<%BANNER%>

Growth of Epitaxial Zirconium Carbide Layers Using Pulsed Laser Deposition


PAGE 1

GROWTH OF EPITAXIAL ZIRCONIUM CARBIDE LAYERS USING PULSED LASER DEPOSITION By JUHYUN WOO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Juhyun Woo

PAGE 3

Dedicated to Jihoon, the parents of us all, and Dr. Craciun.

PAGE 4

ACKNOWLEDGMENTS I would like to express my gratitude to Dr. Valentin Craciun without whose help, guidance and encouragement this work might not have been possible. Also I would like to express my sincere appreciation to my committee members, Dr. Fereshteh Ebrahimi, Dr. Rajiv K. Singh, Dr. Cammy Abernathy, and Dr. Timothy J. Anderson. Especially, I am deeply grateful to Dr. Ebrahimi. A lot of what I have learnt is due to her guidance. Also I would like to thank Gerald Bourne and Kerry Siebein. Gerald Bourne helped me to solve many problems in nanoindentation measurements, and Kerry Siebein helped me to get outstanding high resolution TEM images. The assistance and help of Wayne Acree and Eric Lambers to obtain high quality SEM, XPS, and AES results are gratefully acknowledged. I also would like to thank Woochul Kwak and Sang-yup Kim for their help with AFM and FE-SEM. And I personally thank Dr. Luisa Amelia Dempere for giving me a great opportunity of working in Major Analytical Instrumentation Center and hands-on learning about thin film characterization. Finally I would like to thank my wife Jihoon Yi for her support and patience. My best friend and senior Sang-yup Kim helped me a lot to focus on my research. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLE S............................................................................................................vii LIST OF FIGURES.........................................................................................................viii ABSTRACT.....................................................................................................................xiii CHAPTER 1 INTRODUCTION............................................................................................................1 2 LITERATURE REVIEW.................................................................................................4 2.1 Characteristics of Zirconium Carbide.....................................................................4 2.1.1 Composition and Structure...........................................................................4 2.1.2 Properties and Applications of Zirconium Carbide Thin Films...................5 2.2. Techniques Used for Zirconium Carbide Thin Film Depositions.........................7 2.2.1 Thermal Evaporation (TE)...........................................................................7 2.2.2 Sputtering Deposition (S).............................................................................7 2.2.3 Chemical Vapor Deposition (CVD).............................................................8 2.2.4 Pulsed Laser Deposition (PLD)....................................................................9 2.2.5 Comparison of Techniques.........................................................................12 2.3 Growth and Factors Determining the Quality of Thin Films in PLD...................13 2.3.1 Nucleation and Growth...............................................................................13 2.3.2 Background Gas.........................................................................................15 2.3.3 Vacuum.......................................................................................................16 2.3.4 Laser Fluence..............................................................................................17 2.3.5 Laser Wavelength.......................................................................................17 2.3.6 Target to Substrate Distance.......................................................................18 3 EXPERIMENTAL METHODS......................................................................................24 3.1 Pulsed Laser Deposition System..........................................................................24 3.2 Structural Characterization...................................................................................25 3.3 Film Thickness and Roughness............................................................................26 3.4 Surface Chemistry Analysis.................................................................................27 3.5 Electrical Measurement........................................................................................28 v

PAGE 6

4 GROWTH AND CHARACTERIZATION OF HIGH CRYSTALLINE QUALITY ZIRCONIUM CARBIDE FILMS..............................................................................31 4.1 Introduction...........................................................................................................31 4.2 Experiment............................................................................................................32 4.3 Results and Discussion.........................................................................................33 4.3.1 Laser Fluence and Temperature Effect on Deposited Films......................33 4.3.2 Deposition Rate and Thickness Uniformity of Deposited Films................35 4.3.3 ZrC Films Growth at High Temperature and High Laser Fluence.............38 4.3.3.1 Growth behaviors of ZrC films deposited on Si and sapphire substrate..................................................................................................39 4.3.3.2 TEM analysis of ZrC films grown on Si and sapphire substrate.....43 4.3.4 Surface analysis of ZrC films.....................................................................45 4.4 Summary...............................................................................................................49 5 MECHANICAL PROPERTIES OF ZIRCONIUM CARBIDE FILMS MEASURED BY NANOINDENTATION.......................................................................................85 5.1 Introduction...........................................................................................................85 5.2 Experiment............................................................................................................86 5.3 Nanoindentation Test............................................................................................87 5.4 Result and Discussion...........................................................................................88 5.4.1 Hardness and Youngs Modulus of Substrates...........................................88 5.4.2 Hardness and Elastic Modulus of ZrC Films Deposited on Si (001), Si (111), and Sapphire..........................................................................................90 5.5 Summary...............................................................................................................91 6 CONCLUSION.............................................................................................................101 LIST OF REFERENCES.................................................................................................106 BIOGRAPHICAL SKETCH...........................................................................................111 vi

PAGE 7

LIST OF TABLES Table page 2-1 Characteristics and properties of ZrC reported in literature..........................................5 3-1 Excimer laser operating wavelengths..........................................................................24 3-2 Typical XRR values for the resolution and ranges......................................................27 4-1. Growth conditions of deposited films, showing GIXD spectra in fig. 4-1.................34 4-2. Thickness, density, and surface roughness values of the ZrC films deposited at 700 C for different times.........................................................................................36 4-3. Degree of out-of-plane texture of ZrC films at various growth conditions................37 4-4. Structure information of ZrC Si, and sapphire and lattice mismatch.......................38 4-5. Surface roughness and density of as-grown films......................................................46 4-6. The relative percentage of Zr 3d XPS areas corresponding to Zr-C bonds at different take-off angles...........................................................................................48 4-7. Resistivity of as-deposited ZrC films.........................................................................48 5-1. Poissons ratio and Youngs modulus of Al2O3 (poly), Si, and diamond tip.............88 vii

PAGE 8

LIST OF FIGURES Figure page 2-1. Image of cubic rocksalt (B1) structure of transition metal carbides. Carbon atoms are depicted as the light gray spheres, metals as dark gray spheres.........................19 2-2. Zr-C phase diagram showing wide range congruent compositions............................19 2-3. Schematic diagram of an apparatus for pulsed laser deposition.................................20 2-4. Schematic diagram of the approximate energy range of deposited atoms for various deposition techniques. The shaded box indicates the supposed energy range of atom fluxes considered to be beneficial for film growth...........................21 2-5. Schematic diagram comparing the pressure ranges over various techniques. PLD can operate over the widest range of all the methods..............................................21 2-6. Growth diagram drawn by equation 2-3, showing the dependence of growth behavior on temperature and growth rate.................................................................22 2-7. Relationship of impinging particles fluxes to deposition rate, and gas pressure........22 2-8. ZrC ablation rate, showing linear dependence on laser fluence, and threshold laser fluence of 1.3 J/cm2 .................................................................................................23 3-1. A schematic diagram of symmetric and asymmetric GIXD geometry......................29 3-2. A schematic diagram of omega rocking curve geometry...........................................29 3-3. An example of a schematic reflections on pole figure related to crystal quality........30 4-1. GIXD spectra (incidence beam angle, = 1) of ZrC films deposited under various conditions....................................................................................................51 4-2. Comparison of GIXD spectra obtained from ZrC deposited films at same deposition conditions (a) without pre-ablated target (red) and (b) with pre-ablated target (blue)..................................................................................................51 4-3. XRD spectra obtained from ZrC target; vertical lines represent position and intensity for stoichiometric ZrC, JCPDS PDF# 32-1489.........................................52 viii

PAGE 9

4-4. GIXD spectra obtained from films deposited under residual vacuum and different C2H2 gas pressures...................................................................................................52 4-5. XRR spectra of ZrC films deposited for various times at 700 C and 10 J/cm2 under vacuum...........................................................................................................53 4-6. ZrC film deposition rate at Ts = 700C, Pd = 1.0-6 Torr, 10 J/cm2, and 5Hz........53 4-7. GIXD spectra of ZrC films deposited for various times at 700 C and 10 J/cm2 under vacuum; the standard position of diffraction lines from ZrC (dashed lines) and ZrC0.7 (solid lines) are also shown.....................................................................54 4-8. XRD spectra of ZrC films deposited for various times at 700 C and 10 J/cm2 under vacuum...........................................................................................................54 4-9. XRD spectra of ZrC films deposited at 750 C on various substrates.......................55 4-10. XRD and GIXD spectra (incidence beam angle = 1) of ZrC films deposited on Si (111) substrates....................................................................................................55 4-11. Omega-rocking curves of ZrC (111) or (200) peaks recorded from ZC104 films deposited at 750 C on various substrates................................................................56 4-12. Omega-rocking curves of ZrC (002) peaks recorded from the films deposited on Si (001) substrate at various background gas pressures...........................................56 4-13. Omega-rocking curves of ZrC (111) peaks recorded from the films deposited on sapphire (0001) substrate at various different gas pressures....................................57 4-14. XRD spectra of (002) reflection from the films deposited on Si (001) and calculated lattice parameters....................................................................................57 4-15. Background gas effect on lattice parameter of the films deposited on Si (001) substrate....................................................................................................................58 4-16. The relationship between deposited films on Si (001) and texture degree of the films measured by omega-rocking curve.................................................................58 4-17. Phi-scan of {111} in-plane obtained from (a) ZrC film and Si (001) substrate of sample ZC202 and (b) phi-scan diffractometer configuration.................................59 4-18. (111) pole figures of (a) ZrC film and (b) Si (001) substrate obtained from sample ZC202..........................................................................................................60 4-19. (100) pole figures of (a) ZrC film and (b) Si (111) substrate obtained from sample ZC104..........................................................................................................61 4-20. Pole figures showing (a) (100) pole figures of ZrC film, and (b) (1 1-2 6) pole figure of sapphire (0001) substrate obtained from ZC104.......................................62 ix

PAGE 10

4-21. Projection view for two crystallographic orientations of ZrC grown on sapphire (0001), and orientation relationship between ZrC film and sapphire substrate.......63 4-22. Possible nucleation site (marked as A) for the first monolayer on sapphire (0001)...................................................................................................................................63 4-23. Bright field TEM image obtained from cross-section of ZrC film grown on silicon (111) substrate; the regions showing inhomogeneous random grains are marked......................................................................................................................64 4-24. TEM (a) bright field image and (b) SADP (selected area electron diffraction pattern) obtained from cross-section of ZrC film (sample ZC104) grown on silicon (111) substrate..............................................................................................65 4-25. TEM (a) bright field image and (b) SADP (selected area electron diffraction pattern) obtained from cross-section of ZrC film grown on silicon (001) substrate; diffraction pattern obtained from ZrC film is marked by circles.............66 4-26. Bright field TEM image obtained from cross-section of ZrC film grown on silicon (001) substrate exhibiting clear lattice fringe...............................................67 4-27. Bright field TEM image obtained from cross-section of ZrC film grown on sapphire (0001) substrate showing sharp interface; the regions marked as A and B are magnified in figure 4-28 for observation of twinning....................................68 4-28. High resolution TEM image of (a) A region in fig. 4-26, showing parallel twin to the surface of film; (b) B region in fig. 4-26, showing perpendicular twin to the surface of film....................................................................................................69 4-29. Secondary electron images of ZrC surface grown on Si (001) substrate, at K (a), and K (b)......................................................................................................70 4-30. Secondary electron images of ZrC surface grown on sapphire (0001) substrate, at K (a), and K (b)........................................................................................71 4-31. AFM height images obtained from the surface of ZC106 sample (ZrC (001) layer grown on Si (001) substrate)...........................................................................72 4-32. AFM height images obtained from the surface of ZC202 sample (ZrC (001) layer grown on Si (001) substrate)...........................................................................73 4-33. AFM height images obtained from the surface of ZC208 sample (ZrC (001) layer grown on Si (001) substrate)...........................................................................74 4-34. AFM height images obtained from the surface of ZC107 sample (ZrC (111) layer grown on Si (111) substrate)...........................................................................75 x

PAGE 11

4-35. AFM height images obtained from the surface of ZC208 sample (ZrC (111) layer grown on sapphire (0001) substrate)...............................................................76 4-36. AFM height images obtained from the surface of ZC210 sample (ZrC (111) layer grown on sapphire (0001) substrate)...............................................................77 4-37. Cross-sectional bright field TEM image showing an oxidized layer due to discontinued growth.................................................................................................78 4-38. Cross-sectional bright field TEM image showing distinguishable three layers, also supporting a model used for XRR analysis......................................................78 4-39. Z-contrast image of cross-sectional ZC106 sample for TEM-EDX analysis by line and point scan....................................................................................................79 4-40. High resolution Zr 3d spectra acquired at 45 and 90 take off angles and their fitting for an as-received sample deposited at 600 C under vacuum......................80 4-41. High resolution Zr 3d spectra acquired at 45 and 90 take off angles and their fitting for a sample deposited at 600 C under 1-4 Torr of C2H2 that was sputtered-clean by Ar+ bombardment.......................................................................81 4-42. AES depth profile of an as-deposited ZrC film........................................................82 4-43. AES survey spectrum of a ZrC film (ZC202) sputtered with a 4 kV Ar ion beam..82 4-44. AES survey spectrum of (a) a as-deposited ZrC film under CH4 atmosphere and (b) the ZrC film sputtered for 1 min with a 4 kV Ar ion beam................................83 4-45. Residual gas partial pressure analyzed by RGA before deposition at (a) vacuum and right after introducing (b) C2H2 and (c) CH4.....................................................84 5-1. Tip area function used for calculation of mechanical properties................................92 5-2. Example of erroneous fitting of tip area function in depth below 5nm range............92 5-3. Hardness of Si (001), Si (111), and sapphire (0001) single crystal substrates...........93 5-4. Youngs modulus of Si (001), Si (111), and sapphire (0001) single crystal substrates..................................................................................................................93 5-5. Error range change in Youngs modulus as function of Poissons ratio of substrate according to eq. (5-1)...............................................................................................94 5-6. Load-displacement curves of Si (001) substrate.........................................................94 5-7. Load-displacement curves of Si (111) substrate.........................................................95 5-8. Load-displacement curves of sapphire (0001) substrate............................................95 xi

PAGE 12

5-9. Load-displacement curves of sapphire (0001) substrate before pop-in occurs........96 5-10. Combined modulus of film and substrate for different quality of ZrC (001) grown on Si (001) substrate. FWHM of ZC202 = 2.53, FWHM of ZC210a = 7.27 in omega rocking curve....................................................................................96 5-11. Load-displacement curves of ZrC (001) grown on Si (001) substrate, sample ZC202 in fig 5-10.....................................................................................................97 5-12. Load-displacement curves of ZrC (001) grown on Si (001) substrate, sample ZC202 in fig 5-10.....................................................................................................97 5-13. Measured elastic modulus as a function of relative penetration into a coated specimen (a is contact radius, and t is film thickness). For curve 1, Efilm is less than Esubstrate, and opposite case for curve 2.............................................................98 5-14. Combined modulus of film and substrate for different quality of ZrC (111) grown on sapphire (0001) and Si (001) substrate, sample ZC204 and ZC210c respectively. FWHM of ZC204 = 2.53, FWHM of ZC210c = 6.95 in omega rocking curve............................................................................................................98 5-15. Load-displacement curves of ZrC (111) grown on sapphire (0001) substrate, sample ZC204 in fig 5-14.........................................................................................99 5-16. Load-displacement curves of ZrC (111) grown on Si (111) substrate, sample ZC210c in fig 5-14...................................................................................................99 5-17. Combined hardness of ZrC and substrate as a function of normalized depth; h displacement, t film thickness..............................................................................100 6-1. XRD spectra of thin films deposited from a Co doped ZrC target under various conditions...............................................................................................................105 xii

PAGE 13

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 GROWTH OF EPITAXIAL ZIRCONIUM CARBIDE LAYERS USING PULSED LASER DEPOSITION By Juhyun Woo December 2005 Chair: Valentin Craciun Major Department: Materials Science and Engineering Epitaxial ZrC thin films were grown on Si (001), Si (111), and sapphire (0001) substrate by the pulsed laser deposition technique. It has been found that crystalline films could be grown only by using laser fluences higher than 6 J/cm2 and substrate temperatures in excess of 600 C. For a fluence over 8 J/cm2 and a substrate temperature of 600~700 C, cubic ZrC films exhibiting a (001) texture were deposited under vacuum or low pressure C2H2 atmosphere. Under very low water vapor pressures (10-8~10-9 Torr), high substrate temperatures (700~750 C), and high laser fluence (10 J/cm2), highly textured ZrC films were deposited on single crystalline substrates. Pole figures investigation showed that films were epitaxial, with in-plane axis aligned with respect to those of the substrate. X-ray reflectivity, atomic force microscope, ellipsometry, and scanning electron microscopy confirmed that these films were smooth, with surface roughness values below 1.0 nm and mass densities around the stoichiometric ZrC tabulated value of 6.7 g/cm3. X-ray photoelectron spectroscopy and Auger electron xiii

PAGE 14

spectroscopy investigations showed that the surface of the films contained a significant amount of oxygen and Zr-O bonds, the outmost 1~2 nm of the surface region being mainly ZrO2. However, after the removal of this surface contamination layer, low oxygen atomic concentration below 3 % were measured. Despite of the rather high levels of oxygen contamination, electrical resistivity measured by four probe measurement indicated that the deposited ZrC films were very conductive. The use of a low C2H2 pressure atmosphere during deposition had a small beneficial effect on crystallinity and stoichiometry of the films. Nanoindentation measurements showed higher values of the hardness for higher crystallinity. For the highest crystalline quality, (111) ZrC films deposited on sapphire, values over 450 GPa for the elastic modulus and ~31 GPa for the hardness were measured. xiv

PAGE 15

CHAPTER 1 INTRODUCTION Zirconium carbide (ZrC) is a typical refractory compound that crystallizes in the rock salt (NaCl, B1) ground-state structure under normal conditions. Recently, ZrC is arousing interest because of its several notable properties, characterized by a very high melting temperature of 3530 C [Zai84], excellent thermal stability, exceptional mechanical hardness and strength [Che05a], chemical inertness, and imperviousness to hydrogen attack [Tot71]. In addition, electrical conductivity is comparable to metals [Zai84], and work function for electron emission is low [Mac95]. The common applications are in chemicaland wear-resistant coatings and ultra-high temperature applications so far. However, as a form of thin coating or layer (or film) in a thickness range of micron or submicron, ZrC has more important applications in vacuum electronics or MEMS (micro electro-mechanical system) devices [Cha01, Tem99, and Xie96]. Particularly, the growth of epitaxial ZrC film has special importance because of its anisotropic properties and possibility to be fabricated in various shapes of microstructure by using anisotropic etching. When it is manufactured as a form of micro-structured shapes, its applications are much wider with high efficient and compact devices, using electron emission, such as short responding bright flat displays and electron beam lithography. Also ZrC is a potential good candidate as a diffusion barrier for metallization on silicon because it exhibits lower lattice mismatch and thermal expansion coefficient difference with silicon (Si) than that of zirconium nitride (ZrN). 1

PAGE 16

2 Especially, epitaxial films for this application could very efficiently prevent diffusion of metal atoms (Al or Cu) because there are no grain boundaries for fast atomic diffusion. Despite such many attractive applications and technological interest, only a few studies describing ZrC film growth have been published so far. The growth of ZrC film by thermal evaporation [Tes93], sputtering deposition [Br, Spr86], vacuum plasma spray processes [Var94], chemical vapor deposition (CVD) [Ber95], pulsed laser deposition (PLD) [Ale00], and ion beam deposition [He98] have been reported. However, it appears that it is quite difficult to obtain high crystalline quality ZrC film, because of its high melting temperature, low vapor pressure, and Zr atoms affinity for oxygen. The objective of this study was focused on studying the relationship between the thin film deposition parameters and the structure and properties of the resulting ZrC films. Based on performance considerations, a pulsed laser deposition technique for the growth of ZrC thin film had been chosen, and hence used in this work. In chapter 2, a general background of ZrC material was presented, and growth methods and problems were discussed as well. In chapter 3, the experimental method and characterization tools used for analysis of film properties were explained. Grazing incidence x-ray diffraction (GIXD), symmetrical XRD, and pole figure were used for structure analysis. Omega rocking curve technique was used to check the degree of texture of films. X-ray reflectometry (XRR), atomic force microscope (AFM), scanning electron microscopy (SEM), and ellipsometry were used for surface morphology, thickness and/or roughness measurement. Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS) were used for surface chemical analysis. Four points probe was used to measure electrical sheet resistance and resistivity, while transmission electron microscope (TEM)

PAGE 17

3 was used for microstructure analysis. In chapter 4, the influence of process parameters on the microstructure, crystallinity, and morphology were presented, and also the dependence of substrates and its orientation on growth behavior of ZrC thin film was analyzed and discussed. In chapter 5, the results regarding the hardness and elastic modulus of ZrC films, obtained by nanoindentation technique, were presented. Finally, the overall conclusions drawn from this work and suggestions for future work were summarized in chapter 6.

PAGE 18

CHAPTER 2 LITERATURE REVIEW 2.1 Characteristics of Zirconium Carbide 2.1.1 Composition and Structure Zirconium (Zr, [Kr]4d25s2, group B) is a solid transition metal which has a hexagonal close-packed (HCP) crystal structure at room temperature. Carbon (C, 1s22s2p2, group ) also naturally crystallizes in the same HCP structure. However, the structure of the zirconium carbide (ZrC) does not follow its parent metal structures. That is, ZrC has the Zr on a face centered cubic (FCC) lattice, even though Zr has HCP structure, and carbon atoms occupy octahedral interstitial sites between Zr atoms. In transition metal carbides, their structure depends on the s-p electron count [Oya92a]. With increasing s-p electron count, the metal structure changes from body centered cubic (BCC) to HCP to FCC across the transition series. Therefore, the group B and B metal carbides (MC) crystallize in the rock salt (B1, NaCl-type) structure (fig. 2-1) rather than a hexagonal form because the partially filled bands of the host metals can accommodate a high ratio of sp-electron-rich carbon to metal. In group the stoichiometry of M2C occurs often, while group and retain metal-rich stoichiometries of M3C and M4C, consistent with an attempt to avoid filling anti-bonding levels in the metal bands [Oya92b]. Most of the transition metal mono-carbides form in the rock salt structure, fcc metal with carbon occupying the octahedral interstitial sites. The shortest metal to metal distance is about 30% greater in the metal carbide (MC) structure than in the pure metal 4

PAGE 19

5 structure for the group and group carbides [Oya92a]. At one hundred percent site occupancy, the stoichiometry of transition metal carbide is M1C1. Although the zirconium carbide is stable in solid solutions composition from carbon-deficient ZrC1-x to stoichiometric ZrC (Fig. 2-2) [Bra92], it has been reported that crystalline ZrC film with rocksalt structure can be synthesized with Zr to C atomic ratios from 0.5 up to 1.78 as a vacancy compound [Br, Smi93, and Tes93]. 2.1.2 Properties and Applications of Zirconium Carbide Thin Films The properties of ZrC depend on several factors such as chemical compositions, grain size, defect structures, and porosity. Hence, variations in properties have been observed in the literature. Some of the characteristics and properties of zirconium carbide reported in the literature are listed in table 2-1. Table 2-1 Characteristics and properties of ZrC reported in literature. Structure Rocksalt (B1) [Oya92a, Tot71] Space group Fm3m (225) [JCPDS PDF#: 35-0784] Lattice parameter () 4.698 [Tot71], 4.6930 [JCPDS PDF#: 35-0784] Density (g/cm3) 6.59 [Tot71], 6.73 [Lid05] Hardness 2860 kg/mm2 [Zai84], up to 30.2 GPa [Che05a] Elastic modulus (GPa) ~325 [Che05a] Melting temperature (C) 3532 [Lid05], 3530 [Zai84], 3420 [Shi98] Thermal expansion (-6/C) 6.7 [Tot71] Electrical resistivity (0-5m) 6.2 at 293 K [Zai84], 20.4 at 300 K [Mod85] Work function (eV) 3.38 [Zai84], 3.3~3.4 [Mac95] As shown in table 2-1, ZrC exhibits many unusual properties such as high thermal and electrical conductivity with extreme hardness, and low work function. Recently, most

PAGE 20

6 applications of the zirconium carbide rely on its significant hardness such as cutting tools [Pri92 and Tot71]. However, these combinations of properties are more important in a variety of technological applications. For example, zirconium carbide coatings on molybdenum, niobium, and nickel-based alloy were used to increase the radiation emissivity, indicating a good candidate material for thermophotovoltaic (TPV) radiator [Coc99]. Also, the coatings showed excellent resistance to thermal cycling and acceptable stability during vacuum annealing at 1100 C. In addition, ZrC films improved the field emission stability and beam confinement when they were deposited on field emitter cathodes [Eda96 and Mac95]. The full width at half maximum of energy distribution for particular materials is nearly equal to a half of the work function [Ada74]. Thus, the cathodes of low work function have an advantage because the full width at half maximum in energy distribution of the emitted electrons is narrow. Moreover, ZrC coating layers have much higher temperature stability and are more resistant to the chemical attack by the palladium (Pd) fission product, while silicon carbide (SiC) coating layers lose their mechanical integrity at temperatures over 1700 C [Oga86 and Oga92]. Furthermore, video displays [Mac98a, Mac98b, and Mac98c], microwave application [Mac92, Mac93, Mac94, Mac95, and Xie96], cold cathodes for operation in poor vacuum environment, photocathodes for electron beam lithography [She97], hole injection layers of organic light-emitting diodes (OLEDs), and substrates for epitaxial growth of nitride are good potential fields for ZrC applications.

PAGE 21

7 2.2. Techniques Used for Zirconium Carbide Thin Film Depositions 2.2.1 Thermal Evaporation (TE) The thermal evaporation technique for ZrC films deposition was used by Tessner and Davis in 1993 [Tes93] to widen the emission area of high-temperature thermionic energy converters (TEC) and by Mackie et al. for the use of field emitter arrays (FEAs) [Mac95]. In the evaporation process, the material to be deposited should be heated to a high enough temperature to achieve a sufficient vapor pressure and the desired evaporation. The wire filaments, sheet metal, or electrically conductive ceramic sources are used in evaporation and they are heated by electrical current. Thermal evaporation must be an important technology for the formation of functional coatings on a variety of materials. However, there are limitations regarding the type of material that can be heated. In some cases such as ZrC, it is not possible to achieve the necessary evaporation temperatures without significantly evaporating the source holder. Moreover, chemical reactions between the holder and the material can result in the contamination of the coating. 2.2.2 Sputtering Deposition (S) In the sputtering process, the target material is bombarded with high energy ions that transfer their momentum to the atoms on the target materials which are ejected. These sputtered particles condense on the substrate facing the target. The dc or rf sputtering by an Ar plasma are the most common form. Because of the low sputter yield, magnetron sputtering is often used to increase the deposition rate. Magnets which are positioned behind the target induce the electrons to spiral and increase the degree of ionization of the plasma due to longer path length of electrons.

PAGE 22

8 Alternatively, the configuration consisting of three Ar ions beams can be used to co-deposit ZrC film, which is called tri-ion beam-assisted deposition (tri-IBAD) [He98]. Two beams of Ar ions can be used to sputter from graphite and zirconium. A third Ar ion beam can be used to bombard the growing ZrC film to provide additional energy to enhance film formation processes. Compared to evaporated particles, sputtered particles have considerably higher kinetic energies. As a result, sputtered layers usually have higher adhesive strength and a denser coating structure than evaporated layers. The greatest advantage of sputtering deposition is a large particle source area compared to evaporators, which enables a large area coating with a high degree of uniformity. Nevertheless, sputtering could have a relatively low ratio of energetic ions to neutral species, so that it is possible not to produce the hardest films. 2.2.3 Chemical Vapor Deposition (CVD) In chemical vapor deposition methods, the state of substances used for the growth is in the vapor phase when they are introduced to the vacuum system. The substances must be thermally excited by appropriate high temperatures or with plasma to be deposited. Chemical vapor deposition (CVD) of ZrC can be accomplished by the reaction between zirconium halide and a hydrocarbon in H2 atmosphere at temperatures above 1000C. The zirconium halide vapor can be obtained either by a reaction between a halide vapor and zirconium metal, or by sublimation of ZrCl4. Nuclear fuel particles were coated with ZrC coatings by CVD [Rey74]. Argon (Ar) gas was initially bubbled through dichloromethane (CH2Cl2) that was kept at 0 C. Then the Ar and CH2Cl2 gas mixture was passed through a heated zirconium (Zr) sponge at 600 C to produce ZrCl4 vapors. Preheated methane (CH4) and hydrogen (H2) gases to

PAGE 23

9 600 oC were mixed with the ZrCl4 vapors inside a graphite tube and heated to 1100 C for 3 hours to produce ZrC coatings. Chemical analysis of the coating resulted in a C/Zr ratio of 1.01 which indicated that some free carbon was present in the coatings. The effect of the composition of gas mixtures on the properties of ZrC coatings was studied [Hol77 and Wag76]. Chemical vapor deposition of ZrC coatings was achieved by reacting gaseous mixtures of CH4, H2, ZrCl4, and Ar as carrier gas. The overall reaction is given by )1(,4)1(2244 xHClZrCHxZrClxCHx (2-1) Increasing the amount of CH4 in the coating gas mixture (i.e., increasing C/Zr molar ratio) resulted in ZrC coatings with increased amounts of carbon. The chemical analysis of the coating showed the presence of free carbon, when the C/Zr molar ratio in the coating gas mixture was over 0.21. Organometallic precursor (cyclopentadienylzirconium) was also used to deposit ZrC coatings at temperatures in the range of 300~600 C [Han95]. The chemical composition of the films synthesized at 600 C showed presence of zirconium carbide, as determined by x-ray photoemission spectroscopy (XPS). 2.2.4 Pulsed Laser Deposition (PLD) The technique of pulsed laser deposition (PLD) has been used to deposit high quality films of various materials for more than two decades. The technique uses high energy laser pulses (typically 2~5 J/cm2) to melt, evaporate, and ionize material (ablation process) from the surface of a target. This ablation event produces a transient, highly luminous plasma plume that expands rapidly away from the target surface. The ablated material is collected, and then condenses on a suitably placed substrate.

PAGE 24

10 Applications of the technique range from the superconducting films production and insulating circuit components to medical applications to improve biocompatibility. In spite of this widespread usage, the fundamental processes occurring during the transfer of material from target to substrate are not fully understood yet, and are consequently the focus of much research. The interaction of laser radiation with solid surfaces was under investigation from as early as 1962, when Breech and Cross [Bre62] analyzed the emission spectrum of material vaporized by laser pulses. The first demonstration of PLD in 1965 [Smi65] did not attract significant interest, as the films were inferior to those obtained by other deposition techniques, such as chemical vapor deposition (CVD) or molecular beam epitaxy (MBE). The PLD technique was slowly developing for approximately the next twenty years until Dijkamp and Venkatesan [Dij87] used PLD to grow a film of the high temperature superconducting material Ba2Cu3O7 + YBaCuO (YBCO). The films obtained were found to be superior in quality to those previously grown using other deposition methods and awaked a tremendous interest in the technique. Present day research applications include growing films for magneto-optic storage devices, developing multilayer devices for x-ray optics, and depositing diamond films on components for protection and insulation. A schematic diagram of the basic PLD configuration is shown in fig. 2-3. The general understanding of the process is somewhat simple, which uses short pulses (pulse duration, 10~30 ns) of laser energy to remove material from the surface of a target. The vaporized material, containing neutrals, ions, and electrons, is known as a laser-produced plasma plume, and expands rapidly away from the target surface (velocities typically

PAGE 25

11 ~106 cm/s in vacuum). Film growth occurs on a substrate upon which some of the plume material recondenses. However, in practice, the situation is not so simple, with a large number of variables affecting the properties of the film, such as laser fluence, background gas pressure and substrate temperature. These variables allow the film properties to be improved for individual applications. However, optimization can require a considerable amount of time and effort. Indeed, much of the early research into PLD concentrated on the empirical optimization of deposition conditions for individual materials and applications, without attempting to understand the processes occurring as the material is transported from target to substrate. The technique of PLD was found to have significant benefits over other film deposition methods. The capability for stoichiometric transfer of material from target to substrate, i.e., the almost exact chemical composition of a complex material such as YBCO, can be reproduced in the deposited film. Relatively high deposition rates (~100 /min) can be achieved at moderate laser fluences, with film thickness controlled in real time by simply turning the laser on and off. The fact that a laser is used as an external energy source results in an extremely clean process without internal filaments. Thus deposition can occur in both inert and reactive background gases. The use of a carousel housing a number of target materials enables multilayer films to be deposited without the need to break vacuum when changing between materials. In spite of these significant advantages, industrial adoption of PLD has been slow and most applications have been limited to the research environment. There are basically three main reasons for this. First, the plasma plume created during the laser ablation process is highly forward directed; therefore the thickness of material collected on a substrate is highly non-uniform and the composition

PAGE 26

12 can vary across the film. The area of deposited material is also quite small in comparison to that required for many industrial applications which require area coverage of at least ~8 8 cm2. Second, the ablated material contains macroscopic globules of molten material, up to ~10 m diameter. The arrival of these particulates at the substrate is obviously detrimental to the properties of the film being deposited. Third, the fundamental processes occurring within the laser-produced plasmas are not fully understood. Thus deposition of novel materials usually involves a period of empirical optimization of deposition parameters. To a large extent the first two problems have been solved. Films of uniform thickness and composition can be produced by rastering the laser spot across the target surface and moving the substrate during deposition. Line-focus laser spots have also been used to obtain large area coverage. The particulate material was initially removed from the plume using a mechanical velocity filter, although recently more elaborate techniques, involving collisions between two plasma plumes or off-axis deposition, have been used to successfully grow particulate-free films. The third problem will be resolved by the development of computer simulations to describe PLD. However, a large amount of experimental data is required to support the verification of such models. 2.2.5 Comparison of Techniques Fig. 2-4 [Chr94] compares the energy range of the atomic fluxes for each technique. The solid boxes are typical values for the energy range of the atomic fluxes in practice, and the dashed boxes are the energetic fluxes values which could be controlled. The spread in energy of the atomic flux can be divided into energetic portion to the flux and thermal fluxes. The purely thermal techniques such as CVD (chemical vapor deposition) and MBE (molecular beam epitaxy) are mostly used so far because the energetic

PAGE 27

13 deposition techniques are still under development. The shaded area in fig. 2-4 [Chr94] is the best energy range for deposition flux that was established earlier. Energy of 5 eV to 10 eV per atom promotes surface diffusion and high sticking probability while minimizing damage. The energetic flux over 100 eV can produce deep damage, so it should be avoided. The energy range of sputtering (S) and PLD matches most closely the best energy range indicated. Although the techniques with purely thermal fluxes are in standard use to deposit high quality epitaxial films, the energetic techniques can be employed to lower the growth temperature at which epitaxy can be achieved, especially for high melting temperature materials such as ZrC. Fig. 2-5 [Chr94] shows the range of pressures in which each of the techniques is operable. PLD can relatively operate in a wide pressure range from UHV (ultra high vacuum) to 1 Torr, while other techniques need lower than HV (high vacuum) pressure to minimize contamination, to avoid oxidation of evaporation filaments, or to sustain plasma. Although most materials do not require performing the deposition in a high pressure of background gas, PLD has an advantage over other techniques in operating under high background gas pressure that is particularly useful for multicomponent oxides for example. 2.3 Growth and Factors Determining the Quality of Thin Films in PLD 2.3.1 Nucleation and Growth The atoms accumulated on the surface may diffuse laterally over substrate, also they may encounter other mobile atoms to form mobile or stationary clusters, then they may attach to preexisting film-atom clusters, or they may be reevaporated from substrate [Ven84]. The balance between growth and dissolution processes will be governed by the total free energy of the cluster. Nucleation could be stabilized and destabilized,

PAGE 28

14 depending on the energy stored in forming surface of new face, the adhesion energy released in formation of an epitaxial interface, the strain energy stored in lattice-mismatched epitaxial growth, and electrical energy stored in the formation of a surface dipole. Three modes of film growth at the initial stages are possible [Lew78]. When the cohesive energy of the film atoms is greater than the cohesive binding between the film and substrate atoms, the formation and growth of isolated islands occur (Volmer-Weber Island growth) [Mah99]. This mode can result in an epitaxial film that has a rough surface, or a polycrystalline film. When the cohesive energy between the film and substrate atoms is greater than the cohesive energy of the film (but monotonically decreases as each film layer is added), layer by layer (Frank-Van der Merwe) growth occurs [Mah99]. This mode results in a very smooth epitaxial film. When the monotonic decrease in binding energy with each successive layer is energetically overridden by some factor such as strain energy, island formation becomes more favorable. So mixed (Stranski-Krastanov) growth, island growth after the first monolayer forms, can occur [Mah99]. The growth of thin films by PLD depends on many factors, such as energy, ionization degree, and type of the condensing particles, temperature, and physicochemical properties of the substrate. There are two main thermodynamic parameters that determine the growth mechanism. One is substrate temperature T and the other is supersaturation m. eRRkTmln (2-2)

PAGE 29

15 where k is the Boltzman constant, R is the actual deposition rate, and Re is equilibrium deposition rate at temperature T. The overall film growth process in PLD has been theoretically studied [Met89]. The mean thickness at the moment of 99% covered substrate (N99 [Kas78]) at which a growing thin discontinuous film reaches continuity has been found to be given by kTEERNNaddes323exp5.031099 (2-3) where is the adatom vibrational frequency, N0 is the density of adsorption sites on the substrate, Edes is the activation energy for adatom desorption, and Esd is the activation energy of adatom surface diffusion. The lines of equal mean film thickness at the moment of 99% covered substrate (defined by the condition N99(R, T) = constant) can be drawn in ln R versus 1/T coordinates (fig 2-6). One of line of equal mean film thickness at the moment of 99% covered substrate is drawn as dashed line. Depending on deposition rate and temperature, high-island growth region and low-island growth region are divided in the growth diagram. In the high-island growth region the thin film reaches continuity with a mean thickness. In the low-island growth region a mean thickness does not exceed a few monolayers. At higher R and lower T, continuous growth (amorphous) takes place from island growth [Kas78]. Therefore depending on the experimental conditions, such as energy density of laser and substrate temperature, single crystalline thin film in high-island growth region, polycrystalline film in low-island growth region, or amorphous film can be formed. 2.3.2 Background Gas During pulsed laser deposition, the use of background gas can be divided into passive or active use. The passive use is mostly to compensate for some loss of a

PAGE 30

16 constituent element. For examples, the deposited oxides tend to be deficient in oxygen. Typically 10~300 mTorr of background oxygen in chamber is required for oxide superconductors. Also the introduction of background gas change typical particulate size as the ambient gas pressure varies. Inert or reactive gas can be introduced to form particulates with a desired size or composition for active use of ambient gas during PLD. The decrease in the ambient gas pressure results in a decrease in size and a narrower size distribution [Mat86]. The origins of the formation of the particulates and the mechanisms of enrichment in specific element in the particulates are different in vacuum and in inert ambient gas for PLD processes. The effect of inert ambient gas pressure increases collisions between the ejeted species and the ambient gas as the pressure increases. At a pressure of 1 mTorr, the mean free path is about 5 cm. The mean free path of ejected species becomes 0.05cm at a higher pressure of 100 mTorr. In vacuum, there are no collisions between ejected species virtually, so particulates are predominantly formed from solidified liquid droplets, and the vapor species are deposited as a uniform background film in the same time. However, when the ambient gas pressure increases, the vapor species can have enough collisions. Thus, nucleation and growth of vapor species can occur, before they arrive at the substrates. This suggests that the ultrafine particulates are formed from the vapor species instead of liquid droplets. 2.3.3 Vacuum The quality of the vacuum is a major consideration for determining the deposition rate. Gaseous impurities in the deposition chamber impinge on the growing film and will be incorporated into the film depending on their sticking probabilities. The most common

PAGE 31

17 impurities are H2O, CO, CO2, and H2. Fig. 2-7 [Chr94] indicates the relationship between atomic fluxes on surface with pressure and deposition rate. To avoid contamination of impurities, deposition rate should be controlled depending on deposition pressure. 2.3.4 Laser Fluence The laser fluence is most significant on the particulate size and density. The laser fluence can be changed varying the laser power or the laser spot size. There is threshold laser fluence to ablate target material with laser beam. For example, in the case of ZrC, the threshold laser influence is about 1.3 J/cm2 [Ale00]. The laser fluence is extrapolated and a linear trend up to 13 J/cm2 is maintained (fig. 2-8 [Ale00]). In other word, the saturation of ablation process is not observed. One of the mechanisms that reduce the ablation rate is plasma shielding of the target [Dye89], and it is more often encountered in the laser-ablation deposition using longer wavelength. 2.3.5 Laser Wavelength The laser wavelength is directly related to the effectiveness of the absorption of the laser power into the target. For most metals, the absorption coefficient decreases with decreasing laser wavelength. Thus, the laser penetration depth in metal is larger in the UV range than in the infrared range. For other materials, the variation of absorption coefficient with wavelength is more complex due to various absorption mechanisms, such as lattice vibration, free carrier absorption, impurity centers, or bandgap transition. The primary effect of the laser wavelength on particulate generation is mostly due to the difference in the absorption coefficient when different laser wavelength is used. Larger particulates are generated when using longer laser wavelength [Kau90].

PAGE 32

18 2.3.6 Target to Substrate Distance The effect of target to substrate distance is mainly reflected in the angular spread of the ejected flux. Depending on the position of the substrate, different particulate appearance may occur. The specific effects of target to substrate distance and ambient pressure are related. The plume dimension decreases as the background gas pressure increases due to the increased collisions between the laser-produced plume and the background gas. When the target to substrate distance is smaller than the plume length, there is no remarkable difference in particulate size and density. As the target to substrate distance increases, a few larger particulates appear [Chr94].

PAGE 33

19 Figure 2-1. Image of cubic rocksalt (B1) structure of transition metal carbides. Carbon atoms are depicted as the light gray spheres, metals as dark gray spheres. Figure 2-2. Zr-C phase diagram showing wide range congruent compositions [Bra92].

PAGE 34

20 Figure 2-3. Schematic diagram of an apparatus for pulsed laser deposition.

PAGE 35

21 Figure 2-4. Schematic diagram of the approximate energy range of deposited atoms for various deposition techniques. The shaded box indicates the supposed energy range of atom fluxes considered to be beneficial for film growth [Chr94]. Figure 2-5. Schematic diagram comparing the pressure ranges over various techniques. PLD can operate over the widest range of all the methods [Chr94].

PAGE 36

22 Figure 2-6. Growth diagram drawn by equation 2-3, showing the dependence of growth behavior on temperature and growth rate. Figure 2-7. Relationship of impinging particles fluxes to deposition rate, and gas pressure [Chr94].

PAGE 37

23 Figure 2-8. ZrC ablation rate, showing linear dependence on laser fluence, and threshold laser fluence of 1.3 J/cm2 [Ale00].

PAGE 38

CHAPTER 3 EXPERIMENTAL METHODS 3.1 Pulsed Laser Deposition System Most depositions are conducted using the laser wavelength of 193~400 nm because strong absorption is exhibited in this spectral region by target materials. Absorption coefficients have a tendency to increase as laser wavelength is shorter in this range. The excimer laser wavelengths used in a commercial laser system are listed in table 3-1. A Lambda Physik KrF laser ( = 248 nm) was used for the experiments in this study with a 25 ns pulse duration. Table 3-1 Excimer laser operating wavelengths Excimer Wavelength (nm) F2 157 ArF 193 KrCl 222 KrF 248 XeCl 308 XeF 351 A schematic diagram for chamber was illustrated in fig. 2-3. A stainless steel spherical vacuum chamber is used. The laser entry port is protected by a quartz plate, which is designed to reduce the deposition rate of materials on window. The angle between the target normal and laser beam is 45. A turbo-molecular pump backed by a 24

PAGE 39

25 mechanical pump provided the high vacuum. The system operated between 2.5-6 and 1.5-7 Torr of base pressure, which is critical factor in controlling the quality of film. 3.2 Structural Characterization Symmetric and asymmetric x-ray diffraction spectra (XRD) were collected on a Panalytical MRD Xpert system using Cu K radiation to identify crystal structure and quality. The standard settings for the X-ray generator were 45 kV at 40 mA. Fig. 3-1 compares the grazing incidence angle geometry used for thin film with the conventional -2 symmetric geometry used for bulk analysis. In the grazing incidence XRD geometry for the thin film arrangement, the incident and diffracted beams are prepared nearly parallel using a mirror and a narrow slit on the incident beam (1/8 divergence slit for experiments of this study) and a parallel plate collimator (0.27 divergency) on the detector side. In addition, a very small angle of incidence beam to the sample surface (typically 1 to 3) increases the path length of the X-ray beam through the film. This increases the diffracted beam intensity from the film, while that from the substrate is reduced. During the collection of the diffraction spectrum, only the detector moves along the angular range, while keeping the incident angle fixed. Thus the beam path length and the irradiated area are constant. For more detail observation of highly textured ZrC thin films, omega rocking curves were collected. The geometry is represented in fig. 3-2. The peak broadening along -2 direction is caused by variation in d-spacing with depth for thick films or by use of non-monochromatic beam. On the other hand, the peak broadening in omega direction is due to mosaic spread, lateral incoherence, high uniform dislocation density, or sample curvature.

PAGE 40

26 Also pole figure method was used to confirm not only in-plane texture, but also out-of-plane texture as well as crystal quality and orientation relationship between film and substrate. A schematic diagram of pole figure geometry and relation with microstructure are shown in fig. 3-3. The incidence beam angle and diffracted beam angle is set to specific angle which is satisfying Braggs law. All of reflections in projection area are from specific faces at the Bragg angle, but the geometry of the sample is changing by rotation and by tilting during scanning. Thus, we can obtain the orientation distribution. 3.3 Film Thickness and Roughness For the measurement of film thickness and roughness, mainly x-ray reflectometry (XRR) was employed and compared with spectroscopic ellipsometry (SE, Woollam M-88). XRR method is to record the intensity of the x-ray beam reflected by a sample at grazing angles. The operation mode is the same as regular powders XRD setup which always make the incident angle half of the angle of diffraction. The reflection at the surface and interfaces is due to the different electron densities in the different layers in film, which correspond to different refractive indexes. For incident angles below a critical angle (c), total external reflection occurs. The critical angle for most materials is less than 0.3 in Above c, the reflections from the different interfaces interfere and give rise to interference fringes. The period of the interference fringes and their decrease in intensity are related to the thickness and the roughness of the layers. The reflection spectra can be analyzed with the aid of the WinGIXA software, which is based on the Parratt formalism modified to include roughness [Par54 and Nv80]. The typical ranges for these measurements are between 0.1 and 5 in 2. For the detail analysis of x-ray reflection, WinGixa software was used

PAGE 41

27 for fitting x-ray reflection curves. To get an accurate fit for the acquired data, a three layers model was used, which consists of a topmost contamination layer, then the deposited layer, and an interfacial layer located between the deposited layer and substrate. The measurable ranges and resolution for this measurement are listed in table 3-2 [Auc01]. Table 3-2 Typical XRR values for the resolution and ranges. Parameter Resolution Range Thickness 0.1~5 % 2~200 nm Surface roughness 0.2 nm 0.1~3 nm Interface roughness 0.2 nm 0.1~3 nm Density 1-10% Atomic force microscope (AFM) images were taken with a Digital Instruments Dimension 3100. Images were acquired in the tapping mode with a scan range 5 5 m. Before roughness analysis was performed, an automatic plane fit and a flatten command were applied to the images to level the image and to reduce slope associated with measurement drift. 3.4 Surface Chemistry Analysis For Auger electron spectroscopy (AES) investigation, a Perkin-Elmer PHI 660 instrument was used for the surface analysis. Survey spectra were obtained from as-deposited film, also after removing the topmost layer by Ar ion sputtering. X-ray photoelectron spectroscopy (XPS) spectra were acquired on a PHI model 5100 ESCA system using Mg K X-ray source (1253.6 eV). Argon ion was used to remove surface contaminants and to remove atomic layers into the thin films. The multiplex data were used during peak deconvolution. Background was substracted, and peaks were fitted with

PAGE 42

28 one or more component Gaussian-Lorentzian functions to determine accurate chemical bonding. 3.5 Electrical Measurement The electrical measurement was taken on an Alessi four point probe. A probe head having tungsten carbide tips with a point radius of 0.002" and a probe spacing of 0.05" was used for all measurements. Current was supplied by a Crytronics model 120 current source with a range of applied currents between 30 A to 100 A. Eq. (3-1) was used to determine resistivity (, cm). ItV5324.4 (3-1) Based on the dimensions of the sample and probe head, no geometrical correction factors were applied. The term t is film thickness (cm), and V is the voltage measured at the supplied current (I).

PAGE 43

29 Figure 3-1. A schematic diagram of symmetric and asymmetric GIXD geometry. Figure 3-2. A schematic diagram of omega rocking curve geometry.

PAGE 44

30 Figure 3-3. An example of a schematic reflections on pole figure related to crystal quality.

PAGE 45

CHAPTER 4 GROWTH AND CHARACTERIZATION OF HIGH CRYSTALLINE QUALITY ZIRCONIUM CARBIDE FILMS 4.1 Introduction Zirconium carbide (ZrC) is a refractory compound, characterized by a high melting temperature of 3530 C [Zai84], excellent thermal stability, high mechanical hardness and strength [Che05a], and chemical inertness. In addition, electrical conductivity is comparable to metals [Zai84], and work function for electron emission is low [Mac95]. The common applications are the chemicaland wear-resistant coating, and ultra-high temperature applications. Thin films of ZrC also have important application in vacuum electronics or micromechanics [Xie96, Tem99, and Cha01]. Particularly, the epitaxial growth of ZrC is specially important because it could allow the growth of various types of nanostructure by using anisotropic etching, that would exhibit high performance, efficiency, and compactness, such as short response and high brightness flat displays, and electron beam lithography. ZrC also exhibits lower lattice mismatch and thermal expansion coefficient difference with Si than ZrN, making it a potential good candidate for metallization or diffusion barriers structures for Si-based electron devices. Despite of the considerable attractive applications, relatively few studies describing ZrC film growth have been published so far. The growth of ZrC film by thermal evaporation [Tes93], reactive magnetron sputtering deposition [Br], vacuum plasma spray process [Var94], chemical vapor deposition (CVD) [Ber95], pulsed laser deposition (PLD) [Ale00], and tri-ion beam-assisted deposition [He98] has been reported. However, it seemed to be 31

PAGE 46

32 quite difficult to obtain high crystalline quality ZrC films, because of its high melting temperature, low vapor pressure, and Zr atoms affinity for oxygen. Pulsed laser deposition (PLD) is recognized as a techniques that can overcome these difficulties with respect to other techniques. PLD was successfully employed to grow high crystalline ZrC films in this study. By optimizing the deposition conditions, epitaxial ZrC films on silicon and sapphire substrate were obtained. The structure, stoichiometry, and optical and electrical properties of these films are described in this chapter. 4.2 Experiment The film depositions were conducted in an all-metal vacuum chamber using a KrF excimer laser (=248 nm). First of all, experiments were focused on studying the effects of growth conditions (or process parameters) on the deposited films. Especially an effort was made for growing crystalline ZrC films. Within the range of 2 ~ 10 J/cm2 laser fluences with 5 Hz laser pulse repetition rate, the ZrC films were deposited on Si (001) at the temperature range of 200 ~ 700 oC. Si (001) substrates were cleaned chemically by acetone, methanol, and rinsed in de-ionized water in turn, then dipped in 1 % HF solution for 1 min, blown dry by high purity nitrogen gas, and immediately loaded into the deposition chamber. Depositions were performed under residual vacuum or a low C2H2 atmosphere. The study was continued to produce high quality films by optimizing growth parameters and to study growth behavior on various substrates. At this time, special care was taken to maintain low water vapor pressures below -8 Torr during depositions, as measured with a residual gas analyzer (RGA) attached to the deposition chamber. Additionally, Si substrates were heated to 900 oC and maintained for 20 min under high vacuum to remove the passivation layer on the substrate. The laser parameters used were 10 J/cm2 fluences and 10 Hz repetition rate at substrate temperatures around 750C.

PAGE 47

33 The films surface and interfacial roughness, mass density and thickness were obtained by simulating the measured x-ray reflectivity (XRR) spectra acquired with a Panalytical XPert MRD system. The same instrument was used for structural characterization in symmetric and grazing incidence x-ray diffraction (XRD and GIXD). Pole figure measurements were acquired both from the films and substrates for texture characterization. Omega rocking curve method was used to evaluate quality of epitaxial films. The thickness and optical properties of the films were measured by spectroscopic ellipsometry (SE, Woollam M-88). The chemical composition of the films was investigated by Auger electron spectroscopy (AES, Perkin-Elmer PHI 660). To analyze the bonding structure of ZrC, X-ray photoelectron spectroscopy (XPS, Perkin-Elmer PHI 5100 ESCA) was used. Focused Ion Beam (FIB) Strata DB 235 technique was employed to make cross-sectional samples for high resolution transmission electron microscopy. Bright field image and electron diffraction pattern were obtained from high resolution JEOL TEM 2010F to analyze microstructure and quality of deposited thin films. Details of each experiment are described in chapter 3. 4.3 Results and Discussion 4.3.1 Laser Fluence and Temperature Effect on Deposited Films ZrC films were deposited on Si (001) substrates to study the effect of substrate temperatures and laser fluences on the grown films. The films were grown for 20 minutes at various growth conditions, which resulted in estimated thickness around 400 The deposited films were analyzed by surface sensitive grazing incidence XRD, and the results at 1 of an incidence beam angle are displayed in fig. 4-1. The growth conditions of the films are listed in table 4-1. Under low temperature conditions of 200 ~ 600 C and/or low laser fluence range of 2 ~ 6 J/cm2, GIXD spectra did not show any diffraction

PAGE 48

34 peak (ZC05 and ZC06 in fig. 4-1), indicating that the deposited films under these conditions were amorphous. By increasing the laser fluence to 8 J/cm2 and substrate temperature at 600 C, the deposited films started to show some crystallinity (ZC11 in fig. 4-1). Table 4-1. Growth conditions of deposited films, showing GIXD spectra in fig. 4-1. Pressure during deposition Substrate temperature Laser fluence Repetition rate ZC06 110-3 Torr C2H2 500 C 6 J/cm2 5 Hz ZC05 110-5 Torr C2H2 600 C 3 J/cm2 5 Hz ZC11 110-5 Torr C2H2 600 C 8 J/cm2 5 Hz ZC24 110-4 Torr C2H2 600 C 10 J/cm2 5 Hz ZC26 8.510-4 Torr C2H2 600 C 10 J/cm2 5 Hz ZC25 110-4 Torr C2H2 700 C 10 J/cm2 5 Hz When the ZrC target was not pre-ablated for at least 1 min, GIXD pattern (incidence beam angle = 1) of as-grown films in fig. 4-2 (a) showed some additional diffraction peaks other than those obtained from target in fig. 4-3, which were crystalline with diffraction lines corresponding to cubic and stoichiometric ZrC (a0 = 4.69 JCPDS PDF# 35-0784 [Pcp94]). Those additional diffraction peaks (vertical solid lines) in as-grown films were assigned to substoichiometric ZrC0.7 compound (a0 = 9.38 cubic (space group: P), JCPDS PDF# 32-1489 [Pcp94]), which was probably caused by the oxidation of target surface. So the target was pre-ablated for 1 min hereafter, ahead of the growth processes to remove any surface oxide layer on target. Resulting GIXD spectra in fig. 4-2 (b) showed a stoichiometric Zr1C1 compound (JCPDS PDF# 35-0784 [Pcp94]),

PAGE 49

35 which is indicated as dashed vertical lines. In addition, it should be mentioned here that a low pressure (-5 ~ -3 Torr) of acetylene (C2H2) gas was used as background gas for this study because GIXD spectrum from a film deposited at low C2H2 atmosphere in fig. 4-4 showed a little improvement of crystallinity in the film. Other deposition conditions except background gas pressure were all same (Ts=600 C, 8J/cm2, and 5Hz). At laser fluence at 10 J/cm2, resulting deposited films showed better crystallinity, evidenced by clear peaks in GIXD spectra (ZC24, ZC25, and ZC26 in fig. 4-1). All observed the diffraction lines corresponded to a stoichiometric cubic ZrC (JCPDS PDF# 35-0784 [Pcp94]), which were identical to those from the pure ZrC target in fig. 4-3. The crystallinity of the films deposited at higher substrate temperature was much improved at laser fluence of 10 J/cm2 (from ZC24, 600 C to ZC25, 700 C in fig. 4-1). Therefore it is apparent that the deposition of crystalline films requires the simultaneous use of high substrate temperatures and high laser fluences. 4.3.2 Deposition Rate and Thickness Uniformity of Deposited Films The effect of deposition time was investigated on density and roughness of the ZrC film grown on silicon (001) substrate at 700 C under residual vacuum (1-6 Torr) for 7, 14, and 20 min. The used laser parameters were 10 J/cm2 fluence, and 5 Hz repetition rate. The layer thicknesses and roughness of these films were extracted by fitting the acquired x-ray reflectivity (XRR) spectra using the WinGIXA software from Panalytical, which is based on the Parratt formalism modified to include roughness [Par54 and Nv80]. Instead of using only one ZrC layer model for fitting, a three-layer model was used that resulted in a better fitting: an interfacial layer, a ZrC layer, and a contaminated surface layer such as oxygen and hydrogen oxide. Also, thicknesses measured by spectroscopic ellipsometry were compared with those obtained by XRR data. From the measured x-ray reflectivity

PAGE 50

36 spectra in fig. 4-5, the critical angles indicating the density of films [Rus02 and Zab94] were quite similar for all deposited films, and as high critical angle indicated (c ~ 0.35), high density values around 6.7 g/cm3 were obtained. The thickness, roughness, and density parameters obtained from x-ray reflectivity curves are listed in table 4-2. The thickness change of deposited film was linearly time dependent as shown in fig 4-6. The average deposition rate was 0.34 /sec (or 0.068 / pulse). This was a rather low deposition rate for PLD, implicating rather low vaporized ZrC concentration in a plasma plume. Table 4-2. Thickness, density, and surface roughness values of the ZrC films deposited at 700 C for different times. Deposition time (min) 7 14 20 Thickness () 17 18 18 Surface layer (ZrCxOy) Roughness (, rms) 3.3 5.3 5.3 Thickness () 127 297 376 ZrC layer Density (g/cm3) 6.74 6.69 6.73 XRR Interfacial layer Thickness () 0.6 2.3 2.0 Surface layer Thickness () 10 10 9 ZrC layer Thickness () 125 310 382 Ellipsometry Interfacial layer Thickness () 1.0 1.5 1.0 The GIXD (incidence angle = 1) spectra of these films shown in fig. 4-7 indicate that all crystalline orientations are present in the films. The intensities of these peaks slightly increased with an increase of thickness, indicating some increase in the volume of the randomly oriented grains. However, these diffraction peaks did not appear in the corresponding symmetric XRD spectra displayed in fig. 4-8, that is, the amount of these randomly oriented grains was rather small. According to the XRD results (fig. 4-8),

PAGE 51

37 the deposited films under this condition grew with a (001) texture. It is also clear that, based on XRR data from table 4-2, there are no significant changes in the film density or surface morphology when the deposition time was changed. The thickness uniformity of the deposited films was checked by acquiring XRR spectra at various positions from the same samples that were grown under identical condition. According to the data extracted from these XRR spectra, a ZrC film thickness uniformity of around 5% in the central 1.5.5 cm2 area of the deposited film was estimated. Table 4-3. Degree of out-of-plane texture of ZrC films at various growth conditions. Substrate ID P0 (Torr) Pd (Torr) Ts (C) Texture FWHM (Omega-Rocking) ZC202 3.010-7 6.210-6 C2H2 775 (001) 2.53 ZC104 2.610-6 1.210-5 C2H2 750 (001) 3.37 ZC103 2.410-6 4.010-6 C2H2 750 (001) 4.47 ZC208* 1.510-7 2.210-6 CH4 750 (001) 4.62 Si (001) ZC106 9.010-7 1.010-5 C2H2 750 (001) 5.45 ZC104 2.610-6 1.210-5 C2H2 750 I111/I200=9.08 4.82 ZC107 8.010-7 6.710-6 C2H2 750 1.3 ZC210* 1.410-7 9.010-7 CH4 750 Si (111) ZC208* 1.510-7 2.210-6 CH4 750 ZC107 8.010-7 6.710-6 C2H2 750 (111) 0.46 ZC104 2.610-6 1.210-5 C2H2 750 (111) 0.68 ZC208* 1.510-7 2.210-6 CH4 750 (111) 2.31 Sapphire (0001) ZC204 1.310-7 7.210-6 C2H2 755 (111) 2.53 Titanium was contaminated unintentionally for (*) marked samples.

PAGE 52

38 4.3.3 ZrC Films Growth at High Temperature and High Laser Fluence In this section, ZrC film growth was carefully controlled at higher substrate temperature and higher laser fluence growth conditions based on previous study. Also films were deposited on Si (001), Si (111), and sapphire (0001) single crystal substrates at high temperatures around 750 C to investigate texture, quality, and properties of grown films. The representative growth conditions and results used for ZrC films growth are presented in table 4-3. Special care was taken to maintain low water vapor pressures during depositions by monitoring partial residual gas pressure. Also the Si substrates were heated up to 900 oC and maintained for 20 min under high vacuum to remove the passivation layer on the substrate before deposition. Table 4-4. Structure information of ZrC Si, and sapphire and lattice mismatch. ZrC Silicon Sapphire Unit Cell Cubic (NaCl) Cubic (Dia.) Rhom. (Hex.) Space Group Fm3m (225) Fd3m (227) R 3 c (167) Lattice Parameter () 4.6930 (JCPDS PDF#: 35-0784) [Pcp94] 5.4309 (JCPDS PDF#: 27-1402) [Pcp94] 5.4449 (750 C) [Yim74] a = 4.7592, c = 12.992 (JCPDS PDF#: 43-1484 ) [Pcp94] a = 4.7828 (750 C) [Yim74] Lattice mismatch (%) with ZrC 13.59 at RT 13.41 at 750 C (a ZrCZrC, a ZrC<110>>a Sapphire) Structural information for ZrC and substrates is presented in table 4-4, and the lattice mismatch is calculated with respect to the substrates. For sapphire, since its basal plane has same geometry as ZrC (111) plane, lattice mismatch was calculated with 0211 of sapphire parallel to <110> of ZrC. Lattice mismatches to Si and sapphire are

PAGE 53

39 13.4 % and 39.4 %, respectively. For such a large lattice mismatch, definitely elastic theory can not be applied for epitaxial growth because critical thickness for that big mismatch will be less than 2 [Hu91]. However, the resulted grown films on Si and sapphire were epitaxial. This epigrowth of ZrC could be explained by coincidence-site lattice (CSL) or translational symmetries assumptions [Zur84]. The details were investigated by X-ray techniques and TEM, and oxidation problem was also considered in this section. In addition, electrical resistivity of these films was measured. 4.3.3.1 Growth behaviors of ZrC films deposited on Si and sapphire substrate In fig. 4-9, symmetric theta-2theta XRD spectra acquired from the films (ZC104 samples in table 4-3) grown on Si (001), Si (111), and sapphire (0001) substrates are presented. For the ZrC films deposited on Si substrates, XRD spectra showed that growth planes of the films were the same as Si substrate surface planes, that is, ZrC (001) films were grown on Si (001) substrates, and ZrC (111) films were grown on Si (111) substrates. Besides, ZrC (111) films were grown on the sapphire (0001) basal plane. For all the films grown on Si (001) and sapphire substrates, no other peaks were found other than one peak in theta-2 theta symmetric XRD scan as shown in fig. 4-9, indicating that the films were highly out-of-plane textured. For the films grown on Si (111) substrate, the XRD peaks (top three spectra in fig. 4-10) showed two out-of-plane textures of (111) and (001). Due to high degree of texture, GIXD spectra at incidence angle of 1 (bottom three spectra in fig. 4-10) showed that the films deposited on Si (111) substrates exhibited barely visible humps, indicating a very small amount of grains possessing other crystalline orientations. To check the texture degree, omega-rocking curves of the either ZrC (111) or ZrC (002) peaks were recorded for the ZrC films grown under the same conditions (ZC104

PAGE 54

40 samples) but different substrates. As one can see in fig. 4-11, the films deposited on sapphire exhibited the highest texture with FWHM (full width at half maximum) of the omega-rocking curves of only 0.68. The films deposited on Si (001) exhibited a FWHM of the ZrC (002) peak of 3.37, while those deposited on Si (111) exhibited the rather large FWHM value of 4.82 for the ZrC peak (111). The Growth conditions of samples and measured FWHM of omega-rocking results are presented in table 4-3. The growth conditions of samples were similar except small change of background gas pressures (2.2-6 ~ 12-6 C2H2). However, the FWHM values of omega-rocking curves were dramatically changed from 2.53 (ZC202, lowest) to 5.45 (ZC106, highest) for the films deposited on Si (001), whereas the FWHM values of the films deposited on sapphire (0001) do not seem to be sensitive to the change of background gas pressure (ZC104 and ZC107 in table 4-3). Measured omega-rocking curves for the films deposited on Si (001) and sapphire (0001) are displayed in fig. 4-12 and fig. 4-13, respectively. The reason of the rather high FWHM value for ZC204 sample in fig. 4-13 was due to accidentally discontinued growth. To clarify the causes which affected crystal quality, lattice parameters for the films grown on Si (001) were calculated from ZrC (002) reflection of symmetric XRD in fig. 4-14, because crystal quality could be changed by strain energy in films during growth. As one can see in fig. 4-15, lattice parameter was linearly changed rather rapidly (a = ~1 %) to a small change of background gas pressure (P = ~1.0-5 C2H2). On the basis of the lattice parameter (a = 4.6930) of reference powder diffraction (JCPDS PDF#: 35-0784 [Pcp94]), the closer measured lattice parameter, the better crystal quality in terms of rocking curve measurements, as it is shown in fig. 4-16.

PAGE 55

41 In-plane texture was observed by phi-scan for {111} in-plane of ZrC (001) film (ZC202 in table 4-3) as displayed in fig. 4-17 (a). For cubic structures such as ZrC, four peaks should appear at the x-ray diffractometer configuration of phi-scan as displayed in fig. 4-17 (b), if in-plane texture exists. Details of the in-plane orientation were investigated by pole figure measurement of several crystalline orientations for both the films and the substrates. Typical results for ZrC films deposited on Si (001), Si (111), and sapphire with their substrates are displayed from fig. 4-18 to fig. 4-20. In these figures, the orientation relationships between films ((a) of each fig.) and substrates ((b) of each fig.) are also represented together, by starting the rotation of film and substrate from the same angular position. As one can see in these figures, the films were exhibiting a rather good in-plane texture, therefore, being epitaxial. From (111) pole figures in fig. 4-18, which are obtained from both ZrC film deposited on Si (001) substrate and Si (001) substrate, very clear four in-plane <111> pole reflections were only showed. The angular directions of the four <111> poles are the same as <111> pole directions of their substrates. This fact clearly indicates that (100)ZrC // (100)Si and [100]ZrC // [100]Si orientation relationship between film and substrate, which follows the concept of coincidence-site lattice (CSL), because when there is a good lattice match between film and substrate, high quality epitaxial growth is possible. Also the (100) pole figure of ZrC films grown on Si (111) substrate showed in-plane texture as indicated red circles in fig. 4-19 (a), but rather contained small fraction of randomly oriented inhomogeneous grains (displayed as blue contour), while the pole figure of ZrC films grown on Si (001) substrate showed cube on cube growth.

PAGE 56

42 In case of ZrC films grown on sapphire (0001) substrate, (100) pole figure (fig. 4-20 (a)) from film and (1 1 2 6) pole figure (fig. 4-20 (b)) from sapphire substrate were examined to clarify the orientation relationship between the sapphire substrate and ZrC film. In the cubic system, three in-plane (100) reflections are formed by an angle of 120 with each other for a single crystal having (111) surface plane, thus the presence of six in-plane (100) reflections in fig. 4-20 (a) suggests the existence of two different in-plane crystal orientations, which are 60 rotated relatively to the other crystal orientation on same <111> axis as indicated by red and blue circle in fig. 4-20 (a). This epigrowth was not expected for the interface having big mismatch (~39%) between sapphire <11 2 0> direction and ZrC <110> direction. However, this fact could be well explained by translational symmetries on both sides of the interface. We can define two lattices to match by translational symmetry, instead of comparing the bulk lattice parameters [Zur84]. As orientation relationship between ZrC and sapphire was presented in a schematic diagram (fig. 4-21), the superlattice mismatch is less than 0.6% between five atomic distances of film along the <110> direction and four atomic distances of substrate along the <01 1 0> direction. When superlattice mismatch is less than 1%, lateral movement of 0.5 % of every atom on both sides of the interface can accommodated this superlattice mismatch [Zur84]. For two different crystallographic orientations of ZrC (111) grown on sapphire, it can be seen that there is only one possible nucleation site for atoms of the first monolayer at the interface, which is shown as A in fig. 4-22. However for the second layer, there are two different sites for atomic array, since the binding energy differences for adatoms occupying B sites or C sites are very small. As long as the island growth model is

PAGE 57

43 considered where there are many nucleation sites on the surface of the substrate during the initial stage of growth, some nuclei grow in the ABCABC order; and other nuclei grow in ACBACB order. Thus two twin boundaries can be possibly formed. One is perpendicular to the surface when these crystallites with different orientations come together. The boundary is usually semicoherent or noncoherent depending on the habit plane [Bra66]. Also this fact is responsible for the rather large reflection area in pole figure in fig. 4-20 (a), indicating in-plane mosaicity. The other one is a parallel twin boundary, which is parallel to the surface of the film. During perpendicular growth, the stacking sequence of the (111) atomic planes could be lost by accident. In this case, the crystallites will grow in twin relation to the previous crystallite, and the boundary is coherent. Due to relatively small boundary energy in coherent twin boundary, its normal stacking sequence of (111) plane can be easily lost. 4.3.3.2 TEM analysis of ZrC films grown on Si and sapphire substrate For further understanding of growth behavior and crystal structure of ZrC on Si and sapphire, cross-sectional specimens were prepared by FIB (focused ion beam) process. Before the FIB process, a carbon layer was coated on the surface of films to protect ion beam damage on film and to avoid electron charging. Just before FIB process, Pt was coated again to prevent from high energetic ion beam induced damage. In fig. 4-23, a high resolution bright field image taken from the cross-section of the film grown on Si (111) substrate (ZC104 sample in table 4-3) is displayed. As it was expected from GIXD investigation, some random oriented grains were observed as inhomogeneous particles in continuously well stacked (111) ZrC film matrix rather than completely random oriented polycrystalline or columnar structure. These random grains were grown from very initial stage of growth, as they were located right at the interface,

PAGE 58

44 indicating that chemical contribution to interfacial energy is much higher than that of ZrC grown on Si (001) substrate, assuming structure contribution to the interfacial energy is the same due to same lattice mismatch as ZrC grown on Si (001). An electron diffraction pattern of this sample was taken from the region shown in fig. 4-24 (a). Beam direction was defined by strong diffracted electron beam reflections from fig. 4-24 (b). Therefore, the cut plane was identified as (11 2 ) plane, which is perpendicular to surface plane, meaning strong (111) texture. In addition, diffused electron diffraction pattern identified as (110) plane also can be seen in fig. 4-24 (b), indicating co-existing (001) texture with (111) texture other than random orientations. From ZrC film grown on Si (001) substrate (ZC202 sample in table 4-3) in fig. 4-25 (a), a selected area diffraction pattern (SADP) was obtained in fig. 4-25 (b) to confirm the quality of the epitaxial film, and the orientation relationship between the film and Si substrate. As good epitaxial quality was expected from four clear and small (111) reflections in fig. 4-18 (a), SADP confirmed ZrC (001) was well aligned on Si (001) substrate as a single crystal with cube on cube relationship. In a pair of spots, the inner spots are coming from Si substrate, while the outer spots are from ZrC film. High resolution image in fig. 4-26 of this film showed clean lattice fringes without any Moir lattice fringes. Several cross-sectional TEM pictures taken from ZC104 film grown on sapphire (0001) were analyzed to verify the previously suggested twin structure used to explain the pole figure results. Suggested two types of twin structure were found in the regions marked as A and B from fig. 4-27. In a magnified image (fig. 4-28 (a)) of the region A, twin boundary parallel to the surface was observed as mirror image was observed by

PAGE 59

45 drawn lines along atomic arrangements. A twin boundary perpendicular to the surface was also observed in region B in fig. 4-28 (b). The parallel twin boundary was very coherent by matching lattices of both sides, while the other showed an incoherent perpendicular twin boundary with atomic displacement at boundary. 4.3.4 Surface analysis of ZrC films The density and surface roughness of films were estimated by modeling acquired XRR spectra using the Wingixa software from Panalytical, described in section 4.3.2 and chapter 3. Regardless of the substrate type, similar density values of around 6.7 g/cm3 were obtained (ZC208 and ZC210 samples were excluded for estimation due to Ti contamination), and film surface was atomically flat with rms roughness value of 0.4 ~ 0.7 0.2 nm. The measured density values which are identical to tabulated values [Lid05] suggest that films are compact ZrC. The surface contamination layer density was around 4 ~ 5 0.2 g/cm3, indicating the presence of an oxide or hydroxide compound or a mixture of both. The secondary electron (SE) images obtained by field emission SEM showed that surface morphology was very smooth at K for most deposits, and only at very high magnification ( K), surface roughness was discerned. SE images taken from ZrC film surfaces on Si and sapphire substrate are displayed, respectively in fig. 4-29 (ZC202 sample) and fig. 4-30 (ZC204 sample). Density and roughness values of as-grown films are listed in table 4-5. The rms values obtained from AFM height images from fig. 4-31 to fig. 4-36 confirmed that roughness of surface was less than 1 nm for all samples. But surface morphology was little different depending on growth direction of growing films. That is, ZrC (001) surface showed many small cluster looking features as shown in fig. 4-31 ~ fig. 4-33, whereas ZrC (111) surface was atomically very flat

PAGE 60

46 enough not to be discerned by AFM image as displayed in fig. 4.34 ~ fig. 4.36 under all experimental growth conditions. Table 4-5. Surface roughness and density of as-grown films. Density (g/cm3) Substrate ID Pd (Torr) Ts (C) Thickness () Film Topmost Roughness () ZC106 1.010-5 C2H2 750 714 6.7 4.1 6 ZC107 6.710-6 C2H2 750 376 6.7 4.3 7 ZC202 6.210-6 C2H2 775 1034 6.8 5.8 5 ZC208 2.210-6 CH4 750 368 6.9 4.3 4 Si (001) ZC210 9.010-7 CH4 750 714 6.8 5.6 5 ZC107 6.710-6 C2H2 750 333 6.7 4.3 6 Si (111) ZC208 2.210-6 CH4 750 381 6.9 4.9 7 Sapphire ZC208 2.2-6 CH4 750 383 6.8 5.6 5 (0001) ZC210 9.0-7 CH4 750 656 6.8 5.2 5 As mentioned before in chapter 2, one of problems in ZrC film growth is oxygen contamination due to high zirconium atoms affinity for oxygen. As an example, when laser was stopped during ZrC deposition (discontinuous growth), a bright layer (atomic contrast) was formed due to high oxygen contamination as displayed in a cross-sectional bright field TEM image (fig. 4-37). For quantitative x-ray microanalysis of a ZrC film which was grown on Si (001), a scanning transmission electron microscope (STEM) equipped with an x-ray energy dispersive spectrometer (XEDS) was used. Sample ZC106 was chosen because it showed most distinguishable three different layer structures among samples as shown in fig. 4-38. The result is shown by TEM-EDX analysis in fig. 4-39. From line spectra, carbon and oxygen elements were almost constant through thickness and detected a little more near surface. Oxygen content was 9.6 atomic % near Si/ZrC interface (spot 3), a little lower

PAGE 61

47 value of 5.2 % in bulk (spot 4), and again higher value of 13.8 % at surface layer (spot 5). For carbon content, a higher value (~70%) was detected. But this could not be true because the characteristic x-ray absorption by carbon counted too low, and due to carbon supported TEM grid behind sample and protecting carbon coating near surface. However, the ratio of Zr to carbon was quite constant near interface (spot 3) and in bulk (spot 4), implying stoichiometric composition. For detail surface composition analysis, XPS investigations were performed. These investigations revealed the presence of several C 1s lines on the surface of the deposited ZrC films. Since the binding energies of the XPS peaks are usually referenced to the binding energy of adventitious carbon, this fact could introduce errors in this case. Thus pure ZrO2 films deposited on Si by PLD [How02] were analyzed to accurately determine the position and shape of Zr 3d and O 1s peaks to avoid this problem. In fig. 4-40, high resolution XPS scans of the Zr 3d region acquired at 45 and 90 take-off angles from a sample deposited at 600 C under vacuum (5-6 Torr) are shown. The used laser parameters were 8 J/cm2 fluence, and 5 Hz repetition rate. It is apparent that the surface contains a rather high percentage of Zr-O bonds [Coc98] (peaks denoted by B located at 184.5 eV and D at 186.9 eV, respectively). From a sample that was deposited at 600 C under 1-4 Torr of C2H2, XPS spectra are shown in fig. 4-41 after 5 min sputtering with 4 keV Ar+ ions, which removed the first 4 nm of the outermost layer. For this sample, there is a minimal angle dependence of the ratio of Zr-O to Zr-C peaks, indicating a homogeneous bulk composition. The percentages of the Zr atoms bonded to carbon are shown in table 4-6, which is estimated from the XPS measurements for a series of samples. Based on XRD data in fig. 4-4, the use of a low C2H2 atmosphere

PAGE 62

48 appeared to have a beneficial effect on crystallinity too. However, the oxygen concentration was rather high in the surface region of these films. Table 4-6. The relative percentage of Zr 3d XPS areas corresponding to Zr-C bonds at different take-off angles. As-received Sputtered Deposition Pressure 45 90 45 90 5-6 Torr 20.6 % 31.5 % 44.9 % 45.2 % 1-4 Torr of C2H2 14.6 % 24.8 % 47.2 % 49.1 % 7-4 Torr of C2H2 1.0 % 3.8 % 22.4 % 29.0 % Table 4-7. Resistivity of as-deposited ZrC films I(A) V(mV) 1 (from t1) 2 (from t2) t1(nm) (TEM) t2(nm) (XRR) ZC104i1 1 30 0.302 1.87E-04 1.83E-04 41 40 ((111)Si) 2 30 0.292 1.81E-04 1.76E-04 41 40 3 30 0.282 1.75E-04 1.70E-04 41 40 Z202 1 30 0.115 1.63E-04 1.79E-04 94 103 ((100)SI) 2 30 0.113 1.60E-04 1.76E-04 94 103 3 30 0.106 1.51E-04 1.65E-04 94 103 4 30 0.105 1.49E-04 1.63E-04 94 103 Z204a 1 30 0.115 1.55E-04 1.51E-04 89 87 (Sapphire) 2 30 0.116 1.56E-04 1.52E-04 89 87 3 30 0.12 1.61E-04 1.58E-04 89 87 4 100 0.3855 1.56E-04 1.52E-04 89 87 5 100 0.383 1.54E-04 1.51E-04 89 87 AES investigations were performed to analyzed the bulk composition of the film. The results confirmed that the first 2.0 ~ 3.0 nm (surface region) of the ZrC films were heavily contaminated with oxygen as one can see in fig. 4-42, which was a typical depth profile acquired from an as-deposited film. However, once the topmost layer was

PAGE 63

49 removed by Ar ion sputtering, the oxygen content dramatically decreased to values below 7 ~ 8 % as shown in fig. 4-43 (ZC202). Despite relatively the high levels of oxygen contamination, the deposited ZrC films were very conductive, and similar values were obtained in table 4-7. Under best vacuum conditions (Po = 1.3 ~ 1.5-7 Torr) and using a low pressure of high purity CH4 during deposition, oxygen concentration was less than 2.5 % within the bulk, as shown in fig. 4-44 (ZC107). According to residual gas analysis by RGA (residual gas analyzer), just right before deposition and after introducing C2H2 and CH4 gas, C2H2 gas contained rather high oxygen concentration when compared with pre-vacuum condition as displayed in fig. 4-45. So it is thought that oxygen contamination of the films deposited under C2H2 atmosphere is caused by the limited purity of the used gas. 4.4 Summary In summary, ZrC thin films were deposited on Si and sapphire substrates by the pulsed laser deposition technique. A combination of high laser fluence and high substrate temperature (600 ~ 700 C) was required to obtain crystalline films on Si (001) substrates. Under very low water vapor pressures (~10-8 Torr), high substrate temperatures (~750 C), and high laser fluence (10 J/cm2), epitaxial ZrC films were deposited on single crystalline substrates. The ZrC films grew along the [001] axis on Si (001), while they grew along the [111] axis on Si (111) and sapphire (0001). Pole figure measurements showed that ZrC films exhibited in-plane orientation too, depending on the type of substrates. Grazing incidence x-ray diffraction investigations evidenced the presence of a rather small fraction of randomly oriented crystallites. The films mass density was around the tabulated value of 6.7 g/cm3, while the surface morphology was very smooth, with a roughness value (rms) of 0.4 ~ 0.7 0.2 nm.

PAGE 64

50 C2H2 as background gas increased a chance to bond Zr with C, and improved crystallinity and stoichiometry. Lattice parameter of deposited ZrC film was changed by small variations of C2H2 pressure. There exists an optimum pressure value for good crystal quality. Contamination of mainly Zr-O was found in the surface layers deposited under C2H2 or residual atmosphere, whereas the oxygen content dramatically decreased to values below 7 ~ 8 % after topmost layers were removed by Ar ion sputtering. Despite the high levels of oxygen contamination, the deposited ZrC films were very conductive. Under best vacuum conditions (Po = 1.3 ~ 1.5-7 Torr) and using a low pressure of high purity CH4 during deposition, oxygen concentration was dramatically reduced to less than 2.5 %.

PAGE 65

51 Figure 4-1. GIXD spectra (incidence beam angle, = 1) of ZrC films deposited under various conditions. Figure 4-2. Comparison of GIXD spectra obtained from ZrC deposited films at same deposition conditions (a) without pre-ablated target (red) and (b) with pre-ablated target (blue).

PAGE 66

52 Figure 4-3. XRD spectra obtained from ZrC target; vertical lines represent position and intensity for stoichiometric ZrC, JCPDS PDF# 32-1489. Figure 4-4. GIXD spectra obtained from films deposited under residual vacuum and different C2H2 gas pressures.

PAGE 67

53 Figure 4-5. XRR spectra of ZrC films deposited for various times at 700 C and 10 J/cm2 under vacuum. y = 0.0344x010203040500200400600800100012001400Time (sec)Thickness (nm) Figure 4-6. ZrC film deposition rate at Ts = 700C, Pd = 1.010-6 Torr, 10 J/cm2, and 5Hz.

PAGE 68

54 Figure 4-7. GIXD spectra of ZrC films deposited for various times at 700 C and 10 J/cm2 under vacuum; the standard position of diffraction lines from ZrC (dashed lines) and ZrC0.7 (solid lines) are also shown. Figure 4-8. XRD spectra of ZrC films deposited for various times at 700 C and 10 J/cm2 under vacuum.

PAGE 69

55 Figure 4-9. XRD spectra of ZrC films deposited at 750 C on various substrates Figure 4-10. XRD and GIXD spectra (incidence beam angle = 1) of ZrC films deposited on Si (111) substrates.

PAGE 70

56 10 12 14 16 18 20 22 24 26 Omega 0 2000 4000 6000 8000 10000 12000 counts/s Omega-Rocking CurveZrC (002) on Si (001)ZrC (111) on Sapphire ZrC (111) on Si (111) Figure 4-11. Omega-rocking curves of ZrC (111) or (200) peaks recorded from ZC104 films deposited at 750 C on various substrates. 12 14 16 18 20 22 24 26 Omega 0 5000 10000 15000 counts/s ZC202ZC103ZC104ZC106ZC208Omega-rocking CurveZrC (001) on Si (001) Figure 4-12. Omega-rocking curves of ZrC (002) peaks recorded from the films deposited on Si (001) substrate at various background gas pressures.

PAGE 71

57 10 12 14 16 18 20 22 Omega 0 1000 2000 3000 4000 5000 6000 7000 counts/s ZC104 ZC107 ZC204 Figure 4-13. Omega-rocking curves of ZrC (111) peaks recorded from the films deposited on sapphire (0001) substrate at various different gas pressures 30 35 40 45 Theta 0 5000 10000 15000 20000 25000counts/s ZC104 (blue, 4.701)ZC202 (red, 4.685)ZC103 (green, 4.666)ZC106 (pink, 4.705)ZC208 (black, 4.653)ZC (001) on Si (001)Lattice Parameter (Angstrom) Figure 4-14. XRD spectra of (002) reflection from the films deposited on Si (001) and calculated lattice parameters.

PAGE 72

58 0.4650.4660.4670.4680.4690.4700.4710.4720.00E+004.00E-068.00E-061.20E-051.60E-05C2H2 or CH4Lattice parameter (nm) FWHM=2.53 (ZC202)3.37 (ZC104)4.47 (ZC103)4.62 (ZC208)5.45 (ZC106) ZrC (001) on Si (001) Ref. 0.469 nm (35-0784) Figure 4-15. Background gas effect on lattice parameter of the films deposited on Si (001) substrate. 01234560.4640.4650.4660.4670.4680.4690.4700.4710.472Lattice parameter (nm)FWHM (degree, omega-rocking) ZC202ZC104ZC103ZC208ZC106 ZrC (001) on Si (001) Ref. 0.469 nm (35-0784) Figure 4-16. The relationship between deposited films on Si (001) and texture degree of the films measured by omega-rocking curve.

PAGE 73

59 Figure 4-17. Phi-scan of {111} in-plane obtained from (a) ZrC film and Si (001) substrate of sample ZC202 and (b) phi-scan diffractometer configuration.

PAGE 74

60 Figure 4-18. (111) pole figures of (a) ZrC film and (b) Si (001) substrate obtained from sample ZC202.

PAGE 75

61 Figure 4-19. (100) pole figures of (a) ZrC film and (b) Si (111) substrate obtained from sample ZC104.

PAGE 76

62 Figure 4-20. Pole figures showing (a) (100) pole figures of ZrC film, and (b) (1 1 2 6) pole figure of sapphire (0001) substrate obtained from ZC104.

PAGE 77

63 Figure 4-21. Projection view for two crystallographic orientations of ZrC grown on sapphire (0001), and orientation relationship between ZrC film and sapphire substrate. Figure 4-22. Possible nucleation site (marked as A) for the first monolayer on sapphire (0001).

PAGE 78

64 Figure 4-23. Bright field TEM image obtained from cross-section of ZrC film grown on silicon (111) substrate; the regions showing inhomogeneous random grains are marked.

PAGE 79

65 Figure 4-24. TEM (a) bright field image and (b) SADP (selected area electron diffraction pattern) obtained from cross-section of ZrC film (sample ZC104) grown on silicon (111) substrate.

PAGE 80

66 Figure 4-25. TEM (a) bright field image and (b) SADP (selected area electron diffraction pattern) obtained from cross-section of ZrC film grown on silicon (001) substrate; diffraction pattern obtained from ZrC film is marked by circles.

PAGE 81

67 Figure 4-26. Bright field TEM image obtained from cross-section of ZrC film grown on silicon (001) substrate exhibiting clear lattice fringe.

PAGE 82

68 Figure 4-27. Bright field TEM image obtained from cross-section of ZrC film grown on sapphire (0001) substrate showing sharp interface; the regions marked as A and B are magnified in figure 4-28 for observation of twinning.

PAGE 83

69 Figure 4-28. High resolution TEM image of (a) A region in fig. 4-26, showing parallel twin to the surface of film; (b) B region in fig. 4-26, showing perpendicular twin to the surface of film.

PAGE 84

70 Figure 4-29. Secondary electron images of ZrC surface grown on Si (001) substrate, at K (a), and K (b).

PAGE 85

71 Figure 4-30. Secondary electron images of ZrC surface grown on sapphire (0001) substrate, at K (a), and K (b).

PAGE 86

72 Figure 4-31. AFM height images obtained from the surface of ZC106 sample (ZrC (001) layer grown on Si (001) substrate)

PAGE 87

73 Figure 4-32. AFM height images obtained from the surface of ZC202 sample (ZrC (001) layer grown on Si (001) substrate)

PAGE 88

74 Figure 4-33. AFM height images obtained from the surface of ZC208 sample (ZrC (001) layer grown on Si (001) substrate)

PAGE 89

75 Figure 4-34. AFM height images obtained from the surface of ZC107 sample (ZrC (111) layer grown on Si (111) substrate)

PAGE 90

76 Figure 4-35. AFM height images obtained from the surface of ZC208 sample (ZrC (111) layer grown on sapphire (0001) substrate)

PAGE 91

77 Figure 4-36. AFM height images obtained from the surface of ZC210 sample (ZrC (111) layer grown on sapphire (0001) substrate)

PAGE 92

78 Figure 4-37. Cross-sectional bright field TEM image showing an oxidized layer due to discontinued growth. Figure 4-38. Cross-sectional bright field TEM image showing distinguishable three layers, also supporting a model used for XRR analysis.

PAGE 93

79 Figure 4-39. Z-contrast image of cross-sectional ZC106 sample for TEM-EDX analysis by line and point scan.

PAGE 94

80 Figure 4-40. High resolution Zr 3d spectra acquired at 45 and 90 take off angles and their fitting for an as-received sample deposited at 600 C under vacuum.

PAGE 95

81 Figure 4-41. High resolution Zr 3d spectra acquired at 45 and 90 take off angles and their fitting for a sample deposited at 600 C under 1-4 Torr of C2H2 that was sputtered-clean by Ar+ bombardment.

PAGE 96

82 Figure 4-42. AES depth profile of an as-deposited ZrC film. Figure 4-43. AES survey spectrum of a ZrC film (ZC202) sputtered with a 4 kV Ar ion beam.

PAGE 97

83 Figure 4-44. AES survey spectrum of (a) a as-deposited ZrC film under CH4 atmosphere and (b) the ZrC film sputtered for 1 min with a 4 kV Ar ion beam.

PAGE 98

84 Vacuum0.00E+001.00E-082.00E-083.00E-084.00E-085.00E-086.00E-087.00E-088.00E-080510152025303540mass/chargepressure(torr) H2O, 6.73E-8 OH, 1.68E-8H, 4.57E-9(a) C2H20.00E+005.00E-071.00E-061.50E-062.00E-062.50E-063.00E-063.50E-064.00E-060510152025303540mass/chargepressure(torr) C2H2, 3.63E-6 C2H4, 1.06E-6O2, 1.38E-6C2H, 6.81E-7C, CH, CH2, CH4, H2O, E-8(b)H2O,9.72E-8 CH40.00E+005.00E-071.00E-061.50E-062.00E-062.50E-063.00E-063.50E-060510152025303540mass/chargepressure(torr) CH4, 3.17E-6 CH3, 2.13E-6CH2, 2.68E-7C, CH, E-8(c)H2O, 1.45E-8 Figure 4-45. Residual gas partial pressure analyzed by RGA before deposition at (a) vacuum and right after introducing (b) C2H2 and (c) CH4.

PAGE 99

CHAPTER 5 MECHANICAL PROPERTIES OF ZIRCONIUM CARBIDE FILMS MEASURED BY NANOINDENTATION 5.1 Introduction Zirconium carbide, as mentioned before, is a potentially important material for many applications because its properties including hardness, melting point, corrosion resistance, and abrasion resistance are outstanding. Therefore, ZrC films could be used for MEMS (micro electro-mechanical system) device as wear-resistant or protecting coating, or in electronic device. As one of examples as application in electronic device, ZrC deposited on Si can substitute currently using ZrN as a diffusion barrier in Cu-Si system for metallization application. ZrC also exhibits lower lattice mismatch and thermal expansion coefficient difference with Si compared to ZrN. In addition, epitaxially grown crystal ZrC could prevent copper diffusion into Si substrate along the localized defects in the barrier films. In consequence, the failure of Cu/ZrC/Si films could be avoided by retarding the formation of Cu3Si. Also tip failure by ion bombardment in currently using Si and Mo field emitters is one of problems in stabilizing field emission for a long time. The impact of ion bombardment onto surface creates sharp nanoprotrusions which are resulting in tip failure caused by high local field [Cha99, Cha01]. From this point of view, ZrC is promising material for a field emitter tip application because of its low work function as well as high resistance to ion bombardment. Therefore, it is important to investigate the mechanical properties of deposited thin film ZrC. 85

PAGE 100

86 One of the most challenging for task applications of thin films is to evaluate their mechanical properties. The nanoindentation has been used as a very useful technique to quantify thin-film mechanical properties, such as Young's modulus and hardness, within a submicron scale [Cac99, Kuc00, Now99, Yu98]. In the standard indentation procedure, a Berkovich pyramid-shaped diamond tip using N loads produces submicrometer indentation penetration depths, and allows the properties of thin films to be measured without removing the substrate. Apparently, this technique has become an important tool for material characterizations because one can easily obtain the reduced modulus and hardness values of thin films under different loads from their load-displacement (P-h) curves [Tsu99]. In this study, ZrC films grown on Si (100), Si (111), and sapphire (0001) substrates using pulsed laser deposition were investigated using nanoindentation measurement. The Young's modulus and hardness were calculated from loading-displacement (P-h) curves. 5.2 Experiment The film depositions were conducted in a stainless steel vacuum chamber using a KrF excimer laser. The laser parameters used were 10 J/cm2 fluence and 10 Hz repetition rate. ZrC films were deposited on and Si (001), Si (111), and sapphire (0001) substrates. Depositions were performed under a low C2H2 or CH4 atmosphere. Details for deposition procedures are described in chapter 4. Automated indention pattern grids with indention spacing of 5 m were programmed to run on original substrates and coated substrates with a Hisitron TriboIndenter. Tests were run in displacement control with a total displacement range of 15 to 70 nm. The samples thicknesses used for nanoindentation test were around 70~100 nm. The sample surfaces were carefully cleaned before the measurements. A Berkovich

PAGE 101

87 diamond indenter (a three-sided pyramidal-diamond tip) with a total included angle of 142.3, a half angle of 65.3, and a tip radius of 100~200 nm was used for indentation. A tip area function was generated by curve fitting a plot of area vs. contact depth from indentations on a fused quartz standard with modulus of 72 GPa and a Poissons ration of 0.168. Unloading force displacement curves were analyzed to determine reduced modulus and hardness using a method first described by Doemer and Nix [Doe86] and later refined by Oliver and Pharr [Oli92]. 5.3 Nanoindentation Test The term, the indentation or reduced modulus (Er), is introduced to balance the Young's modulus of specimen (E, film + substrate), and that of the diamond indenter (Ei) [Oli92]. That is, elastic modulus is determined from: iicrEEASE221112 (5-1) dhdPS (5-2) where S is the slope at the beginning of the unloading curve, and Ac is the corresponding projected contact area, in addition, and i are Poisson's ratio of specimen and indenter, respectively. For diamond, Ei = 1,070 GPa and i = 0.07. The Poisson's ratios and Youngs modulus was compared to reference data in table 5-1. To determine the area function, a series of indents at various contact depths were performed on fused quartz specimen, and then the contact area was calculated using eq. (5-1). A fitting procedure is employed to plot the computed area as a function of contact depth (hc) to a sixth order polynomial of eq. (5-3). 1615814413212120cccccchChChChChChCA (5-3)

PAGE 102

88 C0 for a Berkovich tip is 24.5. Area function used for data calculation is displayed in fig. 5-1. Since this calculation is based on the assumption that Youngs modulus of fused (or amorphous) quartz is constant and independent of indentation depth, and mathematical fitting is not perfect in whole range of depth, the result deduced could be erroneous especially in the thin depth region rather than in deep region. An example of an erroneous tip area function calculation is shown in fig. 5-2. Therefore, data below the contact depth of 5 nm were excluded in calculating mechanical properties. Table 5-1. Poissons ratio and Youngs modulus of Al2O3 (poly), Si, and diamond tip. E (GPa) Al2O3 (poly) ~400 [Asm01] ~0.23 [Asm01] Si (111) 185 [Wor65] 0.18 [Wor65] Si (001) 130 [Wor65], 179~202 [Bhu96] 0.279 [Wor65] Diamond tip 1070 0.07 Hardness (H) in depth sensing indentation is determined by applying a maximum indentation load (Pmax), while bulk hardness is calculated by residual contact area. Hardness in nanoindentation measurement is defined by: cAPHmax (5-4) In analyzing data, some erroneous data were removed, and a correction was conducted by shifting data points to set onset of the loading and displacement at zero. 5.4 Result and Discussion 5.4.1 Hardness and Youngs Modulus of Substrates To compare hardness and elastic modulus between substrate and film, hardness and elastic modulus of substrates was measured. Modulus of substrate can be calculated from

PAGE 103

89 eq. (5-1) by replacing E and with Es and s, where Es is Youngs modulus of substrate, and s is Poissons ratio of substrate. Calculated results of hardness and Youngs modulus are shown in fig. 5-3 and fig. 5-4, respectively. The Poissons ratios used in calculation were all 0.2. Error range of calculated Youngs modulus by eq. (5-1) is only 5% in the range of 0 < s < 0.3, when a diamond tip is used, and when s = 0.2 is used (fig. 5-5). The results of elastic modulus for silicon substrate were somewhat higher than theoretically calculated data [Wor65], but very similar to other empirical nanoindentation result [Bhu96] in table 5-1. Load vs. displacement curves of Si (001) and Si (111) substrate are displayed in fig. 5-6 and fig. 5-7, respectively. For sapphire substrate, there was a pop-in phenomenon in load-displacement curve, as it is shown in fig. 5-8. The pop-in phenomenon could be thought as creep behavior or yield of the material. But in displacement control method, it could be thought as yield of material because there is no dwell time at a certain load, contrary to load control method. Total elastic behavior before pop-in in fig. 5-9 supports yield phenomenon. Therefore, hardness measurement can not be correct in the whole range because hardness measurement is base on plastic deformation of materials. To make corrections in the measurement, displacement curve should be shifted to left, to set the onset of load at the extended line from on-load curve after pop-in. However, since knowing that the hardness of sapphire responded in the range of 20~50 GPa before yield was enough, the hardness of bulk sapphire was not obtained in this study. On the other hand, Youngs modulus measurement is based on unloading curve, so this data is still available for analysis of elastic modulus of ZrC film.

PAGE 104

90 5.4.2 Hardness and Elastic Modulus of ZrC Films Deposited on Si (001), Si (111), and Sapphire The elastic modulus of combined film and substrate as a function of contact depth was calculated from the measured reduced modulus using eq. (5-1), assuming that the compliance of specimen (film + substrate) and indenter is combined as springs in series [Oli92]. The value 0.2 as combined Poissons ratio for film and substrate was used for elastic modulus calculation. The values measured from ZrC (001) films on Si (001) substrate and their load-displacement curve are displayed in fig. 5-10 ~ fig. 5-12. As one can see in fig. 5-10, as the contact depth is increased (hence applied force), the values of the elastic modulus are approaching the values measured for the substrates. Since the combined film and substrate modulus is strongly decreasing with the depth of penetration from the value of the modulus of the hard film towards that of the softer substrate system as shown in fig 5-13 [Men97], a value of over 450 GPa of elastic modulus can be estimated for the highest crystalline quality (001) ZrC (film thickness, t ~ 100 nm) on Si (001). For relatively thinner and lower crystalline quality (t ~ 70 nm) samples, a lower value of over 300 GPa of elastic modulus can be estimated. Similarly, a value of ~450 GPa for the highest quality films grown on sapphire and a value of ~300 GPa for films grown on Si (111) can be estimated. The resulting combined elastic modulus data are displayed in fig. 5-14, which are calculated from the load-displacement curves in fig. 5-15 and fig. 5-16. In this case, the substrate effect was very dominant due to thickness of film or high sapphire elastic modulus, and the films moduli barely influenced the measurements. Fig. 5-17 shows the variations of hardness as a function of normalized depth (= indenter displacement / film thickness) for a ZrC films and substrate system. In hard film

PAGE 105

91 on soft substrate system, hardness increases at small indentation depth, and then it reaches at maximum value when normalized depth is 0.25, and it decreases with increasing indentation depth [Che05b]. Fig. 5-17 also shows a similar tendency, so the hardness data can be obtained around 0.25 of normalized indentation depth. The results show that the highest crystal quality films (ZC202: ~27 GPa, ZC204: ~31 GPa) exhibited higher hardness values than those of lower crystalline quality films (ZC210a: ~22 GPa, ZC210c: ~19GPa). This maximum ZrC hardness is comparable to the maximum hardness (30.2GPa) of nanocrystalline ZrC reported recently [Che05a]. 5.5 Summary In summary, epitaxial ZrC films deposited by using the pulsed laser deposition technique were tested for hardness and elastic modulus using depth sensing nanoindentation. In nanoindentation measurement, surface cleanness, accurate area function, and indenter tip stability when approaching near surface were critical problems for obtaining correct data. Both hardness and elastic modulus were predominantly depending on films crystalline quality rather than their textures. A high value of hardness (~31 GPa) and elastic modulus (over 450 GPa) were obtained for high crystalline quality ZrC (111) deposited on sapphire substrate.

PAGE 106

92 Figure 5-1. Tip area function used for calculation of mechanical properties. Figure 5-2. Example of erroneous fitting of tip area function in depth below 5nm range.

PAGE 107

93 01020304050600102030405060Contact depth (nm)Hardness of substrates (GPa) Si (001) Si (111) Sapphire (0001) Figure 5-3. Hardness of Si (001), Si (111), and sapphire (0001) single crystal substrates. 0501001502002503003504004500102030405060Contact depth (nm)Modulus of substrates (GPa) Si (001) Si (111) Sapphire (0001) Figure 5-4. Youngs modulus of Si (001), Si (111), and sapphire (0001) single crystal substrates.

PAGE 108

94 -45-30-15015304500.050.10.150.20.250.3Poisson's ratioError in E (%) Figure 5-5. Error range change in Youngs modulus as function of Poissons ratio of substrate according to eq. (5-1). Figure 5-6. Load-displacement curves of Si (001) substrate.

PAGE 109

95 Figure 5-7. Load-displacement curves of Si (111) substrate. Figure 5-8. Load-displacement curves of sapphire (0001) substrate.

PAGE 110

96 Figure 5-9. Load-displacement curves of sapphire (0001) substrate before pop-in occurs. 10015020025030035040051525354Contact depth (nm)Combined modulus of film+sub. (GPa) 5 ZC202 ZC210a Figure 5-10. Combined modulus of film and substrate for different quality of ZrC (001) grown on Si (001) substrate. FWHM of ZC202 = 2.53, FWHM of ZC210a = 7.27 in omega rocking curve.

PAGE 111

97 Figure 5-11. Load-displacement curves of ZrC (001) grown on Si (001) substrate, sample ZC202 in fig 5-10. Figure 5-12. Load-displacement curves of ZrC (001) grown on Si (001) substrate, sample ZC202 in fig 5-10.

PAGE 112

98 Figure 5-13. Measured elastic modulus as a function of relative penetration into a coated specimen (a is contact radius, and t is film thickness). For curve 1, Efilm is less than Esubstrate, and opposite case for curve 2 [Men97]. 10015020025030035040045050051525354Contact depth (nm)Combined modulus of film+sub. (GPa) 5 ZC204 ZC210c Figure 5-14. Combined modulus of film and substrate for different quality of ZrC (111) grown on sapphire (0001) and Si (001) substrate, sample ZC204 and ZC210c respectively. FWHM of ZC204 = 2.53, FWHM of ZC210c = 6.95 in omega rocking curve.

PAGE 113

99 Figure 5-15. Load-displacement curves of ZrC (111) grown on sapphire (0001) substrate, sample ZC204 in fig 5-14. Figure 5-16. Load-displacement curves of ZrC (111) grown on Si (111) substrate, sample ZC210c in fig 5-14.

PAGE 114

100 10152025303500.20.40.60.8Normalized depth (h/t)Hardness of ZrC (GPa) ZC204 ZC202 ZC210a ZC210c Figure 5-17. Combined hardness of ZrC and substrate as a function of normalized depth; h displacement, t film thickness.

PAGE 115

CHAPTER 6 CONCLUSION Epitaxial ZrC thin films were deposited on Si (001), Si (111), and sapphire (0001) substrates by the pulsed laser deposition technique. First of all, ZrC thin films were grown on chemically cleaned Si (001) substrate with various process parameters except laser pulse repetition rate of 5 Hz. Under the temperature range of 200~600 C and/or laser fluence below 6 J/cm2, resulting ZrC films showed no crystallinity. By increasing the laser fluence at 8 J/cm2 and substrate temperature at 600 C, we obtained ZrC film that showed some crystallinity. However, when the ZrC target was not pre-ablated for at least 1 min, XRD spectra showed as-grown films were a substoichiometric ZrC, which was caused by the oxidation of target surface. A combination of high laser fluence around 10 J/cm2 and high substrate temperature of 600 ~ 700 C was required to obtain high crystalline films. Under these conditions films grew with a (001) texture on Si (001) substrate. Grazing incidence x-ray diffraction investigations, which are much more surface sensitive than symmetric XRD investigations, evidenced the presence of a rather small fraction of randomly oriented crystallites. According to GIXD and XPS investigation, low pressure of C2H2 as background gas gave a little beneficial effect on enhancing crystallinity and stoichiometry. XPS study confirmed that the surface layers contained mainly Zr-O bonds. After a 5 minute Ar+ sputtering process, a homogeneous composition of Zr-C and Zr-O was found in the surface region. The study was then focused on the use of a high laser fluence of 10 J/cm2 and high substrate temperature around 750 C on Si (001), Si (111), and sapphire (0001) substrates. 101

PAGE 116

102 10 Hz repetition rate of laser pulse was used because growth rate was rather low at 5 Hz. To avoid oxidation problem, special care was taken to maintain low water vapor pressures during depositions, below ~-8 Torr, as measured with a residual gas analyzer attached to the deposition chamber. Prior to deposition, the Si substrates were heated up to 900 oC for 20 min under high vacuum to remove the native oxide. Under these conditions, epitaxial ZrC films were deposited on single crystalline substrates. The ZrC films grew along the [001] axis on Si (001), while they grew along the [111] axis on Si (111) and sapphire (0001). Pole figure measurements showed that ZrC films exhibited in-plane orientation too, depending on the type of substrate. For sapphire (0001) substrate case, pole figure analysis revealed parallel in-plane relationship between <11 2 0> of sapphire and <110> of ZrC, which indicates a 30 rotation of ZrC lattice with respect to the sapphire lattice, and six {100} reflections from ZrC film in (100) pole figure showed that twining was occurred during growth. This in-plane orientation relationship was well explained by translational symmetries on both sides of the interface. Grazing incidence x-ray diffraction investigations evidenced the presence of a rather small fraction of randomly oriented crystallites in these films. TEM investigation on cross-sectional specimen showed clear lattice fringes over wide region for all case, corroborating the epitaxial growth indicated by XRD results. For Si (111) substrates only, several Moir fringes were observed, indicating randomly oriented crystallites. For sapphire substrates two types of twin boundary were found; one was perpendicular to surface of film, and the other was parallel to surface. This fact suggests the Stranski-Krastanov (island after monolayer) growth model because these two types of twining are possible when some nuclei grow in the ABCABC order, and other nuclei grow in ACBACB order during the

PAGE 117

103 initial stage of growth. Omega rocking-curves measurement showed that ZrC film had best lateral out-of-plane alignment but relatively rather poor in-plane alignment due to twinning. ZrC (001) films showed good quality both in-plane and out-of-plane. ZrC (111) films grown on Si (111) showed (111) preferred orientation with rather poor quality in epitaxy, which was controlled by moderated C2H2 background gas pressure. Lattice parameter of deposited ZrC film was changed by small variations of C2H2 pressure. There exists an optimum pressure value for good crystal quality. AES surface composition analysis showed oxygen concentration both on surface of film and in bulk (7~8%) was still high. Despite the high levels of oxygen contamination, the deposited ZrC films were very conductive. Under best vacuum conditions (Po = 1.3~1.5-7 Torr) and using a low pressure of high purity CH4 during deposition, oxygen concentration was less than 2.5 %, a very good value for ZrC. Simulation result using a three layer model from XRR spectra showed that the films mass density was around the tabulated value of 6.7 g/cm3, while the surface morphology was very smooth, with a roughness value (rms) of around 0.6 nm. Epitaxial ZrC films deposited by using the pulsed laser deposition technique were tested for hardness and elastic modulus using depth sensing nanoindentation. In nanoindentation measurement, surface cleanness, accurate area function, and indenter tip stability when approaching near surface were critical problems in getting correct data. Both hardness and elastic modulus predominantly depended on crystalline quality rather than their textures. High value of hardness (~31 GPa) and elastic modulus (over 450 GPa) were obtained for high crystalline quality ZrC (111) deposited on sapphire substrate, one of the highest values reported so far.

PAGE 118

104 Finally, as a future work, it is worth noting that films grown at low temperature of only 300 C and a rather low laser fluence of only 2 J/cm2 from a ZrC doped with Co target exhibited sharp diffraction lines, as one can see in fig. 6-1. However, because the films were well textured, exhibiting only three diffraction lines, the crystalline compound formed was not identified. The properties such as electrical resistivity and electron emission of these Co doped ZrC films are worth being investigated further because the rather low processing temperature and laser fluence were required to obtain good crystallinity.

PAGE 119

105 Figure 6-1. XRD spectra of thin films deposited from a Co doped ZrC target under various conditions.

PAGE 120

LIST OF REFERENCES [Ada74] H. Adachi, Y. Shibuya, T. Hariu, and Y. Shibats, J. Phys. D 7, 2150 (1974). [Ale00] L. DAlessio, A. Santagata, R. Teghil, M. Zaccagnino, I. Zaccardo, V. Marotta, D. Ferro, and G. De Maria, Appl. Surf. Sci. 168, 284 (2000). [Asm01] M. Asmani, C. Kermel, A. Leriche, and M. Ourak, J. Euro. Ceram. Soc. 21, 1081 (2001). [Auc01] O. Auciello and A. R. Krauss, In Situ Real-Time Characterization of Thin Films, Wiley-Interscience, New York (2001). [Ber95] H. Berndt, A.-Q. Zeng, H.-R. Stock, and P. Mayer, Surf. Coat. Technol. 74/75, 369 (1995). [Bhu96] B. Bhushan and X. Li, J. Mater. Res. 12(1), 54 (1997). [Bra66] D. G. Brandon, Acta Metall. 14, 1479 (1966). [Bra92] E. A. Brandes and G. B. Brook, Smithells Metals Reference Book, 7th Edition, Butterworth-Heinemann, Oxford, 11-159 (1992). [Bre62] F. Breech and L. Cross, Appl. Spectrosc. 16, 59 (1962). [Br] J. Brckner and T. Mnytl, Suf. Coat. Technol. 59, 166 (1993). [Cac99] D. Caceres, I. Vergara, R. Gonzaalez, E. Monory, F. Calle, E. Munoz, and F. Omnes, J. Appl. Phys. 86, 6773 (1999). [Cha99] F. M. Charbonnier, W. A. Mackie, T. Xie, and P. R. Davis, Ultramicroscopy 79, 73 (1999). [Cha01] F. M. Charbonnier, W. A. Mackie, R. L. Hartman, and T. Xie, J. Vac. Sci. Technol. B19, 1064 (2001). [Che05a] C.-S. Chen, C.-P Liu, and C. -Y. A. Tsao, Thin Solid Films 479, 130 (2005). [Che05b] S. Chen, L. Liu, and T. Wang, Suf. Coat. Technol. 191, 25 (2005). 106

PAGE 121

107 [Chr94] D. B. Chrisey and G. K. Hubler, Pulsed Laser Deposition of Thin Films, Wiley-Interscience, New-York,188 and 342-353 (1994). [Coc98] D. Cocke and M. Owens, Appl. Surf. Sci. 31, 471 (1998). [Coc99] B. V. Cockeram, D. P. Measures, and A. J. Mueller, Thin Solid Films 17, 355 (1999). [Dij87] D. Dijkamp, T. Venkatesan, X.D. Wu, S.A. Shaheen, N. Jisrawi, Y.H. Min-Lee, W.L. McLean, and M. Croft, Appl. Phys. Lett. 51, 619 (1987). [Doe86] M. F. Doerner and W. D. Nix, J. Mater. Res. 1, 601 (1986). [Dye89] P. E. Dyer, R. D. Greenough, A. Issa, and P. H. Key, Appl. Surf. Sci. 43, 387 (1989). [Eda96] K. Edamoto, T. Anazawa, S. Tokumitsu, A. Tanabe, R. Sekine, E. Miyazaki, H. Kato, and S. Otani, Solid State Commun. 97, 435 (1996). [Han95] K. Hanko, G. Vass, and L. Szepes, J. Organometallic Chem. 492, 235 (1995). [He98] X.-M. He, L. Shu, H.-B. Li, H.-D. Li, and S.-T. Lee, J. Vac. Sci. Technol. A16, 2337 (1998). [Hol77] C. M. Hollabaugh, L. A. Wahman, R. D. Reiswig, and R. W. White, Nucl. Tech. 35, 527 (1977). [How02] J. M. Howard, V. Craciun, C. Essary, and R. K. Singh, Appl. Phys. Lett. 81, 3431 (2002). [Hu91] S. M. Hu, J. Appl. Phys. 69(11), 7901 (1991). [Kas78] D. Kashchiev, Thin Solid Films 55, 399 (1978). [Kau90] W. Kautek, B. Roas, and L. Schultz, Thin Solid Films 191, 371 (1990). [Kuc00] S.O. Kucheyev, J.E. Bradby, J.S. Williams, C. Jagadish, M. Toth, M.R. Phillips, and M.V. Swain, Appl. Phys. Lett. 77, 3373 (2000). [Lew78] B. Lewis and J. C. Anderson, Nucleation and Growth of Thin Films, Academic Press, New York (1978). [Lid05] D. R. Lide, Handbook of Chemistry and Physics on CD-ROM version 2005, CRC press (2005). [Mac92] W. A. Mackie, J. L. Morrissey, C. H. Hinrichs, and P. R. Davis, J. Vac. Sci. Technol. A 10, 2852 (1992).

PAGE 122

108 [Mac93] W. A. Mackie, R. L. Hartman, and P. R. Davis, Appl. Surf. Sci. 67, 29 (1993). [Mac94] W. A. Mackie, R. L. Hartman, M. A. Anderson, and P. R. Davis, J. Vac. Sci. Technol. B 12, 722 (1994). [Mac95] W. A. Mackie, T. Xie, and P. R. Davis, J. Vac. Sci. Technol. B 13(6), 2459 (1995). [Mac98a] W. A. Mackie, T. Xie, J. E. Blackwood, S. C. Williams, and P. R. Davis, J. Vac. Sci. Technol. B 16, 1215 (1998). [Mac98b] W. A. Mackie, T. Xie, M. R. Matthews, B. R. Routh, Jr., and P. R. Davis, J. Vac. Sci. Technol. B 16 2057 (1998). [Mac98c] W. A. Mackie, T. Xie, M. R. Matthews, and P. R. Davis, Mater. Res. Soc. Symp. Proc. 509, 173 (1998). [Mah99] S. Mahajan and K. S. S. Harsha, Principles of Growth and Processing of Semiconductors (International Editions), McGraw-Hill, Seoul, 264 (1999). [Mat86] A. Matsunawa, S. Katayama, A. Susuki, and T. Ariyasu, Trans. J. Welding Res. Technol. B3, 968 (1986). [Men97] J. Menik, D. Munz, E. Quandt, and E. R. Weppelmann, J. Mater. Res. 12, 2475 (1997). [Met89] S. Metev and K. Meteva, Appl. Surf. Sci. 43, 402 (1989). [Mod85] F. A. Modine, T. W. Haywood, and C. Y. Allison, Phys. Rev. B 32(12), 7743 (1985). [Nv80] L. Nvot and P. Croce, Rev. Phys. 15, 761 (1980). [Now99] R. Nowak, M. Pessa, M. Suganuma, M. Leszczynski, I. Grzegory, S. Porowski, and F. Yoshida, Appl. Phys. Lett. 75, 2070 (1999). [Oga86] T. Ogawa and K. Ikawa, High Temp. Sci. 22, 179 (1986). [Oga92] T. Ogawa, K. Fukuda, S. Kashimura, T. Tobita, F. Kobayashi, S. Kado, H. Miyanishi, I. Takahashi, and T. Kikuchi, J. Am. Ceram. Soc. 75, 2985 (1992). [Oli92] W. C. Oliver and G. M. Pharr, J. Mater. Res. 7, 1564 (1992). [Oya92a] S. T. Oyama, Catalysis Today 15(2), 179 (1992). [Oya92b] S. T. Oyama, Journal of Solid State Chemistry 96, 442 (1992).

PAGE 123

109 [Par54] L. G. Parratt, Phys. Rev. 95, 359 (1954). [Pcp94] PCPDFWIN Version 1.00, Joint Committee on Powder Diffraction International Centre for Diffraction Data (JCPDS-ICDD) (1994). [Pri92] D. L. Price, B. R. Cooper, and J. M. Wills, Phys. Rev. B 46, 11368 (1992). [Rey74] G. H. Reynolds, J. Nucl. Mater. 50, 215 (1974). [Rus02] C. H. Russell, Mat. Res. Symp. Proc. 716, B1.3.1 (2002). [She97] J. R. Sheats, W. A. Mackie, S. Anz, and T. Xie, Proc. SPIE 3148, 219 (1997). [Shi98] S. Shimada, M. Yoshimatsu, M. Inagaki, and S. Otani, Carbon 36, 1225 (1998). [Smi65] H.M. Smith and A.F. Turner, Appl. Opt. 4, 147 (1965). [Smi93] D. C. Smith, R. R. Rubiano, M. D. Healy, and R. W. Springer, Mater. Res. Soc. Symp. Proc. 282, 643 (1993). [Spr86] W. D. Sproul, J. Vac. Sci. Thchnol. A 4(6), 2874 (1986). [Tem99] D. Temple, Mat. Sci. Engr. R24, 185 (1999). [Tes93] T. C. Tessner and P. R. Davis, J. Vac. Sci. Technol. A 11(1), 1 (1993). [Tot71] L. E. Toth, Transition Metal Carbides and Nitrides, Academic, New York (1971). [Tsu99] T.Y. Tsui and G.M. Phraa, J. Mater. Res. 14, 292 (1999). [Var94] D. J. Varacalle, Jr., L. B. Lundberg, H. Herman, G. Bancke, and W. L.Riggs Surf. Coat. Technol. 68/69, 86 (1994). [Ven84] J. A. Venables, G. D. T. Spiller, and M. Hanbucken, Rep. Prog. Phys. 47, 399 (1984). [Wag76] P. Wagner, L. A. Wahman, R. W. White, C. M. Hollabaugh, and R. D. Reiswig, J. Nucl. Mater. 62, 228 (1976). [Wor65] J. J. Wortman and R. A. Evans, J. Appl. Phys. 36(1), 153 (1965). [Xie96] T. Xie, W. A. Mackie, and P. R. Davis, J. Vac. Sci. Technol. B 14, 2090 (1996). [Yim74] W. M. Yim and R. J. Paff, J. Appl. Phys. 45(3), 1456 (1974).

PAGE 124

110 [Yu98] G. Yu, H. Ishikawa, T. Egawa, T. Soga, J. Watanable, T. Jimbo, and M. Umeno, J. Cryst. Growth 189/190, 701 (1998). [Zab94] H. Zabel, Appl. Phys. A58, 159 (1994). [Zai84] S. Zaima, H. Adachi, and Y. Shibata, J. Vac. Sci. Technol. B 2(1), 73 (1984). [Zur84] A. Zur and T. C. McGill, J. Appl. Phys. 55(2), 378 (1984).

PAGE 125

BIOGRAPHICAL SKETCH Juhyun Woo was born in Deagu, Korea, November 8, 1972. He attended Duk-won High School in Deagu, Korea. After his graduation with a BS in metallurgical engineering from Kyungpook National University in 2000, his education continued with graduate studies at University of Florida in the Department of Materials Science and Engineering. 111


Permanent Link: http://ufdc.ufl.edu/UFE0013064/00001

Material Information

Title: Growth of Epitaxial Zirconium Carbide Layers Using Pulsed Laser Deposition
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013064:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013064/00001

Material Information

Title: Growth of Epitaxial Zirconium Carbide Layers Using Pulsed Laser Deposition
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013064:00001


This item has the following downloads:


Full Text












GROWTH OF EPITAXIAL ZIRCONIUM CARBIDE LAYERS
USING PULSED LASER DEPOSITION















By

JUHYUN WOO


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


2005

































Copyright 2005

by

Juhyun Woo

































Dedicated to Jihoon, the parents of us all, and Dr. Craciun.















ACKNOWLEDGMENTS

I would like to express my gratitude to Dr. Valentin Craciun without whose help,

guidance and encouragement this work might not have been possible. Also I would like

to express my sincere appreciation to my committee members, Dr. Fereshteh Ebrahimi,

Dr. Rajiv K. Singh, Dr. Cammy Abernathy, and Dr. Timothy J. Anderson. Especially, I

am deeply grateful to Dr. Ebrahimi. A lot of what I have learnt is due to her guidance.

Also I would like to thank Gerald Bourne and Kerry Siebein. Gerald Bourne helped

me to solve many problems in nanoindentation measurements, and Kerry Siebein helped

me to get outstanding high resolution TEM images. The assistance and help of Wayne

Acree and Eric Lambers to obtain high quality SEM, XPS, and AES results are gratefully

acknowledged. I also would like to thank Woochul Kwak and Sang-yup Kim for their

help with AFM and FE-SEM. And I personally thank Dr. Luisa Amelia Dempere for

giving me a great opportunity of working in Major Analytical Instrumentation Center and

"hands-on" learning about thin film characterization.

Finally I would like to thank my wife Jihoon Yi for her support and patience. My

best friend and senior Sang-yup Kim helped me a lot to focus on my research.
















TABLE OF CONTENTS



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

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

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

A B S T R A C T .......................................... ..................................................x iii

CHAPTER

1 IN TR OD U CTION ......................................................... ... ........ ........... .... .... 1

2 LITER A TU R E R EV IEW ................................................................. ....................... 4

2.1 Characteristics of Zirconium Carbide.............................................4
2.1.1 Composition and Structure ............................................ ............... 4
2.1.2 Properties and Applications of Zirconium Carbide Thin Films ................5
2.2. Techniques Used for Zirconium Carbide Thin Film Depositions .....................7
2.2.1 Therm al Evaporation (TE) ................................................ .............. 7
2.2.2 Sputtering D position (S) ...................................... .....................................7
2.2.3 Chem ical V apor D position (CV D ) ........................................ ..................8
2.2.4 Pulsed Laser D position (PLD ).................................................................9
2.2.5 Com prison of Techniques.................................................... ................ 12
2.3 Growth and Factors Determining the Quality of Thin Films in PLD .................13
2.3.1 N ucleation and G row th......................................... .......................... 13
2.3.2 B background G as ............................................... ........ .. ...... ............15
2.3.3 Vacuum .................................... ............................... ........16
2.3.4 Laser Fluence .................. .................. ................................ .. 17
2.3.5 Laser W avelength......................................................... .............. 17
2.3.6 Target to Substrate D istance................................ ........................ ......... 18

3 EXPERIM EN TAL M ETHOD S........................................................... ............... 24

3.1 Pulsed Laser D position System ........................................ ....... ............... 24
3.2 Structural Characterization ....................................... ... ..........................25
3.3 Film Thickness and Roughness ........................................ ........................ 26
3.4 Surface Chem istry A analysis ........................................ ........................... 27
3.5 E electrical M easurem ent ............................................... ............................. 28


v









4 GROWTH AND CHARACTERIZATION OF HIGH CRYSTALLINE QUALITY
ZIR CON IU M CA RB ID E FILM S ................................................... .....................31

4.1 Introduction ............... ...... ......... ...... .............. ............. 31
4 .2 E x p erim ent............................. .................................................... ............... 32
4.3 R results and D discussion ........................................... ....................................... 33
4.3.1 Laser Fluence and Temperature Effect on Deposited Films ....................33
4.3.2 Deposition Rate and Thickness Uniformity of Deposited Films ..............35
4.3.3 ZrC Films Growth at High Temperature and High Laser Fluence.............38
4.3.3.1 Growth behaviors of ZrC films deposited on Si and sapphire
substrate ................. ..................... .... ... ...... ................. 39
4.3.3.2 TEM analysis of ZrC films grown on Si and sapphire substrate .....43
4.3.4 Surface analysis of ZrC films .............. ........ .... ......................... 45
4.4 Sum m ary ............... ........ .......................................................... 49

5 MECHANICAL PROPERTIES OF ZIRCONIUM CARBIDE FILMS MEASURED
BY NAN OINDEN TATION ......................................................... ..................... 85

5.1 Introduction ..................................................................... 85
5 .2 E x p e rim e n t...................................................................................................... 8 6
5 .3 N an oin dentation T est......................................... .............................................87
5.4 R result and D discussion .................................................. ................................. 88
5.4.1 Hardness and Young's M odulus of Substrates ............... ..........................88
5.4.2 Hardness and Elastic Modulus of ZrC Films Deposited on Si (001), Si
(111), an d S apphire .......... ..... ....... .. .. ...... .. .............. ... ............ 90
5.5 Sum m ary ............. .... ..............................................................91

6 C O N C L U SIO N ....... ................................................................................. ...... ......10 1

LIST O F R EFEREN CE S ... .... ............................................................ ............... 106

BIOGRAPHICAL SKETCH ........... ................ ......... ............................111
















LIST OF TABLES


Table page

2-1 Characteristics and properties of ZrC reported in literature............... ..................5

3-1 Excim er laser operating w avelengths................................................... ............... 24

3-2 Typical XRR values for the resolution and ranges.................................................27

4-1. Growth conditions of deposited films, showing GIXD spectra in fig. 4-1.................34

4-2. Thickness, density, and surface roughness values of the ZrC films deposited at
700 C for different tim es............................................... .............................. 36

4-3. Degree of out-of-plane texture of ZrC films at various growth conditions ..............37

4-4. Structure information ofZrC, Si, and sapphire and lattice mismatch.....................38

4-5. Surface roughness and density of as-grown films....... ..........................................46

4-6. The relative percentage of Zr 3d XPS areas corresponding to Zr-C bonds at
different take-off angles ................................................ ............................... 48

4-7. Resistivity of as-deposited ZrC films ............................... ...............48

5-1. Poisson's ratio and Young's modulus of A1203 (poly), Si, and diamond tip .............88















LIST OF FIGURES


Figure page

2-1. Image of cubic rocksalt (B 1) structure of transition metal carbides. Carbon atoms
are depicted as the light gray spheres, metals as dark gray spheres.........................19

2-2. Zr-C phase diagram showing wide range congruent compositions..........................19

2-3. Schematic diagram of an apparatus for pulsed laser deposition.............................20

2-4. Schematic diagram of the approximate energy range of deposited atoms for
various deposition techniques. The shaded box indicates the supposed energy
range of atom fluxes considered to be beneficial for film growth .........................21

2-5. Schematic diagram comparing the pressure ranges over various techniques. PLD
can operate over the widest range of all the methods ........................................... 21

2-6. Growth diagram drawn by equation 2-3, showing the dependence of growth
behavior on temperature and growth rate............................. ..............22

2-7. Relationship of impinging particles fluxes to deposition rate, and gas pressure........22

2-8. ZrC ablation rate, showing linear dependence on laser fluence, and threshold laser
fluence of 1.3 J/cm 2 ...................... .................. ..................... .. ...... 23

3-1. A schematic diagram of symmetric and asymmetric GIXD geometry ....................29

3-2. A schematic diagram of omega rocking curve geometry ...........................................29

3-3. An example of a schematic reflections on pole figure related to crystal quality........30

4-1. GIXD spectra (incidence beam angle, co = 1) of ZrC films deposited under
various s conditions ..................................................................... 5 1

4-2. Comparison of GIXD spectra obtained from ZrC deposited films at same
deposition conditions (a) without pre-ablated target (red) and (b) with pre-
ablated target (blue) ................................................ .. ...... .. ........ .... 51

4-3. XRD spectra obtained from ZrC target; vertical lines represent position and
intensity for stoichiometric ZrC, JCPDS PDF# 32-1489................................. 52









4-4. GIXD spectra obtained from films deposited under residual vacuum and different
C 2H 2 gas pressures ............................ ..... ... .. .... ... ............... 52

4-5. XRR spectra of ZrC films deposited for various times at 700 OC and 10 J/cm2
u n d er v acu u m ...................................................... ................ 5 3

4-6. ZrC film deposition rate at Ts = 7000C, Pd = 1.0x10-6 Torr, 10 J/cm2, and 5Hz........53

4-7. GIXD spectra of ZrC films deposited for various times at 700 OC and 10 J/cm2
under vacuum; the standard position of diffraction lines from ZrC (dashed lines)
and ZrCo. (solid lines) are also shown .......... ............................... ............... 54

4-8. XRD spectra of ZrC films deposited for various times at 700 OC and 10 J/cm2
u n d er v acu u m ...................................................... ................ 54

4-9. XRD spectra of ZrC films deposited at 750 oC on various substrates .....................55

4-10. XRD and GIXD spectra (incidence beam angle = 10) of ZrC films deposited on
Si (111) sub states .....................................................................55

4-11. Omega-rocking curves of ZrC (111) or (200) peaks recorded from ZC104 films
deposited at 750 oC on various substrates......... ............. .......................56

4-12. Omega-rocking curves of ZrC (002) peaks recorded from the films deposited on
Si (001) substrate at various background gas pressures................................ .....56

4-13. Omega-rocking curves of ZrC (111) peaks recorded from the films deposited on
sapphire (0001) substrate at various different gas pressures........................ 57

4-14. XRD spectra of (002) reflection from the films deposited on Si (001) and
calculated lattice param eters ............................................................................. 57

4-15. Background gas effect on lattice parameter of the films deposited on Si (001)
su b state ...................................... .....................................................5 8

4-16. The relationship between deposited films on Si (001) and texture degree of the
films measured by omega-rocking curve ...................................... ............... 58

4-17. Phi-scan of { 111 in-plane obtained from (a) ZrC film and Si (001) substrate of
sample ZC202 and (b) phi-scan diffractometer configuration..............................59

4-18. (111) pole figures of(a) ZrC film and (b) Si (001) substrate obtained from
sam ple Z C 202 ........................................................................60

4-19. (100) pole figures of(a) ZrC film and (b) Si (111) substrate obtained from
sam ple Z C 104 ........................................................................6 1

4-20. Pole figures showing (a) (100) pole figures of ZrC film, and (b) (1 1-2 6) pole
figure of sapphire (0001) substrate obtained from ZC104.............................. 62









4-21. Projection view for two crystallographic orientations of ZrC grown on sapphire
(0001), and orientation relationship between ZrC film and sapphire substrate .......63

4-22. Possible nucleation site (marked as A) for the first monolayer on sapphire (0001).
........................................................................................................ . 6 3

4-23. Bright field TEM image obtained from cross-section of ZrC film grown on
silicon (111) substrate; the regions showing inhomogeneous random grains are
m arked .............. .. .. ... ............. ..........................................64

4-24. TEM (a) bright field image and (b) SADP (selected area electron diffraction
pattern) obtained from cross-section of ZrC film (sample ZC 104) grown on
silicon (111) substrate ...................... ................ .............................65

4-25. TEM (a) bright field image and (b) SADP (selected area electron diffraction
pattern) obtained from cross-section of ZrC film grown on silicon (001)
substrate; diffraction pattern obtained from ZrC film is marked by circles............66

4-26. Bright field TEM image obtained from cross-section of ZrC film grown on
silicon (001) substrate exhibiting clear lattice fringe............... ....... ............ 67

4-27. Bright field TEM image obtained from cross-section of ZrC film grown on
sapphire (0001) substrate showing sharp interface; the regions marked as A and
B are magnified in figure 4-28 for observation of twinning ...................................68

4-28. High resolution TEM image of (a) 'A' region in fig. 4-26, showing parallel twin
to the surface of film; (b) 'B' region in fig. 4-26, showing perpendicular twin to
the surface of film ...... .... ............ .. .............. ............ ...... ... ..... ........69

4-29. Secondary electron images of ZrC surface grown on Si (001) substrate, at xlOK
(a), an d x 50K (b ) ................................................................................ 7 0

4-30. Secondary electron images of ZrC surface grown on sapphire (0001) substrate,
at x O1 K (a), and x 50K (b) ............................................ .... .. .......... .. ............ 71

4-31. AFM height images obtained from the surface of ZC106 sample (ZrC (001)
layer grown on Si (001) substrate) ......................... ...... .................. 72

4-32. AFM height images obtained from the surface of ZC202 sample (ZrC (001)
layer grown on Si (001) substrate) ......................... ...... .................. 73

4-33. AFM height images obtained from the surface of ZC208 sample (ZrC (001)
layer grown on Si (001) substrate) ......................... ...... .................. 74

4-34. AFM height images obtained from the surface of ZC107 sample (ZrC (111)
layer grown on Si (111) substrate) ............... ........... ... .. ............... 75









4-35. AFM height images obtained from the surface of ZC208 sample (ZrC (111)
layer grown on sapphire (0001) substrate)..... .................................76

4-36. AFM height images obtained from the surface ofZC210 sample (ZrC (111)
layer grown on sapphire (0001) substrate)..... .................................77

4-37. Cross-sectional bright field TEM image showing an oxidized layer due to
discontinued grow th ................................................. ....... .. ............ 78

4-38. Cross-sectional bright field TEM image showing distinguishable three layers,
also supporting a model used for XRR analysis .......................... ..................78

4-39. Z-contrast image of cross-sectional ZC106 sample for TEM-EDX analysis by
line and point scan ............................................................... .. ..........79

4-40. High resolution Zr 3d spectra acquired at 450 and 900 take off angles and their
fitting for an as-received sample deposited at 600 OC under vacuum .......... ......80

4-41. High resolution Zr 3d spectra acquired at 450 and 900 take off angles and their
fitting for a sample deposited at 600 OC under 1 x 10-4 Torr of C2H2 that was
sputtered-clean by Ar+ bombardm ent....................................... .......................... 81

4-42. AES depth profile of an as-deposited ZrC film.............. .................................82

4-43. AES survey spectrum of a ZrC film (ZC202) sputtered with a 4 kV Ar ion beam..82

4-44. AES survey spectrum of (a) a as-deposited ZrC film under CH4 atmosphere and
(b) the ZrC film sputtered for 1 min with a 4 kV Ar ion beam............. ..............83

4-45. Residual gas partial pressure analyzed by RGA before deposition at (a) vacuum
and right after introducing (b) C2H2 and (c) CH4...... ..... .................................... 84

5-1. Tip area function used for calculation of mechanical properties ..............................92

5-2. Example of erroneous fitting of tip area function in depth below 5nm range............92

5-3. Hardness of Si (001), Si (111), and sapphire (0001) single crystal substrates...........93

5-4. Young's modulus of Si (001), Si (111), and sapphire (0001) single crystal
su b states ............................................................................ 9 3

5-5. Error range change in Young's modulus as function of Poisson's ratio of substrate
according to eq. (5-1) ......................... .... ................ ............................ 94

5-6. Load-displacement curves of Si (001) substrate .............. ..................................... 94

5-7. Load-displacement curves of Si (111) substrate .............. ..................................... 95

5-8. Load-displacement curves of sapphire (0001) substrate ................. ............. .....95









5-9. Load-displacement curves of sapphire (0001) substrate before 'pop-in' occurs........96

5-10. Combined modulus of film and substrate for different quality of ZrC (001)
grown on Si (001) substrate. FWHM of ZC202 = 2.53, FWHM ofZC210a =
7.27 in om ega rocking curve ......................................................... .....................96

5-11. Load-displacement curves of ZrC (001) grown on Si (001) substrate, sample
Z C 202 in fig 5-10 ............ .............................................................. .......... ...... 97

5-12. Load-displacement curves of ZrC (001) grown on Si (001) substrate, sample
Z C 202 in fig 5-10 ............ .............................................................. .......... ...... 97

5-13. Measured elastic modulus as a function of relative penetration into a coated
specimen (a is contact radius, and t is film thickness). For curve 1, Eflm is less
than Esubstrate, and opposite case for curve 2 .................................. ............... 98

5-14. Combined modulus of film and substrate for different quality of ZrC (111)
grown on sapphire (0001) and Si (001) substrate, sample ZC204 and ZC210c
respectively. FWHM of ZC204 = 2.53, FWHM of ZC210c = 6.95 in omega
rocking cu rv e ...................................... .............................. ................. 9 8

5-15. Load-displacement curves of ZrC (111) grown on sapphire (0001) substrate,
sam ple Z C 204 in fig 5-14............................................... .............................. 99

5-16. Load-displacement curves of ZrC (111) grown on Si (111) substrate, sample
Z C 2 10c in fi g 5-14 ....................................................... .. ........ .... ..... ...... 99

5-17. Combined hardness of ZrC and substrate as a function of normalized depth; h -
displacem ent, t film thickness........................................ .......................... 100

6-1. XRD spectra of thin films deposited from a Co doped ZrC target under various
c o n d itio n s ............ .................................................................. 1 0 5















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

GROWTH OF EPITAXIAL ZIRCONIUM CARBIDE LAYERS
USING PULSED LASER DEPOSITION


By

Juhyun Woo

December 2005

Chair: Valentin Craciun
Major Department: Materials Science and Engineering

Epitaxial ZrC thin films were grown on Si (001), Si (111), and sapphire (0001)

substrate by the pulsed laser deposition technique. It has been found that crystalline films

could be grown only by using laser fluences higher than 6 J/cm2 and substrate

temperatures in excess of 600 C. For a fluence over 8 J/cm2 and a substrate temperature

of 600-700 C, cubic ZrC films exhibiting a (001) texture were deposited under vacuum

or low pressure C2H2 atmosphere. Under very low water vapor pressures (10- -10-9 Torr),

high substrate temperatures (700-750 C), and high laser fluence (10 J/cm2), highly

textured ZrC films were deposited on single crystalline substrates. Pole figures

investigation showed that films were epitaxial, with in-plane axis aligned with respect to

those of the substrate. X-ray reflectivity, atomic force microscope, ellipsometry, and

scanning electron microscopy confirmed that these films were smooth, with surface

roughness values below 1.0 nm and mass densities around the stoichiometric ZrC

tabulated value of 6.7 g/cm3. X-ray photoelectron spectroscopy and Auger electron









spectroscopy investigations showed that the surface of the films contained a significant

amount of oxygen and Zr-O bonds, the outmost 1-2 nm of the surface region being

mainly ZrO2. However, after the removal of this surface contamination layer, low oxygen

atomic concentration below 3 % were measured. Despite of the rather high levels of

oxygen contamination, electrical resistivity measured by four probe measurement

indicated that the deposited ZrC films were very conductive. The use of a low C2H2

pressure atmosphere during deposition had a small beneficial effect on crystallinity and

stoichiometry of the films. Nanoindentation measurements showed higher values of the

hardness for higher crystallinity. For the highest crystalline quality, (111) ZrC films

deposited on sapphire, values over 450 GPa for the elastic modulus and -31 GPa for the

hardness were measured.














CHAPTER 1
INTRODUCTION

Zirconium carbide (ZrC) is a typical refractory compound that crystallizes in the

rock salt (NaC1, B 1) ground-state structure under normal conditions. Recently, ZrC is

arousing interest because of its several notable properties, characterized by a very high

melting temperature of 3530 C [Zai84], excellent thermal stability, exceptional

mechanical hardness and strength [Che05a], chemical inertness, and imperviousness to

hydrogen attack [Tot71]. In addition, electrical conductivity is comparable to metals

[Zai84], and work function for electron emission is low [Mac95].

The common applications are in chemical- and wear-resistant coatings and ultra-

high temperature applications so far. However, as a form of thin coating or layer (or film)

in a thickness range of micron or submicron, ZrC has more important applications in

vacuum electronics or MEMS (micro electro-mechanical system) devices [ChaOl,

Tem99, and Xie96]. Particularly, the growth of epitaxial ZrC film has special importance

because of its anisotropic properties and possibility to be fabricated in various shapes of

microstructure by using anisotropic etching. When it is manufactured as a form of micro-

structured shapes, its applications are much wider with high efficient and compact

devices, using electron emission, such as short responding bright flat displays and

electron beam lithography. Also ZrC is a potential good candidate as a diffusion barrier

for metallization on silicon because it exhibits lower lattice mismatch and thermal

expansion coefficient difference with silicon (Si) than that of zirconium nitride (ZrN).









Especially, epitaxial films for this application could very efficiently prevent diffusion of

metal atoms (Al or Cu) because there are no grain boundaries for fast atomic diffusion.

Despite such many attractive applications and technological interest, only a few

studies describing ZrC film growth have been published so far. The growth of ZrC film

by thermal evaporation [Tes93], sputtering deposition [Bru93, Spr86], vacuum plasma

spray processes [Var94], chemical vapor deposition (CVD) [Ber95], pulsed laser

deposition (PLD) [Ale00], and ion beam deposition [He98] have been reported. However,

it appears that it is quite difficult to obtain high crystalline quality ZrC film, because of

its high melting temperature, low vapor pressure, and Zr atoms affinity for oxygen.

The objective of this study was focused on studying the relationship between the

thin film deposition parameters and the structure and properties of the resulting ZrC

films. Based on performance considerations, a pulsed laser deposition technique for the

growth of ZrC thin film had been chosen, and hence used in this work. In chapter 2, a

general background of ZrC material was presented, and growth methods and problems

were discussed as well. In chapter 3, the experimental method and characterization tools

used for analysis of film properties were explained. Grazing incidence x-ray diffraction

(GIXD), symmetrical XRD, and pole figure were used for structure analysis. Omega

rocking curve technique was used to check the degree of texture of films. X-ray

reflectometry (XRR), atomic force microscope (AFM), scanning electron microscopy

(SEM), and ellipsometry were used for surface morphology, thickness and/or roughness

measurement. Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy

(XPS) were used for surface chemical analysis. Four points probe was used to measure

electrical sheet resistance and resistivity, while transmission electron microscope (TEM)






3


was used for microstructure analysis. In chapter 4, the influence of process parameters on

the microstructure, crystallinity, and morphology were presented, and also the

dependence of substrates and its orientation on growth behavior of ZrC thin film was

analyzed and discussed. In chapter 5, the results regarding the hardness and elastic

modulus of ZrC films, obtained by nanoindentation technique, were presented. Finally,

the overall conclusions drawn from this work and suggestions for future work were

summarized in chapter 6.














CHAPTER 2
LITERATURE REVIEW

2.1 Characteristics of Zirconium Carbide

2.1.1 Composition and Structure

Zirconium (Zr, [Kr]4d25s2, group IVB) is a solid transition metal which has a

hexagonal close-packed (HCP) crystal structure at room temperature. Carbon (C, 1s22s2p2,

group IV) also naturally crystallizes in the same HCP structure. However, the structure of

the zirconium carbide (ZrC) does not follow its parent metal structures. That is, ZrC has

the Zr on a face centered cubic (FCC) lattice, even though Zr has HCP structure, and

carbon atoms occupy octahedral interstitial sites between Zr atoms. In transition metal

carbides, their structure depends on the s-p electron count [Oya92a]. With increasing s-p

electron count, the metal structure changes from body centered cubic (BCC) to HCP to

FCC across the transition series. Therefore, the group IVB and VB metal carbides (MC)

crystallize in the rock salt (B1, NaCl-type) structure (fig. 2-1) rather than a hexagonal

form because the partially filled bands of the host metals can accommodate a high ratio

of sp-electron-rich carbon to metal. In group IV the stoichiometry of M2C occurs often,

while group VII and V1II retain metal-rich stoichiometries of M3C and M4C, consistent with

an attempt to avoid filling anti-bonding levels in the metal bands [Oya92b].

Most of the transition metal mono-carbides form in the rock salt structure, fcc

metal with carbon occupying the octahedral interstitial sites. The shortest metal to metal

distance is about 30% greater in the metal carbide (MC) structure than in the pure metal









structure for the group IV and group V carbides [Oya92a]. At one hundred percent site

occupancy, the stoichiometry of transition metal carbide is M1C1. Although the zirconium

carbide is stable in solid solutions composition from carbon-deficient ZrCi-x to

stoichiometric ZrC (Fig. 2-2) [Bra92], it has been reported that crystalline ZrC film with

rocksalt structure can be synthesized with Zr to C atomic ratios from 0.5 up to 1.78 as a

vacancy compound [Bru93, Smi93, and Tes93].

2.1.2 Properties and Applications of Zirconium Carbide Thin Films

The properties of ZrC depend on several factors such as chemical compositions,

grain size, defect structures, and porosity. Hence, variations in properties have been

observed in the literature. Some of the characteristics and properties of zirconium carbide

reported in the literature are listed in table 2-1.


Table 2-1 Characteristics and properties of ZrC reported in literature.
Structure Rocksalt (B 1) [Oya92a, Tot71]
Space group Fm3m (225) [JCPDS PDF#: 35-0784]
Lattice parameter (A) 4.698 [Tot71], 4.6930 [JCPDS PDF#: 35-0784]
Density (g/cm3) 6.59 [Tot71], 6.73 [Lid05]
Hardness 2860 kg/mm2 [Zai84], up to 30.2 GPa [Che05a]
Elastic modulus (GPa) -325 [Che05a]
Melting temperature (oC) 3532 [Lid05], 3530 [Zai84], 3420 [Shi98]
Thermal expansion (x10-6/OC) 6.7 [Tot71]
Electrical resistivity (x 10-lO m) 6.2 at 293 K [Zai84], 20.4 at 300 K [Mod85]
Work function (eV) 3.38 [Zai84], 3.3-3.4 [Mac95]


As shown in table 2-1, ZrC exhibits many unusual properties such as high thermal

and electrical conductivity with extreme hardness, and low work function. Recently, most









applications of the zirconium carbide rely on its significant hardness such as cutting tools

[Pri92 and Tot71]. However, these combinations of properties are more important in a

variety of technological applications. For example, zirconium carbide coatings on

molybdenum, niobium, and nickel-based alloy were used to increase the radiation

emissivity, indicating a good candidate material for thermophotovoltaic (TPV) radiator

[Coc99]. Also, the coatings showed excellent resistance to thermal cycling and

acceptable stability during vacuum annealing at 1100 C.

In addition, ZrC films improved the field emission stability and beam confinement

when they were deposited on field emitter cathodes [Eda96 and Mac95]. The full width at

half maximum of energy distribution for particular materials is nearly equal to a half of

the work function [Ada74]. Thus, the cathodes of low work function have an advantage

because the full width at half maximum in energy distribution of the emitted electrons is

narrow.

Moreover, ZrC coating layers have much higher temperature stability and are more

resistant to the chemical attack by the palladium (Pd) fission product, while silicon

carbide (SiC) coating layers lose their mechanical integrity at temperatures over 1700 C

[Oga86 and Oga92].

Furthermore, video displays [Mac98a, Mac98b, and Mac98c], microwave

application [Mac92, Mac93, Mac94, Mac95, and Xie96], cold cathodes for operation in

poor vacuum environment, photocathodes for electron beam lithography [She97], hole

injection layers of organic light-emitting diodes (OLEDs), and substrates for epitaxial

growth of nitride are good potential fields for ZrC applications.









2.2. Techniques Used for Zirconium Carbide Thin Film Depositions

2.2.1 Thermal Evaporation (TE)

The thermal evaporation technique for ZrC films deposition was used by Tessner

and Davis in 1993 [Tes93] to widen the emission area of high-temperature thermionic

energy converters (TEC) and by Mackie et al. for the use of field emitter arrays (FEAs)

[Mac95]. In the evaporation process, the material to be deposited should be heated to a

high enough temperature to achieve a sufficient vapor pressure and the desired

evaporation. The wire filaments, sheet metal, or electrically conductive ceramic sources

are used in evaporation and they are heated by electrical current. Thermal evaporation

must be an important technology for the formation of functional coatings on a variety of

materials. However, there are limitations regarding the type of material that can be

heated. In some cases such as ZrC, it is not possible to achieve the necessary evaporation

temperatures without significantly evaporating the source holder. Moreover, chemical

reactions between the holder and the material can result in the contamination of the

coating.

2.2.2 Sputtering Deposition (S)

In the sputtering process, the target material is bombarded with high energy ions

that transfer their momentum to the atoms on the target materials which are ejected.

These sputtered particles condense on the substrate facing the target. The dc or rf

sputtering by an Ar plasma are the most common form. Because of the low sputter yield,

magnetron sputtering is often used to increase the deposition rate. Magnets which are

positioned behind the target induce the electrons to spiral and increase the degree of

ionization of the plasma due to longer path length of electrons.









Alternatively, the configuration consisting of three Ar ions beams can be used to

co-deposit ZrC film, which is called "tri-ion beam-assisted deposition (tri-IBAD)"

[He98]. Two beams of Ar ions can be used to sputter from graphite and zirconium. A

third Ar ion beam can be used to bombard the growing ZrC film to provide additional

energy to enhance film formation processes.

Compared to evaporated particles, sputtered particles have considerably higher

kinetic energies. As a result, sputtered layers usually have higher adhesive strength and a

denser coating structure than evaporated layers. The greatest advantage of sputtering

deposition is a large particle source area compared to evaporators, which enables a large

area coating with a high degree of uniformity. Nevertheless, sputtering could have a

relatively low ratio of energetic ions to neutral species, so that it is possible not to

produce the hardest films.

2.2.3 Chemical Vapor Deposition (CVD)

In chemical vapor deposition methods, the state of substances used for the growth

is in the vapor phase when they are introduced to the vacuum system. The substances

must be thermally excited by appropriate high temperatures or with plasma to be

deposited. Chemical vapor deposition (CVD) of ZrC can be accomplished by the reaction

between zirconium halide and a hydrocarbon in H2 atmosphere at temperatures above

10000C. The zirconium halide vapor can be obtained either by a reaction between a

halide vapor and zirconium metal, or by sublimation of ZrC14.

Nuclear fuel particles were coated with ZrC coatings by CVD [Rey74]. Argon (Ar)

gas was initially bubbled through dichloromethane (CH2C12) that was kept at 0 OC. Then

the Ar and CH2C12 gas mixture was passed through a heated zirconium (Zr) "sponge" at

600 C to produce ZrC14 vapors. Preheated methane (CH4) and hydrogen (H2) gases to









600 C were mixed with the ZrC14 vapors inside a graphite tube and heated to 1100 C for

3 hours to produce ZrC coatings. Chemical analysis of the coating resulted in a C/Zr ratio

of 1.01 which indicated that some free carbon was present in the coatings.

The effect of the composition of gas mixtures on the properties of ZrC coatings was

studied [Hol77 and Wag76]. Chemical vapor deposition of ZrC coatings was achieved by

reacting gaseous mixtures of CH4, H2, ZrC14, and Ar as carrier gas. The overall reaction is

given by

xCH4 + ZrCl4 + 2(1 x)H2 = ZrCx + 4HCI, (x < 1) (2-1)

Increasing the amount of CH4 in the coating gas mixture (i.e., increasing C/Zr

molar ratio) resulted in ZrC coatings with increased amounts of carbon. The chemical

analysis of the coating showed the presence of free carbon, when the C/Zr molar ratio in

the coating gas mixture was over 0.21.

Organometallic precursor (cyclopentadienylzirconium) was also used to deposit

ZrC coatings at temperatures in the range of 300-600 C [Han95]. The chemical

composition of the films synthesized at 600 OC showed presence of zirconium carbide, as

determined by x-ray photoemission spectroscopy (XPS).

2.2.4 Pulsed Laser Deposition (PLD)

The technique of pulsed laser deposition (PLD) has been used to deposit high

quality films of various materials for more than two decades. The technique uses high

energy laser pulses (typically 2-5 J/cm2) to melt, evaporate, and ionize material (ablation

process) from the surface of a target. This ablation event produces a transient, highly

luminous plasma plume that expands rapidly away from the target surface. The ablated

material is collected, and then condenses on a suitably placed substrate.









Applications of the technique range from the superconducting films production and

insulating circuit components to medical applications to improve biocompatibility. In

spite of this widespread usage, the fundamental processes occurring during the transfer of

material from target to substrate are not fully understood yet, and are consequently the

focus of much research.

The interaction of laser radiation with solid surfaces was under investigation from

as early as 1962, when Breech and Cross [Bre62] analyzed the emission spectrum of

material vaporized by laser pulses. The first demonstration ofPLD in 1965 [Smi65] did

not attract significant interest, as the films were inferior to those obtained by other

deposition techniques, such as chemical vapor deposition (CVD) or molecular beam

epitaxy (MBE). The PLD technique was slowly developing for approximately the next

twenty years until Dijkamp and Venkatesan [Dij 87] used PLD to grow a film of the high

temperature superconducting material Ba2Cu307 + YBaCuO (YBCO). The films

obtained were found to be superior in quality to those previously grown using other

deposition methods and awaked a tremendous interest in the technique. Present day

research applications include growing films for magneto-optic storage devices,

developing multilayer devices for x-ray optics, and depositing diamond films on

components for protection and insulation.

A schematic diagram of the basic PLD configuration is shown in fig. 2-3. The

general understanding of the process is somewhat simple, which uses short pulses (pulse

duration, 10-30 ns) of laser energy to remove material from the surface of a target. The

vaporized material, containing neutrals, ions, and electrons, is known as a laser-produced

plasma plume, and expands rapidly away from the target surface (velocities typically









~106 cm/s in vacuum). Film growth occurs on a substrate upon which some of the plume

material recondenses. However, in practice, the situation is not so simple, with a large

number of variables affecting the properties of the film, such as laser fluence, background

gas pressure and substrate temperature. These variables allow the film properties to be

improved for individual applications. However, optimization can require a considerable

amount of time and effort. Indeed, much of the early research into PLD concentrated on

the empirical optimization of deposition conditions for individual materials and

applications, without attempting to understand the processes occurring as the material is

transported from target to substrate. The technique of PLD was found to have significant

benefits over other film deposition methods. The capability for stoichiometric transfer of

material from target to substrate, i.e., the almost exact chemical composition of a

complex material such as YBCO, can be reproduced in the deposited film. Relatively

high deposition rates (-100 A/min) can be achieved at moderate laser fluences, with film

thickness controlled in real time by simply turning the laser on and off.

The fact that a laser is used as an external energy source results in an extremely

clean process without internal filaments. Thus deposition can occur in both inert and

reactive background gases. The use of a carousel housing a number of target materials

enables multilayer films to be deposited without the need to break vacuum when

changing between materials. In spite of these significant advantages, industrial adoption

of PLD has been slow and most applications have been limited to the research

environment. There are basically three main reasons for this. First, the plasma plume

created during the laser ablation process is highly forward directed; therefore the

thickness of material collected on a substrate is highly non-uniform and the composition









can vary across the film. The area of deposited material is also quite small in comparison

to that required for many industrial applications which require area coverage of at least

~8 x 8 cm2. Second, the ablated material contains macroscopic globules of molten

material, up to -10 [m diameter. The arrival of these particulates at the substrate is

obviously detrimental to the properties of the film being deposited. Third, the

fundamental processes occurring within the laser-produced plasmas are not fully

understood. Thus deposition of novel materials usually involves a period of empirical

optimization of deposition parameters. To a large extent the first two problems have been

solved. Films of uniform thickness and composition can be produced by rastering the

laser spot across the target surface and moving the substrate during deposition. Line-

focus laser spots have also been used to obtain large area coverage. The particulate

material was initially removed from the plume using a mechanical velocity filter,

although recently more elaborate techniques, involving collisions between two plasma

plumes or off-axis deposition, have been used to successfully grow particulate-free films.

The third problem will be resolved by the development of computer simulations to

describe PLD. However, a large amount of experimental data is required to support the

verification of such models.

2.2.5 Comparison of Techniques

Fig. 2-4 [Chr94] compares the energy range of the atomic fluxes for each technique.

The solid boxes are typical values for the energy range of the atomic fluxes in practice,

and the dashed boxes are the energetic fluxes values which could be controlled. The

spread in energy of the atomic flux can be divided into energetic portion to the flux and

thermal fluxes. The purely thermal techniques such as CVD (chemical vapor deposition)

and MBE (molecular beam epitaxy) are mostly used so far because the energetic









deposition techniques are still under development. The shaded area in fig. 2-4 [Chr94] is

the best energy range for deposition flux that was established earlier. Energy of 5 eV to

10 eV per atom promotes surface diffusion and high sticking probability while

minimizing damage. The energetic flux over 100 eV can produce deep damage, so it

should be avoided. The energy range of sputtering (S) and PLD matches most closely the

best energy range indicated. Although the techniques with purely thermal fluxes are in

standard use to deposit high quality epitaxial films, the energetic techniques can be

employed to lower the growth temperature at which epitaxy can be achieved, especially

for high melting temperature materials such as ZrC.

Fig. 2-5 [Chr94] shows the range of pressures in which each of the techniques is

operable. PLD can relatively operate in a wide pressure range from UHV (ultra high

vacuum) to 1 Torr, while other techniques need lower than HV (high vacuum) pressure to

minimize contamination, to avoid oxidation of evaporation filaments, or to sustain

plasma. Although most materials do not require performing the deposition in a high

pressure of background gas, PLD has an advantage over other techniques in operating

under high background gas pressure that is particularly useful for multicomponent oxides

for example.

2.3 Growth and Factors Determining the Quality of Thin Films in PLD

2.3.1 Nucleation and Growth

The atoms accumulated on the surface may diffuse laterally over substrate, also

they may encounter other mobile atoms to form mobile or stationary clusters, then they

may attach to preexisting film-atom clusters, or they may be reevaporated from substrate

[Ven84]. The balance between growth and dissolution processes will be governed by the

total free energy of the cluster. Nucleation could be stabilized and destabilized,









depending on the energy stored in forming surface of new face, the adhesion energy

released in formation of an epitaxial interface, the strain energy stored in lattice-

mismatched epitaxial growth, and electrical energy stored in the formation of a surface

dipole.

Three modes of film growth at the initial stages are possible [Lew78]. When the

cohesive energy of the film atoms is greater than the cohesive binding between the film

and substrate atoms, the formation and growth of isolated islands occur (Volmer-Weber

Island growth) [Mah99]. This mode can result in an epitaxial film that has a rough

surface, or a polycrystalline film. When the cohesive energy between the film and

substrate atoms is greater than the cohesive energy of the film (but monotonically

decreases as each film layer is added), layer by layer (Frank-Van der Merwe) growth

occurs [Mah99]. This mode results in a very smooth epitaxial film. When the monotonic

decrease in binding energy with each successive layer is energetically overridden by

some factor such as strain energy, island formation becomes more favorable. So mixed

(Stranski-Krastanov) growth, island growth after the first monolayer forms, can occur

[Mah99].

The growth of thin films by PLD depends on many factors, such as energy,

ionization degree, and type of the condensing particles, temperature, and

physicochemical properties of the substrate. There are two main thermodynamic

parameters that determine the growth mechanism. One is substrate temperature T and the

other is supersaturation m.

R
m = kTln- (2-2)
Re









where k is the Boltzman constant, R is the actual deposition rate, and Re is

equilibrium deposition rate at temperature T.

The overall film growth process in PLD has been theoretically studied [Met89].

The mean thickness at the moment of 99% covered substrate (N99 [Kas78]) at which a

growing thin discontinuous film reaches continuity has been found to be given by


N99 = 0.5 exp- 3Es Ed (2-3)
R 3kT

where v is the adatom vibrational frequency, No is the density of adsorption sites on

the substrate, Edes is the activation energy for adatom desorption, and Esd is the activation

energy of adatom surface diffusion. The lines of equal mean film thickness at the moment

of 99% covered substrate (defined by the condition N99(R, T) = constant) can be drawn in

In R versus 1/T coordinates (fig 2-6).

One of line of equal mean film thickness at the moment of 99% covered substrate is

drawn as dashed line. Depending on deposition rate and temperature, high-island growth

region and low-island growth region are divided in the growth diagram. In the high-island

growth region the thin film reaches continuity with a mean thickness. In the low-island

growth region a mean thickness does not exceed a few monolayers. At higher R and

lower T, continuous growth (amorphous) takes place from island growth [Kas78].

Therefore depending on the experimental conditions, such as energy density of

laser and substrate temperature, single crystalline thin film in high-island growth region,

polycrystalline film in low-island growth region, or amorphous film can be formed.

2.3.2 Background Gas

During pulsed laser deposition, the use of background gas can be divided into

passive or active use. The passive use is mostly to compensate for some loss of a









constituent element. For examples, the deposited oxides tend to be deficient in oxygen.

Typically 10-300 mTorr of background oxygen in chamber is required for oxide

superconductors.

Also the introduction of background gas change typical particulate size as the

ambient gas pressure varies. Inert or reactive gas can be introduced to form particulates

with a desired size or composition for active use of ambient gas during PLD. The

decrease in the ambient gas pressure results in a decrease in size and a narrower size

distribution [Mat86].

The origins of the formation of the particulates and the mechanisms of enrichment

in specific element in the particulates are different in vacuum and in inert ambient gas for

PLD processes. The effect of inert ambient gas pressure increases collisions between the

ejeted species and the ambient gas as the pressure increases. At a pressure of 1 mTorr, the

mean free path is about 5 cm. The mean free path of ejected species becomes 0.05cm at a

higher pressure of 100 mTorr. In vacuum, there are no collisions between ejected species

virtually, so particulates are predominantly formed from solidified liquid droplets, and

the vapor species are deposited as a uniform background film in the same time.

However, when the ambient gas pressure increases, the vapor species can have

enough collisions. Thus, nucleation and growth of vapor species can occur, before they

arrive at the substrates. This suggests that the ultrafine particulates are formed from the

vapor species instead of liquid droplets.

2.3.3 Vacuum

The quality of the vacuum is a major consideration for determining the deposition

rate. Gaseous impurities in the deposition chamber impinge on the growing film and will

be incorporated into the film depending on their sticking probabilities. The most common









impurities are H20, CO, C02, and H2. Fig. 2-7 [Chr94] indicates the relationship between

atomic fluxes on surface with pressure and deposition rate. To avoid contamination of

impurities, deposition rate should be controlled depending on deposition pressure.

2.3.4 Laser Fluence

The laser fluence is most significant on the particulate size and density. The laser

fluence can be changed varying the laser power or the laser spot size. There is threshold

laser fluence to ablate target material with laser beam. For example, in the case of ZrC,

the threshold laser influence is about 1.3 J/cm2 [Ale00]. The laser fluence is extrapolated

and a linear trend up to 13 J/cm2 is maintained (fig. 2-8 [Ale00]). In other word, the

saturation of ablation process is not observed. One of the mechanisms that reduce the

ablation rate is plasma shielding of the target [Dye89], and it is more often encountered in

the laser-ablation deposition using longer wavelength.

2.3.5 Laser Wavelength

The laser wavelength is directly related to the effectiveness of the absorption of the

laser power into the target. For most metals, the absorption coefficient decreases with

decreasing laser wavelength. Thus, the laser penetration depth in metal is larger in the

UV range than in the infrared range. For other materials, the variation of absorption

coefficient with wavelength is more complex due to various absorption mechanisms, such

as lattice vibration, free carrier absorption, impurity centers, or bandgap transition.

The primary effect of the laser wavelength on particulate generation is mostly due

to the difference in the absorption coefficient when different laser wavelength is used.

Larger particulates are generated when using longer laser wavelength [Kau90].









2.3.6 Target to Substrate Distance

The effect of target to substrate distance is mainly reflected in the angular spread of

the ejected flux. Depending on the position of the substrate, different particulate

appearance may occur. The specific effects of target to substrate distance and ambient

pressure are related. The plume dimension decreases as the background gas pressure

increases due to the increased collisions between the laser-produced plume and the

background gas. When the target to substrate distance is smaller than the plume length,

there is no remarkable difference in particulate size and density. As the target to substrate

distance increases, a few larger particulates appear [Chr94].






























Figure 2-1. Image of cubic rocksalt (B ) structure of transition metal carbides. Carbon
atoms are depicted as the light gray spheres, metals as dark gray spheres.


Wt%C


0 10 20 30 40 50 60
At% C


Figure 2-2. Zr-C phase diagram showing wide range congruent compositions [Bra92].




































Figure 2-3. Schematic diagram of an apparatus for pulsed laser deposition.










BEST ENERGY
RANGE?

IBAD
0 i I..:::..:.:^A :........
ICBD, PECVD

jj AMBE, IVD

.is


Typical Range
Typical Range


0.1 1 10 100 1000 10000 100000
ENERGY SPREAD OF DEPOSITED ATOMS (eV)

Figure 2-4. Schematic diagram of the approximate energy range of deposited atoms for
various deposition techniques. The shaded box indicates the supposed energy
range of atom fluxes considered to be beneficial for film growth [Chr94].


10-3


PRESSURE (TORR)


Figure 2-5. Schematic diagram comparing the pressure ranges over various techniques.
PLD can operate over the widest range of all the methods [Chr94].


I PLD
------'-- --;::--
I .. IBD
] ALE, CVD, L CVD, M CVD, MBE, TE


PLD

LALL CVM M


IBAD, IBD, ICBD, IVD, TE

| ALE, AMBE, MBE


10-9


..































11T (K1)

Figure 2-6. Growth diagram drawn by equation 2-3, showing the dependence of growth
behavior on temperature and growth rate.


Flux
(i Ulcm2Xs)


1010 1011 1012 1013 1 1014 015 01 1017
I I I I i I I 1


1 1 1 1 1 1 1 1 1 1


Approximt a position Rut.
(Aft)


Approximate G 10-
Preuren
(Torr)


0.05 0.5 5
I I 1


500 5000


I I I I I I


1i- 10l 10-7 10- 10-5 10o4 i"3 10-2
I I I I I I I I


Figure 2-7. Relationship of impinging particles fluxes to deposition rate, and gas pressure
[Chr94].


Continuous Growth




'Low-Island Growth

'-R>Rc, T



High-Island Growth

RTc


1010 1019
I 1


----------------------------------------


J












--25








laser fluency of 1.3 2 [Ale].
S15








Fluence (/cm')

Figure 2-8. ZrC ablation rate, showing linear dependence on laser fluence, and threshold
laser fluence of 1.3 J/cm2 [Ale00].















CHAPTER 3
EXPERIMENTAL METHODS

3.1 Pulsed Laser Deposition System

Most depositions are conducted using the laser wavelength of 193-400 nm because

strong absorption is exhibited in this spectral region by target materials. Absorption

coefficients have a tendency to increase as laser wavelength is shorter in this range. The

excimer laser wavelengths used in a commercial laser system are listed in table 3-1. A

Lambda Physik KrF laser (k = 248 nm) was used for the experiments in this study with a

25 ns pulse duration.


Table 3-1 Excimer laser operating wavelengths
Excimer Wavelength (nm)

F2 157

ArF 193

KrCl 222

KrF 248

XeCl 308

XeF 351


A schematic diagram for chamber was illustrated in fig. 2-3. A stainless steel

spherical vacuum chamber is used. The laser entry port is protected by a quartz plate,

which is designed to reduce the deposition rate of materials on window. The angle

between the target normal and laser beam is 45. A turbo-molecular pump backed by a









mechanical pump provided the high vacuum. The system operated between 2.5x 106 and

1.5 x 107 Torr of base pressure, which is critical factor in controlling the quality of film.

3.2 Structural Characterization

Symmetric and asymmetric x-ray diffraction spectra (XRD) were collected on a

Panalytical MRD X'pert system using Cu Ka radiation to identify crystal structure and

quality. The standard settings for the X-ray generator were 45 kV at 40 mA.

Fig. 3-1 compares the grazing incidence angle geometry used for thin film with the

conventional 0-20 symmetric geometry used for bulk analysis. In the grazing incidence

XRD geometry for the thin film arrangement, the incident and diffracted beams are

prepared nearly parallel using a mirror and a narrow slit on the incident beam (1/80

divergence slit for experiments of this study) and a parallel plate collimator (0.270

divergency) on the detector side. In addition, a very small angle of incidence beam to the

sample surface (typically 1 to 30) increases the path length of the X-ray beam through

the film. This increases the diffracted beam intensity from the film, while that from the

substrate is reduced. During the collection of the diffraction spectrum, only the detector

moves along the angular range, while keeping the incident angle fixed. Thus the beam

path length and the irradiated area are constant.

For more detail observation of highly textured ZrC thin films, omega rocking

curves were collected. The geometry is represented in fig. 3-2. The peak broadening

along 0-20 direction is caused by variation in d-spacing with depth for thick films or by

use of non-monochromatic beam. On the other hand, the peak broadening in omega

direction is due to mosaic spread, lateral incoherence, high uniform dislocation density,

or sample curvature.









Also pole figure method was used to confirm not only in-plane texture, but also

out-of-plane texture as well as crystal quality and orientation relationship between film

and substrate. A schematic diagram of pole figure geometry and relation with

microstructure are shown in fig. 3-3. The incidence beam angle and diffracted beam

angle is set to specific angle which is satisfying Bragg's law. All of reflections in

projection area are from specific faces at the Bragg angle, but the geometry of the sample

is changing by rotation and by tilting during scanning. Thus, we can obtain the

orientation distribution.

3.3 Film Thickness and Roughness

For the measurement of film thickness and roughness, mainly x-ray reflectometry

(XRR) was employed and compared with spectroscopic ellipsometry (SE, Woollam M-

88). XRR method is to record the intensity of the x-ray beam reflected by a sample at

grazing angles. The operation mode is the same as regular powders XRD setup which

always make the incident angle half of the angle of diffraction. The reflection at the

surface and interfaces is due to the different electron densities in the different layers in

film, which correspond to different refractive indexes.

For incident angles 0 below a critical angle (0c), total external reflection occurs.

The critical angle for most materials is less than 0.30 in 0. Above 90, the reflections from

the different interfaces interfere and give rise to interference fringes. The period of the

interference fringes and their decrease in intensity are related to the thickness and the

roughness of the layers. The reflection spectra can be analyzed with the aid of the

WinGIXA software, which is based on the Parratt formalism modified to include

roughness [Par54 and Nev80]. The typical ranges for these measurements are between

0.1 and 50 in 20. For the detail analysis of x-ray reflection, WinGixa software was used









for fitting x-ray reflection curves. To get an accurate fit for the acquired data, a three

layers model was used, which consists of a topmost contamination layer, then the

deposited layer, and an interfacial layer located between the deposited layer and substrate.

The measurable ranges and resolution for this measurement are listed in table 3-2

[AucO 1].


Table 3-2 Typical XRR values for the resolution and ranges.
Parameter Resolution Range

Thickness 0.1-5 % 2-200 nm

Surface roughness + 0.2 nm 0.1-3 nm

Interface roughness + 0.2 nm 0.1-3 nm

Density 1-10%


Atomic force microscope (AFM) images were taken with a Digital Instruments

Dimension 3100. Images were acquired in the tapping mode with a scan range 5 x 5 am.

Before roughness analysis was performed, an automatic 'plane fit' and a 'flatten'

command were applied to the images to level the image and to reduce slope associated

with measurement drift.

3.4 Surface Chemistry Analysis

For Auger electron spectroscopy (AES) investigation, a Perkin-Elmer PHI 660

instrument was used for the surface analysis. Survey spectra were obtained from as-

deposited film, also after removing the topmost layer by Ar ion sputtering. X-ray

photoelectron spectroscopy (XPS) spectra were acquired on a PHI model 5100 ESCA

system using Mg Ka X-ray source (1253.6 eV). Argon ion was used to remove surface

contaminants and to remove atomic layers into the thin films. The multiplex data were

used during peak deconvolution. Background was substracted, and peaks were fitted with









one or more component Gaussian-Lorentzian functions to determine accurate chemical

bonding.

3.5 Electrical Measurement

The electrical measurement was taken on an Alessi four point probe. A probe head

having tungsten carbide tips with a point radius of 0.002" and a probe spacing of 0.05"

was used for all measurements. Current was supplied by a Crytronics model 120 current

source with a range of applied currents between 30 tA to 100 [A. Eq. (3-1) was used to

determine resistivity (p, Q-cm).

4.5324 xV x t
P= (3-1)
I

Based on the dimensions of the sample and probe head, no geometrical correction

factors were applied. The term t is film thickness (cm), and V is the voltage measured at

the supplied current (I).








Conventional 8-28 geometry


Figure 3-1. A schematic diagram of symmetric and asymmetric GIXD geometry.


Ak=Ak

kin

W V


Figure 3-2. A schematic diagram of omega rocking curve geometry.


Subs


Kout


Z::


GIXD geometry





















Random Orientation


Out-of-plane texture


.Sample rotation: 0 (rmttion axjs
parallelto sample surface) o





Sample tilting: q (tilting
direction is perpendicular
to incidence beam)
In-plane texture Perfect crystal
(biaxial)

Figure 3-3. An example of a schematic reflections on pole figure related to crystal
quality.


~83 f














CHAPTER 4
GROWTH AND CHARACTERIZATION OF HIGH CRYSTALLINE QUALITY
ZIRCONIUM CARBIDE FILMS

4.1 Introduction

Zirconium carbide (ZrC) is a refractory compound, characterized by a high melting

temperature of 3530 C [Zai84], excellent thermal stability, high mechanical hardness

and strength [Che05a], and chemical inertness. In addition, electrical conductivity is

comparable to metals [Zai84], and work function for electron emission is low [Mac95].

The common applications are the chemical- and wear-resistant coating, and ultra-high

temperature applications. Thin films of ZrC also have important application in vacuum

electronics or micromechanics [Xie96, Tem99, and Cha01]. Particularly, the epitaxial

growth of ZrC is specially important because it could allow the growth of various types

of nanostructure by using anisotropic etching, that would exhibit high performance,

efficiency, and compactness, such as short response and high brightness flat displays, and

electron beam lithography. ZrC also exhibits lower lattice mismatch and thermal

expansion coefficient difference with Si than ZrN, making it a potential good candidate

for metallization or diffusion barriers structures for Si-based electron devices. Despite of

the considerable attractive applications, relatively few studies describing ZrC film growth

have been published so far. The growth of ZrC film by thermal evaporation [Tes93],

reactive magnetron sputtering deposition [Bru93], vacuum plasma spray process [Var94],

chemical vapor deposition (CVD) [Ber95], pulsed laser deposition (PLD) [Ale00], and

tri-ion beam-assisted deposition [He98] has been reported. However, it seemed to be









quite difficult to obtain high crystalline quality ZrC films, because of its high melting

temperature, low vapor pressure, and Zr atoms affinity for oxygen. Pulsed laser

deposition (PLD) is recognized as a techniques that can overcome these difficulties with

respect to other techniques. PLD was successfully employed to grow high crystalline ZrC

films in this study. By optimizing the deposition conditions, epitaxial ZrC films on

silicon and sapphire substrate were obtained. The structure, stoichiometry, and optical

and electrical properties of these films are described in this chapter.

4.2 Experiment

The film depositions were conducted in an all-metal vacuum chamber using a KrF

excimer laser (X=248 nm). First of all, experiments were focused on studying the effects

of growth conditions (or process parameters) on the deposited films. Especially an effort

was made for growing crystalline ZrC films. Within the range of 2 10 J/cm2 laser

fluences with 5 Hz laser pulse repetition rate, the ZrC films were deposited on Si (001) at

the temperature range of 200 700 C. Si (001) substrates were cleaned chemically by

acetone, methanol, and rinsed in de-ionized water in turn, then dipped in 1 % HF solution

for 1 min, blown dry by high purity nitrogen gas, and immediately loaded into the

deposition chamber. Depositions were performed under residual vacuum or a low C2H2

atmosphere. The study was continued to produce high quality films by optimizing growth

parameters and to study growth behavior on various substrates. At this time, special care

was taken to maintain low water vapor pressures below x 108 Torr during depositions, as

measured with a residual gas analyzer (RGA) attached to the deposition chamber.

Additionally, Si substrates were heated to 900 C and maintained for 20 min under high

vacuum to remove the passivation layer on the substrate. The laser parameters used were

10 J/cm2 fluences and 10 Hz repetition rate at substrate temperatures around 750C.









The films surface and interfacial roughness, mass density and thickness were

obtained by simulating the measured x-ray reflectivity (XRR) spectra acquired with a

Panalytical X'Pert MRD system. The same instrument was used for structural

characterization in symmetric and grazing incidence x-ray diffraction (XRD and GIXD).

Pole figure measurements were acquired both from the films and substrates for texture

characterization. Omega rocking curve method was used to evaluate quality of epitaxial

films. The thickness and optical properties of the films were measured by spectroscopic

ellipsometry (SE, Woollam M-88). The chemical composition of the films was

investigated by Auger electron spectroscopy (AES, Perkin-Elmer PHI 660). To analyze

the bonding structure of ZrC, X-ray photoelectron spectroscopy (XPS, Perkin-Elmer PHI

5100 ESCA) was used. Focused Ion Beam (FIB) Strata DB 235 technique was employed

to make cross-sectional samples for high resolution transmission electron microscopy.

Bright field image and electron diffraction pattern were obtained from high resolution

JEOL TEM 2010F to analyze microstructure and quality of deposited thin films. Details

of each experiment are described in chapter 3.

4.3 Results and Discussion

4.3.1 Laser Fluence and Temperature Effect on Deposited Films

ZrC films were deposited on Si (001) substrates to study the effect of substrate

temperatures and laser fluences on the grown films. The films were grown for 20 minutes

at various growth conditions, which resulted in estimated thickness around 400 A. The

deposited films were analyzed by surface sensitive grazing incidence XRD, and the

results at 10 of an incidence beam angle are displayed in fig. 4-1. The growth conditions

of the films are listed in table 4-1. Under low temperature conditions of 200 600 C

and/or low laser fluence range of 2 6 J/cm2, GIXD spectra did not show any diffraction









peak (ZC05 and ZC06 in fig. 4-1), indicating that the deposited films under these

conditions were amorphous. By increasing the laser fluence to 8 J/cm2 and substrate

temperature at 600 C, the deposited films started to show some crystallinity (ZC11 in fig.

4-1).


Table 4-1. Growth conditions of deposited films, showing GIXD spectra in fig. 4-1.
Pressure during Substrate
e d g S Laser fluence Repetition rate
deposition temperature
ZC06 1x10-3 Torr C2H2 500 C 6 J/cm2 5 Hz

ZC05 1x 105 Torr C2H2 600 C 3 J/cm2 5 Hz

ZC11 1x 105 Torr C2H2 600 C 8 J/cm2 5 Hz

ZC24 1 x 104 Torr C2H2 600 C 10 J/cm2 5 Hz

ZC26 8.5 x 104 Torr C2H2 600 C 10 J/cm2 5 Hz

ZC25 1x 10-4 Torr C2H2 700 C 10 J/cm2 5 Hz



When the ZrC target was not pre-ablated for at least 1 min, GIXD pattern

(incidence beam angle = 10) of as-grown films in fig. 4-2 (a) showed some additional

diffraction peaks other than those obtained from target in fig. 4-3, which were crystalline

with diffraction lines corresponding to cubic and stoichiometric ZrC (ao = 4.69 A, JCPDS

PDF# 35-0784 [Pcp94]). Those additional diffraction peaks (vertical solid lines) in as-

grown films were assigned to substoichiometric ZrCo.7 compound (ao = 9.38 A, cubic

(space group: P), JCPDS PDF# 32-1489 [Pcp94]), which was probably caused by the

oxidation of target surface. So the target was pre-ablated for 1 min hereafter, ahead of the

growth processes to remove any surface oxide layer on target. Resulting GIXD spectra in

fig. 4-2 (b) showed a stoichiometric ZrlC1 compound (JCPDS PDF# 35-0784 [Pcp94]),









which is indicated as dashed vertical lines. In addition, it should be mentioned here that a

low pressure (x 10-5 x 10-3 Torr) of acetylene (C2H2) gas was used as background gas for

this study because GIXD spectrum from a film deposited at low C2H2 atmosphere in fig.

4-4 showed a little improvement of crystallinity in the film. Other deposition conditions

except background gas pressure were all same (Ts=600 C, 8J/cm2, and 5Hz).

At laser fluence at 10 J/cm2, resulting deposited films showed better crystallinity,

evidenced by clear peaks in GIXD spectra (ZC24, ZC25, and ZC26 in fig. 4-1). All

observed the diffraction lines corresponded to a stoichiometric cubic ZrC (JCPDS PDF#

35-0784 [Pcp94]), which were identical to those from the pure ZrC target in fig. 4-3. The

crystallinity of the films deposited at higher substrate temperature was much improved at

laser fluence of 10 J/cm2 (from ZC24, 600 C to ZC25, 700 C in fig. 4-1). Therefore it is

apparent that the deposition of crystalline films requires the simultaneous use of high

substrate temperatures and high laser fluences.

4.3.2 Deposition Rate and Thickness Uniformity of Deposited Films

The effect of deposition time was investigated on density and roughness of the ZrC

film grown on silicon (001) substrate at 700 C under residual vacuum (1 x 10-6 Torr) for 7,

14, and 20 min. The used laser parameters were 10 J/cm2 fluence, and 5 Hz repetition rate.

The layer thicknesses and roughness of these films were extracted by fitting the acquired

x-ray reflectivity (XRR) spectra using the WinGIXA software from Panalytical, which is

based on the Parratt formalism modified to include roughness [Par54 and Nev80]. Instead

of using only one ZrC layer model for fitting, a three-layer model was used that resulted

in a better fitting: an interfacial layer, a ZrC layer, and a contaminated surface layer such

as oxygen and hydrogen oxide. Also, thicknesses measured by spectroscopic ellipsometry

were compared with those obtained by XRR data. From the measured x-ray reflectivity









spectra in fig. 4-5, the critical angles indicating the density of films [Rus02 and Zab94]

were quite similar for all deposited films, and as high critical angle indicated (0c 0.35),

high density values around 6.7 g/cm3 were obtained. The thickness, roughness, and

density parameters obtained from x-ray reflectivity curves are listed in table 4-2. The

thickness change of deposited film was linearly time dependent as shown in fig 4-6. The

average deposition rate was 0.34 A/sec (or 0.068 A / pulse). This was a rather low

deposition rate for PLD, implicating rather low vaporized ZrC concentration in a plasma

plume.


Table 4-2. Thickness, density, and surface roughness values of the ZrC films deposited at
700 C for different times.
Deposition time (min) 7 14 20

Surface layer Thickness (A) 17 18 18
(ZrCxOy) Roughness (A, rms) 3.3 5.3 5.3

XRR Thickness (A) 127 297 376
ZrC layer
Density (g/cm3) 6.74 6.69 6.73
Interfacial layer Thickness (A) 0.6 2.3 2.0
Surface layer Thickness (A) 10 10 9
Ellipsometry ZrC layer Thickness (A) 125 310 382
Interfacial layer Thickness (A) 1.0 1.5 1.0


The GIXD (incidence angle co = 1) spectra of these films shown in fig. 4-7

indicate that all crystalline orientations are present in the films. The intensities of these

peaks slightly increased with an increase of thickness, indicating some increase in the

volume of the randomly oriented grains. However, these diffraction peaks did not appear

in the corresponding symmetric XRD spectra displayed in fig. 4-8, that is, the amount of

these randomly oriented grains was rather small. According to the XRD results (fig. 4-8),









the deposited films under this condition grew with a (001) texture. It is also clear that,

based on XRR data from table 4-2, there are no significant changes in the film density or

surface morphology when the deposition time was changed.

The thickness uniformity of the deposited films was checked by acquiring XRR

spectra at various positions from the same samples that were grown under identical

condition. According to the data extracted from these XRR spectra, a ZrC film thickness

uniformity of around 5% in the central 1.5 x 1.5 cm2 area of the deposited film was

estimated.


Table 4-3. Degree of out-of-plane texture of ZrC films at various growth conditions.
FWHM
Substrate ID Po (Torr) Pd (Torr) Ts (C) Texture (Omega-
Rocking)
ZC202 3.0x10-7 6.2x106 C2H2 775 (001) 2.53

ZC104 2.6x10-6 1.2x105 C2H2 750 (001) 3.37

S ZC103 2.4x10-6 4.0x106 C2H2 750 (001) 4.47
ZC208* 1.5x10-7 2.2x10-6 CH4 750 (001) 4.62
ZC106 9.0x10-7 1.0x105 C2H2 750 (001) 5.45

ZC104 2.6x10-6 1.2x105 C2H2 750 111/I200=9.08 4.82

ZC107 8.0x10-7 6.7x106 C2H2 750 1.3
ZC210* 1.4x10- 9.0x10- CH4 750
ZC208* 1.5x107 2.2x10-6 CH4 750
ZC107 8.0x10-7 6.7x106 C2H2 750 (111) 0.46

S ZC104 2.6x10-6 1.2x105 C2H2 750 (111) 0.68
8 ZC208* 1.5x107 2.2x106 CH4 750 (111) 2.31
ZC204 1.3x10-7 7.2x106 C2H2 755 (111) 2.53
* Titanium was contaminated unintentionally for (*) marked samples.









4.3.3 ZrC Films Growth at High Temperature and High Laser Fluence

In this section, ZrC film growth was carefully controlled at higher substrate

temperature and higher laser fluence growth conditions based on previous study. Also

films were deposited on Si (001), Si (111), and sapphire (0001) single crystal substrates

at high temperatures around 750 C to investigate texture, quality, and properties of

grown films. The representative growth conditions and results used for ZrC films growth

are presented in table 4-3. Special care was taken to maintain low water vapor pressures

during depositions by monitoring partial residual gas pressure. Also the Si substrates

were heated up to 900 C and maintained for 20 min under high vacuum to remove the

passivation layer on the substrate before deposition.


Table 4-4. Structure information of ZrC Si, and sapphire and lattice mismatch.
ZrC Silicon Sapphire
Unit Cell Cubic (NaC1) Cubic (Dia.) Rhom. (Hex.)
Space Group Fm3m (225) Fd3m (227) R3 c (167)
5.4309 (JCPDS a = 4.7592, c = 12.992
4.6930 (JCPDS PDF#: 27-1402) (JCPDS PDF#: 43-1484)
Parameter PDF#: 35-0784) [Pcp94] [Pcp94]
meter) [Pcp94] 5.4449 (750 C) a = 4.7828 (750 C)
[Yim74] [Yim74]
39.45 at RT
13.59 at RT
Lattice mismatch at 39.41 at 750 C
) w13.41 at 750 C
(%) with ZrC (calculated with < 110>Zrc,
(a zrc a ZrC<110>>a Sapphire)



Structural information for ZrC and substrates is presented in table 4-4, and the

lattice mismatch is calculated with respect to the substrates. For sapphire, since its basal

plane has same geometry as ZrC (111) plane, lattice mismatch was calculated with

< 1120 > of sapphire parallel to <110> of ZrC. Lattice mismatches to Si and sapphire are









13.4 % and 39.4 %, respectively. For such a large lattice mismatch, definitely elastic

theory can not be applied for epitaxial growth because critical thickness for that big

mismatch will be less than 2 A [Hu91]. However, the resulted grown films on Si and

sapphire were epitaxial. This epigrowth of ZrC could be explained by 'coincidence-site

lattice (CSL)' or translationall symmetries' assumptions [Zur84]. The details were

investigated by X-ray techniques and TEM, and oxidation problem was also considered

in this section. In addition, electrical resistivity of these films was measured.

4.3.3.1 Growth behaviors of ZrC films deposited on Si and sapphire substrate

In fig. 4-9, symmetric theta-2theta XRD spectra acquired from the films (ZC104

samples in table 4-3) grown on Si (001), Si (111), and sapphire (0001) substrates are

presented. For the ZrC films deposited on Si substrates, XRD spectra showed that growth

planes of the films were the same as Si substrate surface planes, that is, ZrC (001) films

were grown on Si (001) substrates, and ZrC (111) films were grown on Si (111)

substrates. Besides, ZrC (111) films were grown on the sapphire (0001) basal plane. For

all the films grown on Si (001) and sapphire substrates, no other peaks were found other

than one peak in theta-2 theta symmetric XRD scan as shown in fig. 4-9, indicating that

the films were highly out-of-plane textured. For the films grown on Si (111) substrate, the

XRD peaks (top three spectra in fig. 4-10) showed two out-of-plane textures of (111) and

(001). Due to high degree of texture, GIXD spectra at incidence angle of 1 (bottom three

spectra in fig. 4-10) showed that the films deposited on Si (111) substrates exhibited

barely visible humps, indicating a very small amount of grains possessing other

crystalline orientations.

To check the texture degree, omega-rocking curves of the either ZrC (111) or ZrC

(002) peaks were recorded for the ZrC films grown under the same conditions (ZC 104









samples) but different substrates. As one can see in fig. 4-11, the films deposited on

sapphire exhibited the highest texture with FWHM (full width at half maximum) of the

omega-rocking curves of only 0.680. The films deposited on Si (001) exhibited a FWHM

of the ZrC (002) peak of 3.370, while those deposited on Si (111) exhibited the rather

large FWHM value of 4.820 for the ZrC peak (111). The Growth conditions of samples

and measured FWHM of omega-rocking results are presented in table 4-3.

The growth conditions of samples were similar except small change of background

gas pressures (2.2x 106 ~ 12x10-6 C2H2). However, the FWHM values of omega-rocking

curves were dramatically changed from 2.530 (ZC202, lowest) to 5.450 (ZC106, highest)

for the films deposited on Si (001), whereas the FWHM values of the films deposited on

sapphire (0001) do not seem to be sensitive to the change of background gas pressure

(ZC 104 and ZC107 in table 4-3). Measured omega-rocking curves for the films deposited

on Si (001) and sapphire (0001) are displayed in fig. 4-12 and fig. 4-13, respectively. The

reason of the rather high FWHM value for ZC204 sample in fig. 4-13 was due to

accidentally discontinued growth.

To clarify the causes which affected crystal quality, lattice parameters for the films

grown on Si (001) were calculated from ZrC (002) reflection of symmetric XRD in fig. 4-

14, because crystal quality could be changed by strain energy in films during growth. As

one can see in fig. 4-15, lattice parameter was linearly changed rather rapidly (Aa =

~1 %) to a small change of background gas pressure (AP = 1.Ox 10-5 C2H2). On the basis

of the lattice parameter (a = 4.6930) of reference powder diffraction (JCPDS PDF#: 35-

0784 [Pcp94]), the closer measured lattice parameter, the better crystal quality in terms of

rocking curve measurements, as it is shown in fig. 4-16.









In-plane texture was observed by phi-scan for {111 in-plane of ZrC (001) film

(ZC202 in table 4-3) as displayed in fig. 4-17 (a). For cubic structures such as ZrC, four

peaks should appear at the x-ray diffractometer configuration of phi-scan as displayed in

fig. 4-17 (b), if in-plane texture exists. Details of the in-plane orientation were

investigated by pole figure measurement of several crystalline orientations for both the

films and the substrates. Typical results for ZrC films deposited on Si (001), Si (111), and

sapphire with their substrates are displayed from fig. 4-18 to fig. 4-20. In these figures,

the orientation relationships between films ((a) of each fig.) and substrates ((b) of each

fig.) are also represented together, by starting the rotation of film and substrate from the

same angular position. As one can see in these figures, the films were exhibiting a rather

good in-plane texture, therefore, being epitaxial.

From (111) pole figures in fig. 4-18, which are obtained from both ZrC film

deposited on Si (001) substrate and Si (001) substrate, very clear four in-plane <111>

pole reflections were only showed. The angular directions of the four <111> poles are the

same as <111> pole directions of their substrates. This fact clearly indicates that (100)zrc

// (100)si and [100]zrc // [100]si orientation relationship between film and substrate, which

follows the concept of coincidence-site lattice (CSL), because when there is a good lattice

match between film and substrate, high quality epitaxial growth is possible. Also the

(100) pole figure of ZrC films grown on Si (111) substrate showed in-plane texture as

indicated red circles in fig. 4-19 (a), but rather contained small fraction of randomly

oriented inhomogeneous grains (displayed as blue contour), while the pole figure of ZrC

films grown on Si (001) substrate showed cube on cube growth.









In case ofZrC films grown on sapphire (0001) substrate, (100) pole figure (fig. 4-

20 (a)) from film and (1 12 6) pole figure (fig. 4-20 (b)) from sapphire substrate were

examined to clarify the orientation relationship between the sapphire substrate and ZrC

film. In the cubic system, three in-plane (100) reflections are formed by an angle of 1200

with each other for a single crystal having (111) surface plane, thus the presence of six

in-plane (100) reflections in fig. 4-20 (a) suggests the existence of two different in-plane

crystal orientations, which are 60 rotated relatively to the other crystal orientation on

same <111> axis as indicated by red and blue circle in fig. 4-20 (a). This epigrowth was

not expected for the interface having big mismatch (-39%) between sapphire <112 0>

direction and ZrC <110> direction. However, this fact could be well explained by

translationall symmetries' on both sides of the interface. We can define two lattices to

match by translational symmetry, instead of comparing the bulk lattice parameters

[Zur84]. As orientation relationship between ZrC and sapphire was presented in a

schematic diagram (fig. 4-21), the superlattice mismatch is less than 0.6% between five

atomic distances of film along the <110> direction and four atomic distances of substrate

along the <0110> direction. When superlattice mismatch is less than 1%, lateral

movement of 0.5 % of every atom on both sides of the interface can accommodated this

superlattice mismatch [Zur84].

For two different crystallographic orientations of ZrC (111) grown on sapphire, it

can be seen that there is only one possible nucleation site for atoms of the first monolayer

at the interface, which is shown as A in fig. 4-22. However for the second layer, there are

two different sites for atomic array, since the binding energy differences for adatoms

occupying B sites or C sites are very small. As long as the island growth model is









considered where there are many nucleation sites on the surface of the substrate during

the initial stage of growth, some nuclei grow in the ABCABC order; and other nuclei

grow in ACBACB order. Thus two twin boundaries can be possibly formed. One is

perpendicular to the surface when these crystallites with different orientations come

together. The boundary is usually semicoherent or noncoherent depending on the habit

plane [Bra66]. Also this fact is responsible for the rather large reflection area in pole

figure in fig. 4-20 (a), indicating in-plane mosaicity. The other one is a parallel twin

boundary, which is parallel to the surface of the film. During perpendicular growth, the

stacking sequence of the (111) atomic planes could be lost by accident. In this case, the

crystallites will grow in twin relation to the previous crystallite, and the boundary is

coherent. Due to relatively small boundary energy in coherent twin boundary, its normal

stacking sequence of (111) plane can be easily lost.

4.3.3.2 TEM analysis of ZrC films grown on Si and sapphire substrate

For further understanding of growth behavior and crystal structure of ZrC on Si and

sapphire, cross-sectional specimens were prepared by FIB (focused ion beam) process.

Before the FIB process, a carbon layer was coated on the surface of films to protect ion

beam damage on film and to avoid electron charging. Just before FIB process, Pt was

coated again to prevent from high energetic ion beam induced damage.

In fig. 4-23, a high resolution bright field image taken from the cross-section of the

film grown on Si (111) substrate (ZC104 sample in table 4-3) is displayed. As it was

expected from GIXD investigation, some random oriented grains were observed as

inhomogeneous particles in continuously well stacked (111) ZrC film matrix rather than

completely random oriented polycrystalline or columnar structure. These random grains

were grown from very initial stage of growth, as they were located right at the interface,









indicating that chemical contribution to interfacial energy is much higher than that of ZrC

grown on Si (001) substrate, assuming structure contribution to the interfacial energy is

the same due to same lattice mismatch as ZrC grown on Si (001). An electron diffraction

pattern of this sample was taken from the region shown in fig. 4-24 (a). Beam direction

was defined by strong diffracted electron beam reflections from fig. 4-24 (b). Therefore,

the cut plane was identified as (11 2) plane, which is perpendicular to surface plane,

meaning strong (111) texture. In addition, diffused electron diffraction pattern identified

as (110) plane also can be seen in fig. 4-24 (b), indicating co-existing (001) texture with

(111) texture other than random orientations.

From ZrC film grown on Si (001) substrate (ZC202 sample in table 4-3) in fig. 4-

25 (a), a selected area diffraction pattern (SADP) was obtained in fig. 4-25 (b) to confirm

the quality of the epitaxial film, and the orientation relationship between the film and Si

substrate. As good epitaxial quality was expected from four clear and small (111)

reflections in fig. 4-18 (a), SADP confirmed ZrC (001) was well aligned on Si (001)

substrate as a single crystal with cube on cube relationship. In a pair of spots, the inner

spots are coming from Si substrate, while the outer spots are from ZrC film. High

resolution image in fig. 4-26 of this film showed clean lattice fringes without any Moire

lattice fringes.

Several cross-sectional TEM pictures taken from ZC104 film grown on sapphire

(0001) were analyzed to verify the previously suggested twin structure used to explain

the pole figure results. Suggested two types of twin structure were found in the regions

marked as 'A' and 'B' from fig. 4-27. In a magnified image (fig. 4-28 (a)) of the region

'A', twin boundary parallel to the surface was observed as mirror image was observed by









drawn lines along atomic arrangements. A twin boundary perpendicular to the surface

was also observed in region B in fig. 4-28 (b). The parallel twin boundary was very

coherent by matching lattices of both sides, while the other showed an incoherent

perpendicular twin boundary with atomic displacement at boundary.

4.3.4 Surface analysis of ZrC films

The density and surface roughness of films were estimated by modeling acquired

XRR spectra using the WingixaTM software from Panalytical, described in section 4.3.2

and chapter 3. Regardless of the substrate type, similar density values of around 6.7

g/cm3 were obtained (ZC208 and ZC210 samples were excluded for estimation due to Ti

contamination), and film surface was atomically flat with rms roughness value of 0.4 ~

0.7 0.2 nm. The measured density values which are identical to tabulated values

[Lid05] suggest that films are compact ZrC. The surface contamination layer density was

around 4 5 0.2 g/cm3, indicating the presence of an oxide or hydroxide compound or

a mixture of both. The secondary electron (SE) images obtained by field emission SEM

showed that surface morphology was very smooth at x 10 K for most deposits, and only at

very high magnification (x50 K), surface roughness was discerned. SE images taken from

ZrC film surfaces on Si and sapphire substrate are displayed, respectively in fig. 4-29

(ZC202 sample) and fig. 4-30 (ZC204 sample). Density and roughness values of as-

grown films are listed in table 4-5. The rms values obtained from AFM height images

from fig. 4-31 to fig. 4-36 confirmed that roughness of surface was less than 1 nm for all

samples. But surface morphology was little different depending on growth direction of

growing films. That is, ZrC (001) surface showed many small cluster looking features as

shown in fig. 4-31 ~ fig. 4-33, whereas ZrC (111) surface was atomically very flat









enough not to be discerned by AFM image as displayed in fig. 4.34 fig. 4.36 under all

experimental growth conditions.

Table 4-5. Surface roughness and density of as-grown films.
Su e ID Pd (orr) Ts Thickness Densit (g/cm3) Roughness
Substrate ID (C) (A) Film Topmost (A)
ZC106 1.0x105 C2H2 750 714 6.7 4.1 6
ZC107 6.7x10-6 C2H2 750 376 6.7 4.3 7
Si (001) ZC202 6.2x10-6 C2H2 775 1034 6.8 5.8 5
ZC208 2.2x10-6 CH4 750 368 6.9 4.3 4
ZC210 9.0x 107 CH4 750 714 6.8 5.6 5
ZC107 6.7x10-6 C2H2 750 333 6.7 4.3 6
Si (111) ZC208 2.2x10-6 CH4 750 381 6.9 4.9 7

Sapphire ZC208 2.2x10-6 CH4 750 383 6.8 5.6 5
(0001) ZC210 9.0x107 CH4 750 656 6.8 5.2 5


As mentioned before in chapter 2, one of problems in ZrC film growth is oxygen

contamination due to high zirconium atoms affinity for oxygen. As an example, when

laser was stopped during ZrC deposition (discontinuous growth), a bright layer (atomic

contrast) was formed due to high oxygen contamination as displayed in a cross-sectional

bright field TEM image (fig. 4-37).

For quantitative x-ray microanalysis of a ZrC film which was grown on Si (001), a

scanning transmission electron microscope (STEM) equipped with an x-ray energy

dispersive spectrometer (XEDS) was used. Sample ZC106 was chosen because it showed

most distinguishable three different layer structures among samples as shown in fig. 4-38.

The result is shown by TEM-EDX analysis in fig. 4-39. From line spectra, carbon and

oxygen elements were almost constant through thickness and detected a little more near

surface. Oxygen content was 9.6 atomic % near Si/ZrC interface (spot 3), a little lower









value of 5.2 % in bulk (spot 4), and again higher value of 13.8 % at surface layer (spot 5).

For carbon content, a higher value (-70%) was detected. But this could not be true

because the characteristic x-ray absorption by carbon counted too low, and due to carbon

supported TEM grid behind sample and protecting carbon coating near surface. However,

the ratio of Zr to carbon was quite constant near interface (spot 3) and in bulk (spot 4),

implying stoichiometric composition.

For detail surface composition analysis, XPS investigations were performed. These

investigations revealed the presence of several C Is lines on the surface of the deposited

ZrC films. Since the binding energies of the XPS peaks are usually referenced to the

binding energy of adventitious carbon, this fact could introduce errors in this case. Thus

pure ZrO2 films deposited on Si by PLD [How02] were analyzed to accurately determine

the position and shape of Zr 3d and O is peaks to avoid this problem.

In fig. 4-40, high resolution XPS scans of the Zr 3d region acquired at 450 and 900

take-off angles from a sample deposited at 600 OC under vacuum (5 x 10-6 Torr) are shown.

The used laser parameters were 8 J/cm2 fluence, and 5 Hz repetition rate. It is apparent

that the surface contains a rather high percentage of Zr-O bonds [Coc98] (peaks denoted

by B located at 184.5 eV and D at 186.9 eV, respectively). From a sample that was

deposited at 600 OC under 1x 10-4 Torr of C2H2, XPS spectra are shown in fig. 4-41 after 5

min sputtering with 4 keV Ar+ ions, which removed the first 4 nm of the outermost layer.

For this sample, there is a minimal angle dependence of the ratio of Zr-O to Zr-C peaks,

indicating a homogeneous bulk composition. The percentages of the Zr atoms bonded to

carbon are shown in table 4-6, which is estimated from the XPS measurements for a

series of samples. Based on XRD data in fig. 4-4, the use of a low C2H2 atmosphere









appeared to have a beneficial effect on crystallinity too. However, the oxygen

concentration was rather high in the surface region of these films.

Table 4-6. The relative percentage of Zr 3d XPS areas corresponding to Zr-C bonds at
different take-off angles.

Deposition As-received Sputtered
Pressure 450 900 450 900

5x10-6 Torr 20.6% 31.5 % 44.9% 45.2%
1 x10-4 Torr of
14.6 % 24.8 % 47.2 % 49.1%
C2H2
7x10-4 Torr of
7x Toof1.0 % 3.8 % 22.4 % 29.0 %
C2H2____

Table 4-7. Resistivity of as-deposited ZrC films
I(aA) V(mV) P1 P2 tl(nm) t2(nm)
(fromtl) (from t2) (TEM) (XRR)
ZC104il 1 30 0.302 1.87E-04 1.83E-04 41 40
((111)Si) 2 30 0.292 1.81E-04 1.76E-04 41 40
3 30 0.282 1.75E-04 1.70E-04 41 40
Z202 1 30 0.115 1.63E-04 1.79E-04 94 103
((100)SI) 2 30 0.113 1.60E-04 1.76E-04 94 103
3 30 0.106 1.51E-04 1.65E-04 94 103
4 30 0.105 1.49E-04 1.63E-04 94 103
Z204a 1 30 0.115 1.55E-04 1.51E-04 89 87
(Sapphire) 2 30 0.116 1.56E-04 1.52E-04 89 87
3 30 0.12 1.61E-04 1.58E-04 89 87
4 100 0.3855 1.56E-04 1.52E-04 89 87
5 100 0.383 1.54E-04 1.51E-04 89 87


AES investigations were performed to analyzed the bulk composition of the film.

The results confirmed that the first 2.0 3.0 nm (surface region) of the ZrC films were

heavily contaminated with oxygen as one can see in fig. 4-42, which was a typical depth

profile acquired from an as-deposited film. However, once the topmost layer was









removed by Ar ion sputtering, the oxygen content dramatically decreased to values below

7 8 % as shown in fig. 4-43 (ZC202). Despite relatively the high levels of oxygen

contamination, the deposited ZrC films were very conductive, and similar values were

obtained in table 4-7. Under best vacuum conditions (Po = 1.3 1.5 x 107 Torr) and using

a low pressure of high purity CH4 during deposition, oxygen concentration was less than

2.5 % within the bulk, as shown in fig. 4-44 (ZC107).

According to residual gas analysis by RGA (residual gas analyzer), just right before

deposition and after introducing C2H2 and CH4 gas, C2H2 gas contained rather high

oxygen concentration when compared with pre-vacuum condition as displayed in fig. 4-

45. So it is thought that oxygen contamination of the films deposited under C2H2

atmosphere is caused by the limited purity of the used gas.

4.4 Summary

In summary, ZrC thin films were deposited on Si and sapphire substrates by the

pulsed laser deposition technique. A combination of high laser fluence and high substrate

temperature (600 700 C) was required to obtain crystalline films on Si (001) substrates.

Under very low water vapor pressures (-10-8 Torr), high substrate temperatures

(-750 C), and high laser fluence (10 J/cm2), epitaxial ZrC films were deposited on single

crystalline substrates. The ZrC films grew along the [001] axis on Si (001), while they

grew along the [111] axis on Si (111) and sapphire (0001). Pole figure measurements

showed that ZrC films exhibited in-plane orientation too, depending on the type of

substrates. Grazing incidence x-ray diffraction investigations evidenced the presence of a

rather small fraction of randomly oriented crystallites. The films mass density was around

the tabulated value of 6.7 g/cm3, while the surface morphology was very smooth, with a

roughness value (rms) of 0.4 0.7 + 0.2 nm.









C2H2 as background gas increased a chance to bond Zr with C, and improved

crystallinity and stoichiometry. Lattice parameter of deposited ZrC film was changed by

small variations of C2H2 pressure. There exists an optimum pressure value for good

crystal quality.

Contamination of mainly Zr-O was found in the surface layers deposited under

C2H2 or residual atmosphere, whereas the oxygen content dramatically decreased to

values below 7 8 % after topmost layers were removed by Ar ion sputtering. Despite

the high levels of oxygen contamination, the deposited ZrC films were very conductive.

Under best vacuum conditions (Po = 1.3 1.5 x 10-7 Torr) and using a low pressure of high

purity CH4 during deposition, oxygen concentration was dramatically reduced to less than

2.5 %.










counts/s


*2Theta


Figure 4-1. GIXD spectra (incidence beam angle, co = 10) ofZrC films deposited under
various conditions.


counts/s


*2Theta


Figure 4-2. Comparison of GIXD spectra obtained from ZrC deposited films at same
deposition conditions (a) without pre-ablated target (red) and (b) with pre-
ablated target (blue).










counts/s


30 40 50 60


70 80
2Theta


Figure 4-3. XRD spectra obtained from ZrC target; vertical lines represent position and
intensity for stoichiometric ZrC, JCPDS PDF# 32-1489.

counts/s


2Theta


Figure 4-4. GIXD spectra obtained from films deposited under residual vacuum and
different C2H2 gas pressures.











cOunwt/


Figure 4-5. XRR spectra of ZrC films deposited for various times at 700 OC and 10 J/cm2
under vacuum.


50


40


30
ac
S20
1-
10


0


0 200 400 600 800
Time (sec)


1000 1200 1400


Figure 4-6. ZrC film deposition rate at Ts = 7000C, Pd = 1.0 x106 Torr, 10 J/cm2, and 5Hz.










counts/s


I 30
30


50


80
*2eTht


Figure 4-7. GIXD spectra of ZrC films deposited for various times at 700 OC and 10
J/cm2 under vacuum; the standard position of diffraction lines from ZrC
(dashed lines) and ZrCo.7 (solid lines) are also shown.

counts/s


I2Theta


Figure 4-8. XRD spectra of ZrC films deposited for various times at 700 OC and 10 J/cm2
under vacuum.


KUe-


- ---


20 min


I .. ..


I


"'


L.O -


14 min


7 nun





































Figure 4-9. XRD spectra of ZrC films deposited at 750 C on various substrates




c ounts/s
1600
10 ZrC on S

1400 (111)

1200-
Thet Theta
1000- (200)


8oo ZC104 (Blue)

600- ZC107 (Red)

400- ZC-1'.'J (Green)
GIXD
2001. ZtCo<332) (220) (3
a00i AL


? Theta

Figure 4-10. XRD and GIXD spectra (incidence beam angle = 1) ofZrC films deposited
on Si (111) substrates.


countM"












Omega-Rocking Curve


ZrC (002) on S1 (001)


counts/s
12000



10000-



8000



6000



4000



200-


10 12 14 16 18 20 22 24 26
Omega

Figure 4-11. Omega-rocking curves of ZrC (111) or (200) peaks recorded from ZC104
films deposited at 750 C on various substrates.

counts/s


Omega-rocking Curve
ZrC (001) on Si (001)


ZC106
ZC-is


26
"Omega


Figure 4-12. Omega-rocking curves of ZrC (002) peaks recorded from the films
deposited on Si (001) substrate at various background gas pressures.


ZrC (111) on





ZrC (111) on S1(111)


15000





10000





5000


ai




















ZC104 ZC107







Z
F// ZC204


" '^^
^^!ie!'^ ^E ^


14


16


18


20


22


"Omega


Figure 4-13. Omega-rocking curves of ZrC (111) peaks recorded from the films
deposited on sapphire (0001) substrate at various different gas pressures


ZC (001) on Si (001)


Lattice Parameter (Angstrom)

ZC208 (black, 4.653)
ZC103 (green, 4.666)
ZC202 (red, 4.685)
ZC104 (blue, 4.701)
ZC106 (pink, 4.705)


'2Theta


Figure 4-14. XRD spectra of (002) reflection from the films deposited on Si (001) and
calculated lattice parameters.


counts/s
70004


10


12


counts/s
25000-




20000




15000




10000




5000



0-


__~11_~__~1~ ~1~1_1 ~1___~___11~ 1_











ZrC (001) on Si (001)


0.472

- 0.471
E
S0.470

" 0.469

w. 0.468

. 0.467

0.466


4.00E-06


5.45 (ZC106) ,-
*
3.37 (ZC104)

WHM=2.531ZC202)
-'


8.00E-06


1.20E-05


C2H2 or CH4



Figure 4-15. Background gas effect on lattice parameter of the films deposited on Si
(001) substrate.


5 -

g4


0 3


91
2 -
==2
==0

|IJ ^


ZrC (001) on Si (001)


0.464 0.465 0.466 0.467
0.464 0.465 0.466 0.467


0.468 0.469 0.470 0.471 0.472


Lattice parameter (nm)



Figure 4-16. The relationship between deposited films on Si (001) and texture degree of
the films measured by omega-rocking curve.


Ref. 0.469 nm F
(35-0784)


4.47 (ZC103)'

4.62 (ZC208)
4.62 (ZC208)


0.465 -
0.00E+00


1.60E-05


ZC106

ZC208
ZC103

ZC104

ZC202


Ref. 0.469 nm
(35-0784)









~~~~____~~~_________(a)


rsaaa-


10000-


saaa-


n-


-5 ... .
-50


IJ


counts/s


Phiscan (ZC202)
of Sis ubstrate


of ZrC grown on Si (001







Lj


Blue: {111} reflection


Red: {111} reflectionr


-1 .. .
-100


S .
0


50 .. .
50


100 150
"Phi

(b)


Rotation axis


Incidence beam


position)


Figure 4-17. Phi-scan of {111} in-plane obtained from (a) ZrC film and Si (001) substrate
of sample ZC202 and (b) phi-scan diffractometer configuration.


Li









SI


1PEj'


I


{111


-- -- (a)




j/t


Figure 4-18. (111) pole figures of(a) ZrC film and (b) Si (001) substrate obtained from
sample ZC202.


- IrJ





















-- I (a) (b)














Figure 4-19. (100) pole figures of(a) ZrC film and (b) Si (111) substrate obtained from
sample ZC104.






















110 1 (a) 11 61













Figure 4-20. Pole figures showing (a) (100) pole figures of ZrC film, and (b) (1 12 6)
pole figure of sapphire (0001) substrate obtained from ZC104.








Interface transitional symmetry


Savbhire(O0001)


041]' "' -
'" rC ('ii) '-, -. -,.-
[10 1]. [010] .. 60
Are-a 141. A


P/0 [ 11 ,, t000 11 -.
[10 ] f i J" .-


U1 .- UIIU
r(L ACL/


Figure 4-21. Projection view for two crystallographic orientations of ZrC grown on
sapphire (0001), and orientation relationship between ZrC film and sapphire
substrate.


(006) Sapphin 4x4x4


(100) Sapphire 2x2x2


(110) Sapphire 2x2x2


irn 006,


(210) Sapphire 2x2x2
U


,006


< 006


Figure 4-22. Possible nucleation site (marked as A) for the first monolayer on sapphire
(0001).


[0011


c










































Figure 4-23. Bright field TEM image obtained from cross-section of ZrC film grown on
silicon (111) substrate; the regions showing inhomogeneous random grains
are marked.











Carbon co


*
m f


'v. .
,"ii)ZC


(Joiif ZC


(DO2) Z,.
Z, C



B = [112] Circled by red

B = [110] Circled by blue


z'I
Nir


131
131


Figure 4-24. TEM (a) bright field image and (b) SADP (selected area electron diffraction
pattern) obtained from cross-section of ZrC film (sample ZC 104) grown on
silicon (111) substrate.


- ""' P
. l (002|rTaC










SSilicon












Carhon
20 nm
20 no


(b)






B=[110]
S *.'\
01 113

ili

b 111




Figure 4-25. TEM (a) bright field image and (b) SADP (selected area electron diffraction
pattern) obtained from cross-section of ZrC film grown on silicon (001)
substrate; diffraction pattern obtained from ZrC film is marked by circles.






67



































Figure 4-26. Bright field TEM image obtained from cross-section of ZrC film grown on
silicon (001) substrate exhibiting clear lattice fringe.









































Figure 4-27. Bright field TEM image obtained from cross-section of ZrC film grown on
sapphire (0001) substrate showing sharp interface; the regions marked as A
and B are magnified in figure 4-28 for observation of twinning.
























































Figure 4-28. High resolution TEM image of (a) 'A' region in fig. 4-26, showing parallel
twin to the surface of film; (b) 'B' region in fig. 4-26, showing perpendicular
twin to the surface of film.













- -. .- .i ^ *s .< r -- .
--. .- .. ... .' .. *.. .. .. .
... ? : 4 t t

X.-* .S*-. *.- .. .., ,. : 4.. V..F

., .. .
-". .. .. I "- .-: -, -.

: C-





'Zi
3 .. . L
*:.' >-, . .','!. .*,'*, -. ,-..'. .-




, b ,:i N, : .- ^ ,
^ ^^-l ss^ ..-.-. '- ""


Figure 4-29. Secondary electron images of ZrC surface grown on Si (001) substrate, at
xlOK (a), and x50K (b).


(b)




















MAIC S11 15.0kV X50.000 100r;7m WI) 15-8min






















































Figure 4-30. Secondary electron images of ZrC surface grown on sapphire (0001)
substrate, at x10K (a), and x50K (b).


MAIC S1 1 HOW X50,0oo ioown wi) 15.9mm























Image Statistics

Img. Z range 8.370 nm
Img. Mean -0.00002 nm
Img. Raw mean -0.00002 nm
Img. Rms (Rq) 0.487 nm
Img. Ra 0.349 nm
Img. Rmax 8.370 nm
Img. Srf. area 25.028 pm'
Img. Prj. Srf. area 25.000 pm'
Box Statistics

Z range 5.764 nm
Mean 0.005 nm
Raw mean 0.005 nm
Rms (Rq) 0.407 nm
Mean roughness (Ra) 0.292 nm
Max height (Rmax) 5.768 nm
Surface area 7.140 pm'
Proj. Surf. area 7.137 pm'


Digital Instruments NanoScope
Scan size 5.000 pm
Scan rate 1.001 Hz
Number of samples 512
Image Data Height
Data scale 50.00 nm
Engage X Pos 145084.8 um
Engage Y Pos 36323.6 um








U, I view angle
j light angle


110405.003


Figure 4-31. AFM height images obtained from the surface of ZC106 sample (ZrC (001)

layer grown on Si (001) substrate)


Roughness Analysis


5.00


2.50












0
5.00 pm


110405.003


~ ~ad~ ~IA~A~E ~til~rmrmaar ~BC~ 3~1Ed ~pr























Image Statistics

Img. Z range 15.219 nm
Img. Mean 0.00005 nm
Img. Raw mean 0.00005 nm
Img. Rms (Rq) 0.547 nm
Img. Ra 0.281 nm
Img. Rmax 15.219 nm
Img. Srf. area 25.027 pm'
Img. Prj. Srf. area 25.000 pm'
Box Statistics

Z range 4.337 nm
Mean -0.004 nm
Raw mean -0.004 nm
Rms (Rq) 0.335 nm
Mean roughness (Ra) 0.253 nm
Max height (Rmax) 4.304 nm
Surface area 9.100 pm'
Proj. Surf. area 9.090 pm2


Digital Instruments NanoScope
Scan size 5.000 pm
Scan rate 1.001 Hz
Number of samples 512
Image Data Height
Data scale 50.00 nm
Engage X Pos 145084.8 um
Engage Y Pos 36323.6 um








L- [I view angle
jL light angle


110405.001


Figure 4-32. AFM height images obtained from the surface of ZC202 sample (ZrC (001)

layer grown on Si (001) substrate)


Roughness Analysis


5.00


2.50












0
5.00 pm


110405.001


~ ~ad~ ~IA~A~E ~til~rmrmaar ~BC~ 3~1Ed ~pr























Image Statistics

Img. Z range 13.472 nm
Img. Mean -0.00006 nm
Img. Raw mean -0.00006 nm
Img. Rms (Rq) 0.646 nm
Img. Ra 0.439 nm
Img. Rmax 13.472 nm
Img. Srf. area 25.035 pm'
Img. Prj. Srf. area 25.000 pm'
Box Statistics

Z range 7.687 nm
Mean -0.014 nm
Raw mean -0.014 nm
Rms (Rq) 0.564 nm
Mean roughness (Ra) 0.425 nm
Max height (Rmax) 7.675 nm
Surface area 8.372 pm'
Proj. Surf. area 8.360 pm'


Digital Instruments NanoScope
Scan size 5.000 pm
Scan rate 1.001 Hz
Number of samples 512
Image Data Height
Data scale 50.00 nm
Engage X Pos 145084.8 um
Engage Y Pos 36323.6 um








L ijI view angle
L light angle


110405.004


Figure 4-33. AFM height images obtained from the surface of ZC208 sample (ZrC (001)

layer grown on Si (001) substrate)


Roughness Analysis


5.00


2.50












0
5.00 pm


110405.004


~ ~ad~ ~IA~A~E ~til~rmrmaar ~BC~ 3~1Ed ~pr























Image Statistics

Img. Z range 12.936 nm
Img. Mean -0.00006 nm
Img. Raw mean -0.00006 nm
Img. Rms (Rq) 0.582 nm
Img. Ra 0.430 nm
Img. Rmax 12.937 nm
Img. Srf. area 25.055 pm'
Img. Prj. Srf. area 25.000 pm'
Box Statistics

Z range 7.874 nm
Mean 0.004 nm
Raw mean 0.004 nm
Rms (Rq) 0.532 nm
Mean roughness (Ra) 0.413 nm
Max height (Rmax) 7.899 nm
Surface area 3.430 pm'
Proj. Surf. area 3.423 pm'


Digital Instruments NanoScope
Scan size 5.000 pm
Scan rate 1.001 Hz
Number of samples 512
Image Data Height
Data scale 100.0 nm
Engage X Pos 145084.8 um
Engage Y Pos 36323.6 um








Lj I] view angle
r light angle

r r6


lllzcl07.000


Figure 4-34. AFM height images obtained from the surface of ZC107 sample (ZrC (111)

layer grown on Si (111) substrate)


Roughness Analysis


5.00


2.50












0
5.00 pm


lllzc107.000


~ ~ad~ ~IA~A~E ~til~rmrmaar ~BC~ 3~1Ed ~pr























Image Statistics

Img. Z range 34.729 nm
Img. Mean -0.0001 nm
Img. Raw mean -0.0001 nm
Img. Rms (Rq) 1.163 nm
Img. Ra 0.554 nm
Img. Rmax 34.729 nm
Img. Srf. area 25.059 pm'
Img. Prj. Srf. area 25.000 pm'
Box Statistics

Z range 8.238 nm
Mean -0.021 nm
Raw mean -0.021 nm
Rms (Rq) 0.563 nm
Mean roughness (Ra) 0.441 nm
Max height (Rmax) 8.266 nm
Surface area 3.950 pm'
Proj. Surf. area 3.942 pm'


Digital Instruments NanoScope
Scan size 5.000 pm
Scan rate 1.001 Hz
Number of samples 512
Image Data Height
Data scale 100.0 nm
Engage X Pos 145084.8 um
Engage Y Pos 36323.6 um








L I] view angle
r light angle

'a-r


sap.zc208


Figure 4-35. AFM height images obtained from the surface of ZC208 sample (ZrC (111)

layer grown on sapphire (0001) substrate)


Roughness Analysis


5.00


2.50












0
5.00 pm


sap.zc208


~ ~ad~ ~IA~A~E ~til~rmrmaar ~BC~ 3~1Ed ~pr
















Roughness Analysis


5.00


2.50












0
5.00 pm


Image Statistics

Img. Z range 27.595 nm
Img. Mean 0.00001 nm
Img. Raw mean 0.00001 nm
Img. Rms (Rq) 0.898 nm
Img. Ra 0.541 nm
Img. Rmax 27.595 nm
Img. Srf. area 25.081 pm'
Img. Prj. Srf. area 25.000 pm'
Box Statistics

Z range
Mean
Raw mean
Rms (Rq)
Mean roughness (Ra)
Max height (Rmax)
surface area
Proj. Surf. area


Digital Instruments NanoScope
Scan size 5.000 pm
Scan rate 1.001 Hz
Number of samples 512
Image Data Height
Data scale 100.0 nm
Engage X Pos 145084.8 um
Engage Y Pos 36323.6 um








L In view angle
rr- light angle

'a-_


sap.zc210


Figure 4-36. AFM height images obtained from the surface of ZC210 sample (ZrC (111)

layer grown on sapphire (0001) substrate)


sap.zc210


~ ~ad~ ~IA~A~E ~til~rmrmaar ~BC~ 3~1Ed ~pr






























Figure 4-37. Cross-sectional bright field TEM image showing an oxidized layer due to
discontinued growth.


Figure 4-38. Cross-sectional bright field TEM image showing distinguishable three layers,
also supporting a model used for XRR analysis.







79



Silicon













Cat bon






Atomic % 1

Z' 13.00

C 333 43.70 619

0 6.14 696 9.61 524

5i 60A9 3634 1069

100nm Electron Image 1

Figure 4-39. Z-contrast image of cross-sectional ZC106 sample for TEM-EDX analysis
by line and point scan.


































Binding Ecy., eV


Binding Eegy. eV


Figure 4-40. High resolution Zr 3d spectra acquired at 450 and 900 take off angles and
their fitting for an as-received sample deposited at 600 OC under vacuum.











CouNS
4200( -
take off angle-450
soo- Zr-C; 47.2%
4 Ar* min sputtering
o -(-4 am removed)

3000 -



i2 io I










190 188 186 14 112 180 17






take off angle-90
ZrC: 49,1%!
4 Ar+ min sputtering
S(4 am removed)
34 -


l ,' \/
SOot -













1 00 IS 16 1 4 12 I 17
190 Hi 186 184 1r 180 17
BInd nl EnQey, eV


Figure 4-41. High resolution Zr 3d spectra acquired at 450 and 900 take off angles and
their fitting for a sample deposited at 600 OC under 1 x 10-4 Torr of C2H2 that
was sputtered-clean by Ar bombardment.








82








80




60
Zr2








20





0 1 2
Time (mins.)


Figure 4-42. AES depth profile of an as-deposited ZrC film.






















Zr2
55.1


01
fri
7.rl 7.5%






Cl
37.4%
50 250 450 650 850 1050 1250 1450 1650 1850 2050
Kinetc Energy (eV)


Figure 4-43. AES survey spectrum of a ZrC film (ZC202) sputtered with a 4 kV Ar ion
beam.




















... -.I, .... i J.. .,i J. .- i..'.1 r '.#'.'-I. .4 ., r,


9.5%


C1
52.6% 0
37.9%
37.9%


o Io 450 650 850 I(n) IM
Kinetic Energy (eV)


14% low ls0 n05D


(b)







o 2.1%
Zr2


Zrl
34.0%


50 I30 *i 610 lUI 10D 1150
Kinetic Energy (eV)


4liO 1t,50 4!50 i50


Figure 4-44. AES survey spectrum of (a) a as-deposited ZrC film under CH4 atmosphere
and (b) the ZrC film sputtered for 1 min with a 4 kV Ar ion beam.



















Vacuum


8 00E-08

7 00E-08

6 00E-08

5 00E-08

4 00E-08

3 00E-08

2 00E-08

1 00E-08

0 00E+00















4 OOE-06

3 50E-06

3 00E-06

2 50E-06

2 00E-06

1 50E-06

1 00E-06

5 00E-07

0 00E+00















3 50E-06


3 00E-06


2 50E-06


2 00E-06


1 50E-06


1 00E-06


5 00E-07


0 00E+00


---A -

0 5 10 15 20 25 30 35 4(

mass/charge







CH4 (c)


5 10 15 20


25 30 35 40


mass/charge


Figure 4-45. Residual gas partial pressure analyzed by RGA before deposition at (a)

vacuum and right after introducing (b) C2H2 and (c) CH4.


H20, 6.73E-8












OH, 1.68E-


H, 4.57E-9

^ ^


0 5 10 15 20 25 30 35 4(

mass/charge







C2H2 (b)


C2H2, 3.63E-6


C2H4, 1.06E-6

S02, 1.38E-6


C2H, 6.81E-7
H20,9.72E-8
C, CH, CH2, CH4, H20, E-8
A- AI


CH4, 3.17E-6


CH3, 2.13E-6












CH2, 2.68E-

C, CH, E/ H20, 1.45E-8


-

-

-

-

-

-

-

-


-

-

-

-

-

-

-


-


-


-


-


-


-


-














CHAPTER 5
MECHANICAL PROPERTIES OF ZIRCONIUM CARBIDE FILMS MEASURED BY
NANOINDENTATION

5.1 Introduction

Zirconium carbide, as mentioned before, is a potentially important material for

many applications because its properties including hardness, melting point, corrosion

resistance, and abrasion resistance are outstanding. Therefore, ZrC films could be used

for MEMS (micro electro-mechanical system) device as wear-resistant or protecting

coating, or in electronic device. As one of examples as application in electronic device,

ZrC deposited on Si can substitute currently using ZrN as a diffusion barrier in Cu-Si

system for metallization application. ZrC also exhibits lower lattice mismatch and

thermal expansion coefficient difference with Si compared to ZrN. In addition,

epitaxially grown crystal ZrC could prevent copper diffusion into Si substrate along the

localized defects in the barrier films. In consequence, the failure of Cu/ZrC/Si films could

be avoided by retarding the formation of Cu3Si. Also tip failure by ion bombardment in

currently using Si and Mo field emitters is one of problems in stabilizing field emission

for a long time. The impact of ion bombardment onto surface creates sharp

nanoprotrusions which are resulting in tip failure caused by high local field [Cha99,

ChaO 1]. From this point of view, ZrC is promising material for a field emitter tip

application because of its low work function as well as high resistance to ion

bombardment. Therefore, it is important to investigate the mechanical properties of

deposited thin film ZrC.









One of the most challenging for task applications of thin films is to evaluate their

mechanical properties. The nanoindentation has been used as a very useful technique to

quantify thin-film mechanical properties, such as Young's modulus and hardness, within

a submicron scale [Cac99, Kuc00, Now99, Yu98]. In the standard indentation procedure,

a Berkovich pyramid-shaped diamond tip using aN loads produces submicrometer

indentation penetration depths, and allows the properties of thin films to be measured

without removing the substrate. Apparently, this technique has become an important tool

for material characterizations because one can easily obtain the reduced modulus and

hardness values of thin films under different loads from their load-displacement (P-h)

curves [Tsu99].

In this study, ZrC films grown on Si (100), Si (111), and sapphire (0001) substrates

using pulsed laser deposition were investigated using nanoindentation measurement. The

Young's modulus and hardness were calculated from loading-displacement (P-h) curves.

5.2 Experiment

The film depositions were conducted in a stainless steel vacuum chamber using a

KrF excimer laser. The laser parameters used were 10 J/cm2 fluence and 10 Hz repetition

rate. ZrC films were deposited on and Si (001), Si (111), and sapphire (0001) substrates.

Depositions were performed under a low C2H2 or CH4 atmosphere. Details for deposition

procedures are described in chapter 4.

Automated indention pattern grids with indention spacing of 5 [m were

programmed to run on original substrates and coated substrates with a Hisitron

Tribolndenter. Tests were run in displacement control with a total displacement range of

15 to 70 nm. The samples thicknesses used for nanoindentation test were around 70-100

nm. The sample surfaces were carefully cleaned before the measurements. A Berkovich