<%BANNER%>

Deformation Mechanisms in Advanced Structural Ceramics due to Indentation and Scratch Processes

Permanent Link: http://ufdc.ufl.edu/UFE0041123/00001

Material Information

Title: Deformation Mechanisms in Advanced Structural Ceramics due to Indentation and Scratch Processes
Physical Description: 1 online resource (188 p.)
Language: english
Creator: Ghosh, Dipankar
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: amorphization, b4c, ceramics, composite, deformation, dislocation, fib, ftir, grain, hardness, indentation, microcracking, microstructure, modeling, plasticity, processing, scratch, sem, sintering, sliplines, spectroscopy, tem, uhtcs, zrb2
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Mechanical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In this dissertation, mechanical properties of boron carbide (B4C) and zirconium diboride-silicon carbide (ZrB2-SiC) composite have been investigated. To capture the realistic rate effects of ballistic impact, static and dynamic indentations experiments were conducted which in conjunction with spectroscopic techniques revealed that B4C is more prone to damage at high rate of loading. This proved the effectiveness of small scale test set-up under laboratory conditions and also, shed light on better understanding of strength degradation mechanism in B4C. Mechanical studies of the ZrB2-SiC composite revealed the presence of unequivocal evidence for room-temperature dislocation mobility and ductility in ultra-high temperature brittle ceramic. Such behavior can play an important role during service, particularly, at high-temperature. Whether it is armor or nose cone in a space shuttle, a weakness or small crack can lead to a catastrophic failure and therefore, understanding of fundamentals is extremely important for design of improved structural components.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Dipankar Ghosh.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Subhash, Ghatu.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041123:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041123/00001

Material Information

Title: Deformation Mechanisms in Advanced Structural Ceramics due to Indentation and Scratch Processes
Physical Description: 1 online resource (188 p.)
Language: english
Creator: Ghosh, Dipankar
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: amorphization, b4c, ceramics, composite, deformation, dislocation, fib, ftir, grain, hardness, indentation, microcracking, microstructure, modeling, plasticity, processing, scratch, sem, sintering, sliplines, spectroscopy, tem, uhtcs, zrb2
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Mechanical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In this dissertation, mechanical properties of boron carbide (B4C) and zirconium diboride-silicon carbide (ZrB2-SiC) composite have been investigated. To capture the realistic rate effects of ballistic impact, static and dynamic indentations experiments were conducted which in conjunction with spectroscopic techniques revealed that B4C is more prone to damage at high rate of loading. This proved the effectiveness of small scale test set-up under laboratory conditions and also, shed light on better understanding of strength degradation mechanism in B4C. Mechanical studies of the ZrB2-SiC composite revealed the presence of unequivocal evidence for room-temperature dislocation mobility and ductility in ultra-high temperature brittle ceramic. Such behavior can play an important role during service, particularly, at high-temperature. Whether it is armor or nose cone in a space shuttle, a weakness or small crack can lead to a catastrophic failure and therefore, understanding of fundamentals is extremely important for design of improved structural components.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Dipankar Ghosh.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Subhash, Ghatu.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041123:00001


This item has the following downloads:


Full Text

PAGE 1

1 DEFORMATION MECHANISMS IN ADVANCED STRUCTURAL CERAMICS DUE TO INDENTATION AND SCRATCH PROCESSES By DIPANKAR GHOSH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Dipankar Ghosh

PAGE 3

3 DEDICATION I dedicate my dissertation to my loving parents and other family members whose constant guidance, encouragement and inspiration, unselfish help and immeasurable sacrifices have contributed tremendously to build my academic carrier and reach a stage where I am standing today. I find myself wordless to express my deepest gratitude for my parents who taught me the importance of higher education and hard work. I will always appreciate and remember their timely guidance in every way throughout the life to accomplish my academic achievements. They not only supported any decision I took but also always have been a part of that. There wont be any further appropriate time to mention my loving wife who has been with me during this wonderful journey of doctoral work and has always forwarded her helping hands. Also, I dedicate this dissertation to all others whose kind blessings and unconditional love have been with me always.

PAGE 4

4 ACKNOWLEDGMENTS I am indebted to many people who have contributed through their support inspiration, knowledge, encouragement and friendship throughout my PhD to bring this dissertation in a completion. Needlessly, I would like to first thank my PhD advisor Prof. Ghatu Subhash who not only gave me an opportunity to come to United States to pursue higher education but also provided continuous support during the past five years. It is really hard for me to overstate my gratitude to Prof. Subhash without whose inspirational and expert guidance, constant encouragements and endless unselfish help, I never could have finished my doctoral research. He is one of those few professors I met who not only motivate graduate students for performing excellent research but also guide them in the right direction. Prof. Subhash has always been supportive and gave me complete freedom to pursue my research in several directions. His experimental expertise and sound theoretical knowledge of research, and dedication for work are always inspirational for graduate students. I will always be grateful for his insightful discussions and suggestions which not only enriched my breadth of knowledge but also be helpful in pursuing my future endeavors. During the past five years, he has been truly an advisor as well as a good friend to me. I truly acknowledge Dr. Tirumalai S Sudarshan and Dr. Ramachandran Radhakrishnan, Materials Modification Inc. (MMI), Fairfax, VA, for providing me the opportunity to work with the sintering system, plasma pressure compaction, at MMI in Summer 2005 to produce the advanced structural ceramic s and composites which have been investigated in this dissertation. I truly enjoyed the period I stayed at MMI and working with Dr. Sudarshan and Dr. Radhakrishnan enriched my knowledge on advanced ceramic processing.

PAGE 5

5 I acknowledge the help of Prof. Yoke Khin Yap, D epartment of Physics, Michigan Technological University, MI, for the research work on spectroscopic investigation of phase transformation in boron carbide ceramics. He not only gave me the opportunity to work in his group and laboratory but also provided useful guidance to complete the work. I would also like to thank Mr. Chee Huei Lee, graduate student of Prof. Yap, for his assistance in performing the spectroscopic studies. I would also like give my sincere thanks to Mrs. Ruth I. Kramer Mr. St ephen F. Forsell Mr. Edward A. Laitila and Mr. Owen P. Mills (technical stuff members in Materials Science and Engineering Department at Michigan Technological University, MI) for their enormous help in learning several characterization equipments. I am t ruly thankful to Prof. Nina Orlovskaya, Department of Mechanical, Materials and Aerospace Engineering, University of Central Florida, FL, for giving me the access to use Raman spectrometer in her laboratory. I am also grateful for her guidance and fruitful discussions in completing the research on residual stress measurement as well as for the help in preparing a manuscript. I sincerely acknowledge Dr. Gerald Bourne, one of my PhD committee members, for his valuable assistance, guidance and time in perform ing the transmission electron microscopic investigation in ZrB2 ceramics without which this dissertation would not have been completed. He was also helpful in preparing multiple manuscripts during the research. I also take this opportunity to thank my other PhD committee members Professors Gregory Sawyer, Nagaraj Arakere and Fereshteh Ebrahimi for their helpful advice and suggestions in general.

PAGE 6

6 I would like to acknowledge my fellow graduate students and friends Timothy Walter, Mickel Klecka, Shiladitya Pa l, Jiwoon Kwon and several others for their time, suggestions and fruitful discussions. Finally, I sincerely thank my parents and other family members for their love, support and encouragements. I am also grateful to my wife Piyasa who always ha s been helpful to me. This work was supported by U.S. National Science Foundation under grant no. CMS0324461with Ken Chong as the program manager. The use of equipment facilities at Major Analytical Instrumentation Center, University of Florida, is also gratefully acknowledged.

PAGE 7

7 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 9 LIST OF FIGURES ........................................................................................................ 10 ABSTRACT ................................................................................................................... 14 CHAPTER 1 ADVANCED STRUCTURAL CERAMICS AND COMPSOITES .............................. 16 1.1 Introduction ....................................................................................................... 16 1.2 Materials ........................................................................................................... 17 1.2.1 Boron Carbide (B4C) ......................................................................... 17 1.2.2 Zirconium diboride silicon carbide composite (ZrB2SiC) .................. 18 2 PROCESSING OF ADVANCED CERAMICS ......................................................... 21 2. 1 Introduction ...................................................................................................... 21 2.2 Literature on Processing of B4C Ceramics ........................................................ 21 2.3 Literature on Processing of ZrB2SiC Composite .............................................. 22 2.4 Plasma Pressure Compaction Technique ......................................................... 24 2.4.1 Processing of FineGrained B4C Ceramics ............................................. 26 2.4.2 Processing of ZrB25wt%SiC Composite ................................................. 27 2.5 Conclusions ...................................................................................................... 29 3 DYNAMIC INDENT ATION RESPONSE OF B4C CERAMICS ................................ 38 3.1 Introduction ....................................................................................................... 38 3.2 Static and Dynamic Indentations ...................................................................... 38 3.2.1 Static Indentation Fracture ...................................................................... 38 3.2.2 Dynamic Indentation Fracture ................................................................. 40 3.2.3 Dynamic Indentation Tester ..................................................................... 42 3.2.4 Experimental ........................................................................................... 42 3.2.5 Results of Indentation Experiments and Discussion ................................ 43 3.3 Spectroscopic Investigation of Localized Phase Transformation ...................... 47 3.3.1 Localized Phase Transformation ............................................................. 47 3.3.2 Fundamentals of Spectroscopic Techniques ........................................... 49 3.3.2.1 Molecular vibration ......................................................................... 49 3.3.2.2 Raman and infrared spectroscopy ................................................. 50 3.3.2.3 Photoluminescence ........................................................................ 52 3.3.3 Experimental ........................................................................................... 54 3.3.4 Results of Raman Spectroscopy ............................................................. 54

PAGE 8

8 3.3.5 Results of Photoluminescence Spectroscopy .......................................... 57 3.3.6 Results of Infrared Spectroscopy ............................................................ 58 3.4 Conclusions ...................................................................................................... 60 4 SCRATCH INDUCED DEFORMATION IN ZrB2SiC COMPOSITE ........................ 78 4.1 Introduction ....................................................................................................... 78 4.2 Scratch Studies ................................................................................................. 80 4.2.1 Experimental ........................................................................................... 80 4.2.1.1 Constant low load and low velocity scratches ................................ 80 4.2.1.2 Variable highload and highvelocity scratches .............................. 81 4.2.2 Results .................................................................................................... 81 4.2.2.1 Microstructural features of constant load low velocity scratches .... 81 4.2. 2.2 Microstructural features of variable load highvelocity scratches ... 84 4.2.3 Influence of Elastic Stress Field on the Scratch ind uced Deformation .... 87 4.2.4 Discussion ............................................................................................... 92 4.3 Residual Stress Measurement within SiC Grains in ZrB2SiC Composite ......... 97 4.3.1 Micro Raman Spectroscopy .................................................................... 97 4.3.2 Experimental ........................................................................................... 99 4.3.3 Results of MicroRaman Spectroscopy ................................................. 100 4.3.4 Evolution of Residual Stress Field ......................................................... 103 4.3.4.1 Residual stress in SiC grains of as processed composite ............ 103 4.3.4.2 Residual stress in SiC grains located within t he scratch grooves 104 4.3.5 Relationship between Mechanical Residual Stress and Raman Spectroscopy .............................................................................................. 106 4.3.6 Determination of Residual Stress .......................................................... 110 4.4 Conclusions .................................................................................................... 116 5 ROOMTEMPERATURE DISLOCATION ACTIVITY IN UHTCs ........................... 141 5.1 Introduction ..................................................................................................... 141 5.2 Experimental ................................................................................................... 142 5.3 Results and Discussion ................................................................................... 143 5.4 Conclusions .................................................................................................... 151 6 SUMMARY AND FUTURE WORK ....................................................................... 160 6.1 Conclusions .................................................................................................... 160 6.2 Future Work .................................................................................................... 164 6.2.1 Determinat ion of Transition Point of Phase Transformation in B4C ....... 164 6.2.2 Scratch Response of Oxidized Surfaces of ZrB2SiC Composite .......... 164 APPENDIX: DYNAMIC INDENTATION RESPONSE OF B4C CERAMICS .............. 168 LIST OF REFERENCES ............................................................................................. 170 BIOGRAPHICAL SKETCH .......................................................................................... 187

PAGE 9

9 LIST OF TABLES Table page 2 1 Processing conditions in P2C method for sintering of B4C compacts ................ 37 3 1 Comparison of average dynamic and static hardness and fracture toughness values ................................................................................................................. 77 4 1 Material properties ............................................................................................ 140

PAGE 10

10 LIST OF FIGURES Figure page 2 1 Schematic of a P2C equipment. ........................................................................ 30 2 2 Consolidated boron carbide disk and test samples for indentation. .................... 31 2 3 SEM micrographs of sintered B4C samples at various processing temperatures and time. ....................................................................................... 32 2 4 Optical micrographs of the etched sintered B4C samples sintered at 1750oC for A) 2, B) 5 and C) 30 min. The average grain sizes are indicated on the optical micrographs. Darker regions in the micrographs represent mostly grain pull outs during polishing. .......................................................................... 33 2 5 A ZrB25wt%SiC composite sintered using P2C technique. .............................. 34 2 6 A) SEM micrograph of the fragmented surface of ZrB2SiC composite and B) a higher magnificat ion SEM image revealing absence of any porosity within ZrB2 matrix as well as at the ZrB2SiC interface areas. ...................................... 35 2 7 XRD pattern collected from the sintered ZrB2SiC composite. ............................ 36 3 1 Schematic of experimental setup for dynamic indentation hardness measurements. ................................................................................................... 61 3 2 Comparison of static and dynamic hardness values for three grain sizes of B4C. .................................................................................................................... 62 3 3 Optical micrographs of A) static and B) dynamic indents for 2.7 m grain size B4C at 300 gm. Note longer cracks and more severe damage in dynamic indent. ................................................................................................................. 63 3 4 Comparison of static and dynamic indentation fracture toughness for three different grain sizes of boron carbide. ................................................................. 64 3 5 Subsurface damaged region beneath A) static indentation and B) dynamic indentation at a load of 19.6 N for 1.6 m grain s ize boron carbide. .................. 65 3 6 Different normal of mode of vibrations. ............................................................... 66 3 7 A simple schematic illustrating Rayleigh, Stokes and anti Stokes Raman scattering phenomena. ....................................................................................... 67 3 8 A schematic of the infrared absorption spectroscopy. ........................................ 68

PAGE 11

11 3 9 Changes in the net dipole moment of a molecule during asymmetric stretching. ........................................................................................................... 69 3 10 A schematic illustration of photoluminescence phenomenon. ............................ 70 3 11 A) visible Raman spectra and C) uv Raman spectra from the unindented surfa ce and damaged regions beneath the static and dynamic indentations. B) Magnified view of two of the spectra in A. ...................................................... 71 3 12 Crystal structure of boron carbide (B4C). ............................................................ 72 3 13 Origin of D peak and G peak. ............................................................................. 73 3 14 Photoluminescence from the unindented surfaces and damaged regions beneath the static and dynamic indentations. ..................................................... 74 3 15 FTIR spectra from the un indented surfaces and damaged regions beneath the static and dynamic indentations. ................................................................... 75 3 16 Schematic of A) a rhombohedral unit cell of B4C with eight B11C icosahedrons at the corners. B) Amorphization of each unit cell will form two carbon and one boron atoms and a B11C icosahedron. C) One B11C icosahedron and one boron atom will reorganize into one amorphous boron ( a B) cluster, and a carbon atom that form carbon clusters with adjacent carbon. ............................................................................................................... 76 4 1 A schematic of instrumented highvelocity scratch tester. ................................ 118 4 2. Residual scratch profiles on the polished surface of ZrB2SiC composite at various load levels. ........................................................................................... 119 4 3 A) Penetration depth profile and B) the corresponding trace of residual scratch groove at 50 mN. Increase in scratch depth corresponds to the damaged regions as indicated by the white dashed circles in Fig. 43 B). Magnified SEM micrographs of the damaged regions along the scratch path revealed mostly voids, grain pull out, ZrB2 grainboundary fracture and some microcracking. .................................................................................................. 120 4 4 A) Penetration depth profile and B) the corresponding trace of residual scratch groove at 250 mN. Damaged regions along the scratch path are shown by the white dashed circles. .................................................................. 121 4 5 Scratch depth profiles during the nanoscratch experiments at different loads. 122

PAGE 12

12 4 6 A) Micrograph of scratch induced deformation features at a load of 250 mN. B) A magnified view of the region (X) revealing the slip line patterns and micr ocracks along the scratch groove as well as cracks emanating from a SiC particle. C) Another region along the scratch path revealing several sets of slip lines oriented randomly with respect to the scratch direction. ................ 123 4 7 Fracture patterns within the scratch groove at a load of 250 mN: A) grainboundary fracture, microcracking, and scratch debris and B) interfacial cracking between ZrB2 and SiC phases as well as microcracks perpendicular to the direction of the scratch path. ................................................................... 125 4 8 Scratch induced damage at 250 mN, revealing transgranular microcracking in ZrB2SiC composite. ..................................................................................... 126 4 9 A) A high velocity scratch groove, B) high magnificati on images of extensive transgranular microcracks orthogonal to the scratch direction and C) normal force vs. scratch length profile. ......................................................................... 127 4 10 Slip line formation within ZrB2 phase during highvelocity scratch process. ..... 128 4 11 A) Normalized maximum principal stress distribution, B) schematic of the scratch process, C) normalized maximum shear stress distribution in the vicinity of indenter tip and D) plot of maximum principal stress contours and orientation in the wake of the indenter. ............................................................. 129 4 12 Variation of A) normalized maximum principal stress ( 1 ) at 0 y and B) normalized maximum shear st ress ( max ) at 0 y and C) 0.5 y for various values of k. The black triangle indicates the indenter position. ....................... 130 4 13 Micrograph of the exit end region of a scratch at 250 mN revealing numerous slip lines. The white dotted lines indicate the boundary of residual scratch e xit end. ............................................................................................................ 131 4 14 Raman spectra collected from the ZrB2 matrix phase (see the inset also) and the SiC particulate phase present in ZrB2SiC composite. ................................ 132 4 15 Raman spectra collected fr om the SiC grains present within and away from the scratch grooves. ......................................................................................... 133 4 16 A) LO and TO Raman peak positions of SiC grains withi n the scratch grooves at 50, 100, 150 and 250 mN loads. LO and TO peak positions at 0 mN correspond to the Raman spectra collected from SiC grains outside the scratch groove. B) Changes in peak positions to stress free LO and TO positions. .......................................................................................................... 134 4 17 Schematic of the scratch process and different crack systems that evolve during the scratch process are shown. ............................................................. 136

PAGE 13

13 4 18 Distribution of normalized residual stress components; A) R x B) R y and C) R z on the xy plane (B/P = 0.005). The black triangle indicates the position of the indenter whereas the black arrow indicates the scratch direction. .............. 137 4 19 Evolution of residual stress within SiC grains as a function of scratch load. ..... 138 4 20 SEM micrograph of a scratch groove at 250 mN revealing the uncracked SiC grain surrounded by the heavily microcracked ZrB2 matrix. .............................. 139 5 1 Optical micrographs revealing slip lines formed in the vicinity of indented regions of A) ZrB2 and B) HfB2 ceramics. ......................................................... 153 5 2 FIB cut TEM specimen within a constant load (250 mN) scratch groove. ........ 154 5 3 A) A bright field TEM micrograph revealing dense dislocation activity on prismatic planes and B) multiple sets of intersecting slip lines formed on the surface due to scratch load, resembling patterns observed in A). .................... 155 5 4 A) Bright field TEM micrograph at [1210] zone axis and B) the corresponding selected ar ea electron diffraction pattern. C) Bright field TEM image of dislocations in two beam condition with (0001) g vector. ................................. 156 5 5 Electrical conductivity (filled symbols) and electrical resistivity (open symbols) of metals and ceramics. .................................................................................... 158 5 6 Crystal struct ure of transition metal diborides. .................................................. 159 6 1 Cross sectional areas of the ZrB25wt%SiC specimens oxidized for A) 1 B) 5 and C) 15 hours. ............................................................................................... 166 6 2 Evolution of structures in a ZrB2SiC composite during oxidation at 1500oC. ... 167 A 1 Variation of c vs. P in B4C ceramics for dynamic indentations. ......................... 168 A 2 Variation of l vs. P in B4C ceramics for dynamic indentations. .......................... 169

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DYNAMIC INDENTATION AND SCRATCH RESPONSE OF ADVANCED STRUCTURAL CERAMICS By Dipankar Ghosh December 2009 Chair: Ghatu Subhash Major: Mechanical Engineering Plasma pressure compaction technique was used to develop boron carbide (B4C) and zirconium diboridesilicon carbide (ZrB2SiC) composite. B4C ceramics are extensively used as body armor in military and civilian applications, and ZrB2SiC composite has been recognized as a potential candidate for hightemperature aerospace applications. In this dissertation, processing parameters, quasistatic and highstrain rate mechanical response, and fundamental deformation mechanisms of these materials have been investigated. In the case of B4C, the rate sensitivity of indentation hardness was determined using a dynamic indentation hardness tester that can deliver loads in 100 s. By comparing dynamic hardness with the static hardness, it was found that B4C exhibits a lower hardness at highstrain rate, contrary to known behavior in many structural ceramic s. However, these results are consistent with the ballistic testing of B4C armors as reported in recent literature. This behavior was further investigated using a series of spectroscopic techniques such as visible and UV microRaman, photoluminescence and infrared. These studies not only confirmed that structural transformation occurred during indentation experiments similar to that in ballistic testing of B4C but also

PAGE 15

15 suggested a greater degree of structural changes under dynamic loading compared to stati c loading. Due to the potential application as external heat shields in supersonic vehicles, scratch studies were conducted on the ZrB2SiC composite. These studies revealed metal like slip line patterns which are indeed an unusual in brittle solids at room temperature. Utilizing classical stress field solutions under combined normal and tangential loads, a rationale was developed for understanding the formation of scratch induced deformation features. Also, an analytical framework was developed, combini ng the concept of blister field and the secular equation relating Raman peaks to strain, to measure scratch induced residual stress employing microRaman spectroscopy. Transmission electron microscopic investigations confirmed the existence of dislocat ions within the ZrB2 phase. It has been argued here that readily detectable slipline patterns are reflection of metallicity in chemical bonding present in ZrB2 ceramics which has also been suggested in recent literature from chemical bonding and electroni c structure investigations.

PAGE 16

16 CHAPTER 1 ADVANCED STRUCTURAL CERAMICS AND COMPSOI TES 1.1 Introduction Advanced ceramics and their composites with light weight, high strength and superior thermomechanical properties are central to the development of s tructural components for advanced military and aerospace applications. Among these, armor ceramics for personal and vehicle protection against impact threats,1 and ultrahigh temperature ceramics (UHTCs) for leading edge components in hypersonic and reusabl e launch vehicles have received considerable attention in recent years.2 In these applications, demand for materials with controlled and defect free microstructure is paramount. Therefore, the need for nonconventional processing techniques and advanced characterization techniques has increased rapidly. A major disadvantage of the traditional powder processing/sintering methods is that they do not lend themselves to rapid processing due to long processing times and elevated temperature requirements.35 Towards this end, a new processing method called plasma pressure compaction,69 also called spark plasma sintering (SPS), has been used to process two materials: (1) fine grained boron carbide (B4C) ceramics (used for body armor), and (2) a zirconium dibor ide silicon carbide (ZrB2SiC) composite (a potential UHTC ceramic for applications in aerospace and military applications).2 In this dissertation, mechanical response of finegrained B4C ceramics has been investigated employing static and dynamic indentations where as the ZrB2SiC composite has been subjected to scratch experiments. Several issues such as processing of advanced ceramics, localized phase transformation in B4C ceramics, deformationinduced residual stress measurements as well as evolution of

PAGE 17

17 microplasticity in ZrB2SiC composite have been explored throughout this dissertation. These two structural ceramics have different deformation mechanisms and hence, they have been discussed in separate sections and/or chapters. 1.2 Materials 1.2.1 Boron Carbide (B4C) Boron carbide (B4C) ceramic is an excellent candidate material for structural applications at room and high temperatures because of its high melting temperature (2450oC), high elastic modulus (450 GPa), high hardness (Vickers hardness > 25 GPa, next only to diamond and cubic boron nitride), high flexural strength (350500 MPa), low density (2.52 g/cm3) and excellent wear resistance .1016 B4C is used as grinding medium for hard materials, lightweight ceramic armor, wear resistant sandblasting nozzle material, and as neutron absorber in nuclear reactors.10,17 Due to its low density (2.52 g/cm3), high strength and high hardness, B4C was expected to exhibit superior performance as armor ceramic for protection against projectile impac t threats. However, under high velocity and high pressure conditions, B4C has been shown to undergo significant strength degradation.18 Using high resolution transmission electron microscopy (TEM) on the ballistic fragments of B4C, Chen et al .,18 revealed that B4C undergoes localized collapse of the crystal lattice due to solid state phase transformation from crystalline phase to amorphous phase during the impact. The amorphous phase, being weaker than the crystalline phase, is argued to be responsible for the observed loss of impact resistance in B4C armor against high velocity projectile threats. The above fundamental studies are indeed of great value but the costs and time associated with such ballistic experiments and preparation of large B4C tiles (of several

PAGE 18

18 tens of centimeters in dimensions) for impact experiments are expensive. In such situations, new experimental techniques that can induce similar deformation mechanisms in small specimen sizes (of 34 mm in size) can be of high economic value. Impac t experiments not only involve high pressure but also high strain rates of loading which increases with projectile velocity.1921 While the traditional indentation experiments with sharp point indenters can generate a large pressure at a moderate load within the materials, they are unable to capture highstrain rate effect comparable to impact experiments.2226 Towards this end, a dynamic indentation method27,28 can be a better test technique to characterize the deformation mechanisms in armor ceramics. Thi s technique not only delivers load in a controlled manner, similar to static indentation test method, but also produces strain rate of the order of 103/s which are essential to mimic ballistic impact tests in laboratory conditions. Thus, this dynamic indentation method (will be described later in detail) can be effectively used to augment the ballistic impact tests to uncover similar phenomena because of its ability to provide high pressure and high velocity loading into a small localized region. 1.2.2 Zir conium diboride silicon carbide composite (ZrB2SiC) Development of novel light weight and high strength materials which can withstand elevated temperatures (above 2000oC), and provide good thermal insulation properties are crucial to meet future demands of many civilian, defense and aerospace applications.2931 Components (e.g., nozzles, propulsion components, leading edge materials for space vehicles and thermal protection systems) for future high performance aircrafts, hypersonic vehicles, kinetic ener gy interceptors and reusable space planes, etc. operate in severe reactive environments with temperatures well above the melting points of traditional materials. Few materials can withstand such high

PAGE 19

19 temperatures and yet provide adequate mechanical strengt h.32 Ultra high temperature materials (UHTMs) have been identified as the potential candidates for these applications.2,33,34 UHTMs are a class of materials that are chemically and physically stable at temperatures above 2000oC and in reactive atmospheres (e.g., monatomic oxygen). A subclass of these materials is ultrahigh temperature ceramics (UHTCs) which are borides, nitrides and carbides of transitional metals (e.g., Ta, Hf, Zr).2 These ceramic materials (e.g., ZrB2, ZrC, HfB2, HfN, HfC) have high mel ting point above 3200oC. Other materials that may also fall in this UHTMs category are silicon carbide (SiC), graphite and rhenium (Re).29 Among these, Graphite and other carbonbased materials degrade rapidly at temperatures above 800oC due to oxidation. Rhenium has high strength and high melting point but is extremely dense (21 gm/cm3) and oxidizes easily.35,36 In this family of UHTCs, zirconium diboridesilicon carbide (ZrB2SiC) composites2,3740 have drawn a special attention owing to their unique properties such as high melting point (> 3000oC), superior oxidation resistance above 1500oC, excellent thermal shock resistance, low density (6.09 gm/cm3) and good mechanical and chemical stability at elevated temperatures. Most of the recent investigations have been mainly focused on processing and oxidation behavior of ZrB2SiC composites.3748 Mechanical characterization of ZrB2 has been limited to the determination of static hardness2,49,50 fracture toughness,2,47,49,50 Youngs modulus,2,50,51 creep response,52 room and high temperature flexural strength,2,46,47,53 and high temperature arc jet testing.54 While in service (e.g., during takeoff, landing, or reentry in to atmosphere), the structural components in aerospace vehicles are prone to impact by meteorites and

PAGE 20

20 atmospheric debris particles at very high velocities. Under such abrasiondominated wear s cenarios (abrasive wear), sharp particles impacting on the exposed surfaces can result in inelastic deformation and material removal. Therefore, evolution of damage and fundamental inelastic deformation mechanisms should be fully understood to evaluate the suitability of ZrB2SiC composite for the above mentioned applications. Although these composites are intended to be used at high temperatures, investigation of mechanical responses at room temperature constitutes the first step towards a better understanding of their inelastic behavior in service. The rest of the dissertation is organized as follows. In Chapter 2, the processing method to consolidate B4C and ZrB2SiC is discussed. The mechanical response of B4C ceramics under static and dynamic indentat ions as well as spectroscopic investigations of structural phase transformation are discussed in C hapter 3. Then, the scratch induced deformation features in ZrB2SiC composite and residual stress measurements within the particulate phase (SiC), using micr o Raman spectroscopy, are discussed in Chapter 4. The transmission electron microscopic studies of the deformed regions of the ZrB2SiC composite are discussed in Chapter 5. Finally, in Chapter 6 conclusions and future research directions are outlined.

PAGE 21

21 CHAPTER 2 PROCESSING OF ADVANCED CERAMICS 2. 1 Introduction In this dissertation, two advanced structural ceramics, one for armor applications and other for hightemperature aerospace applications, have been developed. Due to their high melting temperature and strong covalent character of chemical bonding, processing of these ceramics is challenging. Use of traditional sintering methods requires high sintering temperature, moderatehigh pressure, and often long consolidation time to produce sintered compacts with high density. In the following, literature on the processing of these ceramics, using conventional sintering techniques, is discussed first. Then, principle of a new and unconventional sintering technique called plasma pressure co mpaction (P2C) is discussed followed by separate discussions on processing of B4C ceramics and ZrB2SiC composite employing P2C method. 2.2 Literature on Processing of B4C Ceramics Processing of pure boron carbide, using traditional sintering techni ques, is challenging due to the difficulty associated with sintering of the starting powder. This is attributed to the high covalent bonding and low self diffusion.10 As a result, high temperatures and high external pressures are required to produce dense B4C ceramics. Traditionally, boron carbide has been consolidated using several sintering techniques such as (i) hot pressing (HP) with and without sintering additives,10,55,56 (ii) hot isostatic pressing (HIP),10,57 (iii) pressureless sintering (PS) in the presence of sintering additives,10,11,1315,58,59 (iv) pressureless sintering in a gaseous atmosphere of hydrogen and helium,60 and (v) microwave sintering.61 Among the above powder processing

PAGE 22

22 techniques, hot pressing and pressureless sintering are the most commonly employed methods to produce boron carbide ceramics with 95 99 % of theoretical density and with grain sizes in the range between 1.5 60 m. However, in hot pressing and pressureless sintering methods, the sintering temperatures used were relatively high (> 2000oC) and the sintering times were on the order of hours. In these sintering techniques, sintering additives are often used which aid the densification of difficult to sinter ceramic powders. However, the addition of sintering additives was found to reduce fracture strength moderately due to formation of weak glassy phases at the grainboundary regions.59 Also, these sintered ceramics with sintering additives are not suitable for nuclear applications10 where high purity boron carbide is r equired for neutron absorption. Although, dense boron carbide ceramic can be produced at lower temperatures using hot isostatic pressing, this method is not suitable for bulk processing.10,57 Boron carbide powder heat treated in a gaseous mixture of hydrogen and helium and then sintered in the presence of pure helium also requires sintering temperature above 2200oC.60 Microwave sintering has also been used to consolidate boron carbide (95 % density) in a short duration of time (12 min) but it also requires high temperatures above 2000oC.61 This limitation is overcome by using P2C technique as described in section 2.3. 2.3 Literature on Processing of ZrB2SiC Composite Ultra high temperature ceramics such as ZrB2 is also a difficult to sinter material due to the highly covalent nature of chemical bonding.2 ZrB2 and its composites have been produced by several conventional consolidation techniques such as (i) pressureless sintering (PS), (ii) hot pressing (HP) and (iii) reactive densification process

PAGE 23

23 such as reactive HP. In these techniques, high sintering temperature ( 2000oC) and long consolidation periods are required to obtain dense ZrB2 ceramics. In PS method, using fine starting powder particles, dense ZrB2 ceramics (~95%) was achieved above 2000oC sint ering temperature in the presence of large amount of sintering additives.62,63 In HP technique, high sintering temperature (>2000oC) and pressure (>25 MPa) along with long consolidation times were required to produce dense (>95% of theoretical density) ZrB2 ceramics.6466 Some recent studies have produced 99% dense ZrB2 ceramics at 1900oC (consolidation time of 45 min and pressure of 32 MPa).67 However, fine milled ZrB2 powder particles (500 nm) were used to reduce the sintering temperature. In reactive HP technique, synthesis and densification processes are combined utilizing in situ reaction and sintering processes to produce dense ceramics.2 Although, this method has the advantage of producing extremely fine reactants due to in situ reaction, long consoli dation times often result in significant grain coarsening at lower temperatures. Therefore, higher sintering temperatures are still required to produce dense ZrB2 ceramics. Using this reactive HP method, dense ZrB2 (99%) was produced at a sintering temperature of 2100oC.68 Some of the recent studies have shown the advantage of using sintering additives to reduce the sintering temperature.2 However, owing to the intended hightemperature applications, use of sintering additives is detrimental as the presence of glassy phase in the sintered compact can reduce the strength at hightemperature considerably. In recent years, researchers have shown that only pure ZrB2 is not suitable for high temperature applications as it is prone to oxidation.2 Towards thi s end, ZrB2SiC composites have received considerable attention due to their improved oxidation

PAGE 24

24 resistance.2,3740 Also, addition of SiC has been shown to improve the densification behavior and mechanical properties of ZrB2 considerably.2,47,50,53 Processi ng conditions to produce dense ZrB2SiC composites mainly depend on the amount of SiC phase and the starting powder particle size. Using HP technique, ZrB2SiC composites (from 1030vol% SiC content) have been sintered in the temperature range of 18001950oC (with pressure and consolidation time ranging from 2050 MPa and 2060 mins, respectively) to produce dense compacts.38,47,50,53,69 Chamberlain et al.,42 utilized reactive HP to produce dense ZrB230vol%SiC composite at 1700oC where a pressure of 40 MPa was applied over a period of one hour. However, as mentioned before, due to the in situ reaction the reactants were extremely fine (below 100 nm) and thus, reduced the sintering temperature considerably. Zhang et al.,70 produced ~98% dense ZrB2SiC composite using reactive HP at 1900oC at a pressure of 30 MPa and over a consolidation period of one hour. In the current work, P2C technique has been utilized to produce a ZrB2SiC composite at significantly lower temperature and processing time which will be discussed in the following section. 2.4 Plasma Pressure Compaction Technique In recent years, plasma pressure compaction (P2C) has evolved as a promising powder consolidation method for sintering of metallic, ceramic and intermetallic particles.69 Unlike the traditional sintering techniques where consolidation times are on the order of several hours, in P2C method various materials, metals, alloys and ceramics, have been successfully sintered to full density in less than 10 minutes. Sintering via t his technique involves plasma activation and localized resistive heating (i.e. Joule heating) of a powder compact through the application of a low voltage direct

PAGE 25

25 current (DC).69 Bonding of the particles is accomplished in two stages: first an insitu elim ination of surface oxide and contaminates due to application of the pulsed DC voltage (~ 5 V), and then consolidation of the purified powder through Joule heating by application of continuous DC voltage. In this technique, powders are first compressed i n a die (graphite, C C, steel, or Mo alloys) by external pressure to establish a current path and then pulsed DC voltage is applied for surface activation. As the effective current path is not established at the beginning, current does not flow freely through the powder compact. Thus, a charge, buildup at the interparticle gaps, causes polarity differences between the particles. As the charges accumulate, voltage difference becomes sufficiently large so that a spark is generated which triggers an ionization process i.e. interparticle plasma. The ions move toward the negatively charged particles where as electrons move toward the positive charged particles. These ionic and electron bombardments (impact of plasma) lead to the removal of oxide layer, impurities, moisture and adsorbed gas present on powder particle surfaces. This process, thus, results in cleaned particle surfaces for subsequent densification. Upon application of a DC voltage, current concentrates at the interparticle contact areas (which establ ish current path) and enormous heat is generated at these areas through Joule heating. This results in softening of those contact areas and result in plastic deformation as the external pressure is increased. Thus rapid densification, over a consolidation period of 510 mins, is facilitated through microwelding and plastic yielding mechanisms. Surface purification and rapid consolidation result in a dense microstructure with minimum grain growth and contamination. A schematic of the P2C equipment is shown in Fig. 2 1.

PAGE 26

26 2.4.1 Processing of FineGrained B4C Ceramics Commercially available B4C powder (Grade HS, H.C. Starck, Germany), with particles of sizes 100 500 nm, was used as the starting material.71 The powder was consolidated using P2C method under high vacuum (200300 mTorr) without the use of any sintering aids. In the current work, the powder compact was consolidated at a constant low voltage (~5 V) direct current and a maximum current density of 4400 amp/cm2. Simultaneously, an external pressu re (88 MPa) was applied to assist powder consolidation. The sintered compacts were produced in the form of a cylindrical disk of 51 mm diameter and 6.4 mm thickness, see Fig. 22. These disks were sintered either at a temperature of 1650oC or 1750oC and the consolidation time was varied between 2 to 30 min. Table 21 shows the processing conditions for all the disks. Densities of the sintered disks were measured using Archimedes method. Microstructural analysis was performed on both fractured and polished s urfaces. Polished specimens were prepared by using standard metallographic principles. The polished surfaces were etched electrolytically using 1% KOH solution at a current density of 0.03 amp/cm2 for 30 seconds.71 From the electrolytically etched specimens, average grain size was determined (by the line intercept method) as per ASTM standard E 11296. Table 21 provides the density measurements and average grain sizes of the B4C ceramics. Sintered B4C compacts produced at 1650oC attained only 9293 % of theoretical density and thus were not considered for further investigation. Upon increasing the sintering temperature to 1750oC a significant improvement in density was observed. Compacts consolidated for 2 min and 5 min at 1750oC attained densities around 96% and 99% of the theoretical value, respectively. When

PAGE 27

27 consolidation time was increased to 30 min at this temperature, no further increase in density was noticed. The improvement in density (or decrease in porosity level) wit h increase in temperature is clearly evident from the SEM micrographs of the fractured surfaces shown in Fig. 23. Microstructures of the polished and electrolytically etched specimens from the three B4C disks produced at 1750oC are given in Fig. 24. Opt ical micrographs revealed nearly equiaxed and finegrained microstructures with average grain sizes around 1.6 m, 2 m and 2.7 m for the samples sintered for 2 min, 5 min and 30 min, respectively. Darker regions in the micrographs represent mostly grain pull outs during polishing. From the density and average grain size measurements it is clear that the processing parameters (temperature, pressure and consolidation time) adopted in the current work were appropriate for the production of dense finegrained boron carbide ceramics. 2.4.2 Processing of ZrB25wt%SiC Composite In the current work, commercially available ZrB2 powder (Grade HS, H.C. Starck, Germany), with particles of sizes between 37 m, and polycarbosilane ( (SiHMe CH2)n ) powder were used as starting materials.72 Polycarbosilane (PCS) has a silicon carbon (Si C) backbone and is used as a preceramic precursor for SiC.69 Initially, the powders were mixed in a shear mixture and the blend was calcined above 1000oC in an atmosphere of flowing argon where the PCS decomposed to cubic SiC (SiC) and amorphous carbon. Then ZrB2 and heat treated PCS mixture were further mixed, and consolidated using the P2C technique at a temperature of 1750oC and at a pressure of 43 MPa over a short consolidation per iod of only 5 min.72 Density of the sintered ZrB2SiC composite was measured using Archimedes method. X ray diffraction (XRD)

PAGE 28

28 analysis was carried out on small specimens to investigate the phase content within the consolidated compact. Some fragmented piec es were examined using scanning electron microscope (SEM) to reveal the average grain size in the composite. Figure 25 shows a ZrB25wt% SiC composite slab of 75mm52mm8mm was produced using (P2C) technique and Fig. 26 reveals low and high magnification SEM micrographs of fractured surfaces of the composite. XRD pattern collected from the sintered composite, shown in Fig. 2 7, revealed the presence of only two crystalline phases; hexagonal ( H ) Z rB2 and cubic (3C ) SiC. Density measurements indicated a composite density 96% of theoretical value (5.84 g/cm3). The theoretical density of the composite was calculated using the rule of mixture. SEM micrographs of the fragmented surface, shown in Fig 26 revealed a well consolidated composite and thus supported the high value of measured density. No porosity was detected within the ZrB2 matrix as well as at the ZrB2SiC interface areas. However, some agglomerated regions (containing 3C SiC and amorphous carbon) were observed in the sintered compact which probably resulted the measured 4% porosity. SEM micrographs revealed that the SiC phase (dark phase) was uniformly distributed within the ZrB2 matrix (gray phase). Nevertheless, microstructural observations indicated that P2C method was successful in producing a uniform and well consolidated ZrB2SiC composite. From the SEM micrographs of the fractured surfaces, average grain size of the ZrB2 phase was estimated to be around 5 m whereas the SiC particles were around 1 m in size.

PAGE 29

29 2.5 Conclusions Boron carbide powder sintered using the P2C technique resulted in near theoretically dense and finegrained boron carbide ceramic disks at significantly lower processing time and temperatures than in convent ional sintering methods. A polycrystalline ZrB25wt.%SiC composite, with a consolidated density above 96% of theoretical density, was produced using the P2C technique at a temperature of 1750 oC and at a pressure of 43 MPa over a short consolidation period of only 5 min. As outlined in C hapter 1, in the following chapters, mechanical responses of P2C processed B4C ceramics and ZrB25wt%SiC composite are discussed in detail. Since, the issues being investigated different, they are discussed in separate chapters. In the following C hapter 3, static and dynamic indentation response as well as phase transformation studies in B4C are discussed.

PAGE 30

30 Figure 21. Schematic of a P2C equipment.

PAGE 31

31 Figure 22. Consolidated boron carbide disk and test samples for indentation.

PAGE 32

32 Figure 23. SEM micrographs of sintered B4C samples at various processing temperatures and time.

PAGE 33

33 Figure 24. Optical micrographs of the etched sintered B4C samples sintered at 1750oC for A) 2, B) 5 and C) 30 min. The average grain sizes are indicated on the optical micrographs. Darker regions in the micrographs represent mostly grain pull outs during polishing. 1.6 m 2 m A B

PAGE 34

34 Figure 25. A ZrB25wt%SiC composite sintered using P2C technique.

PAGE 35

35 Figure 26. A) SEM micrograph of the fragmented surface of ZrB2SiC composite and B) a higher magnification SEM image revealing absence of any porosity within ZrB2 matrix as well as at the ZrB2SiC interface areas. ( a )

PAGE 36

36 Figure 2 7. XRD pattern collected from the sintered ZrB2SiC composite. SiC h ZrB 2 (c) -100 400 900 1400 1900 2400 20 30 40 50 60 70 80 Two Theta (degree) Intensity (a.u.) 3 C SiC H ZrB2

PAGE 37

37 Table 21. Processing conditions in P2C method for sintering of B4C compacts Sample ID Temperature (oC) Time (mins) Pressure (MPa) Density (% theoretical) (g/cc) Grain size ( m) 1650_5 1650 5 88 2.33 0.002 (92.46 %) Not measured 1650_30 1650 30 88 2.35 0.003 (93.25 %) Not measured 1750_2 1750 2 88 2.42 0.002 (96 %) 1.6 1750_5 1750 5 88 2.50 0.004 (99.2 %) 2.0 1750_30 1750 30 88 2.50 0.004 (99.2 %) 2.7

PAGE 38

38 CHAPTER 3 DYNAMIC INDENTATION RESPONSE OF B4C CERAMICS 3.1 Introduction In this work, strain rate sensitivity of boron carbide ceramics has been studied using dynamic and static indentation experiments. These experiments were conducted on the intact as well as on the split specimens (prepared using bondedinterface technique). Hardness measurements were performed on intact specimens where as the split specimens were utilized to probe, using a series of spectroscopic techniques, the subsurface damage regions of both types of indentations for structural phase transformation in B4C. Rest of the C hapter is divided into two sections. In the first part (section 3.2), results from the static and dynamic indentation studies are discussed followed by an indepth discussion, in section 3.3, on spectroscopic investigation of structural phase transformation in B4C. 3.2 Static and Dynamic Indentations 3.2.1 Static Indentation Fracture Since the advent of fundamental pioneering concepts in fracture by Inglis,73 Griffith,74 Irwin,75 and Orowan,76 the field of fracture mechanics has evolved into a mature subject and is regularly practiced in design of structures subjected to a wide range of loads. Advances in materials research and their applications in emerging fields continue to identify new fracture phenomena that challenge researchers to develop new analytical, computational and characterization tools, and provide further insight in to their failure behavior. In recent years, several mechanical characterization techniques such as instrumented nanoindentation, dynamic indentation and microand nanoscratch experiments have been successfully employed on brittle materials to unravel fundamental

PAGE 39

39 mechanisms associated with fracture and material removal at nanoand micro levels during high speed grinding, dynamic wear and ballistic impact phenomena. Among many characterization techniques available, indentation fracture mechanics approach has been one of the most popularly used methods to evaluate the fracture resistance of brittle materials.19,2226,77 79 The field of static indentation fracture mechanics has made significant contributions to our understanding of fracture behavior of brittle solids and materi al removal mechanisms during abrasion and wear. Investigations have ranged from indentation fracture mechanics of monolithic brittle solids to indentation of hard particles embedded in ductile materials (e.g., carbides in tool steels).80 In addition to their ability to extract fundamental fracture characteristics of brittle solids, indentation tests have often provided motivation for development and/or validation of sophisticated numerical models. However, this approach has been mostly applied to static loads and its use for dynamic deformation of materials has been limited. Numerous applications involve rapidly applied loads where dynamic or high strain rate effects become relevant. For example, in highspeed grinding of ceramics, the grinding grit is in c ontact with the work piece for only few tens of microseconds. During impact of a projectile on a confined ceramic target the interaction time between the two contact surfaces is on the order of few microseconds. Similarly, the rate effects may be important in applications such as dynamic wear between two rapidly moving surfaces, meteorite impact on space structures, runway debris impact on airplane structures, atmospheric particles impact on the protective tiles of a reentry vehicle, etc. In all these event s, the interaction between the two contact surfaces can be considered as Hertzian,81 however, the static indentation mechanics principles cannot be fully applied to analyze the dynamic inelastic deformation behavior that is germane to such dynamic events because

PAGE 40

40 the contact time is on the order of few tens of microseconds. Materials under dynamic loads can experience strain rate well in excess of 103/s and, at such high strain rates, the deformation behavior in brittle materials, in general, is characterize d by increased fracture strength27,8290 and fracture toughness.28,91100 3.2.2 Dynamic Indentation Fracture Understanding of projectiletarget interaction is central to the development of superior armor/anti armor materials. For design of efficient c omposite armor systems such as metal/ceramic or polymer/ceramic armor, the interfacial characteristics as well as the deformation and fracture characteristics of individual components must be well understood. Fundamental investigations that have provided k ey insights into the interactions between a metal projectile and a ceramic target have been conducted by several researchers.1,20,21,101114 A survey of candidate ceramic armor materials and the relevant experimental data have been compiled by Holmquist et al .115 Through proper target design, Hauver et al. ,116,117 were able to increase the resistance to penetration of a ceramic target by extended lateral flow at the surface of the ceramic and thus recover intact ceramic target after the impact. This was ach ieved by confining the ceramic within steel case, by providing space for erosion products, and providing a shock attenuator and a steel plate at the entrance to the target. Under these conditions, microstructural analysis of the impacted ceramic targets has revealed similarities of fracture patterns to the ring cracks induced by static indentation of a hard sphere on a ceramic which clearly demonstrated the validity of contact mechanics approach to investigate the damage due to projectile impact on cerami cs. Therefore, modeling efforts to describe impact damage in ceramics have centered around Hertzian contact theory coupled with brittle fracture models in compression.81,118,119

PAGE 41

41 Although the similarities between induceddamage in a confined ceramic target due to projectile impact and that induced due to a static spherical indentation are well recognized, the rate effects that are germane to dynamic impact phenomena are not fully captured in static indentation investigations. In addition, modeling efforts that seek to estimate pressure induced in the target due to impact are limited by the existing elastic perfectly plastic models which fail to describe the stresses accurately in the comminuted region of the impact. Thus, the current contact mechanics approac h to describe the physical phenomena ahead of a projectile suffers from some inherent limitations. To overcome these limitations, several experimental, analytical and numerical approaches have been adopted in the recent literature. Towards these developments, Subhash and co workers27,28,120122 have developed a dynamic indentation test method to investigate the ratedependent indentation properties of ceramics and metals. Here the word dynamic refers to the time of loading which is only around hundred mi croseconds. This technique is based on the momentum trapping principle developed by Nemat Nasser et al.,123 for dynamic compression testing and is modified to impart a single indentation in to the specimen during dynamic indentation. The technique has been successfully utilized to determine the dynamic Vickers indentation hardness of metallic materials,27,122 bulk metallic glasses124,125 and several ceramics.27,28 The above studies on metals are very generic in the sense that the dynamic hardness has been f ound to be three times the dynamic yield strength of a metallic material. This result is similar to the static hardness yield strength relationship proposed by Tabor.126 Recently, such test method has been successfully used to detect negative rate sensitiv ity of hardness in amorphous alloys.125,127 Also, in bulk metallic glasses, dynamic indentation revealed significant difference in shear band propagation phenomena compared to static indentation.125

PAGE 42

42 3.2.3 Dynamic Indentation Tester Subhash and co workers27,28 have adopted a dynamic indentation method to investigate ratedependent indentation properties of metallic and brittle materials. In this technique, elastic stress wave propagation in a slender rod is utilized to deliver the desired load in 150 s duration compared to static indentation where load is applied over 1015 seconds. This technique is parallel to the static indentation technique and therefore, a direct comparison between static and dynamic indentations can be made. The dynamic hardness tester consists of a slender rod with a Vickers indenter mounted at one end and a momentum trap (MT) assembly at the other end as shown schematically in Fig. 31 A high frequency load cell Kistler mounted on a rigid base measures the load. The specimen is sandwiched between the diamond indenter and a load cell. A short striker bar is launched from a gas gun towards the MTend of the incident bar, thus generating a compressive stress pulse followed by a tensile pulse (due to MT) of known duration and amplitude in the incident bar. The MT assembly ensures that only a single compressive pulse reaches the indenter thus causing the indentation and then the tensile pulse retracts the indenter. Furthermore, all the successive wave reflections will be tensile while traveling towards the indenter end thus causing the bar/indenter assembly to retract further away from the specimen. Therefore, single dynamic indentation on the specimen is ensured. The dynamic hardness is calculated based on indentation diagonal size and load similar like the static hardness. 3.2.4 Experimental Static and dynamic indentation studies were conducted on three different grain sizes of B4C ceramic produced by P2C technique as discussed in C hapter 2. Approximately 15 static indentations were performed at each load of 2.94 N (300 gm), 4.9 N (500 gm) and

PAGE 43

43 9.8 N (1000 gm) for 15 seconds on each grain size B4C specimen. Unlike the static indentation tests at fixed loads, the dynamic indentations were conducted utilizing the dynamic indentation tester at loads between 2.94 N to 14.7 N by increasing the velocity of the striker bar in the dynamic indentation tester. Around 2530 tests per specimen type were conducted at load ranges similar to those under static indentation. 3.2.5 Results of Indentation Experiments and Discussion A plot of static and dynamic hardnesses vs. indentation load for the three grain sizes is presented in Fig. 32. A wide scatter in the static hardness values was observed for all the grain sizes. The average static hardness values for 2 m and 2.7 m grain size specimens appeared in the same range (27.45 2 GPa and 27.45 2.5 GPa) for the loads considered in this investigation. On the other hand, the 1.6 m grain size sample, having the smallest grain size, showed relat ively lower average static hardness value of 25.41 1.0 GPa. On the other hand, the dynamic hardness (HVd) was lower than the static hardness for all the grain sizes in this load range. Unlike the static hardness values, the dynamic hardness values exhibited a greater scatter for all the grain sizes as can be seen from Fig. 3 2. The 1.6 m grain size boron carbide showed a significant decrease in dynamic hardness compared to other two grain sizes. This decrease is probably due to the lower density (see Table 2 1) or higher level of porosity in these specimens. For the other two grain sizes, the trends between static and dynamic hardness values were difficult to conclude because of large scatter in the values and lack of sufficient number of data points.71 Cracks extending from the corners of the static and dynamic Vickers indentations were used for fracture toughness ( Kc) measurements. Numerous empirical fracture

PAGE 44

44 toughness relations are available in the literature depending on the nature of crack systems.128 For half penny cracks, the relationship between the half crack length c and load P is expected to be of type c = AP2/3 where as for Palmqvist cracks, the difference between the half crack length c and the half diagonal, a, i.e., l = c a is expected to be linear with respect to load, i.e., l = BP. In the current work, the above quantities were plotted and fitted to an equation of type y = bxn to determine the nature of the crack system. The c vs. P plots yielded n values in the range of 0.620.69 and therefore, matched the equation y = bxn closely. On the other hand, the plots of l vs. P did not result in a linear relationship between l and P as the n values were found to be in the range of 0.680.92. These plots have been presented in Appendix A. Therefore, it was concluded that in the current work static and dynamic indentations resulted in half penny crack system beneath the indentations. Accordingly, Evans and Charles fracture toughness e quation,128 Kc = 0.0824P / c1.5, based on half penny crack system has been used. A comparison of optical micrographs of the top surface of the indented regions of 2.7 m grain size specimen at an indentation load of 2.94 N is shown in Fig. 33. It can be seen that the dynamic indentations resulted in more severe damage compared to the static indentations. Measurements of crack lengths revealed that, in general, the half median crack length ( c) increased linearly with indentation load under both low and high strain rate loads. The average crack lengths were slightly longer under dynamic loads compared to static loads for all the grain sizes. Using the Evans and Charles equation, static and dynamic fracture toughness (s CK and d CK, respectively) values were calculated and plotted in Fig. 34. A wide scatter in the fracture toughness values was observed under both conditions. In general, fracture toughness was lower under dynamic loads compared to static loads.71

PAGE 45

45 It is wellknown that hardness increases with a decrease in grain size (the Hall Petch relationship).129,130 Similarly, a decrease in residual porosity also improves the mechanical properties. But in the current work the 1.6 m grain size boron carbide, having the smallest g rain size, revealed the lowest static hardness which does not follow the well known Hall Petch relationship. No significant variation in hardness was observed between 2 m and 2.7 m grain size boron carbide specimens. The possible reason for such behavior may be the presence of higher level of residual porosity in small grain size boron carbide than that in the other two boron carbide specimens which have more than 99% theoretical density (see Table 21). This increase in porosity may offset the effect of smaller grain size on hardness. Since the larger grain size specimens did not show any variation in static hardness and they contain negligible porosity, it can be surmised that har dness is less influenced by the grain size variation (in the current grain size range) than porosity level at low strain rate loading. Under dynamic loading, hardness was observed to change significantly with grain size variation (Fig. 32 and Table 31). But as mentioned earlier, due to statistically insufficient amount of data, it is difficult to make any definitive conclusion except that the dynamic hardness for 1.6 m grain size was significantly lower than static hardness. Again, the increased level of porosity might have offset the effect of smaller grain size which resulted in the observed drop in dynamic hardness in 1.6 m grain size specimens. Therefore, it can be inferred that hardness of boron carbide is not only influenced by grain size and poros ity level but also by strain rate. A significant lowering of dynamic hardness compared to static hardness in 1.6 m grain size boron carbide indicates that residual porosity could be more detrimental to hardness under dynamic loading than under static loading. Similar conclusion can be made from the

PAGE 46

46 fracture toughness measurements (Fig. 34 and Table 31) where a large drop in fracture toughness was observed for 1.6 m grain size compared to other grain sizes under dynamic loading.71 Clearly, boron carbide revealed lower hardness (Fig. 32) and lower fracture toughness (Fig. 3.4) under dynamic indentation than those under static indentation. But this trend in the loss of mechanical properties of boron carbide at high strain rates contradicts the established trend for many other engineering materials where an increase in hardness, yield strength, fracture toughness and fracture strength have been observed under higher strain rate loading. To investigate the underlying cause for this anomalous behavior in bor on carbide, further studies employing several spectroscopic techniques were conducted in the indented regions beneath the surface which is discussed in section 3 3. For these purposes, subsurface damaged regions of indentations were probed using several sp ectroscopic techniques. To study the subsurface regions of indentations, rectangular specimens were cut into two halves along the length, and these cut surfaces were polished using standard metallographic polishing techniques. They were then bonded with highstrength adhesive and kept clamped for few hours. The top surface containing the bonded interface was polished flat for static and dynamic indentations along the interface. The indentation was performed with one of the diagonals of the indenter aligned parallel to the interface. For these subsurface damage studies, indentations were intentionally performed at slightly higher loads (up to 21 N) so as to cause sufficiently large damage zone beneath the indentation. After indentation, the bonded surfaces were separated and scanning electron microscopy (SEM) was performed to observe the subsurface damage.71

PAGE 47

47 Scanning electron micrographs of the half indents (shown in Fig. 3 5) revealed that the subsurface damages for both types of indentations at similar load levels are significantly different. Under dynamic indentation the extent of damage in the lateral direction was significantly larger than static indentation and several cracks were observed to emanate from the boundary of the damage region as indicated in Fig. 35 (B). Such cracks are not evident under static indentations. These subsurface studies further confirm that boron carbide is more prone to damage under dynamic loading compared to static loading at similar load levels. Moreover, it is clearly seen t hat the damage zones appear to be half penny shaped beneath the indentation as previously assumed for calculation of indentation fracture toughness.71 In the following section, results of spectroscopic studies for structural phase transformation as well as effect of strain rate are discussed in detail. 3.3 Spectroscopic Investigation of Localized Phase Transformation 3.3.1 Localized Phase Transformation The issue of localized amorphization in boron carbide ceramic has already been mentioned which was confi rmed from transmission electron microscopy work.18 It has been suggested that such regions originated due to localized collapse of the crystal lattice as a result of the solidstate phase transformation from crystalline phase to amorphous phase.18,71,131,132 To reveal the exact mechanism behind such localized solidstate phase transformation, it is essential to understand the structure of amorphous B4C. However, the exact structure and the composition of these amorphous regions present within impacted B4C c eramics are still unclear. Ge et al.,131 from their Raman spectroscopic analysis on the nanoindented regions of polycrystalline B4C, have suggested the presence of sp2 hybridized carbon (C) within a B4C. From similar nanoindentation work conducted on single crystal B4C, Yan et al.,132

PAGE 48

48 suggested the presence of sp2 C aromatic rings within a B4C. They also suggested, from the temperature coefficient measurements of Raman bands, that boron (B) atoms can form B clusters or can substitute C atom from the aromat ic rings. However, no direct evidence of B cluster formation was provided. Using Gibbs freeenergy calculations, Fanchini et al.,133 interpreted that amorphization in B4C may trigger from the collapse of B12(CCC) polytype instead of B11C(CBC), and a B4C ma y contain some segregated mixture of amorphous carbon ( a C) and boron icosahedrons (B12). But any experimental evidence of the formation of a C and B12 within amorphous bands is still lacking. Also, B4C is widely accepted to have B11C(CBC) structures inste ad of B12(CCC).134,135 Therefore, the exact formation mechanism of B and C clusters within a B4C is still not well understood. Spectroscopic techniques such as Raman spectroscopy, photoluminescence (PL) and Fourier transformed infrared (FTIR) spectroscopy are widely used to provide information and identification of chemical structures and physical forms.136,137 These methods are sensitive to the molecular vibrations and the electronic transitions. Since, every material has its unique vibrational and electr onic transition characteristics (owing to chemical bonding features), spectroscopic techniques are widely used to identify the unknown structures. Although, Raman spectroscopy has provided some useful information on the phase transformation in B4C and the structure of a B4C,71,131,132 only use of this technique may not be sufficient to unearth all the structural information. Therefore, the aim of the current work is to provide more insight in to the structure of amorphous regions formed within B4C ceramics. Here, spectroscopic evidence for the formation of a C, created by static and dynamic indentations, has been provided along with the observed evidence of aB cluster formation by dynamic indentation. This interpretation was rationalized based on a series of micro spectroscopic techniques including Raman

PAGE 49

49 spectroscopy (both uv and visible), photoluminescence, and Fourier transformed infrared spectroscopy. Before proceeding further with the experimental results and their interpretation, in the following, some of the fundamentals of vibrational and electronic spectroscopy are discussed briefly. 3.3.2 Fundamentals of Spectroscopic Techniques 3.3.2.1 Molecular vibration Raman scattering and infrared absorption are based on the fundamental normal vibrational modes of chemical bonds of molecules, in any form (i.e., solid, liquid or gas).136 There are only certain types of atomic displacements (i.e., nuclear motion) during molecular vibrations which are allowed with well defined frequencies. These are known as the normal modes of vibrations and some of the types of motions which contribute to the formation of a normal mode are given below, also see Fig. 36,136 stretching motion between two bonded atoms (symmetric and asymmetric) bending motion between three atoms co nnected by two bonds out of plane deformation modes that change an otherwise planar structure into a nonplanar one From a simple statement of Hookes law, frequency ( ) of a simple fundamental normal mode of vibration of a molecu lar can be expressed as where c is the velocity of light, k is the force constant and ( ) is the reduced mass expressed as 12 12mm mm ( m1 and m2 are the masses of the atoms constituting the chemical bond). 1 2 k c

PAGE 50

50 3.3.2.2 Raman and infrared spectroscopy Although both Raman and FTIR phenomena are fundamentally related to the vibrations of materials, the former is associated with the inelastic scattering event where as the latter is related to the absorption of light.136,137 During interaction of light with a matter, the photons present within the light can either be absorbed or scattered, or the interaction can be negligible. In vibrational spectroscopy, t he changes in the energies of the incident photons, caused by the nuclear motion, are detected. If the scattering process involves only a distortion of the electron cloud, then the changes in the frequency of the scattered electrons are negligible. This is considered as elastic scattering and called Rayleigh scattering. In contrast, if the scattering process induces nuclear motion, then energy will either be transferred from photon to molecule or vice versa. This is called Raman scattering which is consider ed as inelastic scattering because here the energies (or frequencies) of the incident and scattered photons are different. In any material, there are several electronic states present where each of the electronic states consists of a number of vibrational states. At room temperature, most of the molecules, in general, are present in the lowest energy vibrational level. In Raman spectroscopy, a material is irradiated with a light of single frequency and the resulting radiation due to the scattering process is detected. Figure 37 shows a simplified schematic of the Raman scattering phenomenon. As the material is irradiated with a source of radiation, electrons of the molecules, present in the ground electronic state, interact with the photons of the light so urce. During this event, Rayleigh process, see Fig. 3 7, will dominate since most of the photons scatter this way. However, a fraction of the molecules, initially present at some low energy vibrational energy level (e.g., vo) will not return to the same level, rather, will come back to a higher energy vibrational level (e.g.,

PAGE 51

51 v2). As a result, the scattered photons will have lower energies compared to incident photons. This is called Stokes Raman scattering. In contrast, due to thermal energy at room temper ature, some molecules initially may be present at higher energy vibrational level (e.g., v2) which will return to the lower energy level (e.g., vo) by releasing the extra energy. Therefore, the scattered photons will have higher energy compared to the incident photons. This is called anti Stokes Raman scattering. However, at room temperature, number of the molecules present at an excited vibrational state instead of really low energy ones will be extremely small. Therefore, compared to Stokes scattering, anti Stokes scattering is weak and becomes further weaker as the frequency of vibration increases owing to the decreased population of the excited vibrational states.136 In contrast to Raman scattering, in infrared spectroscopy, a material is subj ected to an infrared light source, consisting of a range of frequencies, and the absorbed frequencies from the light source are detected.136,137 Here at any frequency, within the infrared range, the absorption of light corresponds to a promotion of a molec ule to an excited vibrational state. For example, Fig. 38 illustrates a simplified schematic of the absorption mechanism encountered in the infrared spectroscopy. In any molecule, there are several types of normal modes of vibrations present, however, not all the vibrations are Raman or infrared active. As a result, there are several materials which are either Raman or infrared inactive or both.136,137 Therefore, there are some selection rules which govern the ability of a normal mode to be detected by Raman or infrared spectroscopy. In any molecule, there could be some net dipole moment present as a result of polarized bonds or interaction with an electromagnetic radiation can create dipole moments by induc ing polarization of the bonds. For example, Fig. 39 illustrates some simple schematics of chemical bonds where there is no net dipole moment present

PAGE 52

52 at equilibrium position. However, when exposed to electromagnetic radiation, net dipole moments are induce d depending on the type of bond stretching. During the symmetric stretching, the net dipole moment still remains zero, however, the asymmetric stretching induces a net dipole moment within the molecule. A vibration to be Raman active, polarizability of the molecule must change with vibrational motion. Similarly, a vibration to be IR active, there should be a net change in the dipole moment of the molecule during vibrations of bonds. 3.3.2.3 Photoluminescence Similar to vibrational characteristics, materials can also be identified by their electronic structure i.e., the so called energy gap or band gap. All the electrons from their constituent atoms in a material are not bound with the same energy and thus, several electronic states are available for the electrons leading to a band structure.138 Now, at room temperature, some of the electronic states are occupied (called bonding electronic states or molecular orbitals) where as rest of them (anti bonding electronic states or molecular orbitals) remains empty. Energy gap between the highest occupied electronic state (or valence band) and the lowest unoccupied electronic state (conduction band) is called band gap which can be utilized for characterization of materials except metals and some other materials (e.g, graphite) which do not have any band gap due to overlap of valence and conduction bands. In a simplified description, let us assume that the ground electronic level as the valence band where as the next higher electronic energy level as the conduction band, see Fig. 310. When a material is optically excited, photon(s) having higher energy than the band gap energy can be absorbed which will raise electron(s) from the valence band to the conduction band across the forbidden energy gap. Electrons wi ll initially relax some of the

PAGE 53

53 excess energy by nonradiative decay process and will come to rest at the lowest energy in the conduction band. Eventually, electrons will fall back down to the valence band and the energy difference (or band gap energy) will be emitted from the material as luminescent photons, see Fig. 310. Therefore, energy of the emitted photons is a direct measure of the band gap energy and this process of photon excitation followed by photon emission is called photoluminescence. From the above discussion, it is clear that these spectroscopic techniques are not only useful for characterization of known materials but also they can identify new and unknown phase(s). Apart from the advantage of high spatial resolution, these methods essential ly detect the vibrational and the electronic characteristics of any phase in any form, and therefore, are not only restricted to crystalline structure of materials but also applicable to amorphous forms. It has been well realized that the phase transformat ion in B4C is a result of localized collapse of crystal structure to a disordered state.18 Also, it has been shown that the small collapsed regions do not have any crystallinity and therefore, they are referred as amorphous B4C. However, previous Raman spectroscopic work71,131,132 has detected newer peaks which are not associated with crystalline B4C. Therefore, it has been suggested that the a B4C is not necessarily a disordered state of B4C, rather, its a mixture of newer phase(s) resulted from structural destruction. Therefore, in the current work, apart from visible Raman spectroscopy, uv Raman, PL and FTIR spectroscopic techniques were also utilized which are ideal for detection of newer phases based on the chemical bonding and electronic structure characteristics. It was expected that these studies will shed more light on the disordered state of B4C compared to only visible Raman spectroscopy.

PAGE 54

54 3.3.3 Experimental In the current work, damaged regions beneath the static and dynamic indentations on a polycrystalline B4C, as discussed in section 3.2, were probed using (i) v isible Raman (He Ne laser, 632.8 nm, 1.96 eV), (ii) uv Raman (HeCd laser, 325 nm, 3.80 eV), (iii) photoluminescence (HeCd laser, 325 nm, 3.80 eV) and (iv) FTIR spectroscopy in a re flection mode under a microscope with a focusing diameter of ~10 m, close to the diffraction limit of infrared wavelengths. 3.3.4 Results of Raman Spectroscopy Figures 311 (A) and 3 11 (B) show the visible Raman spectra collected from the unindented pol ished surface and damaged regions, created by static and dynamic indentations, of polycrystalline B4C Raman spectrum from the unindented polished surface in the neighborhood of the indents is consistent with that reported in the literature for single cr ystal and polycrystalline boron carbide ceramics.132,134,135 The various peaks in the spectrum can be related to the crystal structure of boron carbide as follows. Boron carbide has a complex rhombohedral crystal structure containing eight icosahedrons and one linear chain of three atoms, see Fig. 312. Each icosahedron (B11C) consists of 11 Boron (B) atoms and one Carbon (C) atom residing in one of the polar sites.71,132,134,135 The linear chain consists of CBC atoms. The icosahedrons are located at the co rners of the unit cell and one of the longest diagonals along the <111> direction contains the linear chain. O rigin of t he two broad peaks in the lower frequency range (at 275 cm1 and 325 cm1) is not well understood in the field although they often appear in B4C ceramics.134 The appearance of the next two narrow peaks (at 478 cm1 and 532 cm1) have been assigned to the rotation of the CBC chain about an axis perpendicular to the [111] direction and the liberational mode of B11C icosahedron, respectively. Broad peaks in the high frequency

PAGE 55

55 range (between 6001200 cm1) are associated with the B11C icosahedrons. Apart from these characteristic peaks, a small peak near 1580 cm1 was observed in the Raman spectrum of the undamaged surface. This peak is attributed to the so called G peak (graphite peak),139,140 will be described in the following paragraph, due to the presence of free carbon in the boron carbide specimens. Similar spectra were detected from the areas damaged by static indentation as shown in the middle curve in Fig. 311 (A) with a new disorder induced (D) peak (~1335 cm1) as shown in the magnified view of these spectra in Fig. 311 (B). The G and D peaks represent zone center phonons of E2g symmetry and K point phonons of A1g symmetry of graphitic C, respectively.139,140 While the G peak is associated with inplane bondstretching motion of pairs of sp2 bonded C atoms present either in olefinic chains or in aromat ic rings, the D peak corresponds to the breathing mode only from the aromatic rings, see Fig. 313. Significant changes are detected from the indented areas formed by dynamic indentation (top curve of Fig 311 (A)). Intense G and D peaks were detected at ~1590 cm1 and at ~1335 cm1, respectively, as shown in the top spectrum in Fig. 311 (A). The evolution of both D and G peaks in Raman spectra are due to the creation of free C rich phase and thus is well accepted as the indication of localized amorphizat ion in B4C.71,131,132 The evolution of D peak in the Raman spectra for both type of indentations clearly suggests the formation of sp2 hybridized aromatic C rings within the indented areas. Also, the stronger G and D peaks induced by dynamic indentation su ggests a higher level of structural damage or amorphization of the polycrystalline B4C as compared to static indentation.2 This interpretation is also associated with the lower Raman shift of the characteristic breathing modes of the icosahedral B11C struc tures132,134,135 of crystalline

PAGE 56

56 B4C at ~1082 cm1 (for dynamic indentation) and at ~1089 cm1 (for static indentation) due to the strain induced by the amorphization. More structural information can be obtained from the uv Raman spectra in Fig. 311 (C). Similar features like visible Raman spectra were observed in the uv Raman spectra. It is interesting to note that the two low frequency peaks of visible Raman spectra in the range from 250350 cm1 did not appear in the uv Raman spectra. Our results suggest for the need of further investigation of B4C using uv Raman spectroscopy. In addition, the relative intensity of the G peak (IG) is higher than the intensity of the D peak (ID) for both type of i ndentations. An opposite trend was observed in the visible Raman spectra as shown in Fig. 311. The ID/IG ratio is ~0.6 for uv Raman and is ~1.1 for visible Raman. The ID/IG values and deviation of ID/IG observed here (i.e., dispersion) are similar to that detected from hydrogenated and nonhydrogenated aC films with sp2 bonded C clusters.141,142 Also, Dpeak disappears under uv excitation for disordered and nanocrystalline graphite but not for aC.142 Apart from these, the diamondlike carbon films with predominated sp3 bonded C usually have low ID/IG values (< 0.3) and low ID/IG dispersion (< 0.3). This also suggests that a low sp3 hybridized C content in the aC formed within the indented regions. Therefore, Raman spectroscopy confirmed the formation of sp2 bonded aromatic carbon clusters by static and dynamic indentations due to amorphization of B4C. Similar to visible Raman spectroscopy, the results of uv Raman spectroscopy also indicate that dynamic loading results in a greater level of C cluster formation. In addition, dispersion of G peak to higher frequency was observed in the visible (Fig. 311 (A)) and uv Raman spectra (Fig. 311 (C)) collected from the dynamically indented region compared to the static indented region. It was concluded that t he greater level of C cluster formation

PAGE 57

57 under dynamic indentation compared to static indentation causes the observed G peak dispersion.142,143 3.3.5 Results of Photoluminescence Spectroscopy PL experiments were performed to further analyze the structures w ithin the aB4C. Figure 314 presents the PL spectra of B4C obtained from the unindented surface and the indented regions of static and dynamic indentations. A weak PL band at 2.4 eV was detected from the unindented surface (bottom curve and inset). This PL band is stronger for indented regions created by static indentation (middle curve) and strongest for areas damaged by dynamic indentation (top curve). A shoulder peak at ~2.05 eV was observed for the unindented surface (shown clearly in the inset) and indented regions. Apart from these peaks, a new peak at ~3.05 eV was noted in the PL spectra from both type of indented regions. Similar to the 2.4 eV peak, this peak was also observed to be stronger under dynamic indentation compared to static indentatio n. PL band at ~2.05 eV is associated with the optical band gap of B4C.144 But the other two PL signals (at ~2.4 eV and ~3.05 eV) are well above the optical band gap of B4C and therefore, are not associated with the crystalline B4C structures. These PL bands are explained as the following. Amorphous carbon (aC) and hydrogenated aC (a C:H) films have been observed to exhibit a broad PL band from ~1.8 eV to 3.7 eV.145 These aC films contain sp2 bonded carbon clusters embedded in the aC matrix. Their band gaps depend on the size and distribution of the sp2 clusters. Since Raman spectroscopy suggested that the indented areas formed by static and dynamic indentations contain aC structures, we rationalize that the PL spectra detected here are attributed to radiative recombination mechanism of photoexcited el ectrons and holes in localized tail states within sp2 clusters.145 Similar to the Raman spectra, a stronger PL peak for dynamic indentation

PAGE 58

58 compared to static indentation again indicates the presence of more aromatic sp2 C clusters within dynamically indented regions. 3.3.6 Results of Infrared Spectroscopy Discussion so far has suggested the formation of sp2 hybridized aromatic C clusters in the indented areas formed by static and dynamic indentations. However, these results did not provide any information about the existence of B rich phase within the indented regions. Therefore, to further clarify the a B4C structure, FTIR spectra were collected from the unindented surface and damaged regions of static and dynamic indentations. Spectra collected from al l un indented and indented areas are merely similar (except one to be discussed later), see Fig. 315. These spectra exhibited FTIR peaks characteristic of crystalline B4C. The FTIR peak at ~1100 cm1 corresponds to the B C stretching vibration.146 As disc ussed before, B4C crystal contains linear CBC chain along one of the body diagonals of the rhombohedral unit cell, see Fig. 312.71,132,134,135 The end C atoms of the chain have sp3like hybridization where as B exists presumably with an sp2like hybridiza tion.135 The stretching of these CBC chains144 is responsible for the peak around 1590 cm1. It could be observed from Fig. 315 that the intensities of these peaks gradually decrease in succession from the unindented surface to static indentation to dynamic indentation. This trend clearly indicates structural distortion in B4C during indentation as also suggested by Raman spectroscopy. Lower intensity of FTIR peaks from static indentation to dynamic indentation suggests a greater amorphization or structur al disorder The only difference occurred in the FTIR spectra corresponding to the dynamically indented region which exhibited a new IR absorption band between ~760830 cm1 (shown clearly in the inset). This peak covers the known IR signals for free icosa hedral B12 molecules.147,148 Due to the broad spectra distribution, we think that it is more appropriate

PAGE 59

59 to assign this IR band to amorphous B clusters, in consistent with a reported data.148 Therefore, FTIR clearly suggests the formation of aB clusters w ithin the aB4C created by dynamic indentation. However, no such evidence was observed in the FTIR spectrum collected from the damaged region of static indentation. The localized solidstate amorphization in B4C is a shear driven phenomena.18,131 In balli stic impact experiments of B4C, amorphization was observed only above certain impact velocities.18 On the other hand, similar phenomena was observed for nanoindentation,131 static indentation and dynamic indentation.71 In indentation experiments, the sharp edges of the indenter develop large shear stresses compared to impact experiments, and thus such stress state promotes amorphization. In addition to high shear stresses, the high strain rate during dynamic indentation further facilitates the amorphization compared to low strain rate static indentation as revealed by the above spectroscopic results. High strain rateinduced amorphization has also been predicted from MD simulations for metallic nanowires.149 Finally, we present a schematic of the formation of a C and a B clusters. Amorphization of each B4C unit cell (see Fig. 316 (A)), with eight B11C icosahedrons at the corners, will cause the collapse of this structure into two carbon atoms, one boron atom, and a B11C icosahedron per unit cell (Fig. 316 (B)). The possible mechanism could have been accompanied by the collapse of the B11C icosahedron, which is energetically more stable to release the carbon atom and replace with a boron atom. This process will reorganize into one amorphous B12 cluster, and one carbon atom that will form carbon clusters with adjacent carbon atoms (Fig. 316 (C)).

PAGE 60

60 3.4 Conclusions Finegrained boron carbide ceramics were subjected to static and dynamic indentations to study the influence of strain rate on hardness and fracture toughness. The boron carbide ceramics showed a consistent decrease in hardness and fracture toughness as well as a greater extent of damage under dynamic indentations than under static indentations. Presence of residual porosity was observed to lower dynamic hardness significantly compared to static hardness. Similar to indentation studies on intact specimens, subsurface studies employing split specimens also revealed greater extent of damage under dynamic indentation compared to static indentation. Spectroscopic studies revealed greater structural disorder or phase transformation under dynamic loading compared to static loading. In depth analysis of from Raman and PL spectroscopy results revealed that aC clusters were formed within the indented regions Also, FTIR analysis reveals the existence of aB clusters within the dynamically indented region.

PAGE 61

61 Figure 31. Schematic of experimental setup for dynamic indentation hardness measurements. Striker bar Flange Rigid mass Sleeve Digital Oscilloscope Charge Amplifier Indenter Load cell Specimen Momentum Trap Incident Bar

PAGE 62

62 Figure 32. Comparison of static and dynamic hardness values for three grain sizes of B4C. 2.2 2.7 3.2 3.7 4.2 2 4 6 8 10 12 14 16 Load (N) K c (MPa.m0.5) 1.6 m (Static) 1.6 m (Dynamic) 2 m (Static) 2 m (Dynamic) 2.7 m (Static) 2.7 m (Dynamic)

PAGE 63

63 Figure 33. Optical micrographs of A) static and B) dynamic indents for 2.7 m grain size B4C at 300 gm. Note longer cracks and more severe damage in dynamic indent.

PAGE 64

64 Figure 34. Comparison of static and dynamic indentation fracture toughness for three different grain sizes of boron carbide. 2.2 2.7 3.2 3.7 4.2 2 4 6 8 10 12 14 16 Load (N) K c (MPa.m0.5) 1.6 m (Static) 1.6 m (Dynamic) 2 m (Static) 2 m (Dynamic) 2.7 m (Static) 2.7 m (Dynamic)

PAGE 65

65 Figure 35. Subsurface damaged region beneath A) static indentation and B) dynamic indentation at a load of 19.6 N for 1.6 m grain size boron carbide.

PAGE 66

66 Figure 36. Different normal of mode of vibrations. Symmetric stretching vibrations Antisymmetric stretching vibrations Symmetric bending vibrations

PAGE 67

67 Figure 37. A simple schematic illustrating Rayleigh, Stokes and anti Stokes Raman scattering phenomena.

PAGE 68

68 Figure 38. A schematic of the infrared absorption spectroscopy.

PAGE 69

69 Figure 39. Changes in the net dipole moment of a molecule during asymmetric stretching. Equilibrium position Symmetric stretching Symmetric stretching antisymmetric stretching antisymmetric stretching

PAGE 70

70 Figure 310. A schematic illustration of photoluminescence phenomenon. Excitation process E1 = 3 E o Ground electronic state (valence band) 1 2 3 4 o Vibrational energy states E 1 1 2 o Vibrational energy states Vibrational relaxation (nonradiative decay) E2 = Photoluminescenc Excited electronic (conduction band)

PAGE 71

71 Figure 311. A) visible Raman spectra and C) uv Raman spectra from the unindented surface and damaged regions beneath the static and dynamic indentations. B) Magnified view of two of the spectra in A. A visible Raman v isible R uv Raman B C

PAGE 72

72 Figure 312. Crystal structure of boron carbide (B4C). Rhombohedral unit cell B C

PAGE 73

73 Figure 313. Origin of D peak and G peak. Breading vibration (D peak) C C C C C C Sixfold aromatic C ring C C In plane stretching vibration (G peak) Olefinic C C C sp 2 sp 2 C

PAGE 74

74 Figure 314. Photoluminescence from the unindented surfaces and damaged regions beneath the static and dynamic indentations.

PAGE 75

75 Figure 315. FTIR spectra from the unindented surfaces and damaged regions beneath the static and dynamic indentations.

PAGE 76

76 Figure 316. Schematic of A) a rhombohedral unit cell of B4C with eight B11C icosahedrons at the corners. B) Amorphization of each unit cell will form two carbon and one boron atoms and a B11C icosahedron. C) One B11C icosahedron and one boron atom will reorganize into one amorphous boron ( a B) cluster, and a carbon atom that form carbon clusters with adjacent carbon. A B C B C B 11 C Icosahedron = a B cluster

PAGE 77

77 Table 31. Comparison of average dynamic and static hardness and fracture toughness values Grain size ( m) HV s (GPa) HV d (GPa) % Change (HVsHVd) HVs s CK (MPa.m0.5) d CK (MPa.m0.5) % Change ( s CK d CK ) s CK 1.6 25.41 1.00 18.17 2.00 28.49 3.39 0.25 2.69 0.27 20.65 2.0 27.45 2.00 26.15 2.50 4.74 3.61 0.30 3.00 0.29 16.89 2.7 27.45 2.50 23.37 3.00 14.86 3.22 0.16 2.81 0.36 12.73

PAGE 78

78 CHAPTER 4 SCRATCH INDUCED DEFORMATION IN Z r B2SIC COMPOSITE 4.1 Introduction As mentioned in Chapter 1, during service, aerospace structural components such as heat shields, nose cones, leading edges, engine cowl inlets, etc., are often subjected to abrasiondominated wear scenarios (abrasive wear) due to the impact with atmospheric debris particles. As ultra hightemperature ceramics (UHTCs), in recent times, have been identified as the potential candidates for the aforementioned aerospace applications, it is, therefore, important to investigate their fundamental inelastic deformation mechanisms. Among the available ultra high temperature ceramics, zirconium diboridesilicon carbide (ZrB2SiC) composites have received significant attentions owing to their improved oxidation behavior compared to ZrB2 ceramics. So far, most of the recent studies have been mainly focused on the processing and the oxidation behavior of the ZrB2SiC composites.3748 However, investigations on the evaluation of wear characteristics of UHTCs are limited. Similarly, studies on the fundamental deformation and fracture characteristics in polycrystalline ZrB2SiC composites and other UHTCs are also not available. Indentation and scratch experiments have been effectively used to model deformation and damage mech anisms that evolve in ceramics during an abrasion process. While the scratch experiments involve both normal and tangential point loads, indentation experiments only apply a normal point load on the surface. Therefore, in this dissertation, scratch experim ents were preferred for investigation of complex contact interactions and were used for wear

PAGE 79

79 characterization of a ZrB25wt%SiC composite. Two types of experiments were conducted in the current investigation: (i) constant low load and low velocity and (ii) variable highload and highvelocity scratches. The current dissertation mainly emphasizes on the results of low velocity scratch experiments. The highvelocity scratch results have been discussed briefly. During scratch process, materials beneath the s cratch tool undergo elastic plastic deformation. Directly beneath the indenter tool, a plastically deformed groove is formed which remains surrounded by the elastically deformed region. From experimental and theoretical studies, it has been shown that due to the strain incompatibility between these two regions localized residual stresses are induced within the materials (will be discussed in more detail in section 4.3).150 152 During the unloading phase, the accumulated deformationinduced residual stress causes lateral cracking. Initiation of such damage and consequent material removal from the surface due to these contact processes may result in lower mechanical reliability of the components made out of UHTCs. Therefore, quantification of mechanical residual stress due to scratch or abrasion phenomenon is necessary for a fundamental understanding of the material removal process and for design of abrasionresistant materials. However, no such studies in the ZrB2SiC composites, so far, are available in open literature. In the current investigation, microRaman spectroscopic technique has been utilized for the measurement of scratch induced residual stress in the ZrB2SiC composite.

PAGE 80

80 Rest of the C hapter is divided in to two parts. The first part, section 4.2, is mainly focused on the low load and low velocity scratch studies along with some results of highload and highvelocity scratch investigations. This section describes the microstructural features as a result of the scratch process which rationalized in the context of elastic stress field resulting from application of combined normal and tangential point loads. In section 4.3, results from the scratch induced residual stress measurements, employing micro Raman spectroscopy, are discussed. This section describes the Raman spectroscopic measurements which are explained on the basis of sliding blister field model for residual elastic stress field. Since, zirconium diboride ceramic is extremely week in Raman scattering (will be discussed latter in section 4.3), in the current investigation, the scratch induced residual stress measurements were limited to the particulate phase (SiC) of the composite. The following discussion will start with the scratch studies in the ZrB25wt%SiC composite. 4.2 Scratch Studies 4.2.1 Experimental 4.2.1.1 Constant low load and low velocity scratches The scratch studies were conducted in a ZrB25wt%SiC composite, processed via (P2C) technique as discussed in Chapter 2. For scratch experiments, specimens of rectangular dimensions 6 mm3mm4mm were cut from the rectangular slab of the processed composite using a low speed diamond saw. One of the 6 mm 3mm surfaces was first ground succes sively with 120, 240, 320, 400 and 600 grit silicon carbide for around 10 min each and then polished for 15 min using 6 m diamond paste. A MTS nanoindenter XPS

PAGE 81

81 system was utilized to impart scratches on the polished surfaces at constant loads employing a Berkovich nanoindenter (tip radius ~ 100 nm) at a translational speed of 5 m/s. Scratches of length 200 m were conducted at five different constant loads of 50, 100, 150, 200 and 250 mN.72 4.2.1.2 Variable highload and highvelocity scratches The high velocity (~500 mm/s) scratch experiments were conducted in a custom built instrumented scratch tester.153158 A schematic of the scratch tester has been shown in Fig. 41. The specimen was held on top of a high frequency (200 kHz) load cell which was housed in a rigid steel block. This load cell measures the normal force (thrust) during the scratch process. The indenter tool (a diamond conical indenter with 90o included angle) was held in a pendulum holder that was supported by two highprecision ball bea ring bushings housed in a rigid steel frame. A pneumatically driven piston was used to swing the pendulum and cause the diamond tip to transverse a path of circular arc, and create a scratch of variable depth on the specimen surface. Total scratch length a nd maximum depthof cut of each scratch can be varied by changing the length of the tool in the pendulum holder. The residual scratch tracks from both types of experiments were then investigated using a scanning electron microscope ( using JEOL JSM 6400 and JEOL JSM 6335F) to reveal the resulting scratch induced deformation and fracture patterns in the composite. 4.2.2 Results 4.2.2.1 Microstructural features of constant load low velocity scratches Figure 4.2 presents the typical constant low load scratch patterns conducted at five different loads on the polished surfaces of the ZrB25wt%SiC

PAGE 82

82 composite (here after referred to as ZrB2SiC composite).72 Traces of plastically deformed grooves with intact edges along the length were observed at all load levels. T he width of the scratches increased with load. Scratches at 50 and 100 mN, see Figs. 42 (A) and 42 (B), were smooth and without any significant macroscopic damage. But as the load increased beyond 100 mN, damage (to be discussed in more detail later) was visible within the scratch grooves, see Figs. 4 2 (C) 4 2 (E). In Figs. 4 3 and 44, the residual scratch path and the corresponding scratch length vs penetration depth profiles are shown for loads of 50 mN and 250 mN, respectively. Magnified view of the small damaged regions along the scratch groove at 50 mN is also shown in Fig. 43. The fluctuations in the depth profiles, see Figs. 43 (A) and 44 (B), correspond to the regions where either grain pullout has occurred earlier during polishing or severe m icroscopic damage occurred ahead of the indenter tip during the scratch process. Clearly, greater fluctuations in the scratch length vs penetrationdepth profile for 250 mN correspond to greater extent of damage and material removal compared to that at 50 mN, as is evident from the comparison of Figs. 4 3 (A) and 44 (B). The increase in the penetration depth as the scratch load increased from 50 to 250 mN can be seen clearly from Fig. 45. In the following, the deformation mechanisms and the nature of damage that occurred along the scratch grooves at different load levels will be analyzed in more detail. SEM micrographs in Fig. 46 show the deformation features developed within a scratch groove at a load of 250 mN. Figure 46 (A) reveals closely spaced parallel lines and microcracks within the ZrB2 matrix and Fig. 46 (B)

PAGE 83

83 shows a magnified view of region (A) revealing the above deformation features. Cracks emanating from both sides of a SiC particle are seen in Fig. 46 (B). The closely spaced parallel lines, indicative of plastic deformation during the scratch process, were also observed within the scratch grooves at other load levels. These plastic deformation line features are akin to slip line patterns observed in ZrB2 single crystal.159 On the oth er hand, microcracks oriented almost normal to the scratch direction, were mainly observed at scratch loads greater than 100 mN. In Figs. 4 6 (A) and 46 (B), microcracks and slip line patterns were concentrated only on one side of the scratch either because of the asymmetrical orientation of the Berkovich indenter or due to slight inclination of the specimen with respect to the scratch plane. The material that is in contact with the sharpedge of the indenter experiences more severe deformation compared to the material that is in contact with the flat surface (on the opposite side) of the indenter. Accordingly an asymmetry in the deformation pattern was observed on two sides of the scratch. Note that the closely spaced parallel lines (slip lines) within the scratch groove in Fig. 46 (B) were oriented at an angle to the scratch direction and occurred only in the ZrB2 matrix phase. It is interesting to note that the slip line spacing and pattern were unaltered across any microcrack or across the cracks emanat ing from the SiC particle (shown in the white elliptical regions in Fig. 46 (B)). The cracks seem to slightly disturb only the continuity of the slip line patterns. Therefore, it is inferred that the slip lines must have formed before the microcracking during the scratch process. Figure 46 (C) reveals many sets of intersecting slip lines oriented randomly to the scratch direction. Slip lines were

PAGE 84

84 present even well outside the scratch groove. Orthogonal microcracks were also seen at many other locations wi thin the scratch groove where slip lines were absent, see Fig. 46 (C). The mechanistic rationale for the above sequence of formation of these slip lines and microcracks will be explored in the analytical modeling section. Figure 47 (A) shows an image of one of the damaged regions along the scratch groove at a load of 250 mN. Several damage features such as ZrB2 grainboundary fracture, microcracks within the ZrB2 grains and small scratch debris can be observed. Microcracks within a ZrB2 grain and the smal l scratch debris (~ 1 m) clearly indicate transgranular fracture within the ZrB2 matrix. In Fig. 47 (B), microcracks originating from the ZrB2SiC interface as well as microcracks almost orthogonal to the scratch direction are seen. Figure 48 shows anot her region along the scratch groove at 250 mN where no slip bands were present but extensive microcracking normal to the scratch direction occurred within a single ZrB2 grain. It is speculated that the origin of these microcracks is different from those se en in Figs. 46 and 47. Clearly, the material removal mechanisms within the scratch groove of the ZrB2SiC composite were a combination of transgranular and grainboundary modes of fracture. 4.2.2.2 Microstructural features of variable load highvelocit y scratches Figures 49 (A) (C) reveal a high velocity scratch path, extensive transgranular microcracks orthogonal to the scratch direction and the corresponding normal force vs. scratch length profile.160 At the beginning of the

PAGE 85

85 scratch path, the load is small which resulted in a smooth plastically deformed groove without any significant brittle damage. As the load increased, severe macroscopic brittle fracture features such as extended lateral cracking and material removal occurred in the middle of the scratch groove where the load was maximum. A closer view of the scratch groove revealed orthogonal transgranular microcracking and flattened debris, see Fig. 49 (B). Note that in the previous constant low load (50250 mN) and low velocity (5 m/s) scratch experiments were resulted in the formation of smooth plastically deformed grooves with occasional transgranular microcracking. In the variable highvelocity and highload (8 N, maximum load) scratch experiments, the intensity of microcracking was s ignificantly higher compared to previous scratch studies. Microstructural analysis of the regions containing lateral cracks also revealed the predominance of transgranular fracture in the composite. Similar to the low velocity scratch experiments, microplasticity in terms of slip line formation was also observed in these high velocity scratch grooves within the ZrB2 phase, see Fig 410. However, a majority of the regions containing slip lines were probably removed due to extensive lateral cracking and the associated material removal. Thus, these studies revealed that during high velocity scratch process, extensive transgranular microcracking and severe lateral cracking caused significant damage and material removal in the composite. The force profile (Fig. 49 (C)) reveals that the load is minimum at the beginning and reaches maximum at the center of the scratches, and then again falls down to zero towards the exit end. Similarly, the scratch width is also minimum at both ends and maximum in the

PAGE 86

86 middle. The f luctuations in the force vs. distance profile are attributed to the brittle cracking and the associated material removal during the scratch process. It is interesting to note that plastic deformation features (sliplines) were observed in a typical ceramic matrix composite. ZrB2 has a hexagonal closed pack (hcp) crystal structure.2,159 In hcp structures, many slip systems have been observed to operate during plastic deformation.35,159,161 165 Slip occurs on basal ({0001}), prismatic ( {1010} ) and pyramidal ( {1011} ) planes. ZrB2 single crystals have been reported to undergo plastic deformation by prismatic slip during room temperature microindentation and basal slip during uniaxial high temperature experiments.159 The observed line patterns within the ZrB2SiC composite in this study were similar to the residual slipline patterns observed in the above studies on ZrB2 single crystals159 as well as in materials with hexagonal symmetry.165 Therefore, it is inferred that the observed deformation patterns in terms of closely spaced parallel lines within the scratch grooves of ZrB2SiC composite, presented in Fig. 46 and Fig. 47, resulted from activation of slip mechanism during the scratch experiments. But the identification of particular slip systems that were activated during the scratch process in ZrB2 requires more indepth investigation using transmission electron microscopy (TEM) and was not pursued here. Note that these sliplines were present only within the ZrB2 mat rix but did not extend into SiC particles. In general, it has been found that deformation induced defects in SiC particles consisted of stacking faults166 which can only be discerned using TEM and other techniques. To our knowledge, the formation of slip l ines due to dislocation motion in SiC has not been reported.

PAGE 87

87 4.2.3 Influence of Elastic Stress Field on the Scratchinduced Deformation During a scratch event, the material beneath an indenter undergoes elastic plastic deformation.150,151 In the current investigation, it was postulated that the observed transgranular microcracking orthogonal to the scratch direction within the ZrB2 phase might have resulted from the tensile stress field. On the other hand, the shear stress component could cause the obser ved plastic deformation. In the following, we will utilize the elastic stress field solutions to rationalize the evolution sequence of sliplines and microcracking in ZrB2SiC composite. The singlepass scratch process has been well studied in the literat ure.150158 It has been modeled as a sliding microindentation event where the point loads are applied simultaneously in normal ( Fn) and tangential ( Ft) directions on the surface of a specimen. Ahn et al .,150 modeled the scratch process by superposition of Boussinesq field (due to Fn), Cerruti field (due to Ft)167 and blister field solutions168 to rationalize various crack systems that evolve during the scratch event. Recently, Jing et al .,151 utilized a wedge indentation model and estimated the plastic zone size as well as the damage zone size during a scratch process. Subhash et al .,154 studied the influence of Boussinesq and Cerruti stress fields in the presence of a preexisting microcrack and identified the most favorable orientation of cracks that activate during a scratch process. However, in all the above studies, the focus was always on crack activation in the wake of the scratch tip. The evolution of plastic deformation during the scratch process was n ever investigated. In our study, we have noticed both microcracks and slip

PAGE 88

88 bands in the scratch groove. Therefore, we will focus on analyzing the stress components responsible for these inelastic features. A schematic of the scratch process and the relev ant stress fields are shown in Fig. 411. As the indenter moves forward on the specimen surface, it is assumed that a semi cylindrical scratch groove of radius a is created which is surrounded by a semi cylindrical inelastic zone of radius b as shown in Fi g. 4 11 (B). The elastic stress field is now constructed from the superposition of the Boussinesq field and Cerruti field. The Boussinesq field arises due to the point normal load ( Fn) and its components on a x y plane at a depth z = c below the surface ar e given by150,151,154,167 222 2 22 23512 3** 1, 2n n xF zxyzyzx crr (4 1) 2222 22 23512 3* 1, 2nn yF zyxzxzy crr (4 2) and 22 23512***3** 1. 2n n xyF zxyxyzxyz crr (4 3) Similarly, the Cerruti field arises due to the point tangential load ( Ft) and its components on the xy plane at a depth z = c below the surface are given by150,151,154,167 ( 4 4) 3 33 25323223332 (12) ,2 ()()()t t xF xxxxx c zzz

PAGE 89

89 (4 5) and 2 22 25 232233* *2* (12) 2 ()()()t t xyF xy yxyxy c zzz (4 6) where and are the normal and the shear stress components and the superscripts n and t refer to the normal and tangential load directions. The Poissons ratio is denoted by and the normaliz ed variables x y z and r are defined as x x c y y c 1 z z c c and r r c (4 7) Also, 222rxy and 2222xyz define the distance from the point load to the field point in the xy plane and the total distance, respectively. Furthermore, Fn and Ft are related by tnFkF (4 8) where k is defined as the friction coef ficient or as a proportionality constant between normal and tangential point loads.150,151 Now the total normal ( x and y ) and shear stress ( xy ) components due to the combined Boussinesq and Cerruti fields can be expressed as nt xxx nt yyy and nt xyxyxy (4 9) To rationalize the observed microcracking and plastic deformation patterns in our study, the maximum principal stress component ( 1 ) and the 2 22 25 3232233* *2* (12) 2 ()()()t t yF xyxxxyxyc zzz

PAGE 90

90 maximum shear stress component ( max ) have been computed from x y and xy Our intent was to identify the regions with high tensile stress and high shear stress where microcracking and plastic deformation, respectively, can occur during a scratch process. In Fig. 411, along with a schematic of a scratc h event, the distribution of the normalized maximum principal stress component, 2 112/ncF and the normalized maximum shear stress component, 2 maxmax2/ncF see Figs. 411 (A) and 4.11 (C), respectively, on the xy plane in the vicinity of the indenter tip are presented. For clarity, the 2D distributions of the stress components along the x axis for various k values are presented in Fig. 412. The distribution of 1 as shown on the xy plane (see Fig. 411 (A)) as well as along the x direction (see Fig. 412 (A)) indicate a stress singularity at the indenter tip. When 0 k there is no tangential force component and the stress distribution is symmetric about the indenter position ( 0 x 0 y ). For 0 k large tensile principal stress develops in the wake of t he indenter tip and its amplitude increases with k as shown in Fig. 412 (A). Assuming that the maximum principal stress criterion is appropriate for modeI crack propagation in brittle materials, a crack initiates when the maximum principal stress excee ds the fracture stress of the material and modeI cracks open up orthogonal to the direction of maximum principal tensile stress. In the experiments, the magnitude of the maximum principal stress component was not large enough at lower scratch loads (e.g., ~ 50 mN) to cause significant microcracking. But as the load increased, a greater extent of microcracking was

PAGE 91

91 observed as a result of the increased maximum tensile principal stress in the wake of the indenter. These observations are rationalized on the basis of increased tensile stress in the wake of the indenter with increasing k as shown in Fig. 412 (A). The orientation of the principal tensile stress in the wake of the indenter, shown in Fig. 411 (D), causes the crack to open perpendicular to the scratch direction as shown in Figs. 46 4 9. To explain the formation of sliplines, we focus on maximum shear stress distribution during the scratch process. Figure 411 (C) shows the maximum shear stress distribution in the vicinity of the indenter tip. Note that the amplitude of max is symmetric about the x axis and highest in the regions ahead of the indenter position and slightly away from the x axis. The projected locations of the highest maximum shear stress and the maximum tensile principal stress on the scratch plane are shown at the bottom of Fig. 411 (C). The distribution of normalized maximum shear stress along the x direction is plotted for various values of k at 0 y in Fig. 412 (B) and at 0.5 y in Fig. 412 (C). These 2D plots clearly reveal that the maximum shear stress, ahead of the indenter tip, increases with i ncreasing k, similar to the normalized principal stress. Also, note that the maximum amplitude of max occurs not along x axis but slightly away from the axis at 0.5 y Depending on the intensity of loading (i .e., for large k), the maximum shear stress can extend beyond the scratch groove resulting in slip line formation outside the scratch grooves as seen in Fig. 46 (C) and Fig. 413. We postulate that the slip bands originate in these regions due to this high local shear stress during the scratch process. To validate this postulation, the

PAGE 92

92 region near the exit end of each of the scratches was observed in SEM. Recall that this endregion was never in the wake of the scratch tool and thus, was never subjected to high tensile stress shown in Fig. 411 (A) and Fig. 412 (A). On the other hand, this region must have experienced large max because this endregion was ahead of the indenter, see Fig. 411 (C) and Figs. 412 (B) and 4 12 (C). Therefore, only sliplines are expected in this region. Figure 413 shows a SEM micrograph of a scratch end revealing only sliplines present on both sides of the scratch in this region. This observation clearly confirms that slip bands are indeed formed ahead of the scratch tool due to the occurrence of maximum shear stress. Therefore, it is concluded that slip bands initiate in the regions ahead of the indenter tip during the scratch process and then the same region will experience tensile stress in the direction of scratch after the indenter moves away (in the wake). This will cause microcracks to open up orthogonal to the scratch direction as observed in Figs. 46 4 8. The present mechanistic study confirmed the sequence of formation of slip bands and microcracking which is in agreement with the experimental observations. 4.2.4 Discussion In the current investigation on the ZrB2SiC composite, two kinds of transgranular microcracks were observed in the ZrB2 matrix; (i) within the slip bands, see Figs 4 6 (A B) and Fig. 47 (B), and (ii) in regions where no slip bands were observed, see Fig. 4 6 (C), Fig. 47 (A) and Fig. 48. It is believed that the sources for occurrence of these microcracks are different. In brittle

PAGE 93

93 materials, defects can be categorized to two types; (i) processinginduced (inherent) flaws and (ii) deformationinduced defects. As the nomenclature indicates, the former are inherent in the microstructure of the ceramic materials (once ceramics are processed) and the latter result from the mechanical loading. Processing induced defects include inter and intragranular microcracks, grainboundary phases such as sintering aids, inhomogeneous grain size distribution, triple point junctions, pores, agglomerates, impurities and weak interfaces along the second phase particles, etc. These defects are also referred to as strength limiting flaws. On the other hand, mechanical deformationinduced defects include dislocations, twins, stacking faults, and regions of phase transformation and loca lized amorphization, etc. While the number of processinginduced defects remains constant in a given ceramic, the number of deformation induced defects continues to grow with the severity of deformation. Brittle materials, in general, show limited plasticity (i.e., microplasticity) due to limited number of favorable slip systems and low dislocation mobility.169,170 As a result, as soon as plasticity initiates, stress concentration occurs at dislocation pileups which can lead to microcracking to relieve the stress. On the other hand, processinginduced microcracking most commonly occurs as a result of differential thermal expansion between the grains of same phase or multi phase materials.171 In ZrB2SiC composite, thermal residual stress arises because of thermal expansion anisotropy within ZrB2 grains and due to difference in thermal expansion coefficients between ZrB2 and SiC phases.2 It is believed that the microcracks, facilitated by the tensile stress in the wake of the indenter and in

PAGE 94

94 regions where s lip bands did not develop, might have originated from the processing induced defects to relieve the thermal residual stress (e.g., Fig. 4 8). On the other hand, the microcracks inside the slip bands are believed to have originated from the deformationinduced defects to relieve the stress associated with low dislocation mobility in the composite (e.g., Figs. 46 (A) and 46 (B) and Fig. 47 (B)). It is wellknown that microcracking is the preferred mode of deformation in the absence of either well defined s lip systems or low dislocation mobility.169,170 Thus, microcracks outside the slip bands progress easily with further application of load and cause fragmentation as seen in Fig. 47 (A). However, such fragmentation or debris formation was not witnessed in the regions containing microcracks within slip bands. Apart from microcracking, the highvelocity scratch experiments caused lateral crack evolution and significant material removal in the composite which was almost absent in low velocity scratches, see Fig. 4 2 and Fig. 49 (A). As mentioned before, numerous experimental and theoretical investigations have shown that the loading phase of scratch process causes elastic plastic deformation within a brittle material. Therefore, upon unloading residual stress is generated within the material. It has been well known that scratch induced residual stress component perpendicular to the scratch surface causes lateral cracking (will be discussed in detail in next chapter).150 In Fig. 42, subsurface lateral crack gr owth within a constant load scratch groove at 250 mN can be observed. However, due to the low load level used in these experiments, extension of the lateral crack up to the top surface was limited. During high

PAGE 95

95 velocity scratches, the applied scratch load w as significantly higher and resulted in extensive lateral cracking as seen in Fig. 49 (A). The above mechanistic rationale for slipline formation and microcracking in ZrB2SiC composite is also consistent with the well known stickslip mechanism studi ed in sliding interactions.172178 The stick slip mechanism arises from the changes in the frictional force during sliding contact due to the difference in static and kinetic friction coefficients. During a scratch process, pileup of material ahead of the indenter tip resists its motion and as a result, the indenter momentarily sticks to the material (or its motion is temporarily halted) resulting in an increased frictional force.152,172178 This leads to an increased static friction coefficient and hence a greater force is required to overcome this resistance. Once the tangential force is large enough to overcome this frictional resistance, the indenter slips and moves forward. Thus the frictional force (or kinetic friction coefficient) is now decreased. This stickslip mechanism continues during the scratch process which is reflected in the saw tooth behavior of the frictional force as has been observed for several materials.176,178 During the stick period, although the indenter motion is temporally halted, the indenter continuously pushes the material piledup ahead of it causing the material to deform first elastically and then plastically.152,156,158,177 Atomistic scale investigations using MD simulations have predicted dislocation generation in the pileup regions.177,179181 As discussed in the previous section, the analytical model clearly revealed that large shear stress develops just ahead of the indenter during sliding motion

PAGE 96

96 (Fig. 411, Figs. 412 (B) and 412 (C)) and leads to plastic deformation in ZrB2SiC composite as was confirmed by the experimental observations (Fig. 413). Although friction coefficient measurements during the scratch process were not considered in the current investigation, we argue that pileup of material may occur in the ZrB2SiC composite and large shear stress builds up in this region. Also, at a constant scratch load, the frictional force (and hence the friction coefficient) will increase due to the pileup and as a result the induced shear stress will also increase as shown in Figs. 412 (B) and 412 (C). Eventually, the shear stress is relieved by the slip mechanism (dislocation generation) and the formation of slip lines occurs within the ZrB2 phase as shown in Figs. 46 and 47 (B). The changes in frictional coeffi cient may also influence the microcracking observed within the ZrB2 phase, see Figs. 46 4 8. If the stick slip mechanism occurs during the scratch process, then the increase in friction coefficient (when the indenter is temporarily halted) will also inc rease the magnitude of maximum principal tensile stress behind the indenter (see Fig. 412 (A)) which can lead to greater microcracking orthogonal to the scratch direction. The above investigation reveals that the ZrB2SiC UHTC can accommodate m oderate ductility through initiation of slip bands at room temperature. This effect, in turn, can result in higher fracture toughness compared to traditional brittle ceramics. In pure ZrB2, the fracture toughness2 has been observed to be around 3.54.2 MPa.m1/2. In typical brittle ceramic materials such as boron carbide (B4C) fracture toughness71 values are lower than 3 MPa.m1/2. The higher

PAGE 97

97 fracture toughness in ZrB2 ceramics can be a result of moderate ductility as observed in the current study. To investi gate ductility in UHTC ceramics, future studies on the analysis of deformation behavior under elevated temperature are required. Such studies are expected to provide more insight on the deformation behavior these materials experience under more realistic operating environments. 4.3 Residual Stress Measurement within SiC Grains in ZrB2SiC Composite 4.3.1 MicroRaman Spectroscopy Raman spectroscopy has emerged as a useful nondestructive tool for residual stress measurements. The sensitivity of a Raman peak to mechanical stress (or more precisely the strain) was first reported by Anastassakis et al.182 In particular, micro Raman spectroscopy (MRS) is useful in determining local residual stress due to its high spatial resolution (of the order of 1 m).182189 For example, in thin films and MEMS devices, microRaman spectroscopy is largely used for residual stress determination that arises as a result of coefficient of thermal expansion (CTE) mismatch between the substrate and the film materials.183,185,188,189 Indentationand scratchinduced residual stress analyses have also been conducted using microRaman spectroscopy.186,187 In crystalline materials, the atomic vibrational frequencies depend on the interatomic force constants.188 In strainfree crystalline materials, interatomic force constants as well as the vibrational frequencies correspond to the equilibrium atomic spacing. Residual stress resulting from either thermal process (such as sintering) or mechanical deformation (such as indentation, scratch, etc.) causes a definitive residual strain which in turn changes the equilibrium atomic spacing within a

PAGE 98

98 material and thus the interatomic force constants. As a result, Raman scattering wave numbers are also perturbed. Depending upon the tensile or compress ive nature of the residual stress, bond lengths and force constants either increase or decrease compared to the equilibrium values. Accordingly, a Raman peak shifts to lower or higher frequency for tensile or compressive residual stress, respectively,182,183,188,189 though there is no unique general relationship between the Raman spectrum parameters (particularly the wave number shift) and the residual stress state. One way to establish such relationships is the calibration procedure which correlates Raman peak shift in a material with the known applied stress. Then using the calibration curve, unknown residual stresses can be determined from the changes in Raman peak positions for that particular material. However, the resultant calibration curve also depends on the applied stress state and therefore, there is no unique relationship between residual stress and Raman peak shift for a given material. On the other hand, explicit expressions relating Raman peak shift and residual stress have been derived for simple situations such as uniaxial stress state, hydrostatic or equi biaxial stress state which can also approximate the residual stresses in materials.183,188,190 Therefore, expressions for appropriate scenario must be developed to estimate residual stress within Raman active materials utilizing microRaman spectroscopy. Raman spectrum of ZrB2 ceramics is not available in open literature, however, the current investigation has revealed that ZrB2 is extremely week Raman active. On the other hand, SiC is know n to be strong Raman active and

PAGE 99

99 its characteristic Raman peaks, particularly for cubic SiC, are highly sensitive to the residual stress.190192 Therefore, it is possible to determine the magnitude of the residual stresses from the stress/strain sensitive Raman peaks of SiC material. In this dissertation, changes in Raman peak positions of SiC grains, located within the scratch grooves of the ZrB25wt%SiC composite, have been measured as a function of the scratch load. Then, from these Raman spectroscopic measurements, a mechanics based expression has been derived to estimate scratch induced mechanical residual stress within the SiC grains in the c omposite. 4.3.2 Experimental In the current study, a microRaman spectrometer, Renishaw inVia Raman Microscope, was utilized for scratch induced residual stress measurements. The Raman spectrometer consisted of a Si laser (532 nm) to excite the specimen, a single spectrograph fitted with holographic notch filters, and an optical microscope (a Leica microscope with a motorized XYZ mapping stage) rigidly mounted and optically coupled to the spectrograph. The spectrometer was initially calibrated with a Si standard using a Si band position at 520.3 cm1. A 100 objective lens was used to focus the incident beam on the desired SiC grains and to collect the scattered beam from the specimen. The spot size of the incident beam on the specimen was in between 11. 5 m. A maximum 25 mW of the laser power was used. Raman spectra were collected from several SiC grains located within the scratch grooves present in the ZrB25wt%SiC composite. For comparison purposes, Raman spectra were also collected from

PAGE 100

100 several SiC gr ains located outside the scratch grooves. All the Raman measurements were performed at room temperature. 4.3.3 Results of MicroRaman Spectroscopy Figure 414 reveals Raman spectra collected from the ZrB2 and the SiC phases of the composite. It can be clearly seen that the Raman peaks from ZrB2 phase were significantly weaker compared to the Raman peaks from SiC grains. Therefore, the Raman peaks from ZrB2 were not further utilized in the current study for residual stress measurements. Fig. 415 shows the Raman spectra collected from the SiC grains located outside the groove and within the scratch grooves resulting from constant loads at 50, 150 and 250 mN. The Raman spectrum from SiC grains outside the scr atch groove as shown in Fig. 415 is typical of 3C SiC190192 which consists of one transverse optical (TO) peak at 796 cm1 and one longitudinal optical (LO) phonon peak at 972 cm1. These two peak positions have been consistently reported for many anneal ed 3 C SiC thin films and thus can be adopted to correspond to Raman peaks in stress free 3C SiC.190 192 However, in the current study, SiC grains away from the scratch groves showed both the TO and LO peaks at higher wave numbers (around 802.3 cm1 and 978.9 cm1, respectively) compared to stress free 3C SiC. Therefore, it was inferred that compressive residual stress is present within these SiC grains in the as processed composite. The origin of compressive residual stress within the as processed composit e is due to the mismatch in Youngs moduli ( E ) and CTE ( ) between H ZrB2 ( E = 489 GPa, avg ~ 5.9 106 K1)2 and 3 C SiC ( E = 694 GPa, ~ 3.5 106 K1),193,194 as well as the difference between

PAGE 101

101 room temperature (25oC) and sintering temperature (1750oC) of the composite. Because the CTE of polycrystalline ZrB2 is greater than that of 3C SiC, cooling of the consolidated compact from sintering temperature will resu lt in compressive residual stresses within SiC grains. With increasing scratch load, both the TO and LO peaks in the Raman spectra collected from the SiC grains within the scratch grooves progressively shifted to lower wave numbers compared to the Raman spectra collected from the SiC away from the scratches, see Fig. 415. With increasing scratch load, greater peak widening as well as asymmetry in peaks can be seen in these spectra. Raman peak shift to lower wave numbers indicates development of tensile residual stress in these SiC grains. The magnitude of the peak shift (and hence tensile residual stress) increased with scratch load. Figure 416 (A) shows the measured LO and TO peak positions for scratch loads of 50, 100, 150 and 250 mN and Fig. 416 (B) shows the shift in TO peak and LO peak positions compared to the same peaks in stress free 3 C SiC materials (796 cm1 for TO peak and 972 cm1 for LO peak). At 50 mN scratch load, both the TO peak and LO peak were lower than the stress free position indicating the presence of compressive residual stress in SiC grains. As the scratch load increased beyond 50 mN, both the peaks further shifted to lower wave numbers indicating development of a tensile stress state within these grains. In SiC grains, present within the scratch grooves, the total residual stress state ( R ) is the sum of processing induced thermal residual stress ( R t ) and scratch induced mechanical residual stress ( R s ), i.e.,

PAGE 102

102 .RRR ts (4 10) Now, in the SiC grains that exist within the unscratched regions of the composite, RR t which is comp ressive as indicated by the Raman spectroscopy. In contrast, the scratch process induces R s within the SiC grains and its magnitude increases with load. Peak broadening and peak asymmetry of the Raman spectra due to the scratch p rocess, can clearly be observed in Fig. 415. This indicated a structural disorder inside the SiC grains.191 The extent of disorder increased with increasing scratch load. Stacking faults have been identified as the primary defects in SiC whiskers195 and i n polycrystalline SiC.196 It has been reported that an increase in number of stacking faults caused Raman peak shift to lower wave numbers as well as peak broadening in 3C SiC ceramics.191 Based on the bond polarizability model, Rohmfeld et al. ,191 simulat ed the influence of stacking fault distance on the TO phonon mode in 3C SiC. It was shown that as the average stacking fault distance decreased (i.e., structural disorder increased with increase in stacking fault density), the peak shifted to lower frequencies and linebroadening was observed in the TO phonon peak. In the current study, similar observations with increasing scratch load were noted in the Raman spectra collected from the SiC grains residing within the scratch grooves. Therefore, it is argued that as the scratch load increased, a greater number of stacking faults (i.e. structural disorder) was generated within the SiC grains which in turn caused an increase in the residual tensile stress. In the following, the shift in the Raman peak positions with scratch load will be

PAGE 103

103 quantified in terms of residual stress within the SiC grains. The TO peaks were relatively more symmetric compared to the LO peak and therefore, we will consider only the TO peak shifts for residual stress calculation. 4.3.4 Evolution of Residual Stress Field Before embarking on the determination of residual stress from Raman spectra, first the processinginduced residual stress state within the composite197 and residual stress field that evolves in a brittle material due to a sc ratch process have been discussed.150,151 Then, Raman peak shift has been correlated to the appropriate stress component to determine the scratch induced residual stress. 4.3.4.1 Residual stress in SiC grains of as processed composite In the ZrB2SiC com posite, the SiC grains (which are assumed equiaxed and almost spherical as can be seen from Fig. 46) can be assumed as the elastic spheres of uniform size distributed in an infinite elastic continuum of ZrB2 matrix.197 This results in axially symmetric st ress distribution around SiC grains. Let us consider that the grain has an effective radius a and the surrounding matrix has a radius b It has been shown that the ceramic grain will be under a uniform/hydrostatic pressure P which can be expressed as197 () 0.5(1)(12)12 (1)mp mmpp mp pT P V EVE (4 11) where is Poissons ratio, T is temperature change, pV is volume fraction of grains, and subscripts m and p stand for matrix and grain. On the other hand,

PAGE 104

104 radial (rad) and tangential (tan) stresses within the ZrB2 matrix and at a distance r from the center of grain are expressed, respectively, as197 3 31rad p pPa V Vr and 3 tan 312p pPa V Vr (4 12) Thus, from the above equations it can be seen that when mp as is the cas e for ZrB2SiC composites, cooling of the composite from the sintering temperature will induce a uniform compressive stress within the SiC grains. The radial and tangential stresses within ZrB2 will be compressive and tensile, respectively, and both the st resses will be maximum at the ZrB2SiC interface. 4.3.4.2 Residual stress in SiC grains located within the scratch grooves The mechanical residual stress induced by the scratch process needs more in depth analysis. As discussed before, singlepass scratch processes have been widely modeled150,151 as a sliding microindentation event where point loads are simultaneously applied in normal ( Fn) and tangential ( Ft) directions on the surface of a specimen as shown schematically in Fig. 411. Also, see Fig. 417 where various crack systems which evolve as a result of the scratch process are shown and will be further discussed latter in this section. The complete elastic stress field during the scratch process is constructed from the superposition of the Boussines q field, Cerruti field150,151,167 and Yoffes blister field.168 The origin of Boussinesq and Cerruti elastic fields has been already discussed in section 4.2. The blister field is an inelastic stress field related to the residual stress due to loading and unloading of the point normal loads.

PAGE 105

105 222 442626 2232235 884224222426262224 224444426262626882(3) *(262 2 ()() 426762812 61512 288 2)R yyyzx xyxyxy B yzyz yyxyzxyzxyzyzyzxyz xyzyzyzxzxzyzyzzz During the sliding process, due to the misfit strain of deformation between elastic and plastic regions, residual stresses are induced within the material during unloading of the scratch tool. According to sliding blis ter field model (SBFM)150,151 the normal residual elastic stress components (R) along x, yand zdirections are expressed as (4 13) (4 14) (4 15) where 2222xyz and B is the blister field strength per unit length.150,151 Figure 418 shows the 3D distribution of the three normal residual stress components R x R y and R z (all normalized with 22 P c ) for B/P = 0.005 on the xy plane in the vicinity of the indenter tip during a scratch process.150 These normal residual st ress components arise during a scratch process as the indenter moves from x ( x x c ) to its present location at 0 x It is seen from Fig. 418 (C) that R z is highly tensile in the wake of the scratch tool whereas R x and R y are compressive directly behind the indenter along the x axis. R y is slightly tensile 22 42242466 2222225 4222222242422424662() *(22624 2 ()() 24236242)R xyzx xyxyxyyy B yzyz xzxyzxyzyzyzxzxzzz 222 2 4224 2232235 64222242242462(3) *(615 2 ()() 921012536)R zzyyxz xyxy B yzyz yxzxyzyzxzyzz

PAGE 106

106 away from x axis behind the indenter. R z reaches its maximum value just behind the indenter ( 0 x ), see Fig. 418 (C), and its magnitude is significantly larger than R x and R y From these 3D scratch induced residual stress distributions, it is clear that R z is the most dominant residual stress component and is known to contribute to subsurface lateral crack initiation1 50 which was also observed to some extent in the current study. With increasing scratch load, the strength B of the sliding blister field also increases and therefore, the magnitude of R z will also increase causing greater lateral c racking and material removal in brittle materials. Due to these reasons, in the following section, the above residual stress component R z will be correlated to the Raman peak shifts observed in Figs. 415 and 416, and magnitude of residual stress within the SiC grains residing within the scratch grooves will be quantified. 4.3.5 Relationship between Mechanical Residual Stress and Raman Spectroscopy Recall that Raman peak shift is related to induced strain in a deformed specimen. If the Raman wave numbers of optical phonons, in the absence and the presence of strain, are denoted by woj and wj ( 13 j ), respectively, then the strain induced Raman shift, jw is expressed as183,188 2j jjjo jowww w (4 16) where j are the Eigen values of the well known secular equation, relating strain to Raman peak shift for diamond and zinc blende structures,183,188,192,198 as

PAGE 107

107 () () 0 ()RRR R R xyz xy xz R RRR R xy yxz yz R R RRR xz yz zxypq r r rpq r r rpq (4 17) where p q and r are the phonon deformation potentials which describe the change in effective spring constants induced by the strain, and R and R are the residual normal and shear strain tensor components, respectively. 3C SiC has a zincblende crystal structure192 and the relation between residual strain and stress tensor components, for this structure, can be expressed according to Hookes law as189,199 111213 212223 313233 44 55 66000 000 000 00000 00000 00000RR xx RR yy RR zz RR xy xy RR xz xz RR yz yzSSS SSS SSS S S S (4 18) where the S terms represent the compliances and R and R are the residual normal and shear stress components, respectively. For cubic structures, 112233SSS 121321233132SSSSSS and 445566SSS As discussed previou sly, the magnitude of R z is significantly greater compared to the other two normal stress components and therefore, to simplify the residual stress calculations, we neglect all other residual normal ( R x and R y ) and shear stress ( R xy R yz and R xz ) components in the subsequent discussions. Accordingly, solving equation (418), the following relationships will be obtained

PAGE 108

108 13 RR xzS 23 RR yzS 33 RR zzS and 0RRR xyxzyz (4 19) Noting that 3311SS and 132312SSS the normal strain components in equation (4 19) can be rewritten as 12 RR xzS 12RR yzS and 11 RR zzS (4 20) Thus, equation (8) reduces to ()0 0 0()00 0 0()RRR xyz RRR yxz RRR zxypq pq pq (4 21) Solving for and substituting R x R y and R z from equation (420), the following expressions are obtained 1 12 11{()},R zSpqSq 2 12 11{()},R zSpqSq and 3 1112(2).R zSpSq (4 22) Now, substituting the above 1 2 and 3 expressions in equation (416), the relation between Raman shift and residual stress can be expressed as 1 12 11 11 {()}, 2R z owSpqSq w (4 23) 2 12 11 21 {()},2R z owSpqSq w (4 24) and

PAGE 109

109 3 1112 31 {2}. 2R z owSpSq w (4 25) Now the above equations will be used to calculate the residual stress from the observed Raman peak shifts shown in Fig. 416. In strained diamond cubic and zinc blende crystal structures, a maximum of three Raman modes are possible, two TO vibrations and one LO vibration.198 The first two Raman modes ( 1w and 2w ) correspond to TO peaks whereas the third Raman mode ( 3w ) is associated with the LO peak. Therefore, equations (423) and (424) can be used to calculate residual stress from TO peak shift whereas equation (4 25) can be utilized to calculate residua l stress from LO peak shift. The values of p and q for 3 C SiC are not available directly in literature. Therefore, these values have been calculated in the following way. The mode Grneisen parameters for hydrostatic stress ( o ) and uniaxial stress ( s ) are defined as192 2(2) 6o opq w and 2() 2s opq w (4 26) where ow is the Raman peak in strainfree condition. The mode Grneisen parameter for the TO phonon ( TO o ) in 3 C SiC is 1.56,192 but the value of TO s is not readily available. Although, for uniaxial and biaxial stresses, TO s and TO o may differ, it has been assumed here that 1.56TOTO so For the TO peak, assuming ow 796 cm1, one obtains 620.62310/cmTOp and 622.63410/cmTOq Also, for residual stress calculation, the following values of compliances have been used: 132 113.710cm/dyn S and

PAGE 110

110 132 121.0510cm/dyn S .200 The final expression for residual stress calculation from the TOpeak, using equation (423), is expr essed as 1 1 12 112 251.66 {()}R o z TOw ww SpqSq (MPa) (4 27) where TOw is the shift in TO peak position. From equation (427), it is clear that shift of TOpeaks to lower or higher wave number will result in tensile or compressive residual stress, respectively. 4.3.6 Determination of Residual Stress Thermal residual stres ses in the as processed composite were calculated from equations (411) and (412) using the material properties and volume fractions of ZrB2 and SiC phases as given in Table 41, and assuming an average radius of 1 m for SiC phase. The temperature differ ence, T used in equation (4 11) was 1725oC. The calculation revealed that the SiC grains were under a uniform compressive residual stress of 1.731 GPa whereas the compressive radial and tensile tangential stresses at ZrB2SiC inte rfaces were 1.731 GPa and 1.126 GPa, respectively. Magnitudes of these stresses decrease sharply away from the interface. The high tensile tangential stress at the interface can cause radial microcracking in the ZrB2 matrix surrounding the SiC grains and e ventually act as potential sites for the development of larger cracks or critical flaws. Figure 419 shows the magnitude of mechanical residual stress within SiC grains (calculated from equation (427)) due to the scratch process as a function of applied scratch load. At 50 mN, residual stress was still compressive whereas

PAGE 111

111 above 50 mN tensile residual stress was generated within the SiC grains and increased almost linearly in magnitude with scratch load. The highest mechanically induced tensile residual st ress at 250 mN within SiC grains was estimated to be around 2.6 GPa. Such large magnitude of residual stress has been reported on many material systems in the literature. Raman spectroscopic measurements have revealed tensile residual stress as high as 2.1 GPa in diamond films,201,202 up to 2 GPa in carbon thin films203 and up to 1.2 GPa in porous Si films.183 The calculated tensile residual stress (2.6 GPa) at 250 mN in the SiC grains was well above the reported tensile fracture stress of 0.9 GPa for polycrystalline 3 C SiC.204 However, it was observed from the SEM micrographs that the SiC grains present within the grooves after the scratch, from which Raman spectra were collected, were crack free at this scale. For example, Fig. 420 revealed that a SiC grain within a groove formed at 250 mN load was completely intact whereas the surrounding ZrB2 matrix was heavily microcracked due to the scratch process. Similar features were also observed at several locations within the scratch grooves. These observ ations indicate unusually high fracture strength of the SiC grains in the composite compared to a typical polycrystalline SiC ceramic. In this scenario, it may not be appropriate to directly compare the available fracture strength values of polycrystalline SiC ceramics to SiC grains present within the particulate phase distributed in the ZrB2SiC composite. Besides, typical polycrystalline 3 C SiC ceramics contain large elongated grains with a high aspect ratio in the range 26,204 whereas the SiC grains present in the

PAGE 112

112 composite were almost equiaxed. Microstructural investigations did not reveal any porosity within SiC phase. Also, no sintering additive was used during the processing and hence no grain boundary glassy phase is expected. Polycrystalline SiC ce ramics are known to exhibit predominantly intergranular fracture mode as opposed to transgranular fracture.204,205 The density of critical flaws or defects that can act as nucleation sites for crack propagation is expected to be much higher in a monolithic ceramic compared to the defect density within the isolated grains of same size in a composite. Therefore, in the absence of pores, secondary glassy phase and limited SiC SiC grain boundary areas, transgranular fracture is expected to be the predominant mode to relieve the accumulated residual stress within the individual SiC grains. Possible sources of transgranular crack initiation will be the internal flaws present within these individual SiC grains. Fracture stress of a polycrystalline ceramic is limit ed by the critical flaw size which scales down with grain size. This in turn increases the macroscopic fracture strength. Based on this fact, a simple fracture mechanics based (using Griffith criterion) calculation was made to estimate the minimum fracture strength of these individual grains. Thus, if one considers crack initiation/propagation from a preexisting critical flaw, then its maximum size will be equivalent to the grain size which will result minimum fracture strength. For a critical flaw size of c, assuming semielliptical surface flaws, the fracture strength ( f ) and mode I fracture toughness ( CK ) are related by206 1.35C fK c (4 28)

PAGE 113

113 Clearly, estimation of f is dependent on the chosen CK and the c values which are not readily known for this particular SiC phase present within the composite. Depending on the assumptions made on the nature of SiC phase (whether monocrystalline or polycrystalline) distributed within the ZrB2 matrix, the estimated f can differ considerably. If it is considered that the SiC phase in the composite is to be made of several grains, then it is appropriate to consider a critical flaw size of the order of average grain size (~1 m) for the estimation of minimum f The available fracture toughness values for polycrystalline SiC ceramics vary over a broad range (3.46.8 MPa.m1/2) and depend process conditions as well as microstructural features such as average grain size, shape, aspect ratio, etc.205,206 For these values, equation (428) results in a minimum fracture strength between 2.55 GPa. These values provide a conservative estimate of f On the other hand, if one assumes SiC phase to be a single crystal, a more complicated situation arises. In a single crystal, typical processing or deformationinduced defects are vacancies, interstitials, dislocations, stacking faults, twinning etc., which are significantly smaller in size than a grain level microcrack. In the current study, the measured Raman peak shift has been related to the development of deformationinduced stacking faults within the SiC phase. Therefore, it has been conjectured that stacking faults are the main source for crack nucleation. Based on the micrographs of stacking faults presented in Shih et al. ,196 the width of the stacking faults are taken as 50 nm. Using a CK value of 3.2 MPa.m1/2 for a single crystal 3C SiC thin film,207 a

PAGE 114

114 fracture strength of 10.6 GPa was obtained. This value is only 24 times the conservative estimate made earlier bas ed on microscopic flaw size. Therefore, it is argued that due to small size and high fracture strength, SiC grains dispersed within ZrB2 phase were able to sustain the large scratchinduced tensile residual stress without any fracture. Raman stress coeffic ients (defined as the change in Raman peak position per unit residual stress) for diamond, SiC and Si are in the range of 0.44 cm1/GPa.183,187190,192,198,208210 This indicates that small changes in Raman peak positions may result in large residual stress. In the current investigation large peak shifts to lower wave numbers, particularly at 250 mN, were observed (see Fig. 416). From the calculated tensile residual stress, the Raman stress coefficient was estimated to be 3.9 cm1/GPa which is in the abov e range of the reported Raman stress coefficients for various materials. Also, residual stress measurements employing microRaman spectroscopy are highly dependent on spatial location within the grains (due to small spot size of the incident beam) and henc e provides only localized residual stress values within small regions. Therefore, it should be noted that the estimated high tensile residual stress is only a localized value and not necessarily the average over the entire SiC grain(s). From the above discussion, it is clear that the SiC grains in ZrB2SiC composite were initially in a state of residual compressive stress and the scratch process induces large residual tensile stress. Similar to the SiC, tensile residual stress is also expected to dev elop within the ZrB2 matrix, particularly at higher

PAGE 115

115 scratch loads. Although the magnitude of residual stress was unknown in the ZrB2 phase as it is not Raman active, evolution of residual tensile stress state within ZrB2 matrix was evidenced from the onset of lateral crack formation at 250 mN. Although unknown, scratch induced tensile residual stress within ZrB2 at 250 mN is predicted to be close to its fracture strength (around 500 MPa).2 Since the ZrB2 phase is continuous and coarse grained, it cannot sustain high tensile residual stress similar to the individual SiC grains. It is argued that up to 200 mN scratch load, the magnitude of tensile residual stress within the ZrB2 phase was probably below the fracture stress to cause any lateral cracking. But above 250 mN, tensile residual stress can locally become comparable to or higher than the fracture stress of ZrB2 resulting in lateral cracking as seen in the experiments. The above model for scratch in duced residual stress estimation within SiC grains in a ZrB2SiC composite is based on the sliding blister field model.150,151 This analytical model was derived to predict the scratchinduced residual stress state in a homogeneous brittle material. But the scratch induced residual stress state in a composite could be more complex and therefore, application of sliding blister model for residual stress determination may not yield exact values. Another major assumption was that Raman peak shift is related to t he dominant stress component ( R z ) in the scratchinduced residual stress field. Other stress components may also have some correlation to the observed Raman peak shifts in SiC grains. Nevertheless, the calculated values appear reaso nable. Therefore, it was argued that the proposed model can be used for scratch induced residual

PAGE 116

116 stress determination in Raman active materials with diamond cubic or zinc blende crystal structures. 4.4 Conclusions Scratch experiments in constant load mode, using a Berkovich nanoindenter, revealed that both the scratch depth and width increased with scratch load. At lower scratch load levels (50 100 mN), smooth plastically deformed scratch grooves were observed without any significant macroscopic damage. A bove a scratch load of 100 mN, considerable amount of damage and material removal along the scratch path were observed. In high velocity scratch experiments, extensive transgranular lateral cracking caused severe brittle damage and material removal in the composite. ZrB2SiC composite exhibited microplasticity in terms of multiple sets of sliplines oriented at random angles to the scratch direction. This feature is attributed to the activation of multiple slip systems or occurrence of dislocation motion al ong preferred slip systems in randomly oriented grains. Different types of damage modes such as transgranular microcracking in ZrB2 phase (inside the regions containing slip bands as well as in regions without any sliplines), extensive lateral cracking, g rain boundary fracture and interfacial cracking at the ZrB2SiC interface were observed within the scratch grooves. The limited dislocation mobility is assumed to be responsible for the observed microcracks in the vicinity of sliplines. On the other hand, microcracks observed outside the slip band regions are assumed to initiate from pre existing (processinginduced) defects in the composite. Also, the microscopic observations indicated that sliplines occurred first followed by microcracking. An analytica l model, based on the elastic stress field solutions for

PAGE 117

117 a singlepass scratch event, was utilized to rationalize the experimental observations. The analysis revealed that during the scratch process, considerable amount of shear stress develops ahead of the indenter tip whereas tensile stress develops in the wake of the indenter. From the mechanistic analysis, it was concluded that the shear stress resulted in slipline formation ahead of the indenter, whereas orthogonal microcracks (with respect to the scr atch direction) resulted from the tensile stress which develops in the wake of the indenter. The stress analysis also confirmed that the sliplines formed first followed by microcracking. Micro Raman spectroscopic measurements were conducted on 3C SiC grains of as processed ZrB2SiC composite as well as on grains that lie within the scratch grooves. It is found that the TO peak and LO Raman peak in 3C SiC were shifted to increasingly lower wave numbers with increasing scratch loads. An analytical model was developed to relate the scratch induced mechanical residual stresses within the SiC grains to the Raman peak shift in terms of phonon deformation potentials of 3C SiC. Residual stress measurements using microRaman spectroscopy on SiC grains of ZrB2SiC co mposite revealed that these grains were under compressive residual stress in the as processed composite and then experience tensile residual stress due to scratch induced deformation. The magnitude of residual compressive stress in the as processed composite is around 1.731 GPa and the scratch induced tensile residual stress increases linearly with load. At 250 mN scratch load, the magnitude of tensile residual stress can be as high as 2.6 GPa.

PAGE 118

1 18 Figure 41. A schematic of instrumented highvelocity scratch tester. HF load cell Diamond tool Specimen Rigid base F n High precision roller bearings Pneumatic cylinder Tool housing

PAGE 119

119 Figure 42. Residual scratch profiles on the polished surface of ZrB2SiC composite at various load levels.

PAGE 120

120 Figure 43. A) Penetration depth profile and B) the corresponding trace of residual scratch groove at 50 mN. Increase in scratch depth corresponds to the damaged regions as indicated by the white dashed circles in Fig. 43 B). Magnified SEM micrographs of the damaged regions along the scratch path revealed mostly voids, grain pull out, ZrB2 grainboundary fracture and some microcracking.

PAGE 121

121 Figure 44. A) Penetration depth profile and B) the corresponding trace of residual scratch groove at 250 mN. Damaged regions along the scratch path are shown by the white dashed circles.

PAGE 122

122 Figure 45. Scratch depth profiles during the nanoscratch experiments at different loads.

PAGE 123

123 Figure 46. A) Micrograph of scratch induced deformation features at a load of 250 mN. B) A magnified view of the region (X) revealing the slip line patterns and microcracks along the scratch groove as well as cracks emanating from a SiC particle. C) Another region along the scratch path revealing several sets of slip lines oriented randomly with respect to the scratch direction.

PAGE 124

124 Figure 46. Continued Slip lines C Microcracks Scratch groove 5 m

PAGE 125

125 Figure 47. Fracture patterns within the scratch groove at a load of 250 mN: A) grainboundary fracture, microcracking, and scratch debris and B) interfacial cracking between ZrB2 and Si C phases as well as microcracks perpendicular to the direction of the scratch path.

PAGE 126

126 Figure 48. Scratch induced damage at 250 mN, revealing transgranular microcracking in ZrB2SiC composite.

PAGE 127

127 Figure 49. A) A high velocity scratch groove, B) high magnification images of extensive transgranular microcracks orthogonal to the scratch direction and C) normal force vs. scratch length profile.

PAGE 128

128 Figure 410. Slip line formation within ZrB2 phase during highvelocity scratch process.

PAGE 129

129 Figure 411. A) Normalized maximum principal stress distribution, B) schematic of the scratch process, C) normalized maximum shear stress distribution in the vicinity of indenter tip and D) plot of maxim um principal stress contours and orientation in the wake of the indenter.

PAGE 130

130 Figure 412. Variation of A) normalized maximum principal stress ( 1 ) at 0 y and B) normalized maximum shear stress ( max ) at 0 y and C) 0.5 y for various values of k. The black triangle indicates the indenter position.

PAGE 131

131 Figure 413. Micrograph of the exit end region of a scratch at 250 mN revealing numerous slip lines. The white dotted lines indicate the boundary of residual scratch exit end.

PAGE 132

132 Figure 414. Raman spectra collected from the ZrB2 matrix phase (see the inset also) and the SiC particulate phase present in ZrB2SiC composite. 2500 7500 12500 17500 22500 27500 32500 500 700 900 1100 1300 1500 1700 Raman shift (cm-1) SiC ZrB 2 ZrB 2 500 900 1300 1700

PAGE 133

133 Figure 415. Raman spectra collected from the SiC grains present within and away from the scratch grooves.

PAGE 134

134 Figure 416. A) LOand TO Raman peak positions of SiC grains within the scratch grooves at 50, 100, 150 and 250 mN loads. LO and TO peak positions at 0 mN correspond to the Raman spectra collected from SiC grains outside the scratch groove. B) Changes in peak positions to stressfree LO and TO positions. A 945 950 955 960 965 970 975 980 985 0 50 100 150 200 250 300 780 785 790 795 800 805 810 0 50 100 150 200 250 300 Scratch load (mN) Stress free TO peak Stress free LO peak LO peak TO peak Raman shift (cm-1)

PAGE 135

135 Figure 416. Continued -20 -15 -10 -5 0 5 0 50 100 150 200 250 300 Scratch load (mN) Peak shift (cm-1) B

PAGE 136

136 Figure 417. Schematic of the scratch process and different crack systems that evolve during the scratch process are shown.

PAGE 137

137 Figure 418. Distribution of normalized residual stress components; A) R x B) R y and C) R z on the xy plane (B/P = 0.005). The black triangle indicates the position of the indenter whereas the black arrow indicates the scratch direction.

PAGE 138

138 Figure 419. Evolution of residual stress within SiC grains as a function of scratch load. -1.5-1 -0.5 0 0.5 1 1.5 2 2.5 3 050 100 150 200 250 300 Scratch load (mN) Residual stress (GPa) Tensile Compressive

PAGE 139

139 Figure 420. SEM micrograph of a scratch groove at 250 mN revealing the uncracked SiC grain surrounded by the heavily microcracked ZrB2 matrix.

PAGE 140

140 Table 41 Material properties Material E (GPa) (K 1 ) Volume fraction (%) ZrB 2 489 2 5.9 10 6(2) 0.16 2 ~ 91 SiC 694 194 3.5 10 6(195) 0.17 196 ~ 9

PAGE 141

141 CHAPTER 5 ROOMTEMPERATURE DISLOCAT ION ACTIVITY IN UHTCS 5.1 Introduction In C hapter 4, it has been shown that during scratch studies at room temperature readily detectable metal like plastic deformation patterns called sliplines, uncharacteristic of hard ceramics, were observed in a polycrystalline zirconium ZrB25wt%SiC composite .72 Similar features were also noted from indentation experiments studies in ZrB2 and hafnium diboride (HfB2) ceramics.211 Figures 51 (A) and 51 (B) reveal optical micrographs (in Nomarski illumination) of intense slipline or surface step formation in t he vicinity of indented regions in a polycrystalline ZrB2 and a polycrystalline HfB2 ceramics, respectively. Slip line features were also observed in the SEM micrograph as shown in Fig. 51 (C) for the HfB2 ceramic which are similar to those formed in ZrB2 as shown in C hapter 4. Such macroscopic slipsteps are clear evidence of dislocationinduced plastic deformation and dislocation mobility at room temperature in these ultra hightemperature ceramics. At elevated temperatures dislocation assisted plastic d eformation is expected to play a significant role in mechanical deformation compared to room temperature. Therefore, it is important to investigate the nature of dislocations in these materials induced during mechanical deformation. Owing to the extremely high melting temperature of UHTCs as well as high covalent character in chemical bonding, such plastic behavior is unexpected.212 Except the transmission electron microscopic (TEM) observation by Haggerty and Lee159 on the deformed regions of ZrB2 single crystals, dislocations have never been identified either in polycrystalline ZrB2 or in HfB2 ceramics. Therefore,

PAGE 142

142 in this work, employing TEM, the room temperature dislocation activity responsible for the observed sliplines in ZrB2 ceramics has been invest igated. Also, the possible slip systems activated during the mechanical deformation have been identified. Then, an attempt has been made to build up an understanding on the origin of room temperature dislocation mobility in ZrB2 ceramics from crystal structure and chemical bonding perspectives. 5.2 Experimental For identification of deformation features at a finer scale, transmission electron microscopy (TEM) was performed. A site specific cross sectional FIB (FEI Strata DB235) technique213 was employed to prepare TEM specimens from the low velocity scratch groves. These groves were chosen because of the availability of larger smooth areas containing sliplines in the vicinity of scratches. The FIB technique employs a focused Galium ion (Ga+) beam to prepare a thin TEM specimen of < 200 nm thickness (of approximate dimensions 15 m 6 m) from the target material through sputtering action. Here a lift out technique213 was used where an electron transparent thin specimen is cut out from the bulk material and analyzed directly in TEM. First the area of interest (i.e., the region within a scratch groove containing large number of sliplines) was protected from any potential damage from the Ga+ ion beam by depositing a thin layer of platinum (Pt). Then, progress ively deeper trenches were cut on either side to form a thin specimen as shown in Fig. 5 2 illustrating a FIB cut transverse cross sectional sample within a constant load scratch groove of 250 mN. The specimen was then cut through the depth on two

PAGE 143

143 ends (where the specimen is attached to the bulk of the material) so as to relax the residual stress (without cracking) which had developed as a result of elastic plastic deformation during the scratch process. The beam current was then reduced and milling was per formed on both sides of the sample until thinned down to thickness of about 200 nm. The specimen was set free from the scratch groove and analyzed in a TEM (JEOL 200CX). 5.3 Results and Discussion The TEM investigations revealed high density of dislocati ons within the deformed ZrB2 grains in the regions directly beneath the scratch groove. A low magnification bright field image of the TEM specimen is shown in Fig. 53 (A) where as Fig. 53 (B) presents a high magnification bright field TEM image, at [ 0001 ] zone, of a selected area from Fig. 53 (C) revealing highly deformed region beneath the scratch groove. Clearly, the TEM micrograph revealed dense dislocation activity and clustering within the deformed ZrB2 grains. Dislocation s in three different orientat ions were clearly visible, suggesting slip on multiple planes. Similar ly, multiple and intersecting slipbands were observed at a macroscopic scale on the deformed surfaces, see Fig. 53 (C). These dislocations and slip bands, observed from the TEM and SEM images, respectively, appear to be at an orientation of approximately 60o to one another. At [0001] zone axis, the diffraction pattern contains three prismatic g vectors, 101 0 1 100 and 011 0 It should be noted here that t he 112 0 type directions are the closest pack directions contained within the 101 0 type prismatic planes 101 1 type pyramidal planes and ( 0001 ) basal plane. Also

PAGE 144

144 these directions are oriented 90o from the poles of these prismatic pyramidal and basal planes. Howe ver, s uch intersecting dislocations at [ 0001 ] zone axis and their threefold symmetry suggested prismatic or pyramidal slip in ZrB2 during mechanical deformation. To distinguish between prismatic and pyramidal slip, a simple twobeam analysis as described i n the following was performed. These three sets of dislocations observed at [ 0001 ] zone axis were imaged in two beam condition with prismatic g vectors If the dislocations are of prismatic type, then one of the sets of dislocations will go out of contrast f or each of these three diffracting conditions due to = 0 Geometrically, this means that the direction of Burgers vector must lie in the plane responsible for the operating reflection. Therefore, pole of the plane will be perpendicular to the direction of Burgers vector. For example, pole of the ( 1 1 0 0 ) ( ) plane is perpendicular to the [ 112 0 ] direction (closets pack direction). Therefore, will be zero. In contrast, if the dislocations are of pyramidal type, then all these three sets of dislocations will remain in contrast when imaged with prismatic g vectors 0 In the current work, it was indeed observed that one of the three sets of dislocations was always out of contrast for each of the three diffracting conditions with prismatic g vectors i.e., = 0 These observations suggested prismatic slip within ZrB2 grains. As mentioned before that 112 0 type of directions are the closest pack directions, t herefore, it was inferred that Burgers vectors of these dislocations are parallel to the 112 0 directions. In ZrB2 crystal, [ 0001 ] is also a possible close pack direction as the c/a ratio in ZrB2 is 1.114. However, the

PAGE 145

145 p ossibility of [ 0001 ] as the Burgers vector for these three sets of dislocations observed at [ 0001 ] was excluded since these dislocations were clearly visible at [ 0001 ] zone axis, see Fig. 53 (A). Thus, 101 0 112 0 slip system was inferred in this ceramic. The TEM investigati ons also revealed another slip system less commonly reported in HCP materials [161]. Figure 54 (A) and Fig. 54 (B) reveal a bright field image of dislocati ons, at [ 1 2 1 0 ] zone axis, and the corresponding selected area electron diffraction pattern, respectiv ely. Burgers vector of these dislocations was determined in the following way. For this set of dislocations, however, Burgers vector cannot be of [ 1 2 1 0 ] type, closest pack direction in ZrB2, (as was observed for the set of dislocations in Fig. 53 (B)) or parallel to this direction. If the Burgers vector of a set of dislocations is parallel to a zone axis, then those dislocations will be invisible at that zone axis. This is because all the planes at the zone contains the zone axis direction (or parallel) and therefore, for all the g vectors at the zone, = 0 will be zero. Since the dislocations shown in Fig. 54 (A) were visible at [ 1 2 1 0 ] zone axis, their Burgers vector cannot be parallel to this orientation, for they would be invisible. [ 1 2 1 0 ] zone contains ( 0001 ) (101 0 ) and (101 1 ) reciprocal lattice vectors ( g ), corresponding to basal, prismatic and pyramidal planes, respectively. Contrast experiments revealed that these dislocations (as seen in Fig. 54 (A)) remained in contrast when imaged with ( 0001 ) and (1 0 1 1 ) g vectors. For example, Fig. 54 (C) shows a bright field image of the dislocations (seen in Fig. 5 4 (A)) imaged in two beam condition with (0001) g vector. That is for ( 0001 )

PAGE 146

146 and (101 1 ) g vectors, 0 This excluded the possibility of { 0001 } 112 0 basal slip in ZrB2. This is because for basal slip, will be zero with ( 0001 ) g vector and dislocations will be invisible in twobeam condition. On the other hand, these dislocations were completely invisible when imaged with (101 0 ) g vector. As discuss ed above, [ 1 2 1 0 ] direction was excluded as the Burgers vector. Since the only other low index direction perpendicular to the pole of ( 101 0 ) plane is [ 0001 ] the above contrast experiments suggested that the Burgers vector is parallel to [ 0001 ] For further confirm ation, the same set of dislocations was imaged with a ( 112 0 ) g vector (in [ 1 100 ] zone) because { 112 0 } type of planes (secondary prismatic type) only contain [ 0001 ] Burgers vector. It was observed that t hese dislocations went completely out of contrast when imaged with ( 112 0 ) g vector in twobeam condition. Therefore, the same set of dislocations was completely invisible when imaged with (101 0 ) and ( 112 0 ) g vectors. The only direction which is contained by (101 0 ) and ( 112 0 ) planes is [ 0001 ] Both the (101 0 ) and ( 112 0 ) g vectors will dot to zero with [ 0001 ] vector. Therefore, it was concluded that Burgers vectors are parallel to [ 0001 ] direction. This Burgers vector has never been reported in open literature for ZrB2 ceramics. F or dislocations with Burgers vector parallel to [ 0001 ] the only possible close pack slip plane is of prismatic type, either { 101 0 } or { 112 0 } The { 101 0 } primary prismatic planes have a higher planer density (0.727) compared to that (0.682) of the { 112 0 } secondary prismatic planes. Thus, { 0001 } [ 0001 ] is the most p robable slip system compared to { 112 0 } [ 0001 ] in ZrB2. However, due to nearly comparable packing density of the primary and secondary prismatic

PAGE 147

147 planes, slip may also occur on the { 112 0 } planes Nevertheless, the current TEM investigations clearly revealed prismatic slip as the only identified room temperature plastic deformation mechanism in polycrystalline ZrB2. The above TEM studies are also consistent with the work by Haggerty and Lee in ZrB2 single crystals.159 There is also considerable evidence214,215 supporting that prismatic slip is favored over basal slip in ZrB2. Xuan et al.,214 performed Vickers microhardness measurements in ZrB2 single crystals, on basal, ( 0001 ) and prismatic planes, { 101 0 } and { 112 0 } from room temperature to 1000oC. It was observed that over the entire temperature range, hardness was same on both types of prismatic planes but was lower than that on the basal plane. A similar result was also obtained in the work of Nakano et al.215. When a basal plane is indented, only possible slip planes are pyramidal planes. On the other hand, indentation on { 101 0 } and { 112 0 } planes can cause slip on other prismatic planes from { 101 0 } and { 112 0 } families as well as on pyramidal planes. Therefore, it is clear that in ZrB2 prismatic slip is more favorable. The work of Xuan et al.,214 and Nakano et al.,215 also suggested that dislocation motion is probably equally favorable on both { 101 0 } and { 112 0 } prismatic planes. The current TEM investigation clearly revealed that prismatic slip was the only identified mechanism responsible for room temperature plastic deformation in polycrystalline ZrB2. Having determined the nature of dislocations, in the following, it has been attempted to rationalize the room temperature dislocation activity and mobility based on the proper ties, crystal structure, and chemical bonding present in these

PAGE 148

148 ceramics. The nature of chemical bonding plays a dominant role in governing the properties of metals and ceramics. In general, the nonmetallic character of bonding (mainly covalent) in structural ceramics is well reflected in their properties through high elastic modulus, chemical stability at room and high temperatures, extremely low electrical conductivity, and thus, corresponding high electrical resistivity in comparison to metals.216 In contrast, UHTCs have a unique combination of properties which are characteristics of both metals and ceramics.2 As mentioned before, their high melting temperature (> 3000oC), high elastic modulus (> 400 GPa) and superior thermo chemical stability are typical of ceramics. In addition, UHTCs have surprisingly high electrical conductivity (> 106 S/m) and thus, low electrical resistivity (< 105 Ohm cm), comparable to those of metals.2,217225 These values are in contrast to the typical structural ceramics such as aluminum oxide (Al2O3), silicon carbide (SiC), and silicon nitride (Si3N4) as illustrated in Fig. 55. Such electrical properties are an indication of significant metallic nature (i.e., metallicity) in chemical bonding present in these ceramics. ZrB2 and HfB2 have AlB2type ( P 6/ mmm space group) hexagonal crystal structure2 where the unit cell contains a six member graphitelike ring or a net of boron (B) atoms in two dimension which alternates with close packed layers of metal (M) atoms, see Fig. 5 6. Vajeeston et al.,226 studied electronic structure and nature of chemical bonding in AlB2type ceramics using the self consistent tight binding linear muffintin orbital (TB LMTO) method. Zhang et al.,227 utilized the first principles total energy planewave pseudopotential (PW PP) method and investigated the electronic structur es and elastic properties of ZrB2 and HfB2

PAGE 149

149 ceramics. The theoretical analysis of Vajeeston et al.,226 and Zhang et al.,227 have revealed the existence of three types of chemical bonds in this crystal structure: (i) B B, (ii) M B and (iii) M M (M=Zr, Hf).226,227 While the B B bonds are purely covalent, the M B bonds are a mixture of covalent and ionic characters. However, due to small difference in electronegativity values (1.33, 1.30 and 2.04 for Zr, Hf and B, respectively), ionicity is less than 8%. In contrast, the M M bonds are predominantly metallic due to nearly freeelectron nonbonding d orbital states which attribute significant metallicity to these ceramics.226,227 Both of these bonds are known to exhibit strong influence on properties of materials. Here, it is argued that such metallic character of chemical bonding in ZrB2 and HfB2 ceramics is well reflected in their metal like mechanical deformation behavior observed in indentation and scratch studies, see Fig. 51, at room temperature.72,160 In t he following, arguments for the origin of room temperature dislocation mobility in ZrB2 ceramic, may also be applicable for other TMB2 ceramics such as HfB2, are presented. During dislocation motion, chemical bonds constantly distort, break, and reform.228,229 The nature of chemical bonding (i.e., how tightly electrons are bound, their directionality, and symmetry) has a strong influence on dislocation mobility for a given crystal structure at room temperature. Gilman228,229 suggested that local or nonloca l nature of chemical bonding determines mobility at room temperature and proposed a simple but fundamental chemical approach to describe dislocation mobility in pure covalently bonded solids. In this approach, optical band gap, a measure of localized natur e of

PAGE 150

150 chemical bonding, is assumed to be related to glide activation energy. In a similar fashion, qualitatively, dislocation mobility can also be correlated to the local or nonlocal character of chemical bonding in ZrB2 and in other AlB2type UHTCs as hav ing a mixture of covalent and metallic bonds.226,227 Similar to the optical band gap, the electrical conductivity can also reflect the local or nonlocal nature in chemical bonding. While the exact mechanism of high electrical conductivity in ZrB2 and HfB2 ceramics is not known, Zhang et al.,227 suggested that their unusually high conductive nature is related to the Zr Zr and Hf Hf metallic bonding. It is argued that such bonding characteristics can also play a significant role in the room temperature plast ic deformation and the associated dislocation mobility in these ceramics. One of the reasons why such room temperature macroscopic plastic deformation features are not readily observed in other structural ceramics (e.g., Al2O3, ZrO2, SiC, Si3N4) is that they possess strong covalent and ionic bonding instead of direct metal metal bonds.230233 Direct metal metal bonding does not exist in these materials. Strong covalent and ionic bonding makes dislocation mobility extremely difficult, particularly at room temperature where availability of thermal energy is low. Dislocation mobility is strongly influenced by the width of a dislocation core which depends on the type of chemical bonding. While metallic bonding results in wide dislocation core, covalent bonding makes it narrow.230 For wide dislocation core, consisting of nondirectional metallic bonds, the energy barrier for dislocation mobility will be lower compared to that of a narrow core structure made up of highly directional covalent bonds.230 A classic e xample is

PAGE 151

151 cubic SiC which has the required number of independent slipsystems to exhibit room temperature plasticity but due to its pure covalent bonding, the width of dislocation core is narrow and thus, dislocations are immobile at room temperature.230 An accurate description of dislocation core structure and its motion in ZrB2 (or HfB2) crystal structure is complex, and requires a fully atomistic approach. However, for a prismatic slip in ZrB2, the dislocation core structure not only involves B B and Zr B bonds but also the Zr Zr bonds. Existence of metallic bonds can make the dislocation core relatively wider compared to a core made up of only B B or Zr B bonds. Thus, the Zr Zr bonds are assum ed to play a crucial role in enhancing the dislocation mobility at room temperature by lowering the chemical barrier. Due to this nonlocalized nature of Zr Zr bonds and the associated increased dislocation mobility during mechanical deformation, dislocati ons are not restricted to a small material volume (typical of ceramics) rather extend over a relatively larger region allowing the dislocation activity to be readily visible even at room temperature at distances further away from the region of load applica tion as seen in Fig. 51. 5.4 Conclusions ZrB2 and HfB2 ceramics were observed to exhibit metal like slip line patterns at room temperature when deformed by indentations and scratch loads. This deformation pattern indicated significant room temperature di slocation mobility in these ceramics. Indepth TEM analysis in a ZrB2SiC composite revealed that dislocations were indeed generated within the ZrB2 phase as a result of mechanical deformation at room temperature. Prismatic slip was

PAGE 152

152 identified as the major plastic deformation mechanism in polycrystalline ZrB2 ceramic. Finally, a possible role of Zr Zr metallic bonds in room temperature dislocation mobility in ZrB2 ceramics was discussed.

PAGE 153

153 Figure 51. Optical micrographs revealing slip lines formed in the vicinity of indented regions of A) ZrB2 and B) HfB2 ceramics.

PAGE 154

154 Figure 52. FIB cut TEM specimen within a constant load (250 mN) scratch groove.

PAGE 155

155 Figure 53. A) A bright field TEM micrograph revealing dense dislocation activity on prismatic planes and B) multiple sets of intersecting sliplines formed on the surface due to scratch load, resembling patterns observed in A).

PAGE 156

156 Figure 54. A) Bright field TEM micrograph at [1210] zone axis and B) the corresponding selected area electron diffraction pattern. C) Bright field TEM image of dislocations in twobeam condition with (0001) g vector. 1010 () 0001 () 1011 () 1210 B A B

PAGE 157

157 Figure 5 4. Continued

PAGE 158

158 Figure 55. Electrical conductivity (filled symbols) and electrical resistivity (open symbols) of metals and ceramics. HfB 2 TiB 2 ZrB 2 TMB2 Au Zr Ti Hf Metals Cu Ag Al SiC ZrO 2 Al 2 O 3 Si 3 N 4 Structural ceramics 1.00E-15 1.00E-12 1.00E-09 1.00E-06 1.00E-03 1.00E+00 1.00E+03 1.00E+06 1.00E+09 0 100 200 300 400 500 600 Elastic modulus (GPa) Electrical conductivity (S/m) 1.00E-02 1.00E+01 1.00E+04 1.00E+07 1.00E+10 1.00E+13 1.00E+16 1.00E+19 1.00E+22 1.00E+25 Electrical resistivity ( Ohm-cm)

PAGE 159

159 Figure 56. Crystal structure of transition metal diborides. a TM B 0001 [] 1123 [] 1120 [] c TM TM bonding B B bbonding TM TM bonding TM B bonding TM B bonding

PAGE 160

160 CHAPTER 6 SUMMARY AND FUTURE W ORK 6.1 Conclusions Plasma pressure compaction technique has been utilized to process highmelting point and difficult to sinter materials such as boron carbide ceramics71 and zirconium diboridesilicon c arbide composite.72 These materials have high covalent bonding characteristics and hence conventional methods which require high sintering temperature and pressure as well as long consolidation times are not suitable to produce dense compacts. In the current work, use of P2C technique successfully produced high density large sintered ceramic compacts at relatively lower temperature, lower compaction pressure, and over a shorter period of time compared to traditional techniques such as pressureless sintering and hot pressing. Also, it was shown for boron carbide ceramics that grain growth was limited in P2C technique due to lower sintering temperature which aided in the production of finegrained B4C ceramics.71 Boron carbide ceramics with three different grain sizes were subjected to static and dynamic indentation experiments to reveal the influence of grain size and strain rate on mechanical properties. Static indentation experiments were conducted using a traditional Vickers indentation tester where as a dynamic Vickers indentation tester was employed to perform the dynamic hardness measurements. In static indentations load was delivered within a duration of 15 seconds which corresponded to a deformation rate on the order of 105/s. In contrast, dynamic loads were delivered within 100 s resulting in a deformation rate of approximately 103/s. It was consistently observed that for all the B4C ceramics, dynamic hardness and dynamic fracture toughness were lower compared to static hardness and static fractur e toughness.71

PAGE 161

161 To reveal the subsurface region of indentation, bonded interface technique on split specimens was employed. Static and dynamic indentations at similar load levels were then performed along the interface of the bonded specimen. Microstructur al investigation of subsurface regions of indentations revealed that dynamic loads resulted in significantly higher damaged regions beneath the indentations compared to the static indentations which are consistent with the hardness and fracture toughness m easurements.71 For further investigations, these subsurface regions of indentations were probed using a series of spectroscopic techniques such as Raman, photoluminescence and infrared spectroscopy. These studies revealed newer peaks from the subsurface damaged regions which indicated that localized phase transformation due to structural changes occurred within these regions.234 It was also observed that extent of structural changes/disordering was significantly higher in dynamic indentations compared to st atic indentations. From indepth analysis of Raman and PL spectroscopy, it was confirmed that amorphous carbon clusters were formed within the indented regions. Also, FTIR analysis revealed the existence of amorphous boron clusters within the dynamically i ndented region. Thus, the current studies clearly revealed the influence of strain rate on mechanical properties as well as structural changes in boron carbide. Scratch experiments were performed on a ZrB25wt%SiC composite to reveal the contact induced damage and deformation mechanisms which can evolve in this ultra hightemperature ceramic during service. These studies revealed unique metal like plastic deformation features in terms of sliplines within the ZrB2 matrix phase which indicated room temper ature dislocation activity and mobility.72,160 Similar deformation

PAGE 162

162 features were also noted from indentation experiments.160,211,235 Such room temperature deformation patterns are highly unusual in ceramics having high melting temperature.212 Apart from sl ip lines, transgranular microcracks were observed to develop perpendicular to the scratch direction. To rationalize the experimental observations, elastic stress field solutions combining Boussinesq field and Cerruti field solutions for normal and tangenti al point contact loads, respectively, were utilized. From the elastic stress field distribution, it was shown that as the scratch progresses, considerable amount of shear stress is developed ahead of the indenter tip whereas tensile stress is developed in the wake of the indenter. Also, the maximum principal normal tensile stress was oriented in the direction of scratch. Thus, it was concluded that the shear stress ahead of the scratch tool resulted in slipline formation whereas the normal tensile stress i n the wake of the scratch resulted the orthogonal microcracks (with respect to the scratch direction).72 The stress analysis also confirmed that the sliplines were formed first followed by microcracking. Deformationinduced residual stress is detrimental to any brittle materials and the magnitude of residual stress may increase as the deformation progresses. Towards this end, an analytical model was developed which can calculate scratch induced residual stress from the Raman spectroscopic measurements .236 This analytical framework was developed by combining the well known secular equation for diamondlike or zinc blende crystal structure and the concept of blister filed solutions. In this framework, residual stress is calculated from the Raman peak shift using phonon deformation potentials of a material. This model is applicable to materials having diamondlike or zinc blende crystal structure. In the current work, this model was applied to the SiC

PAGE 163

163 particulate phase of the composite which has zinc b lende crystal structure (3 C SiC) and the residual stress was calculated.236 Raman spectra were collected from the SiC grains residing within and outside of the scratch grooves. It was observed that the characteristic Raman peaks of 3C SiC gradually shifted to the lower wave numbers with increasing scratch load indicating accumulation tensile residual stress. The measurements revealed that initially residual stress within the grains in the composite (outside of scratch groove) was compressive and gradually became tensile within increasing scratch load.236 Existence of dislocations within the ZrB2 phase was confirmed through transmission electron microscopic analysis of deformed regions within the scratch grooves.160,211 Prismatic slip was identified as the plastic deformation mechanism in polycrystalline ZrB2. It was argued that similar to having extremely high electrical conductivity which is comparable to metals, formation of typical metal like plastic deformation feat ures at room temperature was also a reflection of metallicity present in chemical bonding in these ceramics. In this context, a possible role of Zr Zr metallic bonds, present in the ZrB2 crystal structure, which can favor dislocation mobility at room tempe rature has been discussed.211 In conclusion, two major findings can be drawn from this thesis: (i) development of a simple technique that can induce structural changes in B4C ceramics and similar to that observed in ballistically investigated ceramic tiles, and (ii) presence of unequivocal evidence for room temperature dislocation mobility in UHTCs.

PAGE 164

164 6.2 Future Work 6.2.1 Determination of Transition Point of Phase Transformation in B4C In this dissertation, it has been shown that localized phase t ransformation or structural changes occurred during static and dynamic indentations. However, the question still needed to be answered is that if there a transition point in terms of indentation load or velocity or pressure, when the transformation initiat es. Also, there are several factors such as microstructure, composition, and presence of graphite phase may have an influence on the extent of phase transformation. However, the use of extremely sharp indenters such as Berkovich and Vickers results in high enough pressure even at a low load to trigger localized phase transformation or structural disordering. Instead, use of spherical indenters with different tip radii can be helpful in capturing a possible existence of transition point for solidstate phase transformation in B4C as a function of load, indenter velocity, etc. 6.2.2 Scratch Response of Oxidized Surfaces of ZrB2SiC Composite It has been shown that the surface morphology of ZrB2SiC composite changes with temperature due to oxidation of both the phases.2,3740 In the temperature range of 800 1200oC oxidation rate of ZrB2 is higher than that of SiC. ZrB2 produces molten B2O3 (melting point ~ 450oC) and crystalline ZrO2 at 1000oC. This process results in a porous layer of crystalline ZrO2 and SiC above the unoxidized ZrB2SiC composite. Thus, above 1000oC, the oxidized surface of the ZrB2SiC composite consists of a porous ZrO2+SiC layer covered by a thin B2O3 amorphous layer.3740 At this temperature range, further oxidation of unreacted ZrB2, present below the porous ZrO2+SiC layer, is controlled by the diffusion of oxygen (O2) through the thin glassy layer of B2O3. Figures 6 1 (A), 6 1 (B) and 61 (C) show the cross sectional areas of the ZrB25wt%SiC

PAGE 165

165 specimens oxidized at 1000oC in ambi ent conditions for 1, 5 and 15 hours, respectively. It can be seen from Fig. 61 that upon oxidation different layers were formed from the top. As the oxidation time increases total thickness of the oxidized layers also increases. Surface morphology of ZrB2SiC composite also changes as the temperature increases. At temperatures above 1200oC the upper B2O3 layer starts to volatilize (i.e., 2323BO(l)BO(g) ) and oxidation of SiC results a thin SiO2rich layer (it may also contained some B2O3 due t o incomplete evaporation or continued oxidation of ZrB2) above the porous ZrO2 layer. Above 1400oC the thickness of the outer SiO2 layer increases while the B2O3 content composition of reduces significantly. Clearly, at this temperature, the structure and the oxidized surface layer is considerably different than that at lower temperatures.3740 For example, Fig. 62 shows a schematic of the cross section of ZrB2S iC specimen oxidized at 1500oC. Thus, it is clear that the composition and morphology of the oxidized surfaces change with temperature. Recall that these ZrB2SiC composites are used as thermal protection systems for leading edges and nose cones of hypers onic vehicles, and reentry vehicles which are subjected to intense erosion upon reentry.2 Presence of outer porous layers of different compositions can drastically affect the overall mechanical response of the structure. Also, the thickness of the oxidiz ed layers will vary with oxidation time. Therefore, it is also important to investigate the influence of oxide layer thickness on the wear properties. Therefore, future work should focus first on developing oxidized surfaces at different temperatures and at different oxidation times. Then these surfaces should be subjected to scratch studies to investigate wear properties of ZrB2SiC composites.

PAGE 166

166 Figure 61. Cross sectional areas of the ZrB25wt%SiC specimens oxidized for A) 1 B) 5 and C) 15 hours. Total oxide layer Total oxide layer Total oxide layer Thin continuous top layer Thin continuous top layer Thin co ntinuous top layer Porous middle layer Porous middle layer Porous middle layer Unreacted bottom layer Unreacted bottom layer Unreacted bottom layer A B C

PAGE 167

167 Figure 62. Evolution of structures in a ZrB2SiC composite during oxidation at 1500oC. ZrB 2 + SiC ZrO 2 + SiO 2 SiO 2 ZrO 2 + ZrB 2

PAGE 168

168 APPENDIX DYNAMIC INDENTATION RESPONSE OF B4C CERAMICS Figures A 1 and A 2 show the plots of c vs. P and l vs. P for dynamic indentations, respectively. Figure A 1 Variation of c vs. P in B4C ceramics for dynamic indentations. y = 10.813x 0.6228 y = 9.8157x 0.6327 y = 9.5952x 0.6902 10 15 20 25 30 35 40 45 50 55 60 65 70 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 c (m) P (N) 1.6 m 2.0 m 2.7 m

PAGE 169

169 Figure A 2 Variation of l vs. P in B4C ceramics for dynamic indentations. y = 6.1082x 0.6819 y = 5.9889x 0.6822 y = 6.1847x 0.7143 10 15 20 25 30 35 40 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 I (m) P (N) 1.6 m 2.0 m 2.7 m

PAGE 170

170 LIST OF REFERENCES 1M. J. Normandia, J. C. Lasalvia, W. A. Gooch, J. W. McCauley, and A. M. Rajendran, Protecting the Future Force: Ceramics Research Leads to Improved Armor Performance, AMPTIAC Quart. 8 [4] 217 (2004). 2W. G. Fahrenholtz, G. E. Hilmas, I. G. Talmy, and J. A. Zaykoski, Refractory Diborides of Zirconium and Hafnium, J. Am. Ceram. Soc. 90 [5] 13471364 ( 2007). 3W. D. Kingery, H. K. Bowen, and D. R. Uhlmann, Introduction to Ceramics Wiley, New York, 1976. 4M. W. Barsoum, Fundamentals of Ceramics McGraw Hill, New York, 1997. 5M. N. Rahaman, Ceramic Processing and Sintering, M. Dekker, New York, 2003. 6B. R. Klotz, K. C. Cho, and R. J. Dowding, Sintering Aids in the Consolidation of Boron Carbide (B4C) by the Plasma Pressure Compaction (P2C) Method, Mat er. Manuf. Process 19, 631639 (2004). 7D. P. II. Harvey, R. Kalyanaraman, and T. S. Sudarshan, Consolidation and Mechanical Behaviour of Nanocrystalline Iron Powder, Mater. Sci. Tech. 18, 95963 (2002). 8T. S. Srivatsan, R. Woods, M. Petraroli, and T. S. Sudarshan, An Investigation of the Influence of Powder Particle Size onMicrostructure and Hardness of Bulk Samples of Tungsten Carbide, Powder Tech. 122 [1] 5460 (2002). 9B. G. Ravi, O. A. Omotoye, T. S. Srivatsan, M. Petrorali, and T. S. Sudarshan, The Microstructure and Hardness of Silicon Carbide Synthesized by Plasma Pressure Compaction, J. of Alloys Compd. 299 [1] 2926 (2000). 10Thevenot, Boron CarbideA Compressi ve Review, J. Eur. Ceram. Soc. 6 205225 (1990). 11H. W. Kim, Y. H. Koh, and H. E. Kim, Densification and Mechanical Properties of B4C with Al2O3 as a Sintering Aid, J. Am. Ceram. Soc. 83 [11] 28632865 (2000). 12K. A. Schwetz, L. S. Sigl, and L. Pfa u, Mechanical Properties of Injection Molded B4C C Ceramics, J. Solid State Chem. 133, 6876 (1997). 13L. S. Sigl, Processing and Mechanical Properties of Boron Carbide Sintered with TiC, J. Eur. Ceram. Soc. 18, 15211529 (1998). 14Z. Zahariev and D. Radev, Properties of Polycrystalline Boron Carbide Sintered in the Presence of W2B5 without Pressing, J. Mat. Sci. Lett., 7 [7] 695696 (1988).

PAGE 171

171 15V. Skorokhod JR, M. D. Vlajic, and V. D. krstic, Mechanical Properties of Pressureless Sintered Boron Carbide Containing TiB2 Phase, J. Mat. Sci. Lett., 15 [15] 13371339 (1996). 16H. Lee and R. F. Speyer, Hardness and Fracture Toughness of Pressureless Sintered Boron Carbide (B4C), J. Am. Ceram. Soc. 85 [5] 12911293 (2002). 17Kaufmann, D. Cronin, M. worswick, G. Pageau, and A. Beth, Influence of Material Properties on the Ballistic Performance of Ceramics for Personal Body Armor, Shock and Vibration, 10, 5158 (2003). 18M. Chen, J. W. McCauley, and K. J. Hemker, Shock induce d Localized Amorphization in Boron Carbide, Science 299 [5612] 15631566 (2003). 19K. A. Iyer, Relationships Between Multiaxial Stress States and Internal Fracture Patterns in SphereImpacted Silicon Carbide, Int. J. Frac. 146 [1 2] 118 (2007). 20M. J. Normandia, Impact Response and Analysis of Several Silicon Carbides, Int. J. Appl. Ceram. Technol. 1 [3] 226234 (2004). 21T. J. Holmquist and G. R. Johnson, Response of Silicon Carbide to High Velocity Impact, J. Appl. Phys. 91 [9] 58585866 (2002). 22B. R. Lawn and A. G. Evans, Elastic Plastic Indentation Damage in Ceramics: the Median/ Radial Crack System, J. Am. Ceram. Soc. 63 [9 10] 574581 (1980). 23D. B. Marshall, B. R. Lawn and A. G. Evans, Elastic Plastic Indentation Damage in Cera mics: the Lateral Crack System, J. Am. Ceram. Soc. 65 [11] 561566 (1982). 24S. S. Chiang, D. B. Marshall and A. G. Evans, The Response of Solids to Elastic/Plastic Indentation: (a) Stresses and Residual Stresses and (b) Fracture Initiation, J. Appl. Phys. 53, 298317 (1982). 25R. F. Cook and G. M. Pharr, Direct Observation and Analysis of Indentation Cracking in Glasses and Ceramics, J. Am. Ceram. Soc. 73 [4] 787817 (1990). 26J. B. Quinn and G. D. Quinn, Indentation Brittleness of Ceramics: A Fresh Approach, J. Mater. Sci., 32, 43314346 (1997). 27G. Subhash, Dynamic Indentation Testing, ASM Handbook on Mechanical Testing and Evaluation, 8 519529 (2000). 28R. J. Anton and G. Subhash, Dynamic Vickers Indentation of Brittle Materials, Wear 2 39, 2735 (2000). 29M. M. Opeka, I. G. Talmy, and J. A. Zaykoski, Oxidationbased Materials Selection for 2000 C + Hypersonic Aerosurfaces: Theoretical Considerations and Historical Experience, J. Mater. Sci., 39, 58875904 (2004).

PAGE 172

172 30D. M. Van Wie, D. G. Drewry JR., D. E. King, and C. M. Hudson, The Hypersonic Environment: Required Operating Conditions and Design Challenges, J. Mater. Sci. 39, 59155924 (2004). 31T. A. Jackson, D. R. Eklund, and A. J. Fink, High Speed Propulsion: Performance Advantage of Advanced Materials, J. Mater. Sci., 39, 59055913 (2004). 32R. Telle, L. S. Sigl, and K. Takagi, Boride Based Hard Materials 802 945, Handbook of Ceramic Hard Materials, Edited by R. Riedel. Wiley VCH, Weinheim, Germany, 2000. 33NSF AFOSR Joint Wor kshop on Future UltraHigh Temperature Materials, Draft Workshop Report, National Science Foundation, April 12 (2004). 34R. Loehman, E. Corral, H. P. Dumm, P. Kotula, and R. Tandon, Ultrahigh temperature ceramics for hypersonic vehicle applications, S andia Report Sandia national Laboratories (SAND 2006 2925). 35G. Subhash, The Constitutive Behavior of h.c.p. Refractory Metals as a Function of Strain Rate Under Uniaxial Compression, JOM 47 [5] 5558 (1995). 36B. J. Koeppel and G Subhash, Influence of Cold Rolling and Strain Rate on Plastic Response of Powder Metallurgy and Chemical Vapor Deposition Rhenium, Metall. Mater. Transac. A 30 [10] 26412648 (1999). 37J. Han, P. Hu, X. Zhang, and S. Meng, Oxidation Behavior of Zirconium DiborideSilicon Carbide at 1800oC, Scripta Mater. 57 825828 (2007). 38A. Rezaie, W. G. Fahrenholtz, and G. E. Hilmas, Evolution of Structure During The Oxidation of Zirconium DiborideSilicon Carbide in Air up to 1500oC, J. Eur. Ceram. Soc. 27, 24952501 (2007). 39W. G. Fahrenholtz, Thermodynamic Analysis of ZrB2SiC Oxidation: Formation of a SiCDepleted Region, J. Am. Ceram. Soc. 90 [1] 143148 (2007). 40A. Rezaie, W. G. Fahrenholtz, and G. E. Hilmas, Oxidation of Zirconium DiborideSilicon Carbide at 1500oC at a Low Partial Pressure of Oxygen, J. Am. Ceram. Soc. 89 [10] 32403245 (2006). 41T. A. Parthasarathy, R. A. Rapp, M. Opeka, and R. J. Kerans, A Model for the Oxidation of ZrB2, HfB2 and TiB2, Acta Mater. 55, 59996010 (2007). 42A. L. Chamberlain, W. G. Fahrenholtz, and G. E. Hilmas, Pressureless Sintering of Zirconium Diboride, J. Am. Ceram. Soc. 89 [2] 450456 (2006). 43S. C. Zhang, G. E. Hilmas, and W. G. Fahrenholtz, Pressureless Densification of Zirconium Diboride with Boron Carbide Additions, J. Am. Ceram. Soc. 89 [5] 15441550 (2006).

PAGE 173

173 44 B. Cech, P. Olivierus, and J. Sejbal, Sintering of Zirconium Diboride with Activating Additions, Powder Metall. 8 [15] 142151 (1965). 45 D. Sciti, M. Brach, and A. Bellosi, Oxidat ion Behavior of a Pressureless Sintered ZrB2 MoSi2 C eramic Composite, J. Mater. Res. 20 [4] 922930 (2005). 46A. Rezaie, W. G. Fahrenholtz, and G. E. Hilmas, Effect of Hot Pressing Time and Temperature on the Microstructure and Mechanical Properties of ZrB2SiC, J. Mater. Sci. 42, 27352744 (2007). 47F. Monteverde, Beneficial Effects of an Ultra Fine SiC Incorporation on the Sinterability and Mechanical Properties of ZrB2, Appl. Phys. A, 82, 329337 (2006). 48H, Wang, C. A. Wang, X. Yao, and D. Fang, Processing and Mechanical Properties of Zirconium DiborideBased Ceramics Prepared by Spark Plasma Sintering, J. Am. Ceram. Soc ., 90 [7] 19921997 (2007). 49R. Telle, L. S. Sigl, and K. Takagi, Handbook of Ceramic Hard Materials Edited by R. Riedel. Wiley VCH, Weinheim, Germany, 802945, 2000. 50A. L. Chamberlain, W. G. Fahrenholtz, G. E. Hilmas, and D. T. Ellerby, High Strength ZrB2based Ceramics, J. Am. Ceram. Soc ., 87 [6] 170172 (2004). 51D. E Wiley, W. R. Manning, and O. Hunter, Elast ic Properties of Polycrystalline TiB2, ZrB2 and HfB2 f rom Room Temperature to 1300o K, J. Less Common Metals 18, 14957 (1969). 52J. J Melendez Martinez, A. Dominguez Rodriguez, F. Monteverde, C. Melandri, and G. de Portu G de, Characterisation and Hig h Temperature Mechanical Properties of Zirconium BorideBased Materials, J. Eur. Ceram. Soc. 22, 25432549 (2002). 53S. Zhu, W. G. Fahrenholtz, and G. E. Hilmas, Influence of Silicon Carbide Particle Size on the Microstructure and Mechanical Properties of Zirconium Diboride Silicon Carbide Ceramics, J. Euro. Ceram. Soc ., 27 [4] 20772083 (2007). 54A. Chamberlain, W. Fahrenholtz, G. Hilmas, and D. Ellerby, Oxidation of ZrB2SiC Ceramics under Atmospheric and Reentry Conditions, Refract. Appl. Trans., 1 [2] 1 8 (2005). 55R. Angers and M. Beauvy, Hot Pressing of Boron Carbide, Ceram. Int., 10 [2] 4955 (1983) 56R. Telle and G. Petzow, Mechanisms in the LiquidPhase Sintering of Boron Carbide with Silicon Based Melts, Mater. Sci. Monogr., 38A 961973 (1987). 57H. T. Larker, L. Hermansson, and J. Adlerborn, Hot Isostatic Pressing and Its Applicability to Silicon Carbide and Boron Carbide, Mater. Sci. Monogr. 38A 795803 (1987).

PAGE 174

174 58S. L. Dole, S. Prochazka, and R. H. Doremus, Microstructural Coarsening During Sintering of Boron Carbide, J. Am. Ceram. Soc. 72 [6] 958966 (1989). 59H. Lee and R. F. Speyer, Pressureless Sintering of Boron Carbide, J. Am. Ceram. Soc. 86 [9] 14681473 (2003). 60H. Lee, R. F. Speyer, and W. S. Hackenberger, Sintering of Boron Carbide Heat Treated with Hydrogen, J. Am. Ceram. Soc. 85 [8] 21312133 (1989). 61J. D. Katz, R. D. Blake, J. J. Petrovic, and H. Sheinberg, Microwave Sintering of Boron Carbide, Met. Powd. Rep. 43, 835837 (1988). 62O. Kida and Y. Segawa. U.S. Patent 4,636,481, January 13, 1987. 63O. Kida and Y. Segawa. U.S. Patent 4,678,759, July 7, 1987. 64D. Kalish, E. V. Clougherty, and K. Kreder, Strength, Fracture Mode, and Thermal Stress Resistance of HfB2 and ZrB2, J. Am. Ceram. Soc. 52 [1] 30 6 (1969). 65R. A. Andrievskii, L. A. Korolev, V. V. Klimenko, A. G. Lanin, I. I. Spivak, and I. L. Taubin, Effect of Zirconium Carbide and Carbon Additions on Some Physicomechanical Properties of Zirconium Diboride, Powder Metall. Metal Ceram. 1 9 [2] 93 4 (1980). 66D. Kalish and E. V. Clougherty, Densification Mechanisms in HighPressure Hot Pressing of HfB2, J. Am. Ceram. Soc. 52 [1] 26 30 (1969). 67R. A. Cutler, Engineering Properties of Borides, 787 803, Ceramics and Glasses: Engineered Materials Handbook, 4 Edited by S. J. Schneider Jr., ASM International, Materials Park, Ohio, 1991. 68A. L. Chamberlain, W. G. Fahrenholtz, and G. E. Hilmas, Reactive Hot Pressing of Zirconium Diboride, J. Eur. Ceram. Soc., In press (2009). 69X. J. Zhou, G. J. Zhang, Y. G. Li, Y. M. Kan, and P. L. Wang, Hot Pressed ZrB2SiCC Ultra High Temperature Ceramics with Polycarbosilane as a Precursor, Mater. Lett. 61, 960 963 (2007). 70G.J. Zhang, Z. Y. Deng, N. Kondo, J. F. Yang, and T. Ohji, Reactive Hot Pressing of ZrB2 SiC Composites, J. Am. Ceram. Soc. 83 [9] 2330 32 (2000). 71D. Ghosh, G. Subhash, T. S. Sudarshan, R. Radhakrishnan, and X L Gao, Dynamic Indentation Response of FineGrained Boron Carbide, J. Am. Ceram. Soc. 90 [6] 18501857 (2007). 72D. Ghosh, G. Subhash, R. Radhakrishnan, and T. S. Sudarshan, Scratch induced Microplasticity and Microcracking in Zirconium Diboride Silicon Carbide Composite, Acta Mater. 56, 3011 3022 (2008).

PAGE 175

175 73C. E. Inglis, Stresses in a Plate Due to the Presence of Cracks and sharp corners, Trans. Inst. Nav. Architects (London), 55, 219 241 (1913). 74A. A. Griffith, The Phenomena of Rupture and Flow in Solids, Philos. Trans. R. Soc. London, 221, 163198 (1920). 75G. R. Irwin, Fracture Dynamics: In Fracturing of Metals American Society of Metals, Cleveland, 1948. 76E. Orowan, Fracture Strength of Solids In Report on Progress in Physics, Physical Society of London, London, 12, 185232 (1949). 77F. C. Frank and B. R. Lawn, On the Theory of Hertzian fracture Proc. Roy. Soc. Lond., A299: 291 (1967). 78B. R. Lawn, Fracture of Brittle Solids Cambridge University Press Cambridge, 1995. 79G. D. Quinn and R. C.Bradt, On the Vickers Indentation Fracture Toughness Test, J. Am. Ceram. Soc. 90 [3] 673680 (2007). 80D. Casellas, J. Caro, S. Molas, J. M. Prado and I. Valls, Fracture Toughness of Carbides in Tool Steels Evaluated by Nanoindentation, Acta Mater., 55, 42774286 ( 2007). 81H. Hertz, On the Contact of Elastic Solids (in Ger.) Zeitschriftfr die Reine und Angewandtemethematik 92, 156171 (1881). English translation in Miscellaneous Papers (translated by D.E. Jones and G. A. Schott); pp. 14662. Macmillan, London, U K, 1896. 82M. A. Meyers, Dynamic Behavior of Materials John Wiley & Sons, Inc., New York 1994. 83J. Lankford, Mechanisms Responsible for StrainRate Dependent Compressive Strength in Ceramic Materials, J. Am. Ceram.Soc. 64, 33 34 (1981). 84G. Ravichandran and G. Subhash, A Micro Mechanical Model for High Strain Rate Behavior of Ceramics, I nt. J. of Solids and Struc., 32, 26272650 (1995). 85G. Subhash and G. Ravichandran, Mechanical Behavior of Hot Pressed Aluminum Nitride Under Uniaxial Compression J. Mater. Sci., 33, 19331939 (1997). 86H. Wang and K. T. Ramesh, Dynamic Strength and Fragmentation of Hot Pressed Silicon Carbide Under Uniaxial Compression, Acta Mater., 52, 355367 (2003). 87G. Ravichandran and G. Subhash, Critical Appraisal of Li miting Strain Rates for Compression Testing of Ceramics in a Split Hopkinson Pressure Bar, J. Am. Ceram. Soc. 77, 263 267 (1994).

PAGE 176

176 88W. Chen and G. Ravichandran, Static and Dynamic Compressive Behavior of Aluminum Nitride Under Moderate Confinement, J. Am. Ceram.Soc. 79, 579584 (1996). 89W. Chen and G. Ravichandran, Failure Mode Transition in Ceramics Under Dynamic Multiaxial Compression, Int. J. Fract., 101, 141159 (2000). 90J. Lankford, W. W. Predebon, J. M. Staehler, G. Subhash, B. J. Pletka C. E Anderson, The Role of Plasticity as Limiting Factor in the Compressive Failure of High Strength Ceramics, Mech. Mater. 29, 205218 (1998). 91S. Suresh, T. Nakamura, Y. Yeshurun, K. H. Yang and J. Duffy, Tensile Fracture Toughness of Ceramic Materials: Effects of Dynamic Loading and Elevated Temperatures, J. Am. Ceram. Soc 73, 24572480 (1990). 92K. H. Yang and A. S. Kobayashi, A Hybrid Procedure for Dynamic Characterization of Ceramics at Elevated Temperatures, In Proc. Conf. Hostile Environments and High Temperature Measurements SEM. 41 44 (1989). 93T. Weerasooriya, P. Moy, D. Casem, M. Cheng and W. Chen, A Four Point Bend Technique to Determine Dynamic Fracture Toughness of Ceramics, J. Am. Ceram. Soc. 89 [3] 990995 (2006). 94W. Bhme and J. F. Kalthoff, The Behavior of Notched Bend Specimens in Impact Testing, Int. J. Fract. 20, R13943 (1982). 95D. Rittel, H. Mairge and H. D. Bui, A New Method for Dynamic Fracture Toughness Testing, Scr. Metall. Material., 26 [10] 15938 (1992). 96C. Bacon, J. Farm and J. L. Lataillade, Dynamic Fracture Toughness Determined from LoadPoint Displacement, Exp. Mech. 34 [3] 21723 (1994). 97W. N. Sharpe and W. Bhme, Dynamic Fracture Toughness Measurements on Small Charpy Specimens A Preliminary Study, J. Test. Eval. 22 [1] 149 (1994). 98C. E. Anderson Jr., C. H. Popelar, A. Nagy and J. D. Walker, A Novel Method for Determining Dynamic Fracture Toughness, 505508, Shock Compression of Condensed Matter 1999, Edited by M. D. Furnish, L. C. Chhabidas, and R. S. Hixson. American Institute of Physics, College Park, MD, 2000. 99G. Weisbrod and D. Rittel, A Method for Dynamic Fracture Toughness Determination Using Short Beams, Int. J. Fract. 104 [1] 89103 (2000). 100C. F. Martins and V. Prakash, Dynamic Fracture Characteristics of LMDPE Utilizing a Modified Split Hopkinson B ar, Proceedings of 2002 Society of Experimental Mechanics Annual Conference Milwaukee, WI, 2002.

PAGE 177

177 101J. Lankford Jr., The Role of Dynamic Material Properties in the Performance of Ceramic Armor, Int. J. Appl. Ceram.Tech. 1 [3] 20510 (2004). 102J.C. La Salvia, M.J. Normandia, H.T. Miller and D.E. MacKenzie, Sphere Impact Induced Damage in Ceramics: I. Armor Grade SiC and TiB2, Advances in Ceramic Armor edited by J. J. Swab. Proceedings of the 29th International Conference on Advanced Ceramics and Comp osites Cocoa Beach, Florida, (2005). 103C. Roberson and P. J. Hazell, Resistance of Silicon Carbide to Penetration by a Tungsten CarbideCored Projectile, in Ceramic Armor and Armor Systems, ed. E. Medvedovski, Ceramics Transsactions 151, 165174 (2003). 104I. M. Pickup, A. K. Barker, R. Chenari, B. J. James, V. Hohler, K. Weber, and R. Tham, Aspects of Geometry Affecting the Ballistic Performance of Ceramic Targets, Ceramic Armour Materials by Design, Edited by J. W. McCauley, A. Crowson, W. A. Gooch, Jr., A. M. Rajendran, S. J. Bless, K. V. Logan, M. Normandia, S. Wax, Ceramic Transactions, 134, 643650, 2002. 105D. L. Orphal, C. E. Anderson, Jr., D. W. Tempelton, Th. Behner, V. Hohler and S. Chocron, Using Long Rod Penetration to Detect the Effect of Failure Kinetics in Ceramics, Proc. 21st. Int. Symp. Ballistics 2 744751 (2004). 106A. M. Rajendran, Modeling the Impact Behavior of a Ceramic under Multiaxial Loading, Int. J. Impact Eng ., 15, 749768 (1994). 107J. C. LaSalvia, E. J. Horwath, E. J. Rapacki, C. J. Shih and M. A. Meyers, Microstructural and Micromechanical Aspects of Ceramic/LongRod Projectiles Interactions: Dwell/Penetration Transitions, In Fundamental Issues and Applications of Shock Wave and HighStrain Rate Phenomena, pp.437446, edited by K. P. Staudhammer, L. E. Murr and M. A. Meyers, Elsevier Science, New York, 2001. 108T. J. Holmquist and G. R. Johnson, A Detailed Computational Analysis of Interface Defeat, Dwell and Penetration for a Variety of Ceramic Targets, Proc. 2 0th. Int. Symp. Ballistics 2 746753 (2002). 109T. J. Holmquist and G. R. Johnson, Modelling Projectile Impact onto Prestressed Ceramic Targets, J. Phys. IV France, 110, 597602 (2003). 110T. J. Holmquist and G. R. Johnson, Modelling PreStressed Cer amic and its Effect on Ballistic Performance, Int. J. Impact Eng., 31 [2], 113 127 (2005). 111R. Renstrm, P. Lundberg and B. Lundberg, Stationary Contact between a Cylindrical Metallic Projectile and A Flat Target Under Conditions of Dwell, Int. J. Impact Eng. 30, 12651282 (2004). 112P. Lundberg, R. Renstrom and L. Holmberg, An Experimental Investigation of Interface Defeat at Extended Interaction Time, 19th International Symposium of Ballistics May, Interlaken Switzerland (2001).

PAGE 178

178 113D. A. Shock ey, A. H. Marchand, S. R. Skagg, G. E. Cort, M. W. Burkett and R. Parker, Failure Phenomenological of Confined Ceramic Targets and Impacting Rods, Int. J. Impact Egg. 9 [3], 26375 (1990). 114D. Sherman, Impact Failure Mechanisms in Alumina Tiles on Fi nite Thickness Support and the Effect of Confinement, Int. J. Impact Eng. 24, 312 328 (2000). 115T. J. Holmquist, A. M. Rajendran, D. W. Templeton and K. D. Bishnoi, A Ceramic Armor Material Database, TARDEC Technical Report # 13754 (1999). 116G. E. Hauver, E. J. Rapacki, P. H. Netherwood, R. F. Benck, ARL TR 3590, September (2005). 117G. E. Hauver, P. H. Netherwood, R. F. Benck and L. J. Kecskes, Ballistic Performance of Ceramic Targets Proceedings of 13th Army Symposium on Solid Mechanics E ds. S. C. Chou, F. D. Bartlett Jr., T. W. Wright and K. Iyer, 23 34, 1993. 118J. C. LaSalvia, A Physically Based Model for the Effect of Microstructure and Mechanical Properties on Ballistic Performance, 26th Annual Inter. Conf. Adv. Ceram.& Comp. Cocoa Beach, January 1318, Florida, (2002). 119J. C. Lasalvia, E. J. Horwath, E. J. Rapacki, C. J. Shih and M. A. Meyers, Microstructural and Micromechanical Aspects of Ceramic/LongRod Projectile Interactions: Well/Penetration Transition, Explomet 2000, Alb uquerque, New Mexico June 1822 (2000). 120B. J. Koeppel and G. Subhash, An Experimental Technique to Investigate the Dynamic Indentation Hardness of Materials, Exp. Tech., USA 21 [3], 16 18 (1997). 121B. J. Koeppel and G. Subhash, Characteristics of Residual Plastic Zone Under Static and Dynamic Indentations Wear 224, 5667 (1999). 122G. Subhash, B. J. Koeppel and A. Chandra, Dynamic Indentation Hardness and Rate Sensitivity in Metals, J. Eng. Mate r. Technol. 121, 1 7 (1999). 123S. Nemat Nasser, J. B. Isaacs, and J. E. Starrett, Hopkinson Techniques for Dynamic Recovery Experiments, Proc. R. Soc. (London) A 435, 371391 (1991). 124H. W. Zhang, S. Maiti, and G. Subhash, Evolution of Shear Bands in Bulk Metallic Glasses under Dynamic Loading, J. Mech. Phys. Solids 56 [6] 21712187 (2008). 125G. Subhash and H. W. Zhang, Dynamic Indentation Response of ZrHf Based Bulk Metallic Glasses, J. Mater. Res. 22 [ 2] 478 485 ( 2007). 126D. Tabor, Hardness of Metals Clarendon Press, Oxford, 1951.

PAGE 179

179 127H. Li, G. Subhash, X. L. Gao, L. J. Kecskes, and R. J. Dowding, Negative Strain Rate Sensitivity and Compositional Dependence of Fracture Strength in Zr/Hf Based Bulk Metallic Glasses, Scripta Mater., 49 [11] 10871092 (2003). 128C. B. Ponton and R. D. Rawlings, Vickers Indentation Fracture Toughness Test Part 1 Review of Literature and Formulation of standardized Indentation Toughness equations. Mater. Sci. and Tech. 5 865872 (1989). 129R. W. R ice, Hardness Grain Size Relations in Ceramics, J. Am. Ceram. Soc. 77 [10] 253953 (1994). 130R. W. Rice, Mechanical Properties of Ceramics and Composites: Grain and Particle Effects, CRC 1st edition May 15 (2000). 131D. Ge, V. Domnich, A. Juliano, E. A. Stach, and Y. Gogotsi, Structural Damage in Boron Carbide under Contact Loading, Acta. Mater., 52 39213927 (2004). 132X. Q. Yan, W. J. Li, T. Goto, and M.W. Chen, Raman Spectroscopy of Pressure Induced Amorphous Boron Carbide, Appl. Phys. Lett. 88, 1319051 1319051 (2006). 133G. Fanchini, J. W. McCauley, and M. Chhowalla, Behavior of Disordered Boron Carbide under Stress, Phys. Rev. Lett. 97, 0355021 0355024 (2006). 134R. Lazzarari, N. Vast, J. M. Besson S. Baroni and A. D. Corso, Atomic Structure and Vibrational Properties of Icosahedral B4C Boron Carbide, Phys. Rev. Lett. 83, 32303233 (1999). 135I. Jimenez, D. G. J. Sutherland, T. v. Buuren, J. A. Carlisle, L. J. Terminello, and F. J. Himpsel, Ph otoemission and X ray absorption Study of Boron Carbide and its Surface Thermal Stability, Phys. Rev. B. 57 13167131174 (1998). 136E. Smith and G. Dent, Modern Raman Spectroscopy A Practical Approach, John Wiley & Sons, Ltd., Clarendon Press, England, 2005. 137E. G. Brame Jr. and J. Grasselli, Infrared and Raman Spectroscopy (Part A), Marcel Dekker, Inc, New York, 1976. 138http://en.wikipedia.org/wiki/Photoluminescence 139 Y. K. Yap, S. Kida, T. Aoyama, Y. Mori and T. Sasaki, Influence of Negative dc Bias Voltage on Structural Transformation of Carbon Nitride at 600 C, Appl. Phys. Lett. 73, 915917 (1998). 140Y. K. Yap, M. Yoshimura, Y. Mori and T. Sasaki, Hybridized Bo ron Carbon Nitride Fibrous Nanostructures on Ni Substrates, Appl. Phys. Lett. 80, 25592561 (2002).

PAGE 180

180 141M. Ramsteiner, and J. Wanger, Resonant Raman Scattering of Hydrogenated Amorphous Carbon: Evidence for pi Bonded Carbon Clusters, Appl. Phys. Lett. 51, 1355 (1987). 142A. C. Ferrari and J. Robertson, Resonant Raman Spectroscopy of Disordered, Amorphous, and Diamondlike Carbon, Phys. Rev. B, 64 0754141 07541413 (2001). 143A. C. Ferrari and J. Robertson, Interpretation of Raman Spectra of Disordered and Amorphous Carbon, Phys. Rev. B 61, 14095 (2000). 144H. Werheit, On Excitons and Other Gap States in Boron Carbide, J. Phys. Conds. Mater. 18, 1065510662 (2006). 145J. Xu, J. Mei, X. Huang, W. Li, Z. Li, X. Li, and K. Chen, The change of Photoluminescence Characteristics of Amorphous Carbon Films due to Hydrogen Dilution, J. Non Cryst. Solids. 338340, 481485 (2004). 146H. Werheita, T. Aua, R. Schmechela, S. O. Shalamberidzeb, G. I. Kalandadzeb, and A. M. Eristavib, IR Active Phonons and Structure Elements of IsotopeEnriched Boron Carbide, J. Solid State Chem. 154, 79 (1999). 147C. L. Beckel, M. Yousaf, M. Z. Fuka, S. Y. Raja, and N. Lu, Lattice Vibrations of the Boron, Phys. Rev. B 44, 25352553 (1991). 148K. Shirai, S. Emura, and S i. Gonda, Infrared Study of Amorphous B1XCX films, J. Appl. Phys. 78, 33923400 (1995). 149H. Ikeda, Y. Qi, T. Cagin, K. Samwer, W. L. Johnson, and W. A. Goddard III, Strain Rate Induced Amorphization in Metallic Nanowires, Phys. Rev. Letts. 82, 29002903 (1999). 150Y. Ahn, T. N. Farris, and S. Chandrasekar, Sliding Microindentation Fracture of Brittle Materials: Role of Elastic Stress Fields, Mech. Mater. 29 [11] 143152 (1998). 151X. Jing, G. Subhash, and S. Maiti, A New Analytical Model for Estimation of Scratch Induced Damage in B rittle Solids J. Am. Ceram. Soc. 90 [3] 885892 (2007). 152H. Wang and G. Subhash, Mechanics of Mixedmode Ductile Material Removal with a Conical Tool and the Size Dependence of the Spec ific Energy, J. Mech. Phys. Solids. 2002;50:1269 1296. 153G. Subhash and R. Bandyo, A New Scratch Resistance Measure for Structural Ceramics, J. Am. Ceram. Soc ., 88 918 (2005).

PAGE 181

181 154G. Subhash, M. A. Marszalek, and S. Maiti, Sensitivity of Scratch Resistance to GrindingInduced Damage Anisotropy in Silicon Nitride, J. Am. Ceram. Soc 89, 25282536 (2006). 155G. Subhash and M. Klecka, Ductile to Brittle Transition Depth during Singl e grit Scratching on Alumina Ceramics J. Am. Ceram. Soc. 90 37043707 (2007). 156H. Wang, G. Subhash, and A. Chandra, Characteristics of Singlegrit Rotating Scratch with a Conical Tool on Pure Titanium, Wear 249 566581 (2001). 157G. Subhash, J. E. Loukus, and S. M. Pandit, Applications of Data Dependent Systems Approach for Evaluation of Fracture Modes during a Singlegrit Scratching, Mech. Mater ., 34, 25 42 (2002). 158H. Wang and G. Subhash, Mechanics of Mixedmode Ductile Material Removal with a Conical Tool and the Size Dependence of the Specific Energy, J. Mech. Phys. Solids 50, 12691296 (2002). 159J. S. Haggerty and D.W. Lee, Plastic Deformation of ZrB2 Single Crystals, J. Am. Ceram. Soc ., 54, 572576 (1971). 160D. G hosh, G. Subhash, and G. R. Bourne, Inelastic Deformation under Indentation and Scratch Loads in a ZrB2 SiC Composite, J. Eur. Ceram. Soc. 29, 3053 3061 (2009). 161J. P. Herth and J. Lothe, Theory of Dislocations Krieger Publishing Company, Florida, 19 82. 162M. H Yoo, J. R. Morris K. M. Ho, and S. R. Agnew, Nonbasal Deformation Modes of HCP Metals and Alloys: Role of Dislocation Source and Mobility, Met. and Mat. Trans A 33A 813822 (2002). 163G. Subhash, G. Ravichandran, and B. J. Pletka, Plast ic Deformation of Hafnium under Uniaxial Compression, Metall. Trans. A 28A 14791487 (1997). 164R. F. Firestone and A. H. Heuer, Creep Deformation of 0 sapphire, J. Am.Ceram. Soc. 59 [1 2] 2429 (1976). 165J. D. O Barceinss saanchez and W. M. Rainforth, On the Role of Plastic Deformation during the Mild Wear of Alumina, Acta Mater., 46 [18] 64756483 (1998). 166S. Rohmfeld, M. Hundhausen, and L. Ley, Raman Scattering in Polycrystalline 3 C SiC: Influence of Stacki ng Faults, Phys. Rev. B 88 [15] 98589862 (1998). 167K. L. Johnson, Contact Mechanics Cambridge University Press, Cambridge, 1985. 168E. H. Yoffe, Elastic Stress Fields Caused by Indenting Brittle Materials, Phil. Mag. A 46 [4] 617628 (1982).

PAGE 182

182 169J. Lankford, W. W. Predebon, J. M. Staehler, G. Subhash, B. J. Pletka, and C. E. Anderson, The Role of Plasticity as a Limiting Factor in the Compressive Failure of High Strength Ceramics, Mech. Mater ., 29, 205218 (1998). 170T. L. Johston, R. S. Stokes, and C. H. Li, Crack Nucleation in Magnesium Oxide Bi crystals under Compression, Phil. Mag. 7 23 (1962). 171R. W. Rice, Mechanical Properties of Ceramics and Composites Marcel Dekker Inc., New York, 2000. 172F. P. Bowden and D. T. Tabor, Friction and Lubrication of Solids Clarendon Press, Oxford, 1950. 173S. V. Prasad and K. R. Mecklenburg, Friction Behavior of Ceramic Fiber reinforced Aluminum Metal matrix Composites Against a 440C Steel Counterface, Wear 162164, 4756 (1993). 174J. A. Ruan and B. Bhushan, Atomic scale and Microscale Friction Studies of Graphite and Diamond using Friction Force Microscopy, J. Appl. Phys. 76, 50225035 (1994). 175F. Varnik, L. Bocquet, J. L. Barrat, and L. Berthier, Shear Localization in a Model Glass, Phys. Rev. Lett. 90, 0957021 0957024 (2003). 176K. Li, Y. Shapiro, and J. C. M. Li, Scratch Test of Sodalime Glass, Acta Mater., 46, 55695578 (1998). 177R. Smith, D. Mulliah, S. D. Kenny, E. McGee, A. Richter, and M. Gruner, Stick Slip and Wear on Metal Surfaces, Wear 259, 459 466 (2005). 178C. Gao, D. KuhlmannWilsdorf, and D. D. Makel, Fundamentals of Stick slip, Wear 162164, 11391149 (1993). 179D. Mulliah, S. D. Kenny, and R. Smith, Modeling of Stick slip Phenomena using Molecular Dynamics, Phys. Rev. B 69, 2054071 2054078 (2004). 180D. Mulliah, S. D. Kenny, R. Smith, and C. F. Sanz Navarro, Molecular Dynamic Simulations of Nanoscratching of Silver (100), Nanotechnology 15, 243249 (2004). 181D. Mulliah, S. D. Kenny, E. McGee, R. Smith, A. Richter, and B. Wolf, Atomistic Modelling of Ploughing Friction in Silver, Iron and Silicon, Nanotechnology 17, 18071818 (2006). 182E. Anastassakis, A. Pinczuk, E. Burstein, F. H. Pollak, and M. Cardona, Effect of Static Uniaxial Stress on the Raman Spectrum of Silicon, Solid State Commun. 8 133138 (1970).

PAGE 183

183 183Y. Kang, Y. Qiu, Z. Lei, and M. Hu, An Application of Raman Spectroscopy on the Measu rement of Residual stress in Porous Silicon, Optics. Laser Eng. 43, 847 855 ( 2005) 184N. Orlovskaya, D. Steinmetz, S. Yarmolenko, D. Pai, J. Sankar, and J. Goodenough, Detection of Temperatureand Stress induced Modifications of LaCoO3 by micro Raman Spectroscopy, Phys. Rev. B 72, 0141221 0141227 (2005). 185J. K. Shin, C. S. Lee, K. R. Lee, and K. Y. Eun, Effect of Residual Stress on the RamanSpectrum Analysis of Tetrahedral Amorphous Carbon Films, Appl. Phys. Lett. 78, 631 6 33 (2001). 186N. Orlovskaya, J. Kuebler, V. Subbotin, and M. Lugovy, Design of Si3N4based Ceramic Laminates by the Residual Stresses, J. Mater. Sci., 40, 54435450 (2005). 187Y. Gogotsi, G. Zhou, S. S. Ku, and S. Cetinkunt, Raman Microspectroscopy Anal ysis of Pressureinduced Metallization in Scratching of Silicon, Semicond. Sci. Technol. 16, 345352 (2001). 188I. De Wolf, Micro Raman Spectroscopy to Study Local Mechanical Stress in Silicon Integrated Circuits, Semicond. Sci. Technol. 11, 139154 (1996). 189J. Chen and I. De Wolf, Theoretical and Experimental Raman Spectroscopy Study of Mechanical Stress Induced by Electronic Packaging, IEEE Transac. Compo. Pack. Tech. 28, 484492 (2005). 190H. Mukaida, H. Okumura, J. H. Lee, H. Diamon, E. Sakum a, S. Misawa, K. Endo, and S. Yoshida, Raman scattering of SiC: Estimation of the Internal Stress in 3C SiC on Si, J. Appl. Phys. 62, 254256 (1987). 191S. Rohmfeld, M. Hundhausen, and L. Ley, Raman Scattering in Polycrystalline 3 C SiC: Influence of St acking Faults, Phys. Rev. B. 88, 98589862 (1998). 192Z. C. Feng, W. J. Choyke, and J. A. Powell, Raman Determination of Layer Stresses and Strains for Heterostructures and its Application to the Cubic SiC/Si System, J. Appl. Phys. 64, 68276835 (1988 ). 193C. M. Su, M. Wuttiga A. Fekade, and M. Spencer, Elastic and Anelastic Properties of Chemical Vapor Deposited Epitaxial 3C SiC, J. Appl. Phys. 77, 56115615 (1995). 194D. N. Talwar and J. C. Sherbondy, Thermal Expansion Coefficient of 3C SiC, Appl. Phys. Lett. 67, 33013303 ( 1995). 195H. J. Choi and J. G. Lee, Stacking Faults in Silicon Carbide Whiskers, Ceram. Inter., 26 [1] 7 12 (2000). 196C. J. Shih, M. A. Meyers, V. F. Nesterenko, and S. J. Chen, Damage Evolution in Dynamic Deformation of Silicon Carbide, Acta Mater. 48 23992420 (2000).

PAGE 184

184 197K. K. Chawla, Ceramic Matrix Composites Kluwer Academic Publishers, Massachusetts. 198F. Cerdeira, C. J. Buchenauer, F. H. Pollak, and M. Cardona, Stress Induced Shifts of First Order Raman Frequencies of Diamondand Zinc BlendeType Semiconductors, Phys. Rev. B 5 580593 (1972). 199J. S. Mitchell, C. A. Zorman, T. Kicher, S. Roy, and M. Mehre gany, Examination of Bulge Test for Determining Residual Stress, Young's Modulus, and Poisson's Ratio of 3C SiC Thin Films, J. Aero. Eng. 16, 4654 (2003). 200M. Miura, H. Murata, Y. Shiro, and K. Lishi, Ionicity Scale and Piezoelectricity of Crystals w ith Zincblendeand Wurtzitetype Structure, J. Phys. Chem. Solids. 42 [10] 931936 (1981). 201D. S. Knight and W. B. White, Characterization of Diamond Films by Raman Spectroscopy, J. Mater. Res. 4 385393 (1989). 202J. G. Kim and J. Yu, Measurement of Residual Stress in Diamond Films Obtained Using Chemical Vapor Deposition, Jpn. J. Appl. Phys. 37, L890L893 (1998). 203C. A. Taylor, M. F. Wayne, and W. K. S. Chiu, Residual Stress Measurement in Thin Carbon Films by Raman Spectroscopy and Nanoindent ation, Thin Solid Films 429, 190200 (2003). 204G. D. Zhan, R. J. Xie, and M. Mitomo, Effect of beta to alpha Phase Transformation on the Microstructural Development and Mechanical Properties of FineGrained Silicon Carbide Ceramics, J. Am. Ceram. Soc. 84 [5] 945950 (2001). 205Y. Zhou, K. Hirao, M. Toriyama, Y. Yamauchi, and S. Kanzaki, Effects of Intergranular Phase Chemistry on the Microstructure and Mechanical Properties of Silicon Carbide Ceramics Densified with RareEarth Oxide and Alumina Addit ions, J. Am. Ceram. Soc. 84 [7] 16421644 (2001). 206Y. W. Kim, M. Mitomo, H. Emoto, and J. G. Lee, Effect of Initial alpha Phase Content on Microstructure and Mechanical Properties of Sintered Silicon Carbide, J. Am. Ceram. Soc. 81 [12] 31363140 (1998). 207H. D. Espinosa, B. Peng, N. Moldovan, T. A. Friedmann, X. Xiao, D. C. Mancini, O. Auciello, J. Carlisle, C. A. Zorman, and M. Merhegany, Elasticity, Strength, and Toughness of Single Crystal Silicon Carbide, Ultrananocrystalline Diamond, and Hydrogenfree Tetrahedral Amorphous Carbon, Appl. Phys. Letts. 89, 0731111 0731113 (2006). 208J. W. III Ager and M. D. Drory, Quantitative Measurement of Residual Biaxial Stress by Raman Spectroscopy in Diamond Grown on a Ti Alloy by Chemical Vapor Deposition, Phys. Rev. B 48, 26012607 (1993).

PAGE 185

185 209E. Anastassakis, A. Cantarero, and M. Cardona, PiezoRaman Measurements and Anharmonic Parameters in Silicon and Diamond, Phys. Rev. B 41, 75297535 (1990). 210R. C. Teixeira, I. Doi, M. B. P. Zakia, J. A. Diniz, and J. W. Swart, Micro Raman Stress Characterization of Polycrystalline Silicon Films Grown at High Temperature, Mater. Sci. Eng. B 112 [2 3] 160164 (2004). 211D. Ghosh, G. Subhash and G. R. Bourne, Room temperature Dislocat ion Activity during Mechanical Deformation of Polycrystalline Ultra Hightemperature Ceramics, Scripta Mater. In press (2009). 212D. Grabco, B. Pushcash, M. Dyntu and O. Shikimaka, Thermal Evolution of Deformation Zones around Microindentations in Differ ent Types of Crystal, Phil. Mag. A. 82, 22072215 (2002). 213L. A. Giannuzzi and F. A. Stevie, Introduction to Focused Ion Beams: Instrumentation, Theory, Techniques and Practice Springer, New York, 2005, pp. 201 212. 214Y. Xuan, C. H. Chen, and S. Otani High Temperature Microhardness of ZrB2 Single Crystals, J. Phys. D: Appl. Phys. 35, L98L100 (2002). 215K. Nakano, T. Imura, and S. Takeuchi, Hardness Anisotropy of Single Crystals of IVa Diborides, Jap. J. Appl. Phys. 12, 186189 (1973). 216W. D. Kingery, H. K. Bowen, and D. R. Uhlmann, Introduction to Ceramics 2nd ed., John Wiley & Sons, New York. 217C. C. Wang, S. A. Akbar, W. Chen, and V. D. Patton, Electrical Properties of Hightemperature Oxides, Borides, Carbides, and Nitrides, J. Mater. Sci 30, 16271641 (1995). 218M. A. Meyers and K. K. Chawla, Mechanical Behavior of Materials Prentice Hall, Inc, New Jersey. 219R. G. Munro, Material Properties of Titanium Diboride, J. Res. Nat. Ins. Std. Tech., 105, 709720 (2000). 220F. Thevenot, Boron Carbide A Comprehensive Review, J. Eur. Ceram. Soc. 6 205225 (1990). 221M. W. Barsoum, Fundamentals of Ceramics McGraw Hill. 222M. M. R. Howlader, C. Kinoshita, T. Izu, K. Shiiyama, and M. Kutsuwada, In Situ Measurement of Elect rical Conductivity of Alumina under Electron Irradiation in a High Voltage Electron Microscope, J. Nuclear Mater. 239, 245252 (1996).

PAGE 186

186 223G. Subhash, G. Ravichandran, and B. J. Pletka, Plastic Deformation of Hafnium under Uniaxial Compression, Metall. Mater. Trans., A 28A 14791487 (1997). 224R. G. Duan, J.D. Kuntz, J.E. Garay, and A.K. Mukherjee, Metal like Electrical Conductivity in Ceramic Nanocomposite, Scripta Mater., 50, 13091313 (2004). 225W. D. Callister, Materials Science and Engineering An Introduction, 5th edition, John Wiley & Sons, Inc, New York. 226P. Vajeeston, P. Ravindran, C. Ravi, and R. Asokamani, Electronic Structure, Bonding, and Groundstate Properties of AlB2type Transitionmetal Diborides, Phy. Rev. B, 63, 0451151 04511512 (2001). 227X. Zhang, X. Luo, J. Han, J. Li, and W. Han, Electronic Structure, Elasticity and Hardness of Diborides of Zirconium and Hafnium: First Principles Calculations, Comp. Mater. Sci., 44, 411421 (2008). 228J.J. Gilman, Why Si Is Hard, Science 261, 14361439 (1993). 229J.J. Gilman, Chemical Theory of Dislocation Mobility, Mater. Sci. Eng. A 409 7 12 (2005). 230R. W. Davidge, Mechanical Behaviour of Ceramics Cambridge University Press; 1st Ed., 1979. 231A. H. Heuer and L. W. Hobbs, Science and Technology of Zirconia, Advances in Ceramics, 3 The American Ceramics Society, Westerville, Ohio, 1981. 232A. Taylo and D. S Laidler, The Formation and Crystal Structure of Silicon Carbide, British J. App. Phys. 1 174181 (1950). 233S. N. Ruddlesden and P. Popper, On the Crystal Structure of the Nitrides of Silicon and Germanium, Acta Cryst. 11, 465468 (1958). 234D. Ghosh, G. Subhash, C. H. Lee, and Y. K. Yap, Straininduced Formation of Carbon and Boron Clusters in Boron Carbide during Dynamic Indentation, Appl. Phys. Lett. 91, 0619101 0619103 (2007). 235G. Subhash,S. Maiti, P. H. Geubelle, and D. Ghosh, Recent Advances in Dynamic Indentation Fracture, Impact Damage and Fragmentation of Ceramics, J. Am. Ceram. Soc. 91 [ 9] 2777 2791 (2008). 236D. Ghosh, G. Subhash, and N. Orlovskaya, Measurement of Scratch induced Residual Stress within SiC Grains in ZrB2SiC Composite using MicroRaman spectroscopy, Acta Mater. 56, 53455354 (2008).

PAGE 187

187 BIOGRAPHICAL SKETCH Dipankar Ghosh was born in Kolkata, West Bengal India. He completed his high school education in 1996 from Sodepur High School in India. He earned his B.S. in Chemistry in 1999 and B.Tech. in Chemical Technology (with specialization in Ceramic Engineering) in 2002 from the University of Calcutta in India. Then, for higher education, he joined Indian Institute of Technology (IIT), Kharagpur (India) where he received M.Tech. in Materials Science and Engineering in 2004. During this period, apart from learning basic s of Mat erials Science and Advanced Ceramics he also conducted research on Colloidal processing of dense and porous ceramics. Upon graduating from IIT, to pursue further higher education, Mr. Ghosh came to United States and star ted his PhD at Michigan Technological University (MTU) Houghton, in 2004 under the supervision of Prof. Ghatu Subhash. At MTU he spent few years as a graduate student in Materials Science and Engineering Department and then he joined the U niversity of Florida, Gainesville, in 2007 where he continued his PhD research under the supervision of Prof. Ghatu Subhash in the Department of Mechanical and Aerospace Engineering. Mr. Ghosh successfully completed his PhD dissertation in 2009. His PhD dissertation has been focused on proces sing and mechanical characterization of advanced structural ceramics. He conducted research on processing of hightemperature ceramics and characterized the processed ceramics using several mechanical and spectroscopic techniques to understand the fundamental deformation mechanisms. He has published 10 peer reviewed papers in prestigious international journals and 4 conference proceedings. Based on his PhD research, he received Second Prize Best Paper Awards at 32nd International Conference on Advanced

PAGE 188

188 C eramics & Composites, American Ceramic Society, Deytona Beach, FL (2008). Mr. Ghosh along with five other graduate engineer students from the School of Engineering, UF, were acknowledged in the Fall, 2009 issue of The Florida Engineer for their significa nt contributions in the respective research areas. He also received an International Student Outstanding Achievement award at 15th Annual International Student Academic Awards Ceremony at UF in Fall, 2009.