• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Ion-beaming channeling in non-centrosymmetric crystals

Dissertation
Author: Dharshana Nayanajith Wijesundera
Abstract:
The focus of this study is to investigate the nature of energetic ion channeling in non-centrosymmetric crystal structures. The objective is to understand the effects of the anisotropic atomic placement in such structures on ion channeling. The specific case of study to achieve this goal is helium ion channeling in wurtzite structured ZnO. The method of geometrically adjusted channeling mapping (GACM) developed and verified under this work, which is a scheme for using ion channeling maps that are corrected for geometric mapping distortions, is used as the main analytical tool. Helium ion axial channeling GACM-based analysis reveals that the anisotropy along the c-axis of wurtzite structure affects the behavior of channeled ions along certain axes and planes, which is observable as asymmetries in channeling maps. It is confirmed that the reverse case, where asymmetries in channeling maps are used to track the orientation of a crystal or a crystalline thin film can be used as a method to determine the surface polarity of an unknown c-plane surface of a wurtzite structured material. Further investigation of much simpler, but analogous c-planar channeling case of ZnO reveals that asymmetries in channeling maps, which correspond to anisotropy in the wurtzite structure is prominent at the ion incidence angles that correspond to ion-atom close encounter probability exceeding that of non-channeling incidence (i.e., in shoulder region of channeling). Simulation-based investigation into these c-planar channeling observations reveals that, shoulder regions are a result of interaction of regions of enhanced ion flux density generated by the focusing effect of channeling with lattice atoms, and that the asymmetry in shoulders occurs because of the relative difference of ion flux density at Zn (or O) atomic planes at the two shoulder angles, caused by the flux masking effect of asymmetrically placed O (or Zn) planes.

Contents 1 Introduction 1 1.1 Objective of the Study 2 1.2 Activities 3 1.3 Outline of the Dissertation 4 2 Background 7 2.1 Ion-beam Channeling 7 2.1.1 Introduction 7 2.1.2 Theoretical Background of Channeling 10 2.2 Stopping Power 21 2.3 Rutherford Backscattering Spectroscopy 23 3 Formulation of Experimental Methods 26 3.1 Azimuthal Dependence and Accurate Determination of the Half-angle in RBS-channeling 26 3.1.1 Introduction 27 3.1.2 Experimental Details 29 3.1.3 Results 31 3.1.4 Summary 40 3.2 Geometric Effects and Geometrically Adjusted Channeling Maps 42 VI

3.2.1 Introduction 42 3.2.2 Theoretical Background 44 3.2.3 Experimental Evaluation and Discussion 48 4 Simulation of Ion Channeling 54 4.1 Simulation Method, Assumptions, Physics, and Algorithm 54 4.1.1 Basic Assumptions 55 4.1.2 Physics 56 4.1.3 Core Algorithm 58 4.2 Crystallographic Implementation in the Code 60 4.2.1 Transformation of Crystallographic Coordinates 62 4.2.2 Calculation of Potential 63 4.3 Examples 64 4.3.1 Motion of He Ions Between two Rows of Si Atoms 64 4.3.2 He Ion Channeling in Si 66 5 Channeling in Wurtzite Structured ZnO 73 5.1 Zinc Oxide 73 5.1.1 Structural Details 73 5.1.2 Crystallographic Polarity 77 5.2 Simulated He Ion Channeling in ZnO near Principal Axes 78 5.3 Polarity Determination of ZnO by Geometrically Adjusted Channeling Maps 81 5.3.1 Introduction 82 5.3.2 Simulation 86 5.3.3 Experimental Results 90 5.4 Interpretation of Observations 95 5.4.1 Planar Channeling in c-Planes 95 vii

5.4.2 A Detailed Look Through Ion Flux Distribution Plots 100 5.5 Summary of Channeling in ZnO 112 6 Conclusion 114 6.1 Overview 114 6.2 Proposed Future Work 116 Bibliography 120 Appendices A Author's Publications 125 Vl l l

List of Figures 2.1 Scattering of an ion of atomic number Z\ by two target atoms of atomic number Z2, aligned along the incoming direction of the ion. The first target atom shadows the second, reducing the probability of close encounter of the ion with the second atom. For the case of an incoming beam of ions, the first atom forms an ion flux void shadow cone behind it 12 2.2 Close encounter probability of individual atoms for 1.0 MeV He ions in cident along the (001) directions of Si and W at room temperature. The sum of the individual probabilities for Si is 3.3 atoms/row while for W, it is 1.2 atoms/row 13 2.3 A schematic comparing the close encounter probability with respect to depth for non-channeling (random) and channeling ion incidence, for the ideal case. Depth scale is reversed to make it analogous to RBS spectra. . 14 2.4 Comparison of experimental RBS spectra for non-channeling (random) and channeling ion incidence. This example is for tungsten (001) probed via 2.0 MeV He ions. Note that the non-channeling spectrum (denoted by R) is divided by 10 for convenient plotting purposes 15 2.5 Two-dimensional illustration of beam-crystal geometry in producing an angular scan. 6 is scanned starting from a negative value, through zero (channeling alignment) to a positive value 18 2.6 An example of a channeling angular scan with actual data for 480 keV protons on W (001) 19 2.7 Detector geometry, d is the diameter of the incident beam, w the width of the detector aperture and LQ the distance between sample and the detector aperture, a is the angle of incidence and j3 is the exit angle, n — (a + j3) is the scattering angle 23 IX

2.8 A Sample RBS spectrum. This is a simulated spectrum is based on 2 MeV 4He ion scattering off a SiC>2 target in a non-channeling (random) direction. Detector placement is at a 165° scattering angle 25 3.1 Schematic drawing of a three-axis goniometer and rotation angles 30 3.2 RBS full scan map of Si (100). The top grayscale indicates the intensity of the RBS yield 32 3.3 RBS yield versus tilt angle about Si (100), independent of azimuthal an gle, obtained from the RBS full scan map 33 3.4 RBS angular scans of Si (100) for azimuthal angles £ = 16.5° to 28.5°. . 34 3.5 RBS angular scans of Si (100) for azimuthal angles £ = 16.5° to 28.5° obtained by cubic spline interpolation of the RBS full scan map 35 3.6 Azimuthally averaged RBS channeling angular scan of Si (100) obtained from the interpolated RBS full scan map. RBS yield is plotted as a func tion of tilt, 6 = ±v/72 + TJ2 36 3.7 RBS random yields and channeling half-angles of Si (100) plotted against the azimuthal angles £ = 16.5° to 28.5° using direct experimental data. The random yield is obtained directly from experimental data by averaging RBS yields in angular scans at large tilt angles 37 3.8 RBS random yields and channeling half-angles of Si (100) plotted against the azimuthal angles C, = 16.5° to 28.5° using interpolated data extracted from the RBS full scan map. The random yield is obtained from the RBS full scan map averaging RBS yields in angular scans at large tilt angles. . 38 3.9 RBS channeling half-angles of Si (100) plotted against the azimuthal an gles C, = 16.5° to 28.5° using direct experimental data. Channeling half- angles are determined using the universal random level 39 3.10 RBS random yields and channeling half-angles of Si (100) plotted against the azimuthal angles £ = 16.5° to 28.5° using interpolated data extracted from the RBS full scan map. Channeling half-angles are determined using the universal random level 39 3.11 Three-axis goniometer configurations and sample (cuboid at the origin of moving coordinate system x,y,z) orientation with respect to fixed coordi nate system X,Y,Z 45 x

3.12 Experimentally obtained Si (111) LLCM with (111) base axis using a (111) polished Si wafer, which plots the backscattering yield versus go niometer angles Ga and % 49 3.13 Experimentally obtained Si (111) LLCM with (100) base axis using a (100) polished Si wafer, which plots the backscattering yield versus go niometer angles Ba and % 49 3.14 Experimentally obtained Si (111) LLCM with (111) base axis using a (100) polished Si wafer, which plots the backscattering yield versus an gles 6x and Gy of the fixed laboratory coordinate system 50 3.15 Experimentally obtained Si (110) LLCM with (110) base axis using a (110) polished Si wafer, which plots the backscattering yield versus go niometer angles 6a and % 52 3.16 Experimentally obtained Si (110) LLCM with (100) base axis using a (100) polished Si wafer, which plots the backscattering yield versus go niometer angles 9a and % 52 3.17 Experimentally obtained Si (110) LLCM with (110) base axis using a (100) polished Si wafer, which plots the backscattering yield versus an gles 6x and 9y of the fixed laboratory coordinate system 53 4.1 Simulation of the path of 2 MeV 4He for a hypothetical case of ions mov ing between two frozen (thermal vibrations are not considered) rows of Si atoms placed 5.43 angstroms apart. The separation between Si atoms in each row is 5.43 angstroms, which corresponds to the lattice constant of Si at room temperature. Each ion is projected parallel to the atomic rows with different impact parameters. The distance traversed along the projected direction is indicated by z. The dependence of wavelength of specular reflections of the ions off the atomic rows is visible. The simu lation neglects electronic stopping power and the total energy loss of a 2 MeV ion in real Si at a depth of 4500 angstroms is roughly 100 keV, which is about 5% 65 4.2 A comparison of simulated channeling angular scans of 2 MeV 4He ions in Si (100), at two different depth ranges, 0 - 60 nm and 60 - 120 nm. Pr is proportional to the close encounter probability. The near surface (0 - 60 nm) channeling angular scan shows higher shoulder compared to the 60 - 120 nm channeling angular scan. Isotropic thermal vibration amplitude of 0.076 angstroms is considered for all lattice atoms 67 XI

4.3 Simulated and RBS experimental channeling angular scans of2MeV4He ions in Si (100), in a depth widow of 0-100 nm. The scan is made across a plane making 22.5° with respect to (110). Scans are normalized to their own average, and not to the random yield. Isotropic thermal vibration amplitude of 0.076 angstroms is considered for all lattice atoms 68 4.4 A simulated Si (001) azimuthal projection channeling map. The map is produced by simulation of 2 MeV 4He scattering within a 20-80 nm depth window. The simulation is carried out only for the azimuthal angles from 0° to 45°, and the rest of the map is constructed considering symmetry in channeling. Isotropic thermal vibration amplitude of 0.086 angstroms is considered. The map is normalized to the azimuthal average of close encounter probability at 2° tilt. All angles are in degrees. The angular resolution for simulation is 0.1° in tilt angle and 1° in azimuthal angle. Moving average smoothing has been applied to the map 69 4.5 A simulated Si (110) azimuthal projection channeling map. The map is produced by simulation of 2 MeV 4He scattering within a 20-80 nm depth window. The simulation is carried out only for the azimuthal angles from 0° to 90°, and the rest of the map is constructed considering symmetry in channeling. Isotropic thermal vibration amplitude of 0.086 angstroms is considered. The map is normalized to the azimuthal average of close encounter probability at 2° tilt. All angles are in degrees. The angular resolution for simulation is 0.1° in tilt angle and 1° in azimuthal angle. Moving average smoothing has been applied to the map 70 4.6 A simulated Si (111) azimuthal projection channeling map. The map is produced by simulation of 2 MeV 4He scattering within a 20-80 nm depth window. The simulation is carried out only for the azimuthal angles from 0° to 60°, and the rest of the map is constructed considering symmetry in channeling. Isotropic thermal vibration amplitude of 0.086 angstroms is considered. The map is normalized to the azimuthal average of close encounter probability at 2° tilt. All angles are in degrees. The angular resolution for simulation is 0.1° in tilt angle and 1° in azimuthal angle. Moving average smoothing has been applied to the map 71 5.1 A ball-and-stick diagram of distorted tetragonal coordination of Zn-0 bonds in wurtzite structured ZnO at 293K. Bond lengths are in angstroms and bond angles are in degrees 75 5.2 A ball-and-stick model of wurtzite structured ZnO viewed toward the c-axis. 75 xn

5.3 A ball-and-stick model of wurtzite structured ZnO viewed along the m- plane normal, a (horizontal), b (vertical) and c indicate the standard lattice vectors. Vector p indicates (1011); —p indicates (1011); p' indicates (TO 11);- p' indicates (1 Oi l ) and q indicates (1230) 76 5.4 A ball-and-stick model of wurtzite structured ZnO viewed toward the (1011) axis (indicated by p) 76 5.5 A simulated Zn yield ZnO (0001) azimuthal projection channeling map. The map is produced by simulation of 2 MeV 4He scattering within a 20- 80 nm depth window. The simulation has been carried out only for the azimuthal angles from 0° to 120°, and the rest of the map is constructed considering 3-fold rotational symmetry about (OOOl). Isotropic thermal vibration amplitude of 0.086 angstroms is considered for both Zn and O. The map is normalized to the azimuthal average of close encounter prob ability at 2° tilt. All angles are in degrees. The angular resolution for simulation is 0.1° in tilt angle and 1° in azimuthal angle. Moving average smoothing has been applied to the map 79 5.6 A simulated 2 MeV 4He ion, Zn yield ZnO (1000) azimuthal projection channeling map at different depth windows. The simulation has been car ried out for azimuthal angles in the full range from 0° to 360°. Isotropic thermal vibration amplitude of 0.086 angstroms is considered for both Zn and O. The map is normalized to the azimuthal average of close encounter probability at 2° tilt. All angles are in degrees. The angular resolution for simulation is 0.1° in tilt angle and 2° in azimuthal angle. Moving average smoothing has been applied to the map 80 5.7 Simulated, 1.5 MeV 4He beam, Zn yield channeling maps of [1011] (Zn polar surface) and [1011] (O polar surface) axes, in a 0-40 nm depth win dow. The asymmetry about the central vertical axis of the Zn polar surface map is inverted in the O polar surface map. The direction of asymmetry is the signature of polarity. Yield scale is normalized to maximum yield of each map. All angles are in degrees 87 5.8 Guidelines for alignment and plotting of ZnO polarity determining chan neling maps, a) of Zn polar surface, and b) of O polar surface 88 xni

5.9 Experimental, 1.5 MeV 4He beam, Zn yield geometrically adjusted chan neling maps of [±1 0 =pl 1] (Zn polar surface) and [±1 0 ^1 -1] (O polar surface) axes, in a 0-40 nm depth window. The asymmetry about the cen tral vertical axis of the Zn polar surface map is inverted in the O polar sur face map. The direction of asymmetry is the signature of polarity. Yield scale is normalized to the maximum yield of each map. All angles are in degrees 91 5.10 Experimentally obtained, 1.5 MeV 4He+ beam, Zn yield ZnO [1011] or [1011] geometrically adjusted channeling maps on Zn polar c-plane sur face, with respect to increasing depth. Positive direction of horizontal axis indicates the direction of increasing tilt from (0001). Yield scale is relative. 93 5.11 Experimentally obtained, 1.5 MeV 4He+ beam, Zn yield ZnO [1011] or [1011] geometrically adjusted channeling maps on O polar c-plane sur face, with respect to increasing depth. Positive direction of horizontal axis indicates the direction of increasing tilt from (0001). Yield scale is relative. 94 5.12 A ball-and-stick diagram of wurtzite structured ZnO viewed parallel to the (0001) planes (c-planes). This is the beam-crystal orientation to obtain a c-planar channeling angular scan. % is the beam incidence angle with respect to the c-planes 96 5.13 A comparison of experimentally obtained and simulated, 1.5 MeV 4He Zn yield (0001) RBS-C planar angular scans, in two depth windows. Asym metry in the shoulders and their reduction with depth is visible. Plots are normalized to RBS spectra at 7° tilt. Tilt angle (incidence angle) is in degrees 97 5.14 Experimentally obtained, 1.5 MeV 4He+ Zn yield (0001) planar angular scans, with respect to increasing depth. Plots are normalized to random RBS yield at 7° incidence angle 98 5.15 Experimentally obtained, 1.5 MeV 4He+ Zn yield ZnO (0001) planar angular scans, with respect to increasing depth. Plots are normalized to random RBS yield at 7° incidence angle 99 5.16 Ion flux distribution in ZnO (0001) planar channel for channeling of 1.5 MeV 4He. Ion incidence is parallel to (0001) planes 101 5.17 Ion flux distribution in ZnO (0001) planar channel for channeling of 1.5 MeV 4He. Angle of incidence +0.42° (i.e., 90°-0.42° from +c-axis). . . 102 xiv

5.18 Ion flux distribution in ZnO (0001) planar channel for channeling of 1.5 MeV4He. Angle of incidence -0.42° (i.e., 90°+0.42° from +c-axis). . . 103 5.19 Ion flux distribution in ZnO (0001) planar channel for channeling of 1.5 MeV4Ue. Angle of incidence -3.00° (i.e., 90°+3.00° from +c-axis). . . 106 5.20 Ion flux distribution in ZnO (0001) planar channel for channeling of 1.5 MeV 4He. Angle of incidence +3.00° (i.e., 90°-3.00° from +c-axis). . . 107 5.21 Mean flux densities within ±0.05 A of an O plane in ZnO (0001) planar channel for channeling of 1.5 MeV 4He, for angles of incidence of ± 0.42°. 109 5.22 Ion flux distribution in ZnO (0001) planar channel for channeling of 1.5 MeV 4He. Angle of incidence: -0.7° to +0.7° in 0.1° steps. Y (lateral spacing) and X (depth) scales are in angstroms. Flux density is in loga rithmic scale. Scales in topmost figure are common to all figures here. . . I l l xv

List of Tables 4.1 Simulation parameter description, definitions, and abbreviations 61 4.2 Simulation parameter description used in simulating MeV He ion chan neling in Si 66 xvi

List of Acronyms 3D - three-dimensional BCA - binary collision approximation GACM - geometrically adjusted channeling mapping GAP - geometric adjustment procedure IBA - ion-beam analysis IOC - initial orientation configuration LLCM - longitude-latitude channeling map LSS - Lindhard-Scharff-Schiott MeV - mega electron volt NRA - nuclear reaction analysis OCV - orientation configuration vector PIXE - particle induced X-ray emission RBS - Rutherford backscattering spectrometry RBS-C - Rutherford backscattering - channeling RCE - resonant coherent excitation rms - root mean square SPCM - stereographic projection channeling map TEM - transmission electron microscopy xvn

Chapter 1 Introduction In the field of energetic particle interaction with solids, channeling is defined as the in fluence of a crystal lattice on the trajectories of energetic particles [1]. As is implied, ion channeling or ion-beam channeling refers to the case in which the energetic particles are a beam of ions. When a beam of energetic ions is incident on a solid, a certain fraction of ions un dergoes large angle coulomb scattering by the atoms of the solid. However, when the direction of incidence of the beam of ions, relative to a crystalline solid is made parallel to a low index axis or a plane of a crystalline solid, a dramatic drop in large angle ion scatter ing yield can be observed. This occurs because, when a beam of energetic ions is incident parallel to a low index crystalline axis or a plane of a crystal, surface atoms of the crystal mask the subsurface atoms that are aligned behind them, reducing the chance of ions to interact with the subsurface atoms. Furthermore, aligned rows or planes of atoms cause ions to go through a series of small angle, correlated scattering events, which further steers 1

ions away from lattice atoms into inter-atomic spaces. These events together dramatically reduce the ion-atom close encounter probability compared to a case of ion incidence in a direction that does not fall along a low index axis or plane. This is the phenomenon of ion-beam channeling (or ion channeling). Ion channeling in a crystalline solid can be experimentally observed via any method that is sensitive to ion-atom close encounter probability. Detection methods that can mea sure ion-atom close encounter probability such as Rutherford backscattering spectrometry (RBS), nuclear reaction analysis (NRA) and inner shell ionization particle induced X-ray emission (PIXE) can be utilized to observe channeling phenomena. Since channeling is an interaction process between energetic ions and a crystalline solid, the process can be strategically used to analyze and extract useful information on crystalline solids. Based on this idea, combined with RBS, NRA, and PIXE, channeling is well evolved into a powerful real-space analytical tool for crystalline material analysis. Material analysis by ion channeling is a major component of the field of ion-beam analysis (IBA)[1,2]. 1.1 Objective of the Study Historically, development of channeling as a material-analyzing tool performed optimally in the early days after its discovery. Rapid advancement of easy-to-use analytical tools based on other physical principles resulted in their overtaking channeling IBA. Relatively small group of participating researchers has been one reason for that. However, channel ing still has its own unique advantages, and needs investigation from a new point of view, 2

fundamentally, to boost it to match analytical requirements in more sophisticated applica tions. At the same time, the relevant technology, methodology and instrumentation should be improved to match those of easy-to-use commercial analytical tools. The theme behind this study is to improve channeling as an analytical tool in terms of methodology and instrumentation, and to apply the improved techniques to an important physical problem, thereby contributing to physics as well as the field of IBA. Staying within that theme, the specific case of the study described in this dissertation is a probe into the nature of ion-beam channeling in zinc oxide using improved IBA techniques. As the crystalline material, the choice of ZnO was made because its wurtzite structured phase is non-centrosymmetric. The nature of channeling in such a non-centrosymmetric binary crystal is of fundamental interest. Furthermore, ZnO is a promising material for optical devices and transparent conductors [3]; and semiconductors with high application potential such as InN and GaN share the same structure with wurtzite structured ZnO. Therefore, understanding of ion channeling behavior in such materials is important; for example, for optimizing ion implantation doping of them. In terms of the contribution to the field of IBA, the improved experimental methods formulated and used in this study are not limited to this particular study. They are designed in a general form, applicable to a variety of problems in the field of IBA. 1.2 Activities The activities carried out to achieve the objective of this study include, 3

1. formulation and experimental verification of the method of geometrically adjusted channeling mapping (GACM), a channeling-based analytical method for crystalline material and crystalline thin film analysis; 2. development of a general purpose code for simulation of ion channeling; 3. helium ion-beam channeling crystallographic analysis of wurtzite structured ZnO via improved, GACM ion-beam analysis methods; 4. formulation of a method for polarity determination of wurtzite structured ZnO or similarly structured materials based on GACM; 5. identification of special features in channeling arising from the structure of ZnO and interpretation of the physics behind them via further experimental observations and simulations. 1.3 Outline of the Dissertation This dissertation consists of six chapters converging to the same focus but designed as independent articles. Chapter 2 covers the basics of ion-beam channeling. An overview of theoretical concepts behind channeling is given as relevant to this study. An introduction is given to Rutherford backscattering spectroscopy, emphasizing its use in experimental studies on channeling as relevant to this study. Chapter 3 covers special experimental methods designed to support this study and to contribute to the field of ion-beam analysis in general. The first half of the chapter dis cusses ambiguities that can arise in conventional ion-beam analysis by channeling. The 4

experimental difficulties and uncertainties that can arise in data when using standard ele ments in channeling such as channeling angular scans are critically discussed. Solutions to such issues as generally applicable to ion-beam analysis are presented. The second half of the chapter discusses the ion-beam analysis method of geometrically adjusted chan neling mapping formulated in this study. This method, which is based on ion channeling maps, is suggested here as a source of physical data in channeling ion-beam analysis. Geometric distortion issues arising when using channeling maps and methods for correct ing them in real time through goniometric instrumentation are discussed, based on work done during this study. Experimental data confirming the usability of the methods are also presented. Chapter 4 covers a description of the simulation code developed for the purpose of predicting and interpreting data. Basic physics behind the implementation of the code that is based on the Vineyard model is discussed in this chapter. Sample simulations of Si channeling maps and channeling angular scans are also presented there to demonstrate the usage of the code. Chapter 5 is the primary focus of this dissertation, which discusses ion-beam channel ing in wurtzite structured ZnO. Experimental and simulation results are presented there highlighting special features arising due to the non-centrosymmetric polar structure of ZnO. Chapter 5 is structured so that it is centered on a method developed under this study for polarity determination of c-plane, wurtzite structured ZnO. The chapter begins with an introduction to the structural details, crystallographic polarity of wurtzite structured ZnO and the importance of surface polarity determination in thin film growth. Then a detailed 5

description of the surface polarity determination method is given along with the confir mation of the method by simulation and experimental results. Next, physical principles behind the polarity determination method are discussed with an explanation of special features in the data, considering a much simpler but analogous planar channeling case of ZnO. Explanations are made by examining simulated ion flux distribution plots for each case. Further, an explanation of the origin of shoulders of channeling angular scans is also given based on those flux distribution maps. Chapter 6 contains a short concluding overview of the achievements of this study and how the introduced methods and the gained understanding can be projected to future work. 6

Chapter 2 Background The purpose of this chapter is to cover the concepts of channeling, which are referred to in this dissertation. 2.1 Ion-beam Channeling 2.1.1 Introduction The influence of a crystal lattice on an energetic ion trajectory, or the channeling effect was first suggested in 1912 by Stark [4, 1]. However, his idea was not pursued at that time possibly because it was overweighed by the rapidly advancing field of X-ray diffraction. After a considerably long pause, the possibility of channeling effect was reconsidered following the observation of strong orientation dependance of sputtering yield of single crystalline materials and abnormally long stopping range of heavy ions in polycrystalline 7

materials [4]. In 1963, Robinson and Oen, performing computer simulations of stopping of Cu ions in various monocrystalline targets found abnormal ranges of ion trajectories when incident along principal axial directions, which confirmed the channeling effect [4]. Subsequent experiments on energetic ion interactions with monocrystalline solids further endorsed the existence of the process of channeling. The real discovery of channeling can be stated as a gradual process in the period of 1960 through 1965 [1,4]. Theoretical basis of channeling was laid by Nelson and Thompson who showed that channeling could result from a series of glancing angle collisions, and they derived an effective transverse potential which governs the trajectories of channeled particles. Lehmann and Leibfried, and Lindhard and Erginsoy made the theory evolve further by introducing the concept of continuum potential. In their work, they were able to obtain expressions for experimentally observable quantities in channeling, such as the angular width of channeling [4]. The mentioned basic theory of channeling can be found in the publication by Lindhard in 1965 [5]. Concept of channeling has found applications in a wide range of fields. Utilization in material science and semiconductor industry as a real space analytical probe is the most prominent of them. In ion-beam modification of materials, for example, in ion implan tation doping of semiconductors, understanding of channeling effects is crucial because they can unexpectedly extend the range of implanted dopant ions, which negatively im pacts fabrication of shallow junctions [6]. Channeling of high energy particles has been proposed for use in accelerators for beam collimation and guiding [7, 8, 9, 10]. Possibility of producing spin polarized particle beams via channeling has also been identified [11]. In addition, resonant coherent excitation (RCE) of relativistic heavy ions when channeling 8

Full document contains 145 pages
Abstract: The focus of this study is to investigate the nature of energetic ion channeling in non-centrosymmetric crystal structures. The objective is to understand the effects of the anisotropic atomic placement in such structures on ion channeling. The specific case of study to achieve this goal is helium ion channeling in wurtzite structured ZnO. The method of geometrically adjusted channeling mapping (GACM) developed and verified under this work, which is a scheme for using ion channeling maps that are corrected for geometric mapping distortions, is used as the main analytical tool. Helium ion axial channeling GACM-based analysis reveals that the anisotropy along the c-axis of wurtzite structure affects the behavior of channeled ions along certain axes and planes, which is observable as asymmetries in channeling maps. It is confirmed that the reverse case, where asymmetries in channeling maps are used to track the orientation of a crystal or a crystalline thin film can be used as a method to determine the surface polarity of an unknown c-plane surface of a wurtzite structured material. Further investigation of much simpler, but analogous c-planar channeling case of ZnO reveals that asymmetries in channeling maps, which correspond to anisotropy in the wurtzite structure is prominent at the ion incidence angles that correspond to ion-atom close encounter probability exceeding that of non-channeling incidence (i.e., in shoulder region of channeling). Simulation-based investigation into these c-planar channeling observations reveals that, shoulder regions are a result of interaction of regions of enhanced ion flux density generated by the focusing effect of channeling with lattice atoms, and that the asymmetry in shoulders occurs because of the relative difference of ion flux density at Zn (or O) atomic planes at the two shoulder angles, caused by the flux masking effect of asymmetrically placed O (or Zn) planes.