact ligament in the crack wake (Green 1998). For crack bridging

to occur, the uncracked ligament, usually an elongated grain, must remain intact as the

crack approaches and it must be energetically favorable for the crack to deflect along the

ligament surface, rather than cutting through it. This can occur if interfacial bonding

occurs due to the presence of a weak interface, which would reduce the stress intensity at

the crack tip and allow the ligament to survive and bridge the crack (Nalla et al. 2004).









The uncracked ligaments span the crack wake and sustain part of the applied load

that would otherwise be used to advance the crack. Thus, the crack bridges shield the

crack tip from part of the applied load. This increases fracture toughness and gives rise to

R-curve behavior (Green 1998 and Nalla et al. 2004).

Several testing methods can be employed to determine if crack bridging is the

dominant toughening mechanism. First, the crack tip and immediate crack tip wake can

be viewed directly with either scanning electron microscopy or transmission electron

microscopy (Yuan et al. 2003). Second, a comparison of elastic compliance of the

cracked specimen can be compared to the theoretical compliance of an ideal bridge- and

microcrack-free crack of the same length. Since crack-bridging increases the modulus, a

reduction in compliance would suggest the presence of crack-bridging (Nalla et al. 2004).

Finally, when the ligaments have a plate-like geometry, the fracture toughness is given by

K 2 =KO2 + 0.81E 2 1 67d
where Kjc is the fracture toughness measured by the SEBP technique, Kjc is the

toughness in the absence of crack bridging, E is Young's modulus, T is the frictional stress

between grains, eL is the bridging rupture strain, aL is the aspect ratio, and d is the mean

grain size. A plot of Kc2 as a function aL1.67 d should yield a straight line if reL2 is a

constant (Flinders et al. submitted).

Microcracking

Stress-induced microcracking can give rise to crack-tip shielding, which may

increase a material's toughness. During microcrack toughening a frontal process zone is

developed in which the microcracks form. Microcrack formation, and subsequent

opening, gives rise to a volumetric increase (Green 1998). The resulting dilation and

reduction in modulus of the frontal process zone, if constrained by the surrounding rigid









material, will shield the crack tip and cause an increase in toughness (Nalla et al. 2004).

In addition, microcracking can lead to crack branching, When crack branching occurs, a

greater applied stress is required to drive the increased number of cracks and thus

toughness will increase (Meyers and Kumarchawla 1999).

Microcracking occurs when residual stresses build-up due to the following

phenomena: phase transformations, thermal expansion anisotropy in one phase materials,

or thermo-elastic property mismatch in multiphase materials (Green 1998). Phase

transformation can cause volumetric expansion and contraction. When this volumetric

change occurs in a confined phase, residual stress can arise and lead to the formation of

microcracks. Thermal expansion anisotropy arises when the crystal symmetry is less than

cubic, such as hexagonal, rhombohedral, monoclinic, tetragonal, and triclinic. Expansion

is different in each crystal direction. Thermal expansion anisotropy results in contraction

of individual grains from neighboring grains at differing rates, depending on crystal

orientation. This produces mechanical stresses at the interface, which can generate

microcracks. Thermo-elastic property mismatch produces residual stresses in a similar

manner as the two aforementioned mechanisms (Case et al. 2005). Areas of low

toughness are particularly susceptible to microcracking: grain boundaries and

intergranular films (Green 1998).

The degree of toughness increase due to microcracking is dependent on the grain

size. As the particle size increases, the toughening increment will increase due to the

creation of a tortuous crack path until the critical grain size is reached. At this point

spontaneous microcracking will occur and the toughness will decrease since microcrack

toughening is only effective in materials that experience microcracking upon crack










propagation. It should also be noted that there is also a critical particle size below which

cracking will not occur (Green 1998).


Spontaneous Microcracking
Microcracking Zone 4






0)
I -____


__--


No Discrete
Microcracking


Particle Size

Figure 2-1. Fracture toughness increase with particle size due to microcracking (Green
1998).

Transformation toughening

Transformation toughening occurs when a stress induced phase transformation of

an unstable phase occurs in the process zone. The phase transformation is associated

with a dissipation of energy and the development of compressive residual stresses that

will oppose crack advance (Hertzberg 1996).

Toughening Mechanisms in SiC

Of the five toughening mechanisms discussed, crack deflection, crack bridging and

microcracking are the most possible mechanisms active in SiC. The necessary conditions

for these mechanisms are similar. Weak interfaces and a residual stress field are needed

for the activation of all three mechanisms. In addition, crack deflection and crack

bridging also rely on the particles, in this case grains, having a high aspect ratio, while









microcracking is dependant on grain size. All of these features can be manipulated by

choosing the right processing conditions and additive systems. The remainder of this

chapter will discuss the effect of sintering parameters on microstructural development.

Structural Properties of Silicon Carbide

Silicon carbide exists in several polytypes, of which the following are the most

common: 3C, 2H, 4H, 6H, and 12R (Fischer et al. 1990). The polytypes consist of

strong, primarily covalently bonded (88% covalent, 12% ionic) Si-C tetrahedra, which

contain a Si-C bond length of 1.89 A. The tetrahedra join at the corners to form the

different SiC polytypes (Ye 2002).









C

Figure 2-2. Si-C tetrahedra, which form the basic structural unit of SiC polytypes.

The structure of the polytypes can be simulated by stacking sheets of close packed

spheres, each of which consists of both a layer of Si atoms and a layer of C atoms. The

sequence in which these sheets are stacked distinguishes one polytype from another. In

Figure 2-3, site A represents the position of the spheres in the first sheet. Subsequent

sheets of spheres can either be stacked in the B or C sites, which represent the valleys

between the spheres. Thus, the sheets can be denoted as A, B, or C-sheets, depending on

the position of the spheres (Ye 2002).

The simplest stacking sequences are as follows: ...ABAB... and ...ABCABC...,

which correspond to the hexagonal wurtzite and cubic zinc-blend structures, respectively.









The hexagonal and rhombohedral polytypes are collectively referred to as a-SiC, whereas

the cubic structure of SiC and is often referred to as P-SiC (Ye 2002).













Figure 2-3. A close packed plane of spheres with the sphere centers denoted by A.
Subsequent planes can be stacked in the sphere valleys denoted by B or C to
form the different SiC structures.

In the Ramsdell notation, the number of sheets needed to define the stacking

sequence is followed by a letter, which represents the crystal structure: C for cubic, H for

hexagonal, and R for rhombohedral. Table 2-1 lists the common polytypes and their

corresponding stacking sequences (Ramsdell 1947).

Table 2-1. Ramsdell Notion of Common SiC Polytypes.
Ramsdell Notation 3C 2H 4H 6H 15R
Type 13-SiC a-SiC a-SiC a-SiC a-SiC
Stacking
Sequence (ABC
Notation) ABC AB ABCB ABCACB ABCBACABACBCACB


The effects of different additive systems on polytype transformation and its

relationship to mechanical properties will be discussed later in this chapter. Chapter 4

will discuss the specific effects of the Al, B, C additive systems used in this body of work

to polytype transformation and its relationship to mechanical properties.

Sintering of Silicon Carbide

Sintering is the most common processing route for polycrystalline SiC; however,

due to the highly covalent nature of the Si-C bond, it is difficult to produce final parts









with densities approaching the material's theoretic density. Several factors contribute to

the limited thermodynamic driving force and sluggish kinetics: high grain boundary

energy, low self-diffusion coefficients, and coarsening (Prochazka 1987, Tanaka 1991).

The thermodynamic driving force for densification and mass transport is the

reduction of free energy in the system due to the formation of interfaces from free

surfaces. In the specific case of grain boundary formation, the grain boundary energy,

YGB, must be less than twice the solid vapor energy, 7sv. For pore shrinkage, where a

three grain junction is formed, the ratio of the grain boundary energy to the solid-vapor

interfacial energy, YGB/ySV, should be less than /3 (Greskovich and Rosolowski 1976). In

a highly covalent material such as SiC, the grain boundary energy is so great that the

reduction of excess free energy is extremely small and densification is practically

inhibited (Prochazka 1987, Tanaka 1991).



77GBv


SiC Grain SiC Grain



SiC Graine C a
SiC Grain

Figure 2-4. Schematic of the competing energy terms during densification.

In addition to having an extremely low thermodynamic driving force for

densification, the kinetics of SiC densification are sluggish (Ye 2002). As previously

mentioned, the high degree of covalency in the Si-C bond limits densification. This

arises from the extremely low self diffusion coefficients of Si and C in SiC (~ 10-13 cm2/s

for Si and 10-11 cm2/s for C), which limits mass transport and makes diffusion-









controlled solid-state sintering infeasible (Hong 1979). In addition, coarsening, not

densification, dominates during the sintering process due to silicon carbide's relatively

high vapor pressure, which causes larger grains with larger radii of curvature to grow at

the expense of smaller grains with smaller radii of curvature. This leads to coarsening

without the elimination of pores (Prochazka 87).

Therefore, to produce final parts with densities approaching the material's

theoretical density, the conditions contributing to the poor thermodynamics and kinetics

must be overcome. This has led to the use of sintering aides, which can lower the grain

boundary energy, increase the solid vapor interfacial energy, and increase the mass

transport. Depending on the additive system, the aforementioned behavior can be

achieved in the absence or presence of a liquid phase, which are referred to as solid-state

sintering and liquid-phase sintering, respectively (Ye 2002). It is also important to note

that the mechanical properties in ceramics are controlled by the flaw size and flaw size

distribution. Sintering aides are commonly used to achieve high density and thus

decrease the flaw size due to porosity (Zhang et al. 1998).

Solid State Sintered SiC

Prochazka was the first to develop pressureless sintering of SiC with small amounts

of B and C. He gave a thermodynamic explanation of the effect of additives on the

sintering of SiC. The Carbon removes the surface oxide by carbothermal reduction,

which increases the solid/vapor interfacial energy. The B segregates to the grain

boundaries and reduces the grain boundary interfacial energy (Prochazka 1975). As

discussed in the previous section, an increase in the solid-vapor interfacial energy and a

decrease in the grain boundary energy will favor pore shrinkage and grain boundary

formation.









Tanaka argued that the C also acted to decrease the grain boundary energy, since a

certain amount of carbon was necessary, regardless of the oxide content of the starting

powder (Tanaka 1984). In addition, B and C have the added effect of increasing self-

diffusion in SiC (Birnie 1986).

While the combined effects of B and C allow for solid state sintering of SiC, the

low fracture toughness, which is 2.5 MPa-m1/2 as measured by the single-edge precracked

beam (SEPB) technique, and exaggerated grain preclude the use of sold-state sintered

SiC as ceramic armor (Flinders et al. submitted).

Liquid-Phase Sintered SiC

Liquid-phase sintering, which can be accomplished via pressureless sintering, gas-

pressure sintering, hot-pressing, and hot-isostatic-pressing, is achieved when the melting

temperatures of the additives are lower than the sintering temperature (Ye 2002). This

condition allows for the formation of a liquid phase that will spread among the grains

during sintering. Upon cooling, the liquid phase may persist as a secondary glassy or

crystalline phase at the grain boundaries and triple grain junctions. Liquid-phase

sintering results in more uniform densification and suppression of the exaggerated grain

growth that may occur in solid state sintering. In addition, the secondary phases can alter

the properties of the sintered ceramic (Kaneko et al. 2000).

Liquid-phase sintering aides

Several additive systems have been used in liquid-phase sintering of SiC.

Aluminum additives have been used in several forms, such as Al-metal, A1203, A14C3, and

A1N (Kaneko et al. 2000), and in combination with several other additives, such as Y203,

B, C, B4C, and CaO (Kim et al. 1995, Gu et al. 1996, Zhou et al. 1999). Each additive









system has different effects on the microstructure and mechanical properties of the final

sintered SiC (Ye 2002).

A1203 is frequently used in combination with Y203 to promote high aspect ratio a-

SiC grains and a YAG (Y3A15012) intergranular phase, which improve the fracture

toughness and creep resistance. Chia and Lau suggested that the main toughening

mechanism was microcracking between the YAG secondary phase and the SiC grains

(Chia and Lau 1991). Nitrogen in the form of AIN is used either alone or in combination

with A1203 and Y203 to retard the P3--a transformation, suppress grain growth, and

promote the formation of oxynitride glass, which have also been found to increase the

fracture toughness (Kim and Mitomo1999). One example of AIN-containing SiC armor

is SiC-N, which is manufactured by Cercom.

Inomata et al. (Inomata et al. 1980) showed that in the Al-B-C additive system a

liquid phase was present at 18000C near the composition of A18B4C7. This non-oxide

system has recently received attention for its ability to produce a three-fold increase in

the fracture toughness of SiC (Chen and Lau 2000). With respect to increased toughness,

the Al-B-C system can achieve high toughness at lower processing temperature than the

YAG system when a comparable amount of liquid phase is used (Flinders et al. 2003).

However, for a given toughness SiC with YAG as a secondary phase is harder than SiC

sintered with a comparable amount of Al-B-C additives (Flinders et al. 2003). The effects

of B and C additives were discussed in the previous section. However, the combined

effect of Al, B, and C differ greatly from those of B and C alone. This additive system

promotes the 3--a phase transformation, the 6H-SiC to 4H-SiC polytype transformation,

the formation of elongated grains, the formation of intergranular films, and the formation

of secondary phases at the triple-grain junctions (Cao et al. 1996).









Al-B-C additive system

The effects of B and C additive were discussed in the previous section, however the

combined effect of Al, B, and C differ greatly from those of B and C alone. Researchers

at the Lawrence Berkeley National Laboratory performed experiments on ABC-SiC in

which they varied the amount of aluminum from 1-3wt%. The boron and carbon

concentrations were held constant at 0.6wt% B and 2 wt% C. Similar to the work of Chia

and Lau, the ABC-SiC exhibited R-curve behavior. They observed the formation of either

crystalline or amorphous secondary phases in the SiC structure as large regions that wet

amongst many matrix grains, triple point junctions, and grain boundaries. The Berkeley

group observed the formation of an amorphous secondary phase at the grain boundaries,

the crystallinity of which is dependent on the aluminum concentration with increasing

crystallinity at higher wt% Al. According to the Berkeley group, the lowered work-of-

adhesion afforded by the amorphous secondary phase that promotes toughening and

intergranular cracking can be controlled by the second phase chemistry (Moberlychan

and De Jonghe 1998).

In a different study at Lawrence Berkeley National Laboratory, two different

experiments on ABC-SiC were performed. In the first set of experiments, aluminum foil

was imbedded in ABC-SiC powder and hot pressed. This allowed for the formation of an

aluminum gradient and the measurement of SiC properties as a function of wt% Al. In

the second set of experiments, the wt% Al was varied from 3-7%. The boron and carbon

concentrations were held constant at 0.6wt% B and 2 wt% C. In both sets of experiments

the researchers observed a lowered densification temperature and an increased number of

triple junctions at higher wt% Al (Zhang et al. 2003).









In the first set of experiments, they observed that the area density, aspect ratio, and

size uniformity of the elongated grains varied as a function of aluminum content. The

aspect ratio is particularly dependent on aluminum concentration. At low concentrations

of approximately lwt% Al, the aspect ratio is reduced to nearly unity, thus grains are

equiaxed. In addition, the researchers observed a decrease in indentation toughness and

a decreased tendency toward intergranular cracking as the wt% Al was decreased (Zhang

et al. 2003).

The second set of experiments was run to better understand the effect of aluminum

concentration on the properties ofABC-SiC. The variation in aluminum content from 3-

7wt% Al altered the microstructure. As with the first set of experiments, the area density,

aspect ratio, and size uniformity of the elongated grains varied as a function of aluminum

content. A nearly linear increase in the aspect ratio is observed up to 6wt% Al while the

length reaches a maximum at 5wt% Al. A bimodal grain distribution was observed at

aluminum contents greater than 3wt%. In addition, the toughness degraded at

concentrations greater than 6wt% Al (Zhang et al. 2003).

The increased aluminum content also had a profound effect on the crystallinity of

the grain boundaries. As previously stated, the Berkeley group found that at aluminum

concentrations lower than 3wt% the second phase was amorphous. Increasing the wt%

Al lead to increased crystallinity of the grain boundaries. At 3wt% Al, 85% of the second

phase films examined were amorphous and at 6wt% Al and above all of the second phase

films examined were crystalline. It is also important to note that increasing the overall

aluminum content raised the concentration of aluminum at the grain boundaries and in

the grain. At aluminum concentrations greater than 5wt% the grains were saturated and

metallic aluminum precipitated at the grain boundaries (Zhang et al. 2003).









According to the Berkeley experiments, increasing the boron content increases the

number density of elongated grains, coarsens the grains, reduces the aspect ratio,

promotes phase transformations, enhances grain boundary diffusion, and produces a

driving force for mass transport. In addition, boron is more effective in promoting the

P3-c transformation (Zhang et al. 2003).

Increased carbon content promotes elongated grains, enhances phase

transformation, enhances grain boundary diffusion, and produces a driving force for mass

transport. It also increases the self-diffusion rate, as well as the grain boundary diffusion

rate of SiC. In addition, the carbon increases apparent density up to 4wt% because of

pore elimination and decreases apparent density after 4wt% because of appearance of

carbon phase in materials. Furthermore, a decrease in grain growth was observed with

increased carbon content. It is important to note that boron and carbon do not form

intergranular films by themselves and in the absence of aluminum will incorporate into

the SiC lattice (Zhang et al. 2003).

The grain morphology is controlled by the B/C ratio and when the B/C ratio favors

elongated grain growth, the aluminum accelerates it and produces higher aspect ratios.

When the Al/B and Al/C ratios are reduced, there are less liquid phases and the aluminum

effects are diminished. When the Al/B and Al/C ratios are increased, boron and carbon

are depleted from the lattice and dissolve into the liquid phases. In addition, at constant

Al/B/C ratios, changing the total amount of additives will still change grain morphology

and phase composition (Zhang et al. 2003).

Microstructural Features of Liquid-Phase Sintered SiC

As mentioned in the previous two sections, the sintering additives used and

processing route can greatly affect the final microstructure of SiC materials. However,










there are several common microstructural features of liquid-phase sintered SiC: SiC

grains, intergranular films at the grain boundaries, and secondary phases at the triple

grain junctions (Moberlychan et al. 1996). However, cases have been reported where

direct grain-to-grain interfaces are present (Kaneko et al. 1999). Figure 2-5 is a

schematic of the aforementioned microstructural features.

Both a-SiC and P-SiC starting powders may be used in liquid-phases sintering of

SiC. Typically a-SiC powders are used when a fine equiaxed microstructure is desired

(Sigl and Kleebe 1993). However, when increased toughness is desired, P-SiC is

generally used as a starting powder since the P-to-ca phase transformation upon sintering

will result in an interlocking microstructure of elongated plate-like grain, which has been

shown to increase fracture toughness via crack-bridging and crack deflection (Gilbert et

al. 1996). The phase transformation can be accelerated by seeding the P-SiC starting

powder with a-SiC (Baud and Theveno 2001), and retarded by incorporating N-

containing sintering aides, such as AIN (Jun et al. 1997).

Crystalline
Secondary
Phase / Large Glass Pocket








Amorphous
Triple Junction



Direct Grain-to-
Grain Boundary Intergranular Film
Figure 2-5. Schematic of the microstructural features of liquid-phase sintered SiC (Ye
2002).









Phase Transformation

As mentioned previously, the Al-B-C additive system promotes the 3->-a phase

transformation, the 6H-SiC to 4H-SiC polytype transformation, and the formation of

elongated grains. The high aspect ratio grains offer reinforcement and increased

toughness by promoting several toughening mechanisms.

Grain Boundary Films in Liquid-phase Sintered SiC

While the presence of intergranular films may appear to be a kinetic phenomenon,

the existence of intergranular films at various experimental conditions suggests that these

films may be in a state of thermodynamic equilibrium. Clarke devised a model to explain

the existence of intergranular films based on a force balance between attractive Van der

Waals forces across the grain boundaries and repulsive steric forces, capillary forces, and

electric double layer forces in the film (Clarke 1987, Chen et al. 1993). The model

predicted the presence of intergranular films on the order of 1-2 nm in several materials,

which agrees with experimental data presented to date. While developing his model,

Clarke considered the case of SiC without additives and concluded that SiC is the only

material where the attractive forces are greater than the repulsive forces (Clarke 1987).

Thus SiC without additives would not have a stable intergranular film.

Varying the width and composition of the intergranular film has a large effect on

the mechanical properties of SiC. In Al-B-C-containing SiC, the intergranular film has

been shown to increase the fracture toughness by a factor of three. Therefore, tailoring of

the intergranular film is an important task.

Triple Junction Phase Formation and Crystallization

The Berkeley group observed that the reduction of large secondary phases resulted

in the appearance of triple junctions. These regions are several nanometers wide and up









to 10 |tm long. They form by heterogeneous nucleation on the basal plane of the SiC.

Typically, only partial crystallization can occur and an amorphous region is observed

between the crystalline secondary phase and the SiC grain (Moberlychan and De Jonghe

1998).

Full crystallization is hindered by several factors: the composition of the

intergranular phase may shift into regions of the phase diagram in which crystallization

capability is low; the volume change due to crystallization of the intergranular phase is

constrained by the surrounding SiC grains, which causes internal stresses that create a

thermodynamic barrier to complete crystallization; and there may be kinetic hindrance to

crystallization (Bonnell et al. 1987, Raj 1981, Kessler et al. 1992).

The crystallization process greatly affects the composition distribution in the triple

junction. The amorphous region between the crystalline secondary phase and SiC grains

has a different composition from the crystalline secondary region. This is due to

undesirable solute rejection to the side and ahead of the solidification front, which allows

the secondary phases to grow free of SiC. Typically, the secondary phases at the triple

junctions are a crystalline ternary or binary containing Al-O-B, Al-B-C, Al-C, or Al-O

(Moberlychan and De Jonghe 1998).

In addition, the extent of triple junction crystallization can affect the mechanical

properties of the final part. Kessler et al. (Kessler et al. 1992) found that the volume

change due to crystallization is strongly dependant on the degree of crystallization with

increased crystallization resulting in increased volume change. This volume change

results in residual stresses whose magnitude is proportional to the volume change of the

constrained phase, or triple junction. Pezzotti and Kleebe (Pezotti and Kleebel999)

found that a nearly 100% fraction of crystallized secondary phase at the triple junction






23


will create substantial tensile stresses due to volume contraction, which drives cracks

toward the triple junction and causes subsequent interface delamination. This

phenomenon can lead to crack brnching and intergranular fracture.

Crack deflection will manifest itself as roughness in the final fracture surface.

When SiC fractures in an intergranular fashion, the fracture surface has a high degree of

surface roughness, which is associated with twist and tilt crack deflection (Moberlychan

et al. 1998).















CHAPTER 3
MATERIALS AND METHODS

Material processing, characterization, and electron microscopy methods used to

investigate the microstructure and chemistry of the SiC-based materials are presented in

this chapter.

Materials Processing of ABC-SiC by Ceramatec Inc.

P-SiC (Superior Graphite grade HSC-059) with surface area in the 15-17 m2/g

range was mixed with Al, B, and C additives dispersed with a polyamine polyester

polymer (Avecia Chemical grade Solsperse 24000) by adding one wt. % of the polymer,

based on solids, to 400 grams of reagent grade toluene. The Al powder (Valimet grade H-

3), which has an average size of 3 [tm, was added in amounts of 0, 0.5, 1, 1.5, and 4 wt.

%, whereas the boron (H. C. Starck, amorphous B grade S-432B) was kept constant at 0.6

wt. %. The carbon was introduced as 4 wt. % Apiezon grade W wax, assuming a 50 %

yield after pyrolysis to give ~2 wt. % C. The slurries were deagglomerated for two hours

with a paint shaker and then rolled overnight before drying. Powders were passed

through a 44 am screen before hot pressing at 28 MPa in stagnant Ar inside graphite dies

at 2100 C for 1 h.

Characterization

Toughness Testing by Ceramactec, Inc.

The hot pressed billets were ground in order to form 3 mm x 4 mm x 45 mm bars,

which were subsequently indented with a 98 N Knoop indenter. The samples were then

pre-cracked to initiate crack growth. The single edge pre-cracked beam (SEPB) tests









were performed with loads ranging from 4 kN to 18 kN using spans of 4, 5, or 6 mm.

The original crack was marked by a dye (Magnaflux Zyglo ZL-60D), using vacuum

infiltration and oven dried at 110C overnight and cooled before testing. All crack planes

were parallel to the hot pressing direction (Flinders et al. submitted).

Microhardness Testing by Ceramatec, Inc.

A Leco microhardness machine was used to perform Vickers and Knoop hardness

measurements on polished SEPB bars under a 9.8 N load. The load was applied at 50

[tm/s, with a 15 second dwell time (Flinders et al. submitted).

Reitveld Analysis by Ceramatec, Inc.

Rietveld analysis was used to determine SiC polytypes present in the densified

samples with X-ray diffraction patterns collected from 30-800 2-theta, with a step size of

0.020/step and a counting time of 4 sec/step (Flinders et al. submitted).

Grain Morphology Characterization by Ceramatec, Inc.

Polished samples of liquid phase sintered SiC with Al content ranging from 0.5 to 6

t. % Al were plasma-etched by evacuating and back-filling with 400 millitorr of CF4-

10% 02 and etching for 20-40 minutes. The SiC sample with 0 wt. % Al was etched in

molten KOH at 5500C for 10-15 seconds. Grain size was determined by the line-

intercept method, where the multiplication constant ranged between 1.5 (equiaxed grains)

and 2.0 (elongated, plate-shaped grains). Typically, 200-300 grains were measured for

each composition in order to get a mean grain size. The aspect ratio of the five most

acicular grains in each of three micrographs was used to estimate a comparative aspect

ratio (Flinders et al. submitted).










Transmission Electron Microscopy (TEM)

Transmission electron microscopy, which uses a finely focused beam of electrons,

is an important part of the experimental procedure used to investigate the microstructure

and chemistry of the SiC-based materials. A wide variety of secondary signals from the

specimen are produced when the electron beam interacts with the matter in the specimen.

In addition to producing signals used for imaging, these secondary signals are also used

to obtain chemical information (Williams and Carter 1996). Figure 3-1 illustrates some

of the interactions between the electron beam and the sample.

Backscattered
El tetro Incident Beam Secondary Electrons
Electrons

\ Characteristic
X-rays

Auger Electrons \ / /
\ / // Visible Light





"Absorbed" ___ Ii Electron-Hole
Electrons Ba Eltrons




SX-rays




DirectT ransmitted Inelasticallv Scattered
Electrons Bea Electrons

Figure 3-1. The above diagram illustrates electron-matter interactions in transmission
electron microscopy (Williams and Carter 1996).

Electron interactions can be divided into two classes: elastic and inelastic-scattering

events. While elastic and inelastic scattering give rise to many useful signals, only the









signals relevant to TEM and the associated characterization techniques will now be

discussed. Elastic scattering events affect the trajectories of the beam electrons inside a

specimen without altering the kinetic energy of the electrons. In TEM, elastically

scattered electrons are the major source of contrast in images and also create the intensity

distributions in diffraction patterns. Inelastic scattering events result in a transfer of

energy from the beam electrons to the atoms of the specimen. Inelastic scattering events

lead to the generation of characteristic X-rays, which are used in energy dispersive

spectroscopy (EDS), and inelastically scattered electrons, which are used in electron

energy loss spectrometry (EELS) (Williams and Carter 1996).

Transmission electron microscopy sample preparation

In order to perform the electron characterization techniques discussed later in the

chapter, the samples must first be prepared for use in the TEM. Transmission electron

microscopy specimens were prepared by the standard mechanical thinning method, which

creates a self-supporting disk. A 3 x 4 mm rectangle with a thickness of 1 mm was cut

from each bulk specimen (3 x 4 x 45 mm bars made for SEPB testing) with a low speed

diamond saw. The samples were then mounted on an aluminum polishing stub with

crystal-bond adhesive and wet polished to a 100 |tm thickness and 3 tm finish. The

samples were then cut into 3 mm diameter discs using an ultrasonic drill and then further

thinned on a precision dimpling machine to 50 |tm using a 3 [tm diamond solution and a

flattening tool. The center of the sample was subsequently thinned to 20 |tm with a

dimpling tool. Then four to five hours of Argon ion milling at 4 keV, I pA and a beam

angle of 120 was carried out until the sample was perforated.









High resolution lattice imaging

The high resolution lattice images presented in this thesis were generated on a

JEOL JEM-2010F FEG with a point-to-point resolution of 1.9 A. The images were

obtained with the Help of Kerry Seibien at the Major Analytical and Instrumentation

Center at the University of Florida. High resolution lattice imaging, or phase-contrast

imaging, requires the selection of several diffracted beams, in addition to the transmitted

beam, to create the image. The collected beams interfere and cause the intensity of the

beam to vary sinusoidally with different periodicities for different values of the

diffraction vector. This in turn causes lattice fringes in an area of crystallinity. At high

magnifications these fringes can be imaged and information on the spacing of the planes

normal to the diffraction vector of the beams can be obtained (Williams and Carter 1996).

Lattice imaging can be used to investigate interfaces such as grain boundaries and

three grain junctions by orienting the interface parallel to the electron beam and tilting the

surrounding grains on-axis. The aforementioned experimental conditions will create

lattice fringes in the surrounding grains. In the absence of a secondary phase, an abrupt

interface of near atomic dimensions can be observed. If an amorphous region exists at

the grain boundary, it will not exhibit lattice fringes and can be imaged directly. Thus,

lattice imaging may be used to determine the existence of amorphous secondary phases at

interfaces (Williams and Carter 1996).

Energy dispersive spectroscopy (EDS)

The EDS data presented in this thesis was generated on the JEOL 2010F equipped

with an Oxford INCA 200 EDS system, which has a probe size of approximately 5 nm.

Energy dispersive spectroscopy produces plots of the X-ray counts, or intensity, versus

the X-ray energy. The plot consists of two types of signals: Bremsstrahlung X-rays, and









characteristic X-rays. Bremsstrahlung X-rays are the radiation which is emitted when

electrons are decelerated due to interaction with the sample. It is characterized by a

continuous distribution of radiation and gives no information on the chemical

composition of the sample. On an EDS plot, Bremsstahlung X-rays make up the

background peaks. The larger Gaussian peaks are the characteristic X-ray peaks.

Characteristic X-rays occur when an incident electron ionizes an inner-shell electron, and

an electron drops down from a higher energy level to fill the vacancy. The radiation

energy produced is equal to the difference between the atomic energy levels. Since the

difference in energy levels is unique to each element, characteristic X-rays give

compositional data (Williams and Carter 1996).

Energy-filtered transmission electron microscopy (EFTEM)

The EFTEM performed in this thesis was done at the Materials Characterization

Facility at the University of Central Florida with the help of Dr. Helge Heinrich on a

Technai F30 equipped with a FEG and a GATAN GIF. Elemental maps can be formed

using EFTEM by imaging with electrons that have lost energy corresponding to the inner-

shell ionization edge of a particular element of interest (Hofer et al. 1997). For

elemental mapping, the intensity of any part of the electron energy loss spectrum can be

selected to form an electron spectroscopic image (ESI) (Williams and Carter 1996);

however, it is necessary to remove the background contribution to the image intensity

(Hofer et al. 1995). The three window technique was employed to create ESIs with the

background contribution removed. This technique involves recording images at two

energy-loss values before and one energy loss value after the ionization energy of the

element under investigation, as seen in Figure 3-2 (Schaffer et al. 2003). Then an







30


extrapolated background image is calculated and subtracted from the ionization edge

image, which gives an elemental map (Hofer et al. 1995).









E, E2 E, E

Figure 3-2. Schematic of the three-window technique. Two pre-edge images are used to
estimate the background and calculate an elemental map.















CHAPTER 4
RESULTS AND DISCUSSION

The discussion in chapter 2 focused on toughening mechanisms in ceramic

materials and the microtructural features necessary for their activation. Per that

discussion it was concluded that the most likely toughening mechanisms active in SiC

depend on weak interfaces and the presence of a residual stress field. Tests performed at

Ceramatec, Inc. showed a change in fracture mode from transgranular to intergranular

between 1 and 1.5 wt. % Al. Therefore, this chapter will focus on characterizing the

changes in microstructure of each of the five SiC materials described in chapter 3 and the

effect of these changes on mechanical behavior.

Transmission Electron Microscopy Characterization

High resolution lattice imaging, energy dispersive spectroscopy, and energy filtered

transmission electron microscopy data were taken for samples of SiC with Al wt. %

content that varied from 0 wt. % Al to 4 wt. % Al, and a constant 0.6 wt. % B and 2 t.%

C. The following sections include a detailed analysis of these samples based upon the

crystallinity, thickness, and composition of the grain boundaries and triple grain

junctions.

Grain Boundary and Intergranular Film Characterization

Based on the high resolution TEM data taken from the samples, shown in Figure 4-

1, the thickness of the grain boundary and the intergranular film (IGF) varies between the

samples with 0-1 wt. % Al, and the samples with 1.5 wt. % Al or greater. In the samples

with 0-1 wt. % Al added, there is a direct transition from grain to grain, which results in a









negligible grain boundary thickness. The samples with 1.5 wt. % Al or greater contain

grain boundaries on the order of 1 nm. The crystallinity of the grain boundary, based on

the lattice image of the samples also varies. While the samples with 0-1 wt. % Al show a

completely crystalline grain boundary, the samples with greater than 1.5 and 4 wt. % Al

show an amorphous grain boundary. These observations are summarized in Table 4-1.

Konishita et al. found that the solubility limit of Al in SiC is 0.2 wt. % Al

(Konishita 1997). Hence, it would be expected that at Al contents greater than 0.2 w.%

Al, the Al would be expelled from the grain and form an intergranular film. Since this is

not the observed phenomenon, the presence, or lack, of the amorphous film can not be

explained based on the solubility limit alone. As Clarke explained in reference to

equilibrium intergranular film thickness, SiC in the absence of additives is one of the

only materials in which the attractive Van der Waals forces across the grain boundaries

are greater than the repulsive steric forces, capillary forces, and electric double layer

forces in the film (Clarke 1987). According to the equilibrium segregation theory of

McLean, the amount of solute present at the grain boundary is directly related to the

solute content in the grains and inversely related to temperature.

CD = ACeQIRT
CD is the solute atomic fraction at the grain boundary. A is a constant related to

vibrational entropy. Q is the free energy of segregation at the grain boundary, which is

related to the energy difference between an atom in the bulk and an atom at the grain

boundary. R is the gas constant and T is the absolute temperature (Konishita 1997).

Therefore, it is possible that there exists a critical level of additives necessary to

promote the formation of an intergranular film and it was not reached at the prescribed









processing temperature until 1.5 wt. % Al. This topic will be revisited in the discussion

of the compositional studies performed on the five SiC materials.


a)












c)


b)












d)


Figure 4-1. The grain boundaries in a) 0 wt. % Al, b) 0.5 wt. % Al, and c) 1 wt. % Al
were completely crystalline and contained no intergranular film. In d) 1.5 wt.
% Al and e) 4 wt. % Al the grain boundaries are amorphous and 1 nm wide.
Please note that the scale marker in a), b), and c) is 2 nm. The scale marker in
d) and e) is 5 nm.









Table 4-1. Grain Boundary Width and Intergranular Film Determination by HRTEM
Al Grain Grain
Content Boundary Boundary
(wt.%) Width Crystallinity
0.0 ~0 nm Crystalline
0.5 ~0 nm Crystalline
1.0 ~0 nm Crystalline
1.5 1 nm Amorphous
4.0 1 nm Amorphous

Triple Grain Junctions

In general the sintering aides will attract impurities in the starting powder, react

with the native oxide on the particle surface, and form a mass transport medium during

densification (Falk 1997). In the specific case of Al-B-C additives, boron and carbon will

not form a secondary phase without aluminum. This assertion has been proven

experimentally by many researchers (Prochazka 1975 and Moberlychan and De Jonghe

1998).

Per the discussion in chapter 2, in SiC sintered without Al the carbon reacts with

the native oxide to form SiO and CO. This reaction depletes the oxygen from the system

(Stobierski and Guberat 2002). However in the presence of Al, 0 will be trapped in the

system because of the highly negative free energy of formation and comparatively low

vapor pressure of aluminum oxides. Thus, the 0 content will increase with increasing Al.

The evidence of this behavior will be presented in later sections of this chapter. This

phenomenon produces a greater volume and crystallinity of secondary phase for higher

wt. % Al compositions, when the B and C compositions are held constant. Figure 4-2

illustrates the formation of triple junction phases once the solubility limit of Al in SiC has

been reached (0.2 wt. % Al). The figure also illustrates that the volume of the triple

junction phase increases with increasing aluminum content.














4,-. .,' _. .....^


~t ~.9


e)
Figure 4-2. Triple grain junctions for ABC-SiC. a) 0 wt. % Al, b) 0.5 wt. % Al, c) 1 wt.
% Al, d) 1.5 wt. % Al, and e) 4 wt. % Al. Please note that the scale marker is
20 nm in a) and b), 10 nm in c), 200 nm in d), and 100 nm in e).


1.5 vvt% Al


I.I: .









Compositional Studies

Energy dispersive spectroscopy (EDS) was performed on the samples using the

HRTEM in scanning transmission electron microscope (STEM) mode. EDS is a semi-

quantitative chemical analysis method. There are several factors to take into

consideration when performing EDS on TEM samples. First, the quality of the data is

dependent upon the electron beam, or probe size with relation to the area of sample being

studied. While the manufacturer lists the minimum probe size as 0.8 nm, the minimum

attainable probe size in practice is closer to 3-5 nm. The grain boundaries in this set of

samples are typically 1 nm wide. Thus, there may be signal coming from the surrounding

grains as well. It should also be noted that the data is averaged through the bulk.

Therefore, if there are overlapping grains, the grain boundary is not oriented parallel to

the beam direction, or the grain boundary does not go all the way through the sample

thickness characteristics X-rays will be generated from the grain as well and affect the

chemical analysis. Therefore, it is more practical to use the difference in characteristic

peaks present and in peak heights between the secondary phases and the grains to identify

variations in composition than to try to use EDS as a quantitative chemical analysis

technique.

In addition to EDS, which uses characteristic X-rays as the signal, EFTEM was

also used to create elemental maps of the samples. Since the signal for EFTEM comes

from the electron energy loss spectrum, EFTEM provides a method of verifying the

presence of species in the samples.

0 wt. % Al

According to the literature, boron and carbon do not form intergranular films by

themselves and in the absence of aluminum will incorporate into the SiC lattice (Zhang et









al. 2003). This can be seen in the EDS data in Figures 4-3 through Figure 4-6. However,

several carbon inclusions were observed in this sample due to excess free carbon. This

phenomenon can be seen in Figures 4-7 and 4-8.

As observed in Figures 4-3 through 4-6, there is very little variation in composition

between the grain, the grain boundary, and the three grain junctions. The presence of a

clean grain boundary combined with the grain boundary width and crystallinity data

suggests the absence of an intergranular film. In addition, the triple grain junctions in

this sample were usually clean except for the occasional carbon inclusion. An EDS

spectra and EFTEM analysis of a triple junction in this sample shows no variation in

composition between the grains and three grain junctions, which suggests the absence of

a secondary phase at the triple junctions. The presence of secondary phases at the grain

boundaries and triple grain junctions generate residual stresses that promote several

toughening mechanisms and intergranular fracture. The absence of these secondary

phases may explain the low toughness of solid state SiC.

0.5 wt. % Al

While this sample is above the solubility limit of Al in SiC, the grain boundaries

are free of secondary phase and the composition does not change between the grains and










ull Scale 252 Cursor -U 675 kV (0 ds) -400 .... cale 252 dts Cursor -0 675 key (0 cts)
Grain Boundary G 400nm ,
Grain
Figure 4-3. This is a series of micrographs and EDS spectra from a grain boundary in the
0 wt. % Al sample. There is no variation observed in composition between
the grains and grain boundary.




















Grain
boundary in the 0 wt. % A sample.













Figure 4-4. EFTEM elemental maps of a grain boundary in the 0 wt. % Al sample.


J S, Spectrum 4

S i '.



0 05 1 15 2
ull Scale 239 .,s Cursor 0 007 keV (721 ds) keV

Triple Junction


Grain Boundary


S' SpEcrum Z

Si




1i1 SEde 29 s Cursor D C07 ke (761 Ms) keV
Grain
Figure 4-5. The above figure is a micrograph of a three grain junction in the 0 wt. % Al
sample and the corresponding EDS spectra. The EDS data shows no
secondary phase present in the three grain junction, and no variation in
composition between the three grain junction and surrounding grains.







































Figure 4-6. EFTEM elemental maps of a triple junction in the 0 wt. % Al sample. There
is no change in composition between the three grains and the three-grain
junction.


C 4 Spedrum 1






0 05 1 15 2
ull Scale 634 ds Cursor -0 023 keV (495 cs) e

SiC Grain

c lSpectrum 2






0 05 1 15 2
'ullScale d113 C ,rsor -D D04 0eV(1D008dcs]) kev


Carbon Indusion SiC Grain

Figure 4-7. The above STEM image and corresponding EDS spectra show the presence
of carbon inclusions between SiC grains in the 0 wt. % Al samples.


grain boundaries. This can be seen in Figures 4-9 and 4-10. This data in combination


with the clean, narrow, crystalline grain boundary observed in the HRTEM images


' Spedrum
































Figure 4-8. EFTEM elemental maps of a C inclusion surrounding a pore in the 0 wt. %
Al sample.

testifies to the assertion that no intergranular film exist at the grain boundaries. Unlike

the 0 wt. % Al sample, the 0.5 wt. % Al sample shows a small 0 peak at the grain

boundary. This may be due to the diminished effectiveness of the C additive at removing

the native oxide in the presence of Al.

Since the solubility limit of Al in SiC has been reached in this sample, rejection of

Al from the grains and the formation of a region of secondary phase are expected. While

this was not observed at the grain boundaries, it was observed in the triple junctions as an

Al-O rich phase. EELS spectrum taken from these regions did not show the

characteristic peaks for Al or 0 in A1203 (see Figure 4-13); however, the spectrum did

show the peaks of elemental Al and 0. This would suggest that the region is amorphous.

The triple junctions may retain the second phase since the energy to create a film free

grain boundary is less than that to create a clean triple junction. In addition, since the




















Grain Boundary


Figure 4-9. The above a) STEM image and b)-e) corresponding EDS spectra are taken
from b), e) the grain boundary and c), and d) the surrounding grains.


GB


100 nm


A 10 n


0 10066


B 106n


Figure 4-10. EFTEM elemental maps of a grain boundary in the 0.5 wt. % Al sample.

triple junctions start out as pores, which are larger in volume, the liquid may be pushed

into the pores as the SiC-SiC grain boundaries form first.

In addition to the triple junction phase, transition metal inclusions and B-C rich

inclusions, which are larger than the triple junctions, were observed. These can be seen

in Figure 4-14.











S, Sedrum
C
C 0


LJ
0 1 1 5 2
W Sul, 201 cts Cur- 000HV W


SCo



05
r 1 Scale 201 ds Curo 0 000 keVkV


:(Grain
Triple

Figure 4-11. The above STEM image and corresponding EDS spectra are taken from the
triple junction and the surrounding grains in the 0.5 wt. % Al samples. The
secondary phase is confined to the triple junction and does not extend into the
grain boundaries. The scale marker is 100 nm.






















Figure 4-12. EFTEM elemental maps of an Al-O rich inclusion in the 0.5 wt. % Al
sample.

1 wt. % Al

Although the change from transgranular to intergranular fraction occurs between 1

and 1.5 wt. % Al, there is still very little variation in composition between typical grains

and grain boundaries in this sample as shown in Figures 4-15 and 4-16. However, the









O-k


0 Al k
Al-k






Elemental 0 Peak Elemental Al Peak
Figure 4-13. EELS spectra taken from a triple junction in the 0.5 wt. % Al sample shows
does not show crystalline A1203.











200nm
B-C Rich Grain SiC







Transition Metal Inclusion
Figure 4-14. A STEM image and corresponding EDS spectra showing metal and B-C
rich inclusions. The scale bar is 200 nm.

overall counts of O in the sample are greater than those of the 0.5 wt. % sample with the

majority of the increase in the triple junctions. This lends credence to the assertion that

the Al is retaining the 0 in the system.

The nature of the triple grain junction of this sample differs from that of the 0.5 wt.

% Al sample as shown in Figures 4-17 and 4-18. While the increase in volume of

secondary phase is insignificant, the change in composition is great. The secondary









phase in the 0.5 wt. % Al sample was found to be amorphous; however, the 1 wt. % Al

sample contains crystalline A1203 in the center of the triple junction and an amorphous

phase separating the triple junction phase from the grains. In addition, the concentration

of aluminum and oxygen increases with the distance into the triple junction. This is in

agreement with the Berkeley group's assertion that the crystallization process greatly

affects the composition distribution in the triple junction and that the amorphous region

between the crystalline secondary phase and SiC grains has a different composition from

the crystalline secondary region (Moberlychan and De Jonghe 1998).

In addition to the secondary phases observed at the triple junctions, grains of

transition metals were also observed. These grains, which are shown in Figure 4-19, are

the result of transition metals present in the starting powder.




S- Spe tru n1
S




ullScale 84 Cursor -0U 017 keV (623 cts) keV
Grain Grain Boundary



Figure 4-15. A STEM image of two grains and a grain boundary from the 1 wt. % Al
sample and the corresponding EDS spectra. The composition does not vary
between the grains and grain boundary.

1.5 wt. % Al

The change in fracture mode from transgranular to intergranular fracture is evident

at 1.5 wt. % Al. This sample differs from those at lower wt. % Al, which fracture

transgranularly, in that there is now an intergranular film that is rich in Al and 0 present

































Figure 4-16. EFTEM elemental maps of a grain boundary and transgranular crack in the
1 wt. % Al sample. There is no variation in composition across the grain
boundary or at the site of the crack.


C Al I-----














Q 05 1 1'5 2
Grainc







Scae 19 ur -0 006kV (746cA) keV


Spectrum1
C O




S 05 1 1 5 2
ull Scale 619 ts Cursor 0 006 keV (970 cs) keV

Triple Junction Center


Triple Junction Edges
Figure 4-17. A STEM image of a triple junction in 1 wt. % Al showing that aluminum
and oxygen segregate on the triple point and that the composition changes as a
function of depth into the triple junction. The scale marker is 100 nm.

at the grain boundaries (Figures 4-20 and 4-21), and the triple junction is filled with

A1203 (Figures 4-22 through 4-24). In addition, there is an increase in 0 at the triple


































Figure 4-18. EFTEM elemental maps of a triple junction in the 1 wt. % Al sample.
Notice that the very center of the triple junction is Si and C free.


si Spectrum 2
S ,.






0 05 1 1 5 2 1*- I ,, 3cale 72 cts Cursor -0 020 keV (255 cts) keV
ull Scale 72 ds Cursor -0 020 keV (632 cts) keV : ..


F '



VV
F2
0 2 4 6 8 10 12 14 16 18 2
ull Scale 163 cts Cursor -0 357 eV (0 cts) keV
Secondary Phase
Figure 4-19. A STEM image and corresponding EDS spectra of grains and
contaminations observed in the 1.0 wt. % Al sample.

junctions with respect to the lower wt. % Al samples. Moreover, the appearance of Al-O


rich grains is observed in this sample (Figure 4-25).




















Grain Boundary


Q Al



0 05 1 15 2
Full Scale 294 cts Cursor -0 006 keV (726 ds) keV

Grain


Figure 4-20. A STEM image of two grains and a grain boundary from 1.5 wt. % Al ABC
sample and the corresponding EDS spectra.

- C


Figure 4-21. EFTEM elemental maps of a grain boundary in the 1.5 wt. % Al sample.
Note the formation of an Al-O rich intergranular film. Scale marker is 20 nm.

4 wt. % Al

The 4 wt. % Al containing sample also exhibits high concentrations of Al and 0 at

the grain boundaries (see Figures 4-26 and 4-27). The triple junctions are high in

aluminum, and oxygen. In addition, this sample had aluminum-rich grains that may be

Al4C, carbon precipitates, and a B-C rich grains that may be B4C. In addition, the XRD











performed by Ceramatec showed that a crystalline phase of AlsB4C7 at 3 wt. % Al. This

phase was observed as an Al-B-C rich grain in the EFTEM images.

Materials Characterization Performed at Ceramatec, Inc.

This section will detail the material characterization performed at Ceramatec, Inc.

Determination of fracture mode, grain morphology, and phase assemblage will be

discussed.


A| Spectrum 1






0 05 1 15 2
Full Scale 601 cts Cursor -0 017 keV (360 cts) keV

Triple Junction


C Al i



0 05 1 15 2
ull Scale 326 cts Cursor -0 006 keV (421 cts) keV

Grain


Figure 4-22. A STEM image and EDS spectra of a triple junction in the sample
containing 1.5 wt. % Al.


Figure 4-23. EFTEM elemental maps of a triple junction in the 1.5 wt. % Al sample.
Notice that the entire triple junction is void of Si and C.


r

















Al- k


S -k

;7


Characteristic Al-k Peak in A1203 Characteristic O-k Peak in A1203
Figure 4-24. EELS spectra taken from the triple junction in the 1.5 wt. % Al sample
shows that the triple junction is filled with A1203.


ullScale152st Cursor -017keV(488a s) keV

SiC Grain






Sp 05 1 k1
ull Scale 3550cs Cursor -0 17 keV (298 cs) kceV


AI-O-rich Grain AI-O-rich Grain
Figure 4-25. A STEM image and corresponding EDS spectra of two grains and a grain
boundary in 1.5 wt. % Al sample. This site shows two different grains. The
upper grain is a SiC grain and the lower grain is that of an Al-O rich grain.
The presence of aluminum-rich grains were only observed for 1.5 wt. % Al
and 4.0 wt. % Al.


s RI















X02 4 9 2 :
__. -Grain Boundary
B-C Rich Grain n ..... Grain BoundaryE.







0 2 4 6 a
See71Bd3 ci o&t.: OjXM key
SiC Grain
Figure 4-26. The above STEM image and corresponding EDS spectra are of a SiC-SiC
grain boundary, two SiC grains, and a B-C rich inclusion in the 4 wt. % Al
sample.
Fracture Mode Determination by Ceramactec, Inc.
The fracture mode was determined by inspecting the fracture surface of samples
after SEPB testing. Figure 4-33 shows the fracture surfaces. From this data it is apparent
that there is a change in fracture mode from transgranular to intergranular between 1 and
1.5 wt. % Al. The samples develop some degree of mixed mode fracture above 3 wt. %
Al. While SEPB measurements give more accurate values of toughness, often indent
surfaces are more instructive as to the fracture mode. Thus, the change to mixed mode
can be seen more clearly in the Vickers hardness indents in Figure 4-34.



















I, ,, ,

il SOce1 DOacl CuLsor DODV eY
B-C Rich Grain


S3


S4


'
r


SiC Grain Boundary


111 Dctl e Ciss .[ODeV


Al4C Grain
Figure 4-27. The above STEM image and corresponding EDS spectra are of a SiC-SiC
grain boundary, a B-C rich inclusion, and an A14C inclusion in the 4 wt. % Al
sample.












G a- '(n, -rain Eondar.,











u, ,01D3,i CW Or D DOeMV tl ScMt 1 DO da Cuwo DO.DiOeV
AI-O-C Rich Grain Grain Boundary

Figure 4-28. The above STEM image and corresponding EDS spectra are of a SiC-SiC
grain boundary, two SiC grains, and an Al-O-C rich inclusion in the 4 wt. %
Al sample. Note that the area of the grain boundary that is farthest from the
inclusion contains Al, while the area closest to the inclusion is depleted in Al.


N


1Electron Imeae 1

















Carbon Inclusion
Carbon Inclusion


AIS-O-C Rich Grain
AI-O-C Rich Grain


/


s.coeincms cuws% Donev
SiC-SiC Grain
Boundary


hr *


hu scale im CuioA DlmnteY


ua Isc U cis C.n 0 DOBe, V
SiC Grain


AI-O-C Rich Grain
Figure 4-29. The above STEM image and corresponding EDS spectra are of a SiC-SiC
grain boundary, a SiC grain, and a carbon inclusion imbedded in an Al-O-C
rich inclusion in the 4 wt. % Al sample.


0 1 2 3 4 5
ull Scale 100 cts Cursor: 0,000 keV keVI


Grain Boundary


n|l 'ii 1111.1 i:ls .u r i r ir 11l.,' I l 'i i.' h.

Grain


Figure 4-30. The above STEM image and corresponding EDS spectra are of a SiC-SiC
grain boundary, and two SiC grains in the 4 wt. % Al sample. Note the
presence of Al at the grain boundary.


.'s
s'... .B I.
.... i ] ... .. ..





























Figure 4-31. EFTEM elemental maps of a grain boundary between two SiC grains in the
4 wt. % Al sample.


Figure 4-32. EFTEM elemental maps of a multiple grain junction in the 4 wt. % Al
sample. The circled grain is Al-B-C rich and may be A18B4C7. The area
between the grains is Al-O rich.









Relationship Between Hardness and Toughness by Ceramactec, Inc.

The hardness of the SiC samples initially diminishes as Al is added; however, once

the fracture mode changes to intergranular the hardness increases again and reaches a

maximum at 4 wt. % Al. The toughness remains constant until the fracture mode changes

to intergranular. From this point, the toughness increases dramatically and reaches a

maximum at 2 wt. % Al. This relationship can be seen in Table 4-2 and Figure 4-35.

Grain Morphology Characterization by Ceramatec, Inc.

Several possible toughening mechanisms in SiC are dependant on the size and

aspect ratio of the grains. The data on the size and aspect ratio of the grains is presented

in Table 4-3. Figure 4-36 illustrates how aspect ratio and grain size change as a function

of Al content. In general, the grain size increases as the aspect ratio increases. Figure 4-

37 presents the SEM micrographs upon which the grain morphology calculations were

performed.

XRD and Reitveld Analysis by Ceramatec, Inc.

Per the discussion in chapter 2, the transformation from P-SiC to a-SiC and the

formation of secondary phases can affect the toughness. When SiC transforms from a

cubic to hexagonal structure, the grains grow anisotropically into a three dimensional

structure of interlocking elongated grains. The formation of this network of interlocking

grains has been shown to offer some degree of hardness in other studies. In addition, the

formation of secondary phases has been known to produce residual stresses, which are

needed to initiate several toughening mechanisms in ceramics. It is therefore important

to understand the phase assemblage in the SiC samples under study.

Figure 4-38 is the XRD data collected from the samples containing 0-6wt. % Al.

The formation of the secondary phase A18B4C7 begins at 3 wt. % Al. It is important to
















































Figure 4-33. SEPB fracture surfaces for samples with 0-6 wt. % Al. There is a clear
change in fracture mode between transgranular and intergranular between 1
and 15 wt. % Al. The scale marker is 10 [tm (Flinders et al. submitted).

note that A1203, Al4C, and B4C do not appear in the XRD spectra. This would suggest

that the volume fraction of these species is much less than that of A18B4C7.

Figure 4-39 and Table 4-4 show the results of the Reitveld analysis. Unlike the

Berkeley group who processed at 19000C, all of the 3-SiC has transformed to ca-SiC.







56




-*















MI ..
I ... r i i t. ... .. T V .
.......... .4 ,

















in '
!: ''1,4 'g,: t !:ki :




q -A de:; ,,
































Figure 4-34. Comparison of HVi (left) and HTKi (right) for various SiC materials. A
change in fracture mode is evident between 1 and 1.5 wt. Oo Al (Flinders et al.
submitted).
W6 .e,. ......
--: :-,.!:S


VE..l A., -5 U ."
U ., I. .... O'
.%:,-- ...... ... ....2 .. .



.= ....... ." ... ... ........ .
N +g Aj>i+ +..
'V +4.4







Figure~ ~ ~ ~ ~ ~~~~~~~~~E: 4-34. Coprio ofHI(et adH.(ih),o aiu iCmtras
chageinfrctremoe s eidntbewen ad .5 wt. % Al(lidr e l
submitted).:"







57


























































materials. The onset of the transgranular fracture mode is accompanied by a
high degree of crack branching and crushing. A change toward mixed mode
'- S








































fracture occurs at 4 wt. % Al at which point the cracks are straight and there is
no crushing (Flinders et al. submitted).
74 v

no.crun ( ea... s..s









However, this data does agree with the Berkeley group assertion that the 4H content

increases up to 3 wt. % Al and then decreases thereafter.

Correlation Between Microstructure and Mechanical Properties

Toughness

Per the discussion in chapter 2, the viable toughening mechanisms in SiC crack

deflection, crack bridging, and microcracking depend on at least one of the following:

weak interfaces, residual stresses, high aspect ratio grains, and large grains. This section

will analyze the observed effect that these properties have on SiC sintered with Al-B-C

additives and suggest a possible toughening mechanism.

Weak interfaces

The high resolution lattice imaging and compositional analysis presented earlier in

this chapter support the formation of a weak intergranular film between 1 and 1.5 wt. %

Al. This film was primarily composed of an amorphous film that contained Al and 0.

Konoshita et al. concluded that the segregation of Al and 0 atoms to grain boundaries

will weaken the grain boundary strength and therefore provide an energetically favorable

crack path (Konoshita et al. 1997).

Table 4-2. Toughness and Hardness Measurements of SiC with 0-6 wt. % Al.
Al Density Fracture Toughness HK1 Hardness
(g/cc) Mode (MPa-m1/2) (GPa)
0.0 3.170.01 T 2.60.2 20.40.4
0.5 3.170.01 T 2.60.1 19.40.2
1.0 3.170.01 T 2.70.1 19.20.3
1.5 3.130.0 1 I 6.10.3 13.70.2
2.0 3.130.01 I 6.70.4 14.30.3
3.0 3.140.01 I 6.20.4 16.60.7
4.0 3.120.01 I 6.10.2 16.90.8
6.0 3.070.01 I 5.60.3 15.00.6
T = Transgranular Fracture
I = Intergranular Fracture

































1 2 3 4 5 6


Al Content (wt. %)
Figure 4-35. The above graphs plot Knoop hardness and SEPB toughness as a function
of Al content. The change in hardness and toughness with changing Al
content is inversely related (Flinders et al. submitted).

Table 4-3. Characterization of SiC with 0-6 wt. % Al (Flinders et al. submitted)


x Density
(g/cc)
0.0 3.17+0.01


Grain Size

4.70.4


Aspect
Ratio
5.41.1


Fracture
Mode
T


Toughness
(MPa-m1/2)
2.60.2


HK1 Hardness
(GPa)
20.40.4


0.5 3.170.01 4.60.2 5.91.3

1.0 3.170.01 4.10.5 5.40.9

1.5 3.130.01 5.20.4 3.90.6

2.0 3.130.01 5.80.5 5.01.3

3.0 3.140.01 11.91.2 4.51.0

4.0 3.120.01 5.30.5 5.11.0

6.0 3.07+0.01 7.1+0.6 8.7+2.7


T 2.6+0.1

T 2.7+0.1

I 6.1+0.3

I 6.7+0.4

I 6.2+0.4

I 6.1+0.2

I 5.6+0.3


19.4+0.2

19.2+0.3

13.7+0.2

14.3+0.3

16.6+0.7

16.9+0.8

15.0+0.6







60




Grain Size and Aspect Ratio vs. Aluminum Content

14


12 -


10 )
o--




2--
0










2 Grain Size
0 Aspect Ratio
0






0 1 2 3 4 5 6 7
Aluminum Content (wt.%)


Figure 4-36. Change in aspect ratio and grain size with Al content.

The effect of the weak intergranular film on toughness is apparent. Upon formation

of the intergranular film at 1.5 wt. % Al, the toughness more than doubled. In addition,

the formation of the intergranular film was accompanied by a change in fracture mode

from transgranular to intergranular. It is therefore apparent that the intergranular film

plays a large role in increasing the toughness due to a change in fracture mode. However,

since all three of the previously mentioned toughening mechanisms could be affected by

this intergranular film, or at least weak interfaces, no conclusion can be drawn with

respect to the dominant toughening mechanism.

Residual stresses

The analysis of residual stresses due to thermal expansion anisotropy in single

phase systems and thermo-elastic property mismatch in multiphase systems is important


























































Figure 4-37. Polished and chemically or plasma-etched surfaces of SiC samples with 0-
wt. % Al (Flinders et al. submitted).











2000

1800

1600

1400

1200

1000


30 32 34 36 38 40 42 44 46


2-Theta (degrees)


Figure 4-38. X-ray diffraction patterns SiC with 0-6 wt. % Al. Note that starting powder
was beta-3C and that in addition to SiC polytypes, A18B4C7 phase is noted at
high Al contents (Flinders et al. submitted).


SiC Al 0.6 wt% B 2 wt% C, 21000C / lhr


0 1 2 3 4 5 6
Al Content (wt. %)


Figure 4-39. Polytypes from Rietveld analysis for SiC-0.6 wt. % B-0.2 wt.
with Al contents ranging from 0 to 6 wt. %.


% C samples









Table 4-4. Reitveld Analysis for SiC Samples with 0-6 wt. % Al (Flinders et al.
submitted).
Al Content (wt. %) Phase Assemblage (wt. %)
3C 4H 6H 15R
0 0 0 93.1 6.9
0.5 0 0 87.5 12.5
1 0 22.4 68.2 9.4
1.5 0 34 58.9 7.1
2 0 40.7 51.2 8.1
3 0 89 6.4 4.6
4 0 81.4 14.5 4.1
6 0 52.3 37.6 10


in determining crack behavior. As mentioned in chapter 2, in multiple phase systems

such as one containing grains and intergranular films or triple junction phases, the

difference in thermal expansion mismatch between the grains and secondary phases can

give rise to residual stresses. If the difference in thermal expansion between the grains

and secondary phases is negative (am < ap), a compressive radial stress is developed at

the matrix-particle boundary and a tensile circumferential stress is developed in the

matrix. The particle will be in tension and the crack will be attracted to the particle.

When the difference in thermal expansion is positive (am > ap), the crack will be

deflected between particles and a more tortuous crack path will be created, which will

increase toughness (Hertzberg 1996). If one considers a material with an intergranular

film in which the intergranular film is the matrix and the grains are the particles, this

second case can be used to explain the origin of intergranular fracture.

With respect to intergranular films a local compressive tangential residual stress

around the grains can cause diminishing stress intensity directly at the tip of the deflected









crack and are a result of a difference in thermo-elastic properties between the

intergranular film and grains (Sternitzke 1997).

Crystallizing triple junctions can also cause residual stresses, the magnitude of

which is dependent on the volume change upon crystallization (Pezzotti and Kleebe

1999). The extent of triple junction crystallization can affect the mechanical properties of

the final part. Kessler et al. found that the volume change due to crystallization is

strongly dependant on the degree of crystallization with increased crystallization resulting

in increased volume change. This volume change results in residual stresses whose

magnitude is proportional to the volume change of the constrained phase, or triple

junction (Kessler et al. 1992). Pezzotti and Kleebe found that a nearly 100% fraction of

crystallized secondary phase at the triple junction will create substantial tangential tensile

stresses at the interface between the grains and triple junction due to volume contraction,

which drives cracks toward the triple junction and causes subsequent interface

delamination. This phenomenon can lead to crack splitting and intergranular fracture

(Pezzotti and Kleebe 1999).

It is therefore necessary to determine the local residual stresses between the grains

and grain boundaries, and between the grain and triple junctions. For the sake of

simplicity, the residual stresses with respect to the grain boundaries will be calculated

with the assumption that the grains are fibers in an infinite matrix and are bonded to the

grain boundaries. This is obviously a simplification considering the fact that the matrix,

or intergranular film, is narrower than the particle, but serves as an approximation and

means of relative comparison. Superposition of stress fields would be expected,

however, they are ignored in this calculation. The grain boundary film is an amorphous

glass containing Al and 0, and is most likely an aluminosilicate glass with a low silica









volume fraction. The properties of an aluminosilicate with 64.4 % A1203 and 36.6 %

SiO2 were used for the amorphous intergranular film. Thus, the residual stress due to a

fiber in an infinite matrix can be calculated as follows (Green 1998):

Fiber:

7rr = 000 =-P
Matrix:

_Pa2 (r -b 2)
arr 2(b2 2)

-Pa (r + b2)
00 r2(b2 -a2)
Where a =fiber radius, b= the radius of the fiber plus the radius of the matrix

P E (a,, a )(To )(1 + )
(1 + Vf )E (1 2o) + Ef (1 + V,)
430GPa70GPa(9.7 pm/mC-4.5 [m/mC)(1373K-298K)(l+0.17)
(1+ 0.17)(70GPa)(1 -.34) + 430GPa(1 + 0.22)
= 340MPa

Thus, the stresses in the fiber are compressive and the stresses in the matrix are

tensile in the radial direction. This stress state will cause matrix cracking. This residual

stress state can therefore explain the transition to intergranular fracture when an

intergranular film is present.

It is also important to understand the effect of the triple junction phases on residual

stress. For the sake of simplicity, the effect of superposition of stresses will be ignored,

and the triple junction phase is assumed to be a spherical particle in an infinite matrix.

The properties of alumina will be used for the triple junction and those of SiC will be

used for the matrix. The stress state will be calculated as a function of temperature. The

residual stress due to a particle in an infinite matrix can be calculated as follows (Green

1998):











Particle:


'rr = O 00 -P


Matrix:


-r3P
,rr = -2oo00 = 3
r


P= 2EmEp(am -ap)(To -TA)
2Em(1 -2p)+ E(l+ 0,)

2(430GPa)(370GPa)(4.5 |tm/mC-8.2 [tm/mC)(T7-298K)

2(430GPa)(1-2*0.22)+370GPa(1+. 17)

= -2.72x 06Pa(To-298K)


To may be one of several temperatures: the processing temperature, the melting

point of alumina, the meting point of mulite, or 2/3 of the homologous temperature.

Therefore, the stress, P, is plotted vs. temperature to determine the sign of P, which allow

for the determination of the residual stress fields in the secondary phases at the triple

junctions and in the surrounding SiC grains.


Stress vs. Temperature
0.








-4-4

-5 -


-6

-7
0 500 1000 1500 2000 2500 3000
Temperature

Figure 4-40. Plot of Stress vs. Temperature for a particle in an infinite matrix.









From the graph, the calculated load can be between -3 GPa using 2/3 of the

homologous temperature and -6 GPa using the processing temperature. The entirety of

this range is negative, thus the stresses in the particle are tensile and the stresses in the

matrix are tensile in the radial direction and compressive in the circumferential direction.

As the crack is attracted to regions of tension, this stress state will cause the crack to be

driven toward the triple junction and cause delamination.

The combined effect of tensile stresses in the grain boundary phase and the triple

junction phase will cause the crack to propagate within the intergranular film and around

the triple junctions. This behavior was observed in the 1.5 wt. % Al sample amd is

shown in Figure 4-42.

Triple Junction in Tension Crack Attracted to
ST Triple Junction
Crack Tip '- /








Triple Junction in Compression Crack Deflected from
Triple Junction

Crack Tip









Figure 4-41. Effect of residual stress fields around triple junctions on crack propagation
(Pezzotti and Kleebe 1999).

























Figure 4-42. Intergranular cracking in the 1.5 wt. % sample. The figure on the left
shows cracking through the intergranular film at the grain boundaries. The
figure on the right shows a missing triple junction due to the crack deflecting
around the triple junction, but within the amorphous secondary phase. Note
that the image on the right is slightly out of focus, which makes the crack
appear to be a broad region that is lighter in color.

Residual stresses contribute to all three toughening mechanisms discussed earlier in

the section. Therefore, it would be difficult to conclude a specific toughening mechanism

at this point. However, a probable origin of the intergranular cracking has now been

proposed.

High aspect ratio grains

Both crack deflection and crack bridging depend on the aspect ratio of the grains

(Green 1998). If either of these mechanisms were active, an increase in toughness with

aspect ratio would be expected. In the case of crack deflection, the SiC grains are taken

to be the particles and thus the volume fraction of high aspect ratio grains is close to

unity. Faber and Evans found that after the volume fraction reaches 0.2 no additional

increase in toughening will be observed with increased volume fraction of elongated

grains (Green 1998). In addition, the crack wake on the indented samples does not show









the deflection of the crack and therefore, it is unlikely that crack deflection is the main

toughening mechanism in the samples.

If crack bridging were to occur, the toughness would increase as a function of

aspect ratio according to the following equation (Flinders et al. submitted):

KIc2 oK 2 2 1.67
K = Kic + 0.81Er 'L aL d,

However, when the square route of toughness is plotted against aspect ratio raised

to the 1.67, Figure 4-43, the slope is negative. In addition, no crack ligaments are

observed in the crack wake in the hardness indents. Thus, the evidence points away from

crack bridging.

Large grains

Microcracking occurs when residual stresses build-up due to the following: phase

transformations, thermal expansion anisotropy in one phase materials, or thermo-elastic

property mismatch in multiphase materials (Green 1998). The previous section on

residual stresses showed that residual stresses existed at the interfaces between the

intergranular film and SiC grains, and the triple junctions and SiC grains. These stresses

were also shown to cause cracking at the interface and debonding of the triple junctions,

thus satisfying one of the conditions necessary for microcracking.

The degree of toughness increase due to microcracking is dependent on the grain

size. As the particle size increases, the toughening increment should increase due to the

creation of a tortuous crack path until the critical grain size is reached. Baud and

Theveno found that the assertion that the toughness increases with grain size, or more

precisely the square route of grain size, holds true for liquid-phase sintered SiC (Baud

and Theveno 2001). However, the composition of the starting powers was all the same

and only grain size varied in the materials that Baud and Theveno. investigated. For the







70


case of the SiC sintered with Al-B-C additives under investigation, not only does the

grain size change, but the types and volume fraction of secondary phases change. This

would in turn affect the magnitude of the residual stresses developed and therefore the

degree of toughening gained by microcracking. Therefore, there may be a synergistic

effect between the grain size and the volume fraction of secondary phases, which would

result in the plot of toughness vs. grain size showing no correlation between the two.

Therefore, the possibility that microcracking is the dominant toughening mechanism

should not be dismissed.

Indeed, an inspection of the Vickers indents reveals a high degree of crushing and

crack branching in the 1.5 wt. % sample. The 4 wt. % Al sample experiences a much

lower degree of crack branching and crushing than the 1.5 wt. % sample. This may be

due to the formation of additional secondary phases in this sample such as A18B4C7, Al4C,

Toughness2 vs. Aspect Ratio1 67

50 -

45 -

40 -

35 -

30 -



20

1--
10 -
1-

5 -
0 -----------------
0
0 5 10 15 20 25 30 35 40 45

aL1 67
Figure 4-43. Plot of toughness vs. aspect ratio shows a negative slope and argues against
crack bridging. The three data points at the bottom of the graph correspond to
the samples that fail transgranularly.







71


Toughness vs. Square Route of Grain Size



7-


0A
7 S
4'4

& 4-
3-


2

1


0 1 2 3 4 5

Figure 4-44. Plot of toughness vs. grain size. Toughness generally increases with
increased grain size.

B4C, and an Al-O-C rich phase. Not only are more species present, but they are present

in much greater volume fraction than the secondary phases in the 1.5 wt. % Al sample.

As shown in the calculation performed in the residual stresses section, the secondary

phases surrounded by the intergranular film will create compressive stresses in the

secondary phase, which will deflect the crack. If these phases are present in high enough

concentration, it could explain the change in fracture mode from intergranular to mixed

mode at this composition, and the drop in toughness.

Hardness

The hardness initially drops when small amounts of aluminum are added with the

largest drop in toughness occurring when the fracture mode changes to intergranular

fracture. From this point, the hardness begins to increase again, reaching a maximum at 4

wt. % Al, and then decreases again. The large drop in hardness upon the formation of an

intergranular film is due to the increased ability of the SiC grains to move under the











indentation load, which results in lower hardness (Flinders et al. 2003). The increase in

hardness up to 4 wt. % may be explained in a similar fashion to the drop in hardness.

The increased number of secondary phases that exist in a state of compression would

resist grain movement under the indentation load. By building-up compressive stresses

in the material, a greater applied load is needed to deform the material.

Effect of grain size on hardness

While the grain size had little effect on toughness, it does affect the hardness.

Often a Hall-Petch type relationship is applied to correlate the hardness with the grain

size; however, this model is not as useful for ceramic as it is for metals. The relationship

was originally designed to explain an increase in strength due to dislocation pile-up at


Hardness vs Grain Size

21 -


20 -


19 -


18 -
a.
16 -



15 -
16 -








13
3 5 7 9 11 13
Grain Size (microns)

Figure 4-45. Plot of the effect of grain size on hardness. The plot shows that hardness
increases with grain size for the samples that fail via intergranular fracture.







73


grain boundaries, since the number of grain boundaries increases with decreasing grain


size. Yet, the hardness does scale inversely with grain size. This effect is more likely due


to the reduction of the inherent size of the flaws when the grain size is reduced than to an


increase in dislocation piles.


Effect of aspect ratio on hardness

By plotting hardness vs. the aspect ratio, it observed that the aspect ratio has no


effect on the hardness. This behavior can be seen in Figure 4-46.





Hardness vs. Aspect Ratio

21 -


20 -


19 -


18 -

S17 -O Intergranular Fracture
17 I E -OI -Transgranular Fracture


16 -


15 --_ _-_ |-


14 -


13 -
3 4 5 6 7 8 9 10 11 12
Aspect Ratio

Figure 4-46. The plot of hardness vs. aspect ration shows that the hardness is not
dependant on aspect ratio.















CHAPTER 5
SUMMARY AND CONCLUSIONS

Conclusions

With the use of high resolution lattice imaging, energy dispersive spectroscopy, and

energy filtered transmission electron microscopy, it was determined that an amorphous

intergranular film forms between 1 and 1.5 wt. % Al. The film is composed primarily of

an Al-O rich phase and contains an equilibrium film thickness of approximately 1 nm.

The residual tensile stresses in this film allow for cracking through the film, promote the

change in fracture mode from transgranular to intergranular fracture, and increase the

toughness values by a factor of two.

The addition of aluminum to the samples promoted the formation of secondary

phases at the triple junctions. These secondary phases were Al-O rich and the oxygen

content and crystallinity increased as the Al content was increased, with the formation of

alumina occurring at 1.5 wt% Al. Additional secondary phases were formed as discrete

grains and inclusions at Al contents greater than 1.5 wt. % Al. These phases may include:

Al4C, an Al-O-C phase, A18B4C7, and B4C. The crystallization of the secondary phases

generated a compressive stress field in the particles that caused the crack to deflect

through the amorphous phase around the crystalline triple junctions in the 1.5 wt. % Al

sample, which has a lower volume fraction of secondary phase. However, as the volume

fraction of secondary phases increased, as in the 4 wt. % Al sample, the amount of

residual compressive stress in the samples increased and caused a simultaneous increase

in hardness and decrease in toughness.









Suggested Future Work

While some conclusions were made to possible toughening mechanisms in SiC

sintered with Al-B-C additives, additional investigation into the toughening mechanisms

is needed. In-situ cracking studies performed in a scanning electron microscope in

secondary electron mode could be used to monitor the advance of the crack and image

the crack tip and crack wake. This study would produce potential identification of the

toughening mechanism.

In addition, a comparison of the elastic compliance of the cracked specimen can be

compared to the theoretical compliance of an ideal bridge- and microcrack-free crack of

the same length. Since crack-bridging increases the modulus, a reduction in compliance

would suggest the presence of crack-bridging. Conversely, a decrease in the compliance

would suggest microcracking (Nalla et al. 2004).















LIST OF REFERENCES


Baud, S., and Theveno, F., "Microstructural and Mechanical Properties of Liquid-phase
Sintered Seeded Silicon Carbide," Mater. Sci. Phys., 67, 165-174, 2001.

Birnie, D. P., "A Model for Silicon Self-Diffusion in Silicon Carbide: Anti-Site Defect
Motion," J. Am. Ceram. Soc., 69 [1], C33-C35, 1986.

Biswas, K. Rixecker, G., Wiedmann, I., Schweizer, M., Upadhyaya, G. S., and
Aldinger, F., "Liquid Phase Sintering and Microstructure-Property Relationships of
Silicon Carbide Ceramics with Oxynitride Additives," Mater. Chem. Phys., 67,
180-191, 2001.

Bonnell, D. A., Tien, T. Y., and Ruhle, M., "Controlled Crystallization of the Amorphous
Phase in Silicon Nitride Ceramics," J. Am. Ceram. Soc., 70, 460-465, 1987.

Cao, J. J., MoberlyChan, W. J., DeJonghe, L. C., Gilbert, C. J., and Ritchie, R. 0., "In
Situ Toughened Silicon Carbide with Al-B-C Additions," J. Am. Ceram. Soc., 79
[2], 461-69, 1996.

Case, E. D., Smith, I. 0., and Baumann, M. J., "Microcracking and Porosity in Calcium
Phosphates and Implications for Bone Tissue Engineering," Mater. Sci. Eng. A,
390, 246-254, 2005.

Chen, D., Sixta, M. E., Zhang, X. F., De Jonghe, L. C., and Ritchie, R. 0., "Role of the
Grain Boundary Phase on the Elevated-Temperature Strength, Toughness, Fatigue
an Creep Resistance of Silicon Carbide Sintered with Al, B, and C," Acta Mater.,
48, 4599-4608, 2000.

Chia, K. Y., and Lau, S. K., "High Toughness Silicon Carbide," Ceram. Eng. Sci.Proc.,
12[9-10], 1845-1861, 1991Clarke, D. R., "On the Equilibrium Thickness of
Intergranular Glass Phases in Ceramic Materials," J. Am. Ceram. Soc., 70 [1], 15-
22, 1987.

Clarke, D. R., Shaw, T. M., Philipse, A. P., and Horn, R. G., "Possible Electrical
Double-Layer Contribution to the Equilibrium Thickness of Intergranular Glass
Films in Polycrystalline Ceramics," J. Am. Ceram. Soc., 76 [5], 1201-1204, 1993.

Falk, L. K. L., "Microstructral Development During Liquid Phase Sintering of Silicon
Carbide Ceramics," J. Eur. Ceram. Soc., 17, 983-994, 1997.

Fischer, G. R., and Barnes, P., "Toward a Unified View of Polytypism in Silicon
Carbide," Phil. Mag. B., 61, 217-236, 1990.









Flinders, M., Ray, D., and Cutler, R. A., "Toughness-Hardness Trade-Off in Advanced
SiC Armor," Ceram. Trans., 151, 37-48, 2003.

Flinders, M., Ray, D., Anderson, A., and Cutler, R. A., "Liquid Phase Sintered Silicon
Carbide as Protective Armor," J. Am. Ceram. Soc., submitted.

Gilbert, C. J., Cao, J. J., Moberlychan, W. J., De Jonghe, L. C., and Ritchie, R. 0.,
"Cyclic Fatigue and Resistance-curve Behavior of an In-situ Toughened Silicon
Carbide with Al-B-C Additions," ActaMetal. Mater., 44 [8], 3199-3214, 1996.

Gooch, W. A. Jr., "An Overview of Ceramic Armor Applications," Ceram. Trans., 134,
3-21, 2002.

Green, D. J., An Introduction to the Mechanical Properties of Ceramics, Cambridge
University Press, Cambridge, UK, 1998.

Greskovich, C., and J. H. Rosolowski, "Sintering of Colavent Solids," J. Am.Ceram. Soc.,
59, 336-343, 1976.

Gu, H., Pan, X., Tanaka, I., Cannon, R. M., Hoffmann, M. J., Mullejans, H., and Ruhle,
R., "Structure and Chemistry of Intergranular Films in Ca-Doped Si3N4," Materials
Science Forum, 207-209, 729-732, 1996.

Hertzberg, R. W., Deformation and Fracture Mechanics of Engineering Materials. John
Wiley & Sons, Inc., New York, NY, 1996.

Hofer, F., Warbichler, P., and Grogger, W., "Imaging of Nanometer-sized Precipitates in
Solids by Electron Spectroscopic Imaging," Ultramicroscopy, 59, 15-31, 1995.

Hofer, F., Grogger, W., Kothleitner, G., and Warbichler, P., "Quantitative Analysis of
EFTEM Elemental Distribution Images," Ultramicroscopy, 67, 83-103, 1997.

Hong, J. D., "Self-Diffusion in Alpha and Beta Silicon Carbide", Proc. Of the 4h Int. Met.
Mod. Ceram., Mat Sci. Monographs, 6, Energy and Ceramics, P. Vincenzini (Ed.)
Elsevier, 1979.

Huang, J. L., Hurford, A.C., Bruner, S. L., Cutler, R. A., and Virkar, A. V., "Sintering
Behavior and Properties of SiCAlON Ceramics," J. Mater. Sci., 21, 1448-1456,
1986.

Inomata, Y., Tanaka, H., Inoue, Z., and Kawabata, H., "Phase Relation in SiC-Al4C3-B4C
System at 18000C," J. Ceram. Soc. Jpn. (Yogyo Kyokaishi), 88 [6], 353-355, 1980.

Jun, H-W., Lee, H-W., Kim, G-H., and Song, H. S., "Effect of Sintering Atmosphere on
the Microstructure Evolution and Mechanical Properties of Silicon Carbide
Ceramics," Ceram. Sci. Eng. Proc. 18 [4], 487-504, 1997.









Kaneko, K., Yoshiya, M., Tanaka, I., and Tsurekawa, S., "Chemical Bonding of Oxygen
in Intergranular Amorphous Layers on High-purity P-SiC Ceramics," Acta Mater.,
47, 1281-1287, 1999.

Kaneko, K., Honda, S., Nagano, T., and Saitoh, T., "Analytical Investigation of Grain
Boundaries of Compressive Deformed Al-doped Sintered P-SiC" Mater. Sci. Eng.
A, 285, 136-143, 2000.

Kessler, H.,, Kleebe, H-J., Cannon, R. W., and Pompe, W., "Influence of Internal
Stresses on Crystallization of Intergranular Phases in Ceramics," Acta Metall.
Mater., 40 [9], 2233-2245, 1992.

Kim, Y. W., Mitomo, M., and Hirotsuru, H., "Grain Growth and Fracture Toughness of
Fine-Grained Silicon Carbide Ceramics," J. Am. Ceram. Soc., 78 [11], 3145-3148,
1995.

Kim, Y-W., and Mitomo, M., "Fine-Grained Silicon Carbide Ceramics with Oxynitride
Glass," J. Am. Ceram. Soc., 82 [10], 2731-2736, 1999.

Konoshita, T., Munekawa, S., and Tanaka, S-I., "Effect of Grain Boundary Segregation
on High-temperature Strength of Hot-pressed Silicon Carbide," Acta Mater., 45 [2],
801-809, 1997.

Meyers, M. A., and Kumarchawla, K., Mechanical Behavior of Materials, Prentice-Hall,
Inc., Upper Saddle River, NJ, 1999.

MoberlyChan, W. J., Cannon, R. M., Chan, L. H., Cao, J. J., Gilbert, C. J., Ritchie, R. 0.,
and De Jonghe, L. C., "Microstructural Development to Toughen SiC," Mat. Res.
Soc. Symp. Proc., 410, 257-262, 1996.

Moberlychan, W. J., Cao, J. J., and De Jonghe, L. C., "The Roles of Amorphous Grain
Boundaries and the P-a transformation in Toughening SiC," Acta Mater., 45 [5],
1625-1635, 1998.

Moberlychan, W. J., and De Jonghe, L. C., "Controlling Interface Chemistry and
Structure to Process and Toughen Silicon Carbide," Acta Mater. 45 [7], 2471-2477,
1998.

Nalla, R. K., Krunzic, J. J., and Ritchie, R. 0., "On the Origin of the Toughness of
Mineralized Tissue: Microcracking or Crack Bridging?" Bone, 34, 790-798, 2004

Pezzotti, G., and Kleebe, H. J., "Effect of Residual Microstresses at Crystalline
Multigrain Junctions on the Toughness if Silicon Nitride," J. Eur. Ceram. Soc., 19,
451-455, 1999.

Prochazka, S., "The Role of Boron and Carbon in the Sintering of Silicon Carbide," pp.
171-181 in Special Ceramics 6. Edited by P. Popper. British Ceramic Research
Association, Stoke-on-Trent, U. K., 1975.









Prochazka, S., "Why is It Difficult to Sinter Covalent Substances?" General Electric
R&D Center, Schenectady, USA, Technical Information Series, Class 1 GECI
Program 89CRD025, 1987.

Raj, R." Fundamental Research in Structural Ceramics for Service Near 20000C," J. Am.
Ceram. Soc., 76 [9], 2147-2174, 1993.

Ramsdell, R. I., "Studies on Silicon Carbide," Amer. Mineral., 32, 64, 1947.

Schaffer, B., Grogger, W., and Hofer, F., "Width Determination of Si02-films in Si-based
Devices Using Low-loss EFTE: Image Contrast as a Function of Sample
Thickness," Micron, 34, 1-7, 2003.

Sigl, L. S., and Kleebe, H. J., "Core/Rim Structure of Liquid-Phase-Sintered Silicon
Carbide," J. Am. Ceram. Soc.,76, 773-776, 1993.

Sternitzke, M., "Review: Structural Ceramic Nanocomposites," J. Eur. Ceram. Soc., 7,
1061-1082, 1997.

Stobierski, L., and Guberat, A., "Sintering of Silicon Carbide I. Effect of Carbon,"
Ceramics International, 29, 287-292, 2003.

Tanaka, H., "Sintering of Silicon Carbide," in Silicon Carbide Ceramics-], Ceramics
Research and Development in Japan Series, Eds. S. Somiya und Y. Inomata,
London, Elsevier Applied Science, 213-236, 1991.

Williams, D. B., and Carter, C. B., Tansmission Electron Microscopy, Plenum Press,
New York, NY, 1996.

Ye, Haihui, "Microstructure and Chemistry of Grain-Boundary Films and Triple Junction
Phases in Liquid-Phase Sintered SiC Ceramics," Ph.D. Dissertation, Max-Planck-
Institut fur Metallforschung, Stuttgart, Germany, 2002.

Yuan, R., Kruzic,. J. J., Zhang, X. F., De Jonghe, L. C., and Ritchie, R. 0., "Ambient to
High-temperature Fracture Toughness and Cyclic Fatigue Behavior in Al-
containing Silicon Carbide Ceramics" Acta Mater., 51, 6477-49, 2003.

Zhang, Y. H., Edwards, L., and Plumbridge, W. J., "Effect of Crystallization on Fatigue
Crack Growth in a Sic-Reinforced Silicon Nitride Composite at 12000C," Acta
Mater., 46 [4],1327-132, 1998.

Zhang, X. F., Yang, Q., and De Jonghe, L. C., "Microstructure Development in Hot-
pressed Silicon Carbide: Effects of Aluminum, Born, and Carbon Additives," Acta
Mater., 51, 3849-3860, 2003.

Zhou, Y., Tanaka, H., Otani, S., and Bando, Y., "Low-Temperature Pressureless
Sintering of a-SiC with A14C3-B4C-C Additions," J. Am. Ceram. Soc., 82 [8], 1959-
1964, 1999.






80


Zhou, Y., Hiroa, K., Yamauchi, Y., and Kanzaki, S., "Tailoring the Mechanical
Properties of Silicon Carbide Ceramics by Modification of the Intergranular Phase
Chemistry and Microstructure," J. Eur. Ceram. Soc., 22, 2689-2696, 2002.















BIOGRAPHICAL SKETCH

Samantha Crane moved from her childhood home in Roslyn Heights, NY, to

Weston, FL, at the age of 15. She graduated high school from The University School of

Nova Southeastern University in 1999. In the summer of 1999 she enrolled in the Honors

College at the University of Florida. She graduated cum laude from the University of

Florida with her Bachelor of Science in materials science and engineering in May 2003

and will receive her Master of Science in materials science and engineering in August

2005. Samantha has a particular interest in nuclear materials, which began with her

senior research project on burnable poison rod assemblies. She has pursued this interest

by participating in two summer internships with the United States Nuclear Regulatory

Commission (NRC) in Rockville, MD. Samantha will be returning to the NRC for full-

time employment in the fall of 2005.