The microsecond pulsed glow discharge : developments in time-of-flight mass spectrometry and atomic emission spectrometry

MISSING IMAGE

Material Information

Title:
The microsecond pulsed glow discharge : developments in time-of-flight mass spectrometry and atomic emission spectrometry
Physical Description:
x, 242 leaves : ill. ; 29 cm.
Language:
English
Creator:
Oxley, Eric, 1976-
Publication Date:

Subjects

Subjects / Keywords:
Glow discharges   ( lcsh )
Mass spectrometry   ( lcsh )
Atomic emission spectroscopy   ( lcsh )
Chemistry thesis, Ph.D   ( lcsh )
Dissertations, Academic -- Chemistry -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references (leaves 233-241).
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Eric Oxley.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 029223182
oclc - 51024062
System ID:
AA00022672:00001

Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
        Page iv
    Table of Contents
        Page v
        Page vi
        Page vii
        Page viii
    Abstract
        Page ix
        Page x
    Chapter 1. Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Chapter 2. The glow discharge
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
    Chapter 3. Glow discharge modes
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
    Chapter 4. Quantitative depth analysis using microsecond pulsed glow discharge atomic emission spectrometry
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
    Chapter 5. Introduction to the Grimm-type glow discharge axial time-of-flight mass spectrometer system
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
    Chapter 6. Ion transport diagnostics in a microsecond pulsed Grimm-type glow discharge time-of-flight mass spectrometer
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
    Chapter 7. Direct pin sample analysis using a conventional Grimm-type glow discharge source and time-of-flight mass spectrometer
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
    Chapter 8. Thin film analysis with a microsecond pulsed glow discharge time-of-flight mass spectrometer
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
    Chapter 9. Concluding remarks
        Page 229
        Page 230
        Page 231
        Page 232
    References
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
    Biographical sketch
        Page 242
        Page 243
        Page 244
Full Text












THE MICROSECOND PULSED GLOW DISCHARGE:
DEVELOPMENTS IN TIME-OF-FLIGHT MASS SPECTROMETRY
AND ATOMIC EMISSION SPECTROMETRY













By

ERIC OXLEY












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 2002





























Dedicated to my parents, Betty & Larry, and Paige, my love.














ACKNOWLEDGMENTS

Many individuals have contributed to the work in this dissertation. They have been not only scientific colleagues, but also good friends. First, I wish to thank my advisor, Dr. W.W. Harrison, for his continued guidance and support. His compassion and leadership have helped me grow as a scientist and as a person. Fortunately, his cynical sense of humor has made these years of growth very enjoyable.

I am also grateful that I could work closely with Dr. Jim Winefordner and his research group. Special recognition is given to Dr. Benjamin Smith, senior scientist. Dr. Winefordner and Dr. Smith always had an open door for my relentless questions, and were very important to my research progression. I would also like to thank my mentor, Dr. Chenglong Yang, for his help during my early days in the laboratory. Mike Herlevich and Dr. David Myers, who directly contributed to many of my research efforts, also deserve special recognition. I also thank Dr. Fred King, my undergraduate research advisor, who supported my early research endeavors at West Virginia University and who is responsible for my pursuit of graduate research.

I would like to extend appreciation to all members of the Harrison and Winefordner groups whom I have had the privilege to meet. Dr. Kristofor Ingeneri and Dr. Matthew Mohill, my predecessors in the Harrison group, deserve special recognition. I also mention my close friends, Nathan Pixley and Dimitri Pappas, who provided an endless source of amusement in the laboratory.










I would like to mention two visiting scientists, Dr. Arne Bengtson and Dr. Volker Hoffmann, for their helpful discussions and insightful suggestions.

A special word of gratitude goes to my family. I would certainly not be the person that I am today without the continued love and support of my parents, Betty and Larry. I have been blessed with their compassion throughout life's endeavors, and for this I am indebted forever. I owe special thanks to my love, Paige, for bestowing her incessant love and support. I honor all three of you with this dissertation.

This research has been supported by LECO Corporation and the United States Department of Energy, Basic Energy Sciences.































iv











TABLE OF CONTENTS

paqe

ACKNOWLEDGEMENTS .................................................................. .... iii

A B S T R A C T ..................................................................................... .... ix

CHAPTERS

1 INTRODUCTION ........................................................................... ..... 1

Intro d uctio n ................................................................................. ..... 1
Solid Sample Elemental Analysis .................................................... ..... 4
G low D ischarge ........................................................................... ..... 6
The Pulsed Glow Discharge ........................................................... ..... 7
Dissertation Scope ....................................................................... ... 10

2 THE GLOW DISCHARGE ............................................................... ... 13

Intro d uctio n ................................................................................. .... 13
General Discharge Characteristics ....................................................... 14
Glow Discharge Spatial Regions ...................................................... ... 17
Cathode Dark Space ................................................................ ... 19
N egative G low ......................................................................... ... 2 1
Faraday Dark Space ................................................................ ... 23
Glow Discharge Processes ............................................................ ... 23
Cathodic Sputtering .................................................................. ... 23
Sputtering process ............................................................... ... 25
S putter rate ........................................................................ ... 27
S putte r yield ....................................................................... ... 28
Collisional Phenomena ............................................................. ... 29
Electron behavior ................................................................ ... 30
Sputtered species excitation and ionization ............................... ... 30
E xcitatio n ........................................................................... ... 3 1
Io n iza tio n ........................................................................... ... 34

3 GLOW DISCHARGE MODES .......................................................... ... 36
Introd uctio n .................................................................................... 36
Source Configurations ................................................................... ... 36
Hollow Cathode Geometry ......................................................... ... 37
Diode (Coaxial) Geometry ......................................................... ... 39
Grimm Geometry ..................................................................... ... 42
Operational Modes ....................................................................... ... 45
D irect C urrent ......................................................................... ... 45


V










Radio Frequency ..................................................................... ... 48
P u lse d ................................................................................... ... 5 0
Microsecond Pulsed Glow Discharge Advantages ............................... ... 52
Additional Control Parameters .................................................... ... 52
Enhanced Sputtering Rates ....................................................... ... 53
Enhanced Ionization and Emission .............................................. ... 54
Temporal Resolution ................................................................ ... 54
Pulsed Glow Discharge Spectroscopies ............................................ ... 55
Atomic Fluorescence/Absorption Spectrometry .............................. ... 57
Atomic Emission Spectrometry ................................................... ... 57
Time-of-Flight Mass Spectrometry ............................................... ... 59

4 QUANTITATIVE DEPTH ANALYSIS USING MICROSECOND PULSED GLOW DISCHARGE ATOMIC EMISSION SPECTROMETRY ................... ... 64

Introd uction ................................................................................. ... 64
Experimental ............................................................................... ... 65
Instrumentation ........................................................................ ... 65
Sample Material ...................................................................... ... 68
Sputter rate protocol ............................................................ ... 68
Additional samples ............................................................... ... 69
Results and Discussion ............................................................. ... 70
Optimization of Conditions ......................................................... ... 70
Effect of voltage and pressure ................................................ ... 71
Effect of pulse width and pulse frequency ................................. ... 75
Effect of sputtering time ........................................................ ... 75
Sputtering Rates and Penetration Rates ....................................... ... 78
Calibration Curve Correction ...................................................... ... 82
Quantification Overview ............................................................ ... 84
Principle of Quantification .......................................................... ... 88
Concluding Remarks ..................................................................... ... 96

5 INTRODUCTION TO THE GRIMM-TYPE GLOW DISCHARGE AXIAL TIME-OF-FLIGHT MASS SPECTROMETER SYSTEM ............................ ... 97

Introd uction ................................................................................. ... 97
Time-of-Flight Mass Spectrometry Background .................................. ... 98
Operating Characteristics .......................................................... ... 98
Basic Principles ....................................................................... ... 99
Design Considerations ................................................................ 104
Sampling geometry .............................................................. .. 104
Io n o ptics ............................................................................. 104
D etecto r ............................................................................... 105
The Grimm-type glow discharge source and Renaissance timeof-flight mass spectrometer ...................................................... 107
Grimm-type Glow Discharge Source ............................................ .. 108


Vi










Source description .............................................................. .. 108
Source operation ................................................................ .. ill
Renaissance Time-of-Flight Mass Spectrometer ............................. .. 112
Mass spectrometer description .................................................. 112
Mass spectrometer modifications ............................................ .. 115
Overall Operation .................................................................... .. 116
Performance of the Glow Discharge Time-of-Flight Mass Spectrometer..... 116
Ion Detection .......................................................................... .. 120
S ensitivity .............................................................................. .. 122
Resolving Power ..................................................................... .. 124
Linear Dynamic Range ................................................................ 124
Isotopic Accuracy ....................................................................... 127
Signal-to-Noise Considerations ..................................................... 127

6 ION TRANSPORT DIAGNOSTICS IN A MICROSECOND PULSED GRIMM-TYPE GLOW DISCHARGE TIME-OF-FLIGHT MASS SPECTROMETER .............................................................................. 131

Introduction ................................................................................... 131
Experimental ................................................................................. 133
Glow Discharge Source and Time-of-Flight Mass Spectrometer ........ .. 133
Sample Material ...................................................................... .. 133
Results and Discussion ................................................................. .. 134
Background .............................................................................. 134
Parametric Study ....................................................................... 135
Gas flow rate ...................................................................... .. 135
P ressure .............................................................................. 139
Other operating parameters ..................................................... 142
Source Design ........................................................................ .. 142
Initial source design ............................................................. .. 142
Gas directing sleeve design ..................................................... 149
Sampler/Sleeve combination design ........................................ .. 153
Application of Temporal Resolution ............................................. .. 156
Concluding Remarks ....................................................................... 161

7 DIRECT PIN SAMPLE ANALYSIS USING A CONVENTIONAL GRIMMTYPE GLOW DISCHARGE SOURCE AND TIME-OF-FLIGHT MASS SPECTROMETER .............................................................................. 162

Introduction ................................................................................... 162
Experimental ................................................................................. 166
Glow Discharge Source and Time-of-Flight Mass Spectrometer ........ .. 166
Pin Sample Holder ..................................................................... 166
Sample Material ...................................................................... .. 169
Results and Discussion ................................................................. .. 170
Sample Holder Design .............................................................. .. 170


Vii










Source/Plasma Configuration ..................................................... .. 170
Sputtering Rate Comparison ...................................................... .. 171
M atrix Ion S ignal ........................................................................ 174
Detection Limit and Isotope Ratios .............................................. .. 177
Molecular Interferences ............................................................ .. 179
Temporal Resolution ................................................................ .. 186
P in Length ................................................................................ 189
Pin Sample Application ............................................................. .. 191
Concluding Remarks ....................................................................... 197

8 THIN FILM ANALYSIS WITH A MICROSECOND PULSED GLOW DISCHARGE TIME-OF-FLIGHT MASS SPECTROMETER ....................... .. 198

Introductio n ................................................................................... 198
E xperim ental ................................................................................. 200
Glow Discharge Source and Time-of-Flight Mass Spectrometer ........ .. 200
Sample Material ...................................................................... .. 201
Results and Discussion ................................................................. .. 202
Sample Fabrication .................. .............................................. .. 202
Gold Thickness ....................................................................... .. 202
Scanning electron microscope ................................................ .. 202
Weight measurements ............................................................ 204
Parametric Studies .................................................................. .. 206
Pulse voltage ........................................................................ 206
Source pressure .................................................................... 208
Pulse frequency .................................................................... 210
P ulse w idth .......................................................................... 2 11
Thickness Limit ....................................................................... .. 211
Calibration Curve ....................................................................... 216
Quantitative Conversion .............................................................. 216
Simultaneous Mass Spectra Collection ........................................... 218
Multiple Sample Layers ............................................................... 222
Additional Application ............................................................... .. 226
Concluding Remarks ....................................................................... 227

9 CONCLUDING REMARKS ................................................................ 229

REFERENCES ................................................................................... 233

BIOGRAPHICAL SKETCH .................................................................... 242








Viii

















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


THE MICROSECOND PULSED GLOW DISCHARGE:
DEVELOPMENTS IN TIME-OF-FLIGHT MASS SPECTROMETRY
AND ATOMIC EMISSION SPECTROMETRY


By

Eric Oxley

August 2002

Chair: Willard W. Harrison
Major Department: Chemistry

The focus of this dissertation is the study of the microsecond pulsed glow discharge (GD) source through developments with time-of-flight mass spectrometry (TOFMS) and atomic emission spectroscopy (AES). The GD is an electrical plasma long known for its ability to convert solid samples directly into the atomic state for subsequent analysis. Applying a transient pulse to a GD source provides exciting new features and applications that can be evaluated. The microsecond pulsed GD shows great potential as an ion source for mass spectrometry (MS) and as a photon source for atomic emission (AE) measurements. New directions for these atomic techniques have been evaluated and will be described throughout this dissertation.



ix










After successfully completing a microsecond pulsed GD-AES depth profiling project, research efforts were shifted to MS studies. The conversion of a commercial inductively coupled plasma (ICP) time-of-flight mass spectrometer to a GD-TOFMS system is described. A Grimm-type GD ion source was chosen, which has found limited use for mass spectrometric applications, despite its effectiveness as an AE source. The sluggish development of the Grimm-source for mass spectrometry is attributed to its inherent ion transport complexity. The performance of this GD-TOFMS, including ion transport efficiency, is reported.

A limitation of the Grimm configuration is its lack of versatility only flat samples that can mount externally on the source are possible. Non-flat samples, such as pins and wires, require a probe-type configuration. A sample holder has been developed that allows pin sample analysis directly on a Grimm-type source. Easy sample interchange and reproducible sample placement inherent advantages of the Grimm source are conserved with this sample holder.

Depth profiling is an important feature of GD-AES studies, but has not been considered a practical application for GDMS. The feasibility of Grimm GDTOFMS depth profiling is determined by establishing operating conditions that simultaneously afford sufficient ion transport and well-defined craters. This was achieved by merging our established ion transport conditions with operating parameters yielding good depth resolution. Samples consisting of very thin layers, as well as multi-layered samples, were analyzed with our GD-TOFMS.















CHAPTER
INTRODUCTION

Introduction

Mass spectrometry is an analytical technique in which gaseous ions are separated based on their mass-to-charge ratios. The versatility of this technique makes it one of the most widely applicable tools available to an analytical chemist. Indeed, mass spectrometry has significantly evolved since its initial
2
development by Aston' and Dempster, who made the first measurements of ionic masses and abundances. These experiments followed developments by Thomson, who demonstrated the existence of elemental isotopes by measuring
3
intense streams of particles, which he termed positive rays. These pioneering endeavors in mass spectrometry gave impetus to probe the isotopic composition of many types of samples. Resulting mass spectra displayed only single lines at characteristic mass-to-charge (m/z) values, which proved less complex than convoluted atomic emission spectra. The analysis of petroleum-based samples was a driving force of mass spectrometry at this time, and the spectral simplicity of isotopic spectra gave reason to actively pursue mass spectrometry research eventually resulting in the eventual fabrication of the sophisticated mass spectrometers that are available today.

Many types of mass spectrometers have evolved from the pioneering experiments by Thomson. Present day mass spectrometers can be separated



1







2

based on their means of mass selection: dispersion, filtration, and trapping. Sector-based mass spectrometers,4' which were developed from Aston's prototype instrument,' employ an electromagnetic and electrostatic field to disperse different mass-to-charge ions spatially. Today, these devices have evolved into high-resolution, yet expensive, devices. Quadrupole mass spectrometers6'7 use a band-pass principle to select ions of a particular mass window, and have made tandem mass spectrometry experiments possible.8 Ion trap mass spectrometers'1 selectively store ions prior to their measurement. Ion cyclotron mass spectrometers, which also utilize a trapping mechanism to store ions, have achieved ultra-high resolution measurements. Finally, time-offlight mass spectrometers,3,14 which are based on a dispersion-in-time phenomenon, have developed into a leading form of mass spectrometry due to

(1) the soaring popularity of biological mass spectrometry research and (2) improvements in the speed and quality of electronic devices.

Improving these mass spectrometric devices has coincided with the development of sophisticated ion sources. The current popularity in biological mass spectrometry has shifted attention from atomic ionization sources to sources capable of producing intact molecular species. Two of the most popular molecular-based ion sources include electrospray injection (ESI)15 and matrixassisted laser desorption mass spectrometry (MALDI). 16 These ion sources are generally credited with bringing mass spectrometry to the forefront of analytical research.







3

Developing ion sources that are capable of producing atomic species is certainly also important for analytical measurements, particularly mass spectrometric research. These sources provide a wealth of information about samples, including elemental quantitation, ionic speciation, isotopic composition, 17
and spatial distribution. Samples exist in a wide variety of types (i.e., matrices), complicating the choice of an appropriate atomic ionization source. Ion sources, particularly those capable of analyzing solid samples, have been plagued by problems that arise with the production of atomic vapor from solid samples. However, solid sample analysis is crucial for many divisions of science, including geological and environmental studies, which is why sample introduction methods continue to be at the forefront of spectrochernical research.

Solid samples can be treated in several ways prior to analysis by mass spectrometry. Some ionization sources, such as the inductively coupled plasma (lCp),18 require sample digestion and extraction prior to analysis. Indeed, these added steps can be tedious and time-consuming. Since sample turnaround time is of prime importance for many production laboratories, this added step can be a concern. Sample digestion is typically accomplished through dissolution in an acidic solution, which dilutes the concentration of the analyze and can result in the introduction of contaminants. Surface chemistry studies are also negated with dissolution steps since the sample is physically converted to aqueous form.







4

Solid Sample Elemental Analysis

Despite a wealth of sample preparation methods available, the ability to directly analyze solid samples is a substantial advantage since contamination and dilution concerns are avoided. Some examples of atomic ionization sources that can directly analyze solid samples include secondary ion bombardment, laser ablation, and plasma-based processes. All of these are capable of converting solid samples into gaseous ions for subsequent analysis by mass spectrometry; however, all of them suffer from inherent limitations.

Secondary ion mass spectrometry (SIMS) utilizes a beam of ions to analyze sample surfaces, such as metals, alloys, and semiconductors. 19 Primary ions, such as argon and cesium ions, are generated by an electron impact ion gun and then accelerated to the sample surface at approximately 5 20 keV. The collision of these ions with the sample surface leads to sample removal through a phenomenon termed sputtering. While the sputtered sample material is primarily composed of sample atoms, only a small portion (-1 %) of the ejected sample material is in ionic form (i.e., secondary ions). SIMS is capable of probing lateral and depth resolution of the sample surface and is characterized by high sensitivities; however, matrix effects and sensitivity variations are 20
common limitations.

Laser ablation techniques, such as laser induced breakdown spectroscopy (LIBS), utilize a laser beam focused onto a sample surface to ablate material for 21,22
subsequent analysis. The impingent laser energy, usually originating from an excimer or pulsed Nd:Yag laser, is transferred to the sample surface, resulting in







5

the ejection of molecules, atoms, and ions. While laser ablation is a versatile technique capable of analyzing conductive and insulating materials with little sample preparation, this technique suffers from poor precision and severe interference effects.

Laser-based sources have also been coupled to inductively coupled plasma mass spectrometers (ICPMS) to eliminate dissolution steps required for 23,24
solid sample ICP analysis. In laser ablation inductively coupled plasma mass spectrometry (LA-ICPMS), the sample material is ablated from the sample surface and transported to the ICP plasma and subsequently analyzed. The use of LA-ICPMS allows sample introduction and sample ionization steps to be optimized independently, affording better control of the analysis. While this independent control can be an advantage, it can also contribute to the poor 25
accuracy and precision of the technique. Future research in sample transport efficiency and sample ionization may help resolve these problems.

Another means of converting solid samples into gaseous ions for mass spectrometry is the use of plasma sources. Spark sources were among the first 26
plasma-type sources coupled to mass spectrometry. This source converts solid samples into atomic constituents by applying a high-potential (-30 kV) between two electrodes that are typically fabricated from the sample material. A repetitive high current discharge forms on the sample surface and electron bombardment releases sample material for subsequent mass spectral analysis. Spark sources are highly sensitive, yet erratic sampling devices. This erratic behavior hinders the precision of the source, and renders it unable to perform depth-profiling of the







6


sample surface. Also, a spark produces a large kinetic energy distribution of ions, which can be difficult to focus in mass spectrometry applications.



Glow Discharge

Another plasma-based technique capable of directly producing atomic species from solid samples is the glow discharge (GD). The desire to develop better means of solid elemental analysis by mass spectrometry has resulted in a growing interest in glow discharge research. This versatile source lends itself to a range of analytical possibilities, including bulk metal and alloy analysis, depthresolved analysis, and non-conductive sample analysis. While GD devices have many inherent advantages, stability and reproducibility are considered their best attribute compared to other competitive sources.2

Also adding to the versatility of the GD is its ability to produce representative atomic populations of the sample, as well as excited state and ionic populations of those atoms. Since Coburn and colleagues demonstrated the analytical utility of the GD source in 1 970s,29 this versatile source has seen application in atomic emission spectrometry (GD-AES),30,31 atomic absorption spectrometry (GD-AAS)2.33 atomic fluorescence spectrometry (GD-AFS),34,35 and mass spectrometry (GDMS).36.37

A glow discharge plasma is formed when a sufficiently high voltage is applied between two electrodes that are immersed in a gaseous environment, such as argon. Typically, the cell is arranged so that high voltage is applied directly to the sample (cathode), while the anode is held at ground potential. The







7

potential difference between the two electrodes causes breakdown of the argon atoms (Aro), forming argon ions (Ar+) and free electrons (e). Acceleration of the argon ions toward the sample surface leads to sputter removal of surface atoms

(MO). A simple glow discharge, along with characteristic phenomena of this device, is represented in Figure 1-1.



The Pulsed Glow Discharge

Although most glow discharge devices are operated in a continuous, direct current (dc) mode, aspects of operating a GD in a pulsed mode have been
38
examined. Supplying a pulsed voltage to the sample material generates "packets" of sample atoms that expand as they diffuse across the cell with each pulse, as shown in Figure 1-2. Advantages are obtained when this transient pulse is applied to a GD, such as enhanced sputtering, excitation, and ionization; fewer problems with thermal effects; and temporal discrimination of analytical signals through gated detection.

One method of operating a glow discharge is to operate it in the microsecond-pulsed regime. In this mode, a voltage up to 3.0 kV can be applied to the sample for approximately 10 to 20 microseconds at a frequency between 100 and 1000 Hz. The highly energetic plasma formed during the application of this transient, high voltage yields more excited and ionized sample atoms than are produced by a dc source. Signal increases from 10- to 100-fold have been 38
reported using the pulsed mode of operation.























meow

















400ow






















IJL

























CL




E










06





CL


m ol:








10

Dissertation Scope

The research endeavors described in this dissertation focus on developments of the microsecond pulsed glow discharge source. This research encompasses experiments in both glow discharge atomic emission spectrometry (GD-AES) and glow discharge mass spectrometry (GDMS). Fundamental studies, such as ion transport diagnostics, as well as applications, such as depth profiling, of the microsecond pulsed GD source are shown. This work is separated in five distinct, yet related, sections.

The first project undertaken was an evaluation of the microsecond pulsed GD source for performing depth profiling measurements. This experiment was performed on a commercial atomic emission spectrometer (SA-2000 Surface Analyzer, LECO Corporation, St. Joseph, MI, USA), which was modified for our pulsed GD work. The pulsed mode of operation afforded the analysis of thin sample layers that are difficult to examine using a dc source. Optimal conditions of the glow discharge source, including voltage, pressure, pulse width, and pulse frequency were utilized. The analysis of thin coatings of varying depths of copper deposited on a steel substrate is shown.

After successful completion of the AES depth profiling project, research efforts shifted to microsecond pulsed glow discharge mass spectrometry studies. The conversion of a commercial inductively coupled plasma (ICP) time-of-flight mass spectrometer (TOFMS) to a GD-TOFMS system is described. A Grimmtype GD source was designed and coupled to the modified TOEMS. The software was also modified for pulsed-source compatibility, allowing the







11

instrument to trigger the power supply and synchronize timing events between the mass spectrometer and the glow discharge source. This new GD-TOFMS system was evaluated, and its figures of medt, such as mass resolving power, are shown.

The next project describes a fundamental study employing the GDTOWS system. Ion transport from the source to the mass spectrometer, an inherent limitation of a Grimm source and mass spectrometer combination, was evaluated. Fundamental studies of ion transport are described, and methods of improving this transfer are shown. Some operating conditions of the source were found to significantly alter the efficiency of the transfer step. The configuration of the Grimm-type source was also a significant factor in maximizing ion transport. An example shown in this dissertation is the ability to identify trace amounts of magnesium from background ions using an optimized source configuration.

After establishing conditions that yield sufficient ion transport, applications of the GD-TOFMS were evaluated. A limitation of the Grimm configuration is its lack of versatility only flat samples that can mount externally on the source are possible. Non-flat samples, such as pins and wires, typically require a probetype configuration. A sample holder has been designed that allows pin sample analysis directly on a Grimm-type source. Easy sample interchange and reproducible sample placement inherent advantages of the Grimm source are conserved with this sample holder. An evaluation of this sample holder for pin sample analysis is shown.







12

The final project undertaken was a depth profiling application similar to the initial AES project described above. Although depth profiling is an important feature of GD-AES, it has not been considered a practical application for GDMS. This section describes the establishment of operating conditions that simultaneously afford sufficient ion transport and well-defined craters for successful depth profiling with our GD-TOFMS system. Conditions were chosen by merging our previously established conditions for sufficient ion transport (TOFMS) with those that yielded good depth resolution (AES). Shown is depth profiling of thin gold films on silicon wafers, as well as samples consisting of multiple layers, such as spent computer hard discs.















CHAPTER 2
THE GLOW DISCHARGE

Introduction

A glow discharge cell consists of two electrodes immersed in a gaseous medium, typically argon. The gas is insulating by nature, but becomes conductive when sufficient potential is applied across the electrodes. Atoms of the gas break down electronically into argon ion-electron pairs, allowing current to flow through the system. The resultant glow discharge is a cool plasma that attains temperatures up 39
to approximately 800 K.

An electrical discharge, such as a glow discharge, consists of many complex processes involving electrons, atoms, and ions of a wide range of energies. Collisions account for a significant portion of the chaotic processes that occur within a glow discharge. Sputtering, for example, is a collisional event that converts sample material into gaseous atoms. These diffusive sample atoms can collide with other energetic species, resulting in the production of photons or ions that can provide information about the plasma. Spatial zones of a glow discharge plasma are formed (or collapsed) through collisional events. The processes within a glow discharge are complex, and understanding them has been a goal of researchers for many decades.

This chapter will describe the most fundamental processes that comprise a glow discharge, and will be divided into the three separate, but related



13







14

sections. The first segment will deal with general discharge characteristics and their formation; the next section will discuss the spatial zones that comprise a glow discharge; and the final portion will treat the major species within a plasma and the processes responsible for their formation.



General Discharge Characteristics

Different types of glow discharge can be attained by altering the applied voltage and current. Discharges are generally separated into four classifications based on their voltage-current characteristics: Townsend discharge, normal glow discharge, abnormal glow discharge, and the arc discharge .40 For analytical purposes, the discharge in the abnormal region is the most important. This region will be the central focus of this dissertation, though insight can be gained about the glow discharge by examining other regions as well.

Consider the example of a potential dropped across two electrodes submerged in an argon medium. An electrical discharge will result when the potential is applied, given the proper operating conditions. The type of discharge that forms is largely dependent on these operating conditions, as illustrated in Figure 2-1 .40 This figure shows the voltage-current characteristics of the discharge, where Ub is the breakdown voltage and U, is the normal operating voltage.

The left-most region in Figure 2-1, the Townsend discharge regime (A-B), is generally operated at low pressures (sub-millitorr). Only a small number of ions and free electrons are produced at the breakdown voltage (Ub); hence, this regime is characterized by a low operating current and is not self-sustaining. An external







15








Norn~l glow discharge


Voltage Townsend dark discharge Abnormal glow discharge


A B
Ub

:F
Arc discharge D E

Un







:G H
I I I I I ,

10' 10-1 106 104 101 10
Current/A Figure 2-1. Voltage-current characteristics of a glow
discharge.40







16

source, such as X-rays and ultraviolet light, is needed to facilitate free electron and ion production that is needed for sustaining the plasma. The small number of collisional processes within the Townsend discharge leads to little photon production and subsequently low luminosity.

A transition region (C-D, Figure 2-1) is encountered when the operating voltage is increased beyond the Townsend threshold. Increased collisions in the transition region result in large energy exchanges, and subsequently higher currents. 17 The potential then remains nearly constant for large variations in current, which is termed the normal discharge region (D-E). The discharge in this region forms only on a small portion of the cathode, and subsequently only small current fluctuations are found. This glow discharge region is noted for its luminous glow resulting from strong visible light emission.

In the abnormal glow discharge (E-F), the plasma begins to encompass the entire sample surface, and the current begins to increase. Sustaining the abnormal discharge requires that the applied voltage be increased as a function of current. For analytical applications requiring an emission or ion source, the information-rich abnormal region is generally chosen owing to its high population of sample atoms and ions. 17,40

If the current is increased beyond the abnormal region, the discharge enters a second transition region (F-G), which is marked by a large drop in resistance. The discharge voltage abruptly decreases and establishes high currents, even for small operating voltages. Increasing the current beyond this transition region can result in severe heating of the cathode as a result of elevated current densities. Extreme







17

currents are a result of intense gas bombardment and subsequent thermal vaporization of the cathode. This region, termed the arc discharge (G-H), currently 41-43
finds limited application in analytical spectroscopy. The spectrometrist wishing to use a glow discharge for analysis is often dismayed at the sight of an arc discharge and its erratic, unstable nature.



Glow Discharge Spatial Regions

The abnormal glow discharge, hereafter called simply the glow discharge, comprises several alternating dark and luminous regions, as shown in Figure 22 17,40,44
The formation of these regions is dependent on the configuration of the glow discharge cell employed, particularly the distance separating the electrodes. For long electrode distances, a highly luminescent positive column region is most prominent; however, it is not required to sustain a glow discharge. This region contracts as the distance between the electrodes is decreased, until eventually collapsing. Other regions of the discharge are also diminished by decreasing the electrode separation, with exception to the cathode dark space, which must exist to sustain a glow discharge. Since most analytical glow discharge devices employ short electrode distances, they generally exhibit only the cathode dark space and negative glow regions. An obstructed gloW,45 a discharge in which most of the regions have collapsed, occurs when the electrode separation is only a few times 40,46
larger than the thickness of the cathode dark space. Further decreasing the electrode separation yields a distorted dark space region and an unstable, or extinguished, plasma. For brevity, only the three prominent regions of the glow







18











0
c ad

u
CL








-rA

lie






iz CL


m lie
-j Tu







19

discharge will be discussed in the following sections: the cathode dark space, negative glow, and Faraday dark space regions.



Cathode Dark Space

The cathode dark space is a region of low luminosity that appears adjacent to the cathode (sample) surface, as depicted in Figure 2-2. Electrons are repelled away from this region due to the negative cathode potential. This electron repulsion creates a positive space, where most of the potential difference between the two 47
electrodes is dropped. Hence, the cathode dark space is often called the cathode fall region.

Figure 2-3 shows a profile of potential distribution as a function of electrode distance for the three major regions of the glow discharge. The large potential fall inherent of the cathode dark space region is represented in this figure. Electrons are accelerated away from this region and consequently have too much energy for excitation reactions. Indeed, an excitation cross-section shows a dip at these high energies '48,49 "Which accounts for the low luminosity exhibited by the cathode dark space.

Despite the lack of excitation and ionization processes occurring within the cathode dark space, a glow discharge cannot exist without this region. Electrons that are accelerated away from this region are responsible for ionizing the fill gas

(Ar) in other regions of the discharge. The electrons created from these collisions may cause further ionization, while the resulting argon ions are attracted to the cathode. The continuous cycle of these complex processes leads to a self-







20


























.2


........ ...


96el4U.. 0

T2







21

sustaining discharge, which originated with the acceleration of electrons in the cathode dark space region.



Negative Glow

The negative glow region is the most visually striking region of the glow discharge. This large, bright region is adjacent to the cathode dark space and is analytically the most important region of the glow discharge. Two types of electrons enter the negative glow: fast, highly energetic electrons and slow, thermal electrons.

Fast secondary electrons are those that have fortuitously traversed the cathode dark space without losing significant energy due to collisions. These electrons, typically called group I electrons, are only capable of ionizing collisions dues to their inherent high energies. Electrical (Langmuir) probe experiments show that group I electrons attain electron temperatures of -20 25 eV and number densities on the order of 106 CM-3 in the negative glow region.50

Low-energy electrons, often termed thermal electrons, are separated into two groups. Group 11 electrons are secondary electrons of gas-phase ionization collisions that have electron temperatures of -2 10 eV, and number densities of approximately 107 108 CM-3 .50 Group III electrons are actually electrons from group I or group 11 that have experienced several elastic and inelastic collisions within the plasma. These electrons typically have electron temperatures of only 0.05

- 0.6 eV, and number densities in the range of 109 1011 CM-3. 50 Unlike their energetic counterparts, group 11 and group III electrons may either excite atomic species or ionize excited state species.







22

The number of electrons in the negative glow is generally matched by a 47
similar number of positive ions. The combination of these free ions and electrons results in an essentially field-free region, as represented in the potential distribution of Figure 2-3 by a linear segment just above ground potential (Vo). Langmuir probe experiments have demonstrated that plasma potentials within the negative glow region vary less than 1 V for discharge voltages between 800 and 1000 V.51 Electrons are not significantly accelerated in this field-free region, and as such are capable of causing excitation collisions, resulting in a bright 'glow.' The color of this glow is largely dependent on the fill gas and to a lesser degree on the sample material.

The size of the negative glow region is affected by the discharge pressure and the distribution of electron energies. High operating pressures tend to increase the number of collisions within the discharge, which compresses the spatial distribution of the negative glow. Conversely, a wide range of electron energies can extend the length of this region.

The largest population of electrons exists at the interface between the cathode dark space and the negative glow a result of electron multiplication by ionization processes. Accordingly, the greatest emission intensity is seen at this edge of the negative glow, which fades as a function of cathode distance. Mass spectrometers are typically configured to sample the high population of ions that exist at this negative glow interface.







23

Faraday Dark Space

The Faraday dark space region is found just beyond the anode end of the negative glow region. Electrons comprising the Faraday dark space are thermal electrons that have lost most of their energy due to excitation and ionization
47
collisions. In many cases, the separation between the cathode and anode is so small that this region of low luminosity is not observed.



Glow Discharge Processes

The glow discharge is a deceptively simple analytical device full of chaotic processes involving different species of varying energy. An overview of these basic processes will be beneficial to understanding the experimental results in subsequent chapters. The present section will describe an overview of the mechanisms within the spatial regions described in the previous section. For brevity, complex processes, such as multi-body collisions and sputtered clusters, will not be treated. Readers wanting a detailed description of plasma processes should consult other discussions '40,44.47 some of which are based on computer modeling resu ItS.51-55 The fundamental processes that will be described in this section, such as sputtering, excitation, and ionization, are summarized in Figure 2-4.



Cathodic Sputtering

An energetic particle that strikes a solid surface can result in an initiation of many processes. This phenomenon, termed sputtering, is a large contributor to the analytical utility of the glow discharge. Atomization of the sample occurs through ion






24










--------- ---Metastable Penning Ionization
Fon'ationI





Negative
Glow
Electron Ionization I
Initial ~ '
IBreakdown



Dark Space NiI







Figure 2-4. Major processes that occur in a glow discharge.







25

bombardment of the sample surface, and provides an atomic population for 47
subsequent excitation and ionization processes. This section gives an overview of the sputtering process, as well as quantitative principles for characterizing sputtering.



Sputtering process. The impact of a particle can lead to considerable damage to the sample through atom rearrangement or particle implantation, as illustrated in Figure 2-5. The sputtering process illustrated in this figure is simplified, as the incident argon ion typically undergoes several charge exchange reactions with neutral gas atoms before impacting the cathode surface. 56 Therefore, argon atoms can also contribute to the sputter yield, to a certain degree. Upon impact, the kinetic energy of the incident particle (e.g., Ar*) is transferred to the sample (M) through collisional events, while the potential energy of the particle results in electron ejection. The incident argon ion is either backscattered from the sample surface or penetrates the sample and transfers its energy to the surface. If the impinging ion has sufficient velocity (>30 eV), material may be sputtered (i.e., ejected) from the sample surface in the form of atoms or molecules in neutral (MO) or 47
ionized states (M+ or M-), or electrons may be ejected. Sputtered particles generally possess low energies (few eV) and can be ejected at many possible angles.

Positively charged ions (M'), which make up approximately 1% of the total sputtered particle flux, are redeposited back onto the sample surface, as dictated by the negative cathode potential .57 Negative ions (M-), conversely, are accelerated






26






LI









00




II ........
Ib

IC I 4J p
SZ
It IC

t~i WV
z7







27

away from the cathode surface. Neutral sample atoms comprise the majority of the 58
ejected material, and typically possess kinetic energies approximately 5 10 eV. Upon ejection, most of these sample atoms lose their momentum to elastic collisions with discharge species due to the small mean free paths at the cathode surface (-0.1 mm for pressures of 0.1 10 torr). Weight loss studies by Harrison and Bruhn have shown that up to 95% of these atoms may be deposited back onto the cathode surface .59 The atoms that are not redeposited can diffuse into the negative glow region and undergo excitation or ionization processes, which are essential for atomic emission and mass spectrometry measurements, respectively.



Sputter rate. The amount of sample ejected for a given analysis time is defined as the sputter rate (q ng/s):

q = AWIt (Eqn. 2-1)

where AW (ng) is the amount of sample lost due to sputtering and t (s) is the total time of sputtering. Typically, Equation 2-1 is termed the net sputtering rate for glow discharge since it recognizes that the glow discharge inherently involves redeposition of sputtered material 60

Sputtering rates vary as a function of many parameters, including voltage, current, and pressure of the discharge. The type of fill gas may also affect sputtering, since the mass of the sputtering agent will depend on the fill gas. The type of sample can also alter the sputtering rate; multi-component alloys of the same matrix exhibit sputtering rates that are dependent on the concentration of other minor constituents comprising the sample. For example, pure copper has a







28

measured sputtering rate of approximately 0.67 ng/pulse for a microsecond pulsed glow discharge source .60 Adding a constituent to the sample with a different sputtering rate, such as zinc (1.74 ng/pulse), can affect the overall sputtering rate. Weight loss measurements show that brass (70% copper and 30% zinc) has a sputtering rate of approximately 0.83 ng/pulse, which is higher than pure copper. 60

Ideally, all elements (and alloys made of the same matrix) would exhibit the same sputtering rates. However, sputtering rates vary from element to element, which can adversely affect calibration curves. One way to correct an erroneous calibration curve is to normalize it based on established sputtering rates .60 This procedure will be described in more detail in Chapter 4.



Sputter yield. Another way to characterize the sputtering efficiency of a sample is to calculate sputter yield. The sputter yield (S atoms/ion) of a sample is defined as the number of sputtered atoms per incident sputtering particle.0 Boumans has experimentally developed an expression for the sputter yield :61 S =(I x 10-6*q *N *e) /(M ) (Eqn. 2-2)

where, q (ng/sec) is the sputtering rate, N (mole-) is Avogadro's number, e (C) is the charge of an electron, M (no units) is the atomic weight, and i+ (A) is the ion current. The bombarding ion current is related to the total discharge current by: i= i/(1 + ) (Eqn. 2-3)

where, i (A) is the total discharge current, and y is the number of secondary electrons ejected by one ion.







29

Collislionall Phenomena

Collisions account for many of the processes within the glow discharge, and contribute to the self-sustaining nature of the plasma. Collisions that occur within the glow discharge can be broadly separated into two types, elastic and inelastic.

Elastic collisions are the simplest collisions, in that kinetic energy is conserved. Excitation and ionization processes within the plasma are generally not attributed to these types of collisions, since energy is not transferred. For excitation or ionization to occur, an impinging electron must have more energy than the energy needed to remove an electron from the atom (called the ionization energy). If the energy of the electron is insufficient, it will be deflected from the atom. Since electrons and atoms have significantly different masses, the energy transfer function suggests that the amount of energy transferred for an electron-atom elastic collision
44
is negligible. Despite their inability to cause ionization, elastic collisions serve to 62
redistribute the kinetic energy and help to thermalize the plasma.

The other type of collision within a glow discharge plasma is an inelastic collision. This type of collision, which involves energy transfer between plasma species, is responsible for (1) forming species of analytical interest and (2) maintaining the self-sustaining nature of the plasma. Since inelastic collisions comprise many of the processes that occur in the glow discharge, this section will be separated into three parts. A general description of electron behavior will be discussed first, followed by the collisional processes leading to the excitation of sputtered sample atoms. A discussion about the ionization of these sputtered sample atoms will conclude the section.







30

Electron behavior. Electrons within a glow discharge are accelerated away from the cathode and may eventually collide with gas atoms and cause ionization. However, many of the electrons that are formed from collisional processes are lost due to recombination .40 Electrons can also impart their energy to an atom, thereby raising the internal energy of the atom to an excited state. When this excited atom relaxes to a ground state, a photon is released which contributes to the characteristic glow of the discharge. These photons also prove analytically useful for probing optical characteristics of the discharge.

Electrons are typically accelerated with enough kinetic energy (up to 2000 eV at 2 kV) to ionize atoms within the discharge. Argon, for example, has an ionization energy of 15.76 eV, and is primarily ionized by collisions with electrons. This process, called electron ionization, is illustrated in Figure 2-4. Electron ionization is a key process within the discharge, as evidenced by the detection of doubly charged argon (Ar +2) ions by mass spectrometry; only electrons have enough energy to form 62
these ions, which have an ionization energy of 27.63 eV.



Sputtered species excitation and ionization. Authorization of a sample in a glow discharge is accomplished by sputtering the surface with argon ions (and atoms). Atoms ejected from the sample surface can diffuse into the negative glow region of the discharge where they may undergo a series of collisions with electrons, metastable atoms, and other ions. These collisions are responsible for excitation and ionization of the sample material. For excitation and ionization to occur, however, there must be an inelastic collision between an atom and a particle with







31

kinetic or potential energy. Table 2-1 lists the three most probable excitation and ionization mechanisms that can occur in the negative glow region. The role of these processes is largely dictated by operating conditions of the discharge.



Excitation. Generally, electrons in the virtually field-free negative glow region possess low kinetic energy. These thermal electrons have lost a considerable amount of their energy due to multiple collisions. The number of energetic electrons can be shown as an electron energy distribution function (EEDF), as illustrated in Figure 2-6. The abundance of thermal electrons will likely result in excitation mechanisms that preferentially involve energy transfer to levels 53-4 eV above ground state .50 These ground state transitions are primarily responsible for the atomic transitions that occur in the ultraviolet-visible region, and are responsible for the emission (i.e., glow) from the negative glow. The photons that are released can be measured by glow discharge atomic emission spectrometry (GD-AES).

Electronic excitations involving electron impact are also credited with populating argon metastable atoms. Argon has two long-lived (approximately one millisecond) excited states with energies of 11.55 and 11.72 eV, respectively. Collisions between these metastable gas atoms and sample atoms may lead to energy transfer and subsequent excitation of the sample atom, termed a Penning collision .63 This process is illustrated in Table 2-1. A third possible excitation 64.65
mechanism within the negative glow region is asymmetric charge exchange.





32



Table 2-1. Common excitation and ionization processes in the glow discharge.


Electron Ionization / Excitation


(+st or

%F (slow)


Penning Collisions



+-... or






Charge Exchange

.+ +0~








33
















so.


40
El 5.44 mm
30 0 7.26 mm
4: .o 9 9.07 mm
. g0o 13-6 mm
20 0 18.1 mm

10 0


0 I I I
0.0 1.0 2.0 3.0 4.0
Electron energy, eV

Figure 2-6. Electron energy distribution as a function of
distance from the cathode.47







34

Most researchers have found that this excitation mechanism is too selective to 63
contribute significantly to excitation processes.



Ionization. The excitation mechanisms described in the previous section are also responsible for ionization mechanisms within the negative glow region. Once ions are formed they can be detected with a number of mass spectrometric techniques. The goal of an atomic mass spectrometrist is to produce (and detect) a 47
maximum number of these ions with as much elemental uniformity as possible.

Electron ionization is not considered the dominant ionization mechanism, but does account for a small portion of ionization within the negative glow. This mechanism, which is shown in Table 2-1, is essentially unselective since any atom can be ionized by an electron with sufficient energy. The cross-section curve for electron ionization is similar in shape and magnitude for all elements of the periodic table .52,66 However, based on the EEDF shown in Figure 2-6, few electrons will have sufficient energy to ionize metals with ionization energies between 5 10 eV. Therefore, it is unlikely that electron ionization accounts for a significant portion of ionization within a glow discharge.

Penning ionization is generally considered the dominant ionization mechanism in the glow discharge. This form of ionization consists of a collision between an argon metastable atom and a sputtered atom, as shown in Table 2-1. The argon metastable atom should be able to sufficiently ionize most elements based on its energies (e.g., 11.55 and 11.72 eV). Like electron ionization, Penning ionization is unselective for elements with lower ionization energies than the







35


metastable atom energy. Separate experiments by Harrison and Coburn have demonstrated that Penning ionization is the dominant ionization mechanism in the glow discharge .67-69 Elemental quantitation studies have suggested that Penning ionization accounts for approximately 50 to 95% of sample ionization in the
70
discharge.

Asymmetric charge exchange is another mechanism that may ionize sample atoms in the negative glow region, as shown in Table 2-1. However, this highly selective, and controversial process, is not credited with being a significant ionization mechanism. Asymmetric charge exchange occurs when an electron is transferred from a sputtered atom to an argon ion, and is likely only if the difference in energy between the two species is small.















CHAPTER 3
GLOW DISCHARGE MODES

Introduction

A subtle advantage of the glow discharge is its operational simplicity. One must only apply to the cathode a sufficient voltage to establish a discharge. The combination of sputtering, excitation, and ionization of the sample constituents leads to photon and ion production, which can be measured by appropriate techniques. The simplicity of the glow discharge has led many researchers to develop source configurations and modifications for particular applications. There is also a growing interest in testing alternative power sources to drive the glow discharge. This chapter will discuss the range of glow discharge configurations, as well as the operational modes, that have found analytical utility. Often, the sample characteristics (e.g., type, shape, etc.) dictate the selection of a particular source configuration and operational mode. One specific operational mode, the microsecond pulsed GD, offers special opportunities and advantages, which will be described. A discussion of analytical methods that probe the sample atoms and ions produced with a microsecond pulsed glow discharge will conclude the chapter.



Source Configurations

The diverse applications of the glow discharge have led to the development of a variety of source configurations. All of these configurations have the ability to



36







37

efficiently analyze different sample shapes, ranging from cylindrical wire samples to large metal alloys, collectively making them well-suited for a range of applications. The glow discharge configurations that will be described in this section include the hollow cathode, diode, and Grimm design. Emphasis will be placed on the latter configuration (i.e., Grimm-type design) since it is the central aspect to many of the research investigations described in the following chapters. Each configuration has 71
its own strengths and weaknesses, which are summarized in Table 3-1. While many configurations have been successfully implemented, an optimum configuration that can efficiently analyze all sample types still has yet to be developed.



Hollow Cathode Geometry

The hollow cathode discharge is the oldest glow discharge configuration, 72
dating back to the 1930s. Hollow cathode sources have enjoyed reasonable success for atomic emission measurements (e.g., hollow cathode lamps), but hollow cathode sources designed for mass spectrometric applications are rarely
73
implemented. The plasma in a hollow cathode is confined to a hollow cavity within the sample, which forces sputtering to occur on the sides of the cavity walls. Highly energetic atoms and electrons are trapped within this hollow, and extended residence times led to enhanced excitation and ionization of the sample atoms. Hollow cathode configurations have demonstrated detection limits down to the 74
subnanogram range, owing to efficient ionization capabilities. Typical operating conditions for a hollow cathode configuration are voltages between 200 500 V, 75
currents of 10 100 mA, and argon pressures between 0.1 10 torr.







38







Table 3-1. Comparison of the glow discharge configurations.71


Source Advantages Disadvantages


*Efficient excitation and *Ion extraction difficult
ionization processes *Complicated sample Hollow *Great sensitivity geometry
C oHigh sputter rate 'Sample alteration

'Amenable to powders required


'Amenable to pins, wires, 'Depth profiling difficult Coaxial discs, etc. -Sample placement
Cathode *Ion extraction easy imprecise
*Ionization dominated by 'Powders require
Penning process conversion


'Depth profiling possible
'External sample mounting 'Non-flat samples Grimm 'Fast sample turnaround difficult
Source 'Precise sample placement Ion transport difficult
'Compacted powders
possible







39

The high population of ions formed within the cavity provided impetus to 76
design a hollow cathode source for mass spectrometry. A hollow cathode configuration that has been designed exclusively for mass spectrometry applications is shown in Figure 3-1. This figure also demonstrates the confined nature of a hollow cathode plasma. While the confined plasma does provide an enhanced population of sample atoms and ions, the intrinsically confined plasma does present a concern. For mass spectrometry applications, the ions must be extracted from the hollow well and transported to the sampling region; these additional steps can prove intricate. However, the potential advantages for hollow cathode mass spectrometry studies, such as enhanced detection limits, do provide justification for developing methods to circumvent these concerns.



Diode (Coaxial) Geometry

The most popular glow discharge configuration for mass spectrometry, 77,78
though typically not used for atomic emission, is the diode geometry. This design, typically called a "pin-type" source, is capable of analyzing sample types (e.g., pins, wires, discs, etc.) that can be mounted on the end of a direct insertion probe (DIP). The probe and sample assembly is introduced into the glow discharge chamber through a vacuum interlock, as illustrated in Figure 3-2. The discharge forms on the end of the sample, and ions are extracted from the negative glow region of the plasma. Diode geometries typically employ conditions of 500 1000 V, 1 5 mA, and 0.4 2.0 torr Ar pressures .75 The diode configuration is arranged so that the sample serves as the cathode, while the surrounding discharge chamber






40













E





_________IT- El_U.8



CL




IV 0 c E L
6 =0
rz






41






01-N m
JOU 4 E


































IA







42

becomes the grounded anode. The only commercially available glow discharge mass spectrometer, a VG 9000 double focusing spectrometer,9 employs a probe configuration for sample introduction. Limitations of this geometry include problems with sample placement, thermal effects, and redeposition on source components.' Further, glow discharge applications that require layer analysis, such as depth profiling, are difficult using a probe configuration.



Grimm Geometry

A source that is widely employed for atomic emission measurements and is gaining popularity for mass spectrometry applications is the Grimm-type configuration. This source design, which employs a flat cathode geometry rather than a probe assembly, was introduced by Grimm as an atomic emission source in the late 1960s.80 The sample is pressed against an O-ring on the flat cathode plate, which ensures adequate sealing and proper vacuum conditions, as shown in Figure 3-3. This type of source utilizes a constricted plasma, in which the discharge is laterally restricted to the sample surface and confined to the size of the surrounding anode. Restriction of the plasma to the sample surface is achieved by maintaining a distance between the anode tip and sample that is smaller than the mean free path of the electrons.81 The plasma is also confined in circumference by the cylindrical anode, which allows planar sputtering of the sample. Typical operating conditions of the Grimm-type source are approximately 500 150OV, 3 30 mA, and 2 6 torr

r. 75






43











VOo



78I 72.
-)






E








00
0 ~ -- -o G







44



The easy sample interchange provided by the Grimm source is a significant 82-85
advantage over other source configurations. Samples are mounted externally on the source, affording quick sample turn-around time and precise sample placement. Sample positioning concerns, which can plague probe-type configurations, are avoided with the Grimm-type source.80,86 Another well established feature of the Grimm configuration is its ability to obtain surface and indepth analysis by properly controlling the discharge parameters to obtain planar sputtering .60,87,88 The ability of the Grimm-type source to analyze thin sample layers with pulsed operation has been demonstrated '60 ",'o and is described for atomic emission and mass spectrometry in Chapter 4 and Chapter 8 of this dissertation, respectively.

One of the aforementioned features of the Grimm-type configuration its ability to analyze flat samples can also be a concern. A flat, smooth sample surface is required to ensure adequate sealing. Other sample shapes, such as pins and wires that have a smaller diameter than the diameter of the anode, are less adaptable. Another limitation of the Grimm configuration, at least for mass spectrometric applications, is ion transport. Transport efficiency, which is crucial to the success of mass spectrometry, is difficult to optimize with this flat cathode geometry. Methods of circumventing these two limitations (i.e., ion transport and sample adaptability) have been studied, and will be described in Chapter 6 and Chapter 7, respectively.







45

Operational Modes

The electrical characteristics of a glow discharge largely determine its performance, and the diverse applications of the glow discharge have led to the development of many operation modes. The glow discharge may be powered by direct current (dc), radio frequency (rf), or pulsed operation. Emphasis will be placed on the microsecond pulsed mode since it is the central aspect to many of the studies described in the following research chapters. Not surprisingly, all of these operational modes have inherent benefits, such as the ability to temporally separate analytical signals or the ability to analyze non-conductive samples. The operational mode of choice largely depends on the nature of the sample to be analyzed.



Direct Current

The direct current (dc) mode is unquestionably the preferred method of powering the glow discharge. The inexpensive power supplies that drive these continuous discharges are dependable, rugged, and easy to operate. A glow discharge operated with a dc voltage creates a constant supply of sputtered atoms that diffuse through the glow discharge chamber, as depicted in Figure 3-4 .38 The discharge is self-cleaning due to the inherent sputtering mechanism. Atmospheric exposure and human contact can deposit impurities (e.g., oxide layers, water vapor, etc.) on the sample surface. However, these contaminants are automatically removed via the sputtering process. Often, the discharge is unstable during the first few moments after ignition due to the electrical nature of these contaminants. An induction period, called a pre-burn, is used to clean the surface and stabilize the







46

discharge prior to signal collection. This pre-burn time can range from seconds to minutes depending on the sample and experimental conditions.

After the pre-burn period, the dc discharge provides a stable plasma with a steady-state production of sample atoms, as shown in Figure 3-4. A sputtered atom gradient is produced, as shown by the concentration contour. Despite the steadystate nature of the plasma, the number densities of the species comprising the plasma can vary significantly.62 The number densities of these species have been predicted mathematically and have shown good correlation with experimental data.91 The results of these modeling studies show that electron impact and Penning 92-94
ionization are the dominant ionization mechanisms of the dc GD.

Since the glow discharge establishes a stable concentration gradient, reproducible results can be obtained by consistently sampling the same region of the discharge. For atomic emission measurements, the negative glow region of the discharge is probed because of its high photon population. Absorption and fluorescence measurements, however, are sampled in the dark space which 62
provides a low emission background. Mass spectrometry measurements require physical transport of ions to the mass analyzer, which can limit their means of sampling.

The direct current mode of operation has two significant disadvantages: power limitations and conductivity requirements. The dc glow discharge is operated at approximately 1 kV with currents less than 50 mA. This yields operating powers 62
that are generally lower than ten wafts. Conversely, techniques such as laser ablation, inductively coupled plasma (ICP), and electrothermal vaporization (ETV),







47

















4~46







CC




0 0


o C


CC 84 U>
U'Q







48


operate in the kilowatt power regime. Some samples require the use of high power, which is not provided by the cool plasma formed by the glow discharge.

The electrical nature of the sample also presents a concern for a GD operated in the dc mode only conductive samples are amenable. This is a significant disadvantage compared to techniques such as laser ablation that can 24
readily produce atom populations of insulating materials. The conductivity requirement of the dc glow discharge is not a problem for the metallurgical industry, but many samples of analytical interest, such as glasses, soils, and ceramics, are inherently nonconductive. Two methods have been proposed to circumvent this limitation. One method is to incorporate a conducting matrix within the sample. The insulating sample is crushed into a powder and mixed with a conductive matrix. 95 This method, however, adds a time-consuming sample preparation step and can introduce impurities into the sample. A secondary conductive cathode can also be employed to analyze non-conductive samples with a dc GID source. In this technique, a conductive ring is placed above the non-conductive sample surface, 96
leading to sample sputtering.



Radio Frequency

The inability of the dc source to directly analyze nonconductive samples provided impetus to develop the radio frequency (rf) glow discharge. This source, originally proposed by Wehner, 97 uses a rapid oscillating power at typically 13.5 MHz to produce a net bias voltage that induces sputtering on insulating samples. Matching networks are needed to couple the rf power to the sample surface.







49


The response attained when a dc source is used to analyze an insulating surface is analogous to charging a capacitor. When a negative voltage is applied to an insulating sample, the potential on the surface initially drops, but then decays rapidly to more positive potentials due to ion neutralization reactions on the sample surface .98 A discharge formed on the sample surface will indeed be short-lived. The rapid oscillating power of the rf discharge, however, provides a source of electrons for ion neutralization while maintaining a negative potential on the cathode.","

Sample thickness is an important consideration for rf-operated GD sources, unlike other modes of operation. The thicknesses of an insulating sample can affect the amount of power that is coupled to the sample. Non-coupled (i.e., reflected) power can hinder the efficiency of an rf plasma. Reflected power can also turn the rf GD source into a radio transmitter, which can cause considerable electronic pickup if the source and other components are not properly shielded.

Radio frequency sources are able to analyze conductive samples, adding versatility to this mode of operation. The sputtered atom profile is similar to that of a dc source, which was represented in Figure 3-4. One disadvantage, however, is that the sputtering efficiency of an rf discharge is often less than obtained with a dcoperated source. The rf source yields less sputtering, and subsequently fewer sample photons and ions, than a dc-operated source for the analysis of conductive samples. While the reasons for this sputtering defIciency are not completely understood, it may be attributed to a lower average dark space potential, which draws argon ions onto the cathode (sample) surface.99 The sputter rate for insulating samples is typically an order of magnitude less than the corresponding







50


sputtering rate for a conductive sample.99 Additionally, the mechanisms responsible for excitation and ionization within the rf discharge are not fully understood. The cost and complexity of the rf source, in addition to its reduced analytical signals, are responsible for its minimal commercial application.



Pulsed

The previous operational modes have employed a continuous power regime; a steady-state discharge is formed, producing an continuous gradient of sputtered atoms, as shown in Figure 3-4. These techniques are limited by the power level that can be tolerated by the sample.99 A sample being analyzed via glow discharge must efficiently dissipate heat that is produced during the sputtering process. A steadystate source may not allow proper heat dissipation under extreme operating conditions. Small samples such as pins and wires can overheat, leading to thermal effects, such as melting.

One method of extending the glow discharge to higher power levels is by applying a transient voltage to the sample. Indeed, one thrust for developing a pulsed source was to combat the thermal effects that are encountered with steadystate sources. Pulsing an analytical source is not novel; laser and NMR devices are well-established pulsed sources that have inherent advantages. Likewise, the glow discharge can benefit from operating with a duty cycle, by gaining the ability to employ elevated potentials. The application of high transient voltages is possible only if the power is not supplied too long as to alter the sample. Piepmeir first reported enhanced emission intensities by pulsing a hollow cathode device.'00 This







51

initial foray paved the way for future pulsed glow discharge experiments by Harrison and colleagues in the mid-1970s.10'
38
The principle of a pulsed glow discharge source is shown in Figure 3-5. A high voltage, negative pulse is applied to the sample, resulting in a sputter release of atoms. The voltage is terminated and then applied again some microseconds later, producing another 'packet' of sample atoms. As these packets diffuse across the discharge chamber, they collide with other species, causing them to slow and diffuse. In a manner similar to the dc mode, the resulting photons and ions can be detected via optical emission and mass spectrometric techniques, respectively. The benefits of pulsing a glow discharge will be described in the next section of this chapter.

Concerns about applying a pulsed voltage to a sample may arise in the reader's mind. For example, the application of this intermittent power could lead to the formation of multiple unstable plasmas. However, the individual plasmas formed are just as robust as a steady-state plasma, given the proper operating conditions. Accordingly, the analytical signal produced by each pulse is reproducible. Indeed, some limitations of the pulsed glow discharge are encountered. Pulse generators are more complex and expensive than dc sources, and special equipment is needed to gate and collect transient signals. In summary, the advantages that are gained by pulsing a GD significantly offset the trivial disadvantages inherent with this exciting technique.







52

Microsecond Pulsed Glow Discharge Advantages

Operating a glow discharge with a transient voltage, as shown in Figure 3-5, affords advantages compared to continuous discharges. One advantage already described is reduced sample heating through heat dissipation during the off-time of the plasma. However, this solitary advantage is probably not sufficient to justify the additional requirements of a pulsed GD, such as pulsed high voltage supplies and gated electronics. Luckily, many other advantages are found with the pulsed mode of operation, which will be described in this section.
02,103 1 3 104
Two pulsed modes, the millisecond-' and microsecond-regmie, have been successfully implemented for GD application. Both regimes offer significant advantages compared to continuous operation, but the smaller duty cycle of the microsecond pulsed mode provides additional opportunities. Despite being a technique in its infancy, the microsecond pulsed GD has demonstrated advantages and possibilities that have propelled it to the forefront of glow discharge research in our laboratory. The pulsed advantages described in this section will consider only the microsecond pulsed regime; also, reference to the 'pulsed mode' hereto will refer only to the microsecond pulsed mode.



Additional Control Parameters

The operating conditions of the glow discharge that must be controlled, regardless of the power mode, are voltage, current, and pressure. Two additional control parameters are available to a GD operated in the pulsed mode: pulse width and frequency. The pulse width is the time (in gs) for which the voltage is applied to







53

the sample. Typical pulse widths are between 10 and 30 ;is. The frequency at which the intermittent voltage is applied to the sample is termed the pulse frequency. Pulse frequencies are chosen between 100 and 1000 Hz. Higher frequencies can be applied, though frequencies less than 100 Hz typically yield unstable plasmas.

The ability to control these two parameters offers an advantage for the microsecond pulsed mode. The sample removal rate can be altered by controlling the applied pulse width and frequency, while depth resolution is unaffected. For the analysis of thin films or multiple-layered samples, the ability to control the rate of sputtering is a significant advantage. Thin films that would be removed immediately with dc operation show extended sputter removal times with a microsecond pulsed glow discharge source and properly chosen conditions .60,81 90 The ability to control the sample removal rate for thin film analysis is demonstrated in Chapter 4.



Enhanced Sputtering Rates

The intermittent nature of the pulsed GID allows higher operating voltages than with a continuous, dc GD. This added benefit is a result of the reduced sample heating inherent of a pulsed discharge. Typical voltage thresholds are 600 to 1200 V and 1000 to 3000 V for a dc and microsecond pulsed glow discharge, respectively. Likewise, current levels are considerably higher for a pulsed discharge with nominal values from tens to hundreds of milliamps, compared to only tens of milliamps for comparable dc methods.

The application of higher voltages can result in an enhanced production of electrons, which in turn, increases the amount and efficiency of ionization, leading to







54

elevated ion signals. The higher potential also enhances the acceleration of impinging argon ions that are responsible for sputtering the sample surface, yielding enhanced sputter rates. The pulsed glow discharge has shown 14-fold enhancements in sputtering rates over the dc discharge when the time scale is
62,83
normalized. The average sputtering rate, however, will be lower for the pulsed system when factoring in the duty cycle of the pulse.



Enhanced Ionization and Emission

The pulsed glow discharge not only creates a higher population of sputtered atoms than a dc discharge, but also leads to higher energy species. The combination of these two factors can yield enhanced excitation and ionization processes, producing more photons and ions for atomic emission and mass spectrometry, respectively. Emission signals from glow discharges and hollow cathode lamps using the pulsed mode have displayed enhancements up to 3 orders of magnitude over their dc operated counterparts. 105 Similarly, GD mass spectrometry experiments have shown ion signal enhancements by employing a
37
transient source.



Temporal Resolution

Advantages of the pulsed discharge described thus far have given substantial reason to operate a glow discharge in a transient mode. Enhancements in (1) heat dissipation, (2) sputtering control, and (3) atomic and ion signals are all noteworthy







55

benefits. However, the most significant advantage of a pulsed glow discharge source is the ability to achieve time-resolved measurements.

An intermittent GD consists of repetitive, high-voltage pulses that create a cascade of physical and chemical interactions with each pulse. As voltage is applied to the cell, argon atoms break down electrically, yielding sputtering agents (i.e., Ar+ and Aro) that convert the solid sample into a gaseous mixture. A delay separates the initial formation (and detection) of these argon atoms and ions, and sputtering of the sample atoms and subsequent ionization. Typically, an induction period of 0.5 1 ms is found between the measurement of argon species and the initial 38
measurement of sputtered species. This induction period results in special temporal responses for different elements. Analytical advantages can be obtained by delaying the detection of these diverse temporal responses for each element. Temporal advantages have been obtained for atomic emission techniques and mass spectrometry, both of which are described in the next section.



Pulsed Glow Discharge Spectroscopies The temporal elemental differentiation resulting from time-dependent phenomena of the pulsed glow discharge can be beneficial for a variety of analytical techniques. Figure 3-6 illustrates analytical methods that have been used in conjunction with the pulsed glow discharge source. All of these techniques benefit from the pulsed glow discharge source to some degree. In the case of atomic






56









?t u0 E E
0
2 E
LU CL








..............







57


emission, the advantages are small, but potentially useful. For mass spectrometry, however, the gain is substantial.



Atomic Fluorescence/Absorption Spectrometry

Few sputtered sample atoms travel far from the cathode within a nominal pulse width of 10 or 20 gs, as collisions serve to impede their progression across the glow discharge cell. These atoms can achieve long lifetimes, often in the millisecond regime, as a result of these collisional events .38 At this extended timescale most of the non-sputtered species, such as electrons and argon ions, have been reduced, providing a minimal background for spectroscopic measurements in the post-pulse period. Atomic fluorescence measurements have shown signal-tobackground improvements by measuring the sputtered atoms after pulse termination.106 Likewise, atomic absorption measurements can benefit by taking measurements within the dark period of the plasma.



Atomic Emission Spectrometry

The use of a pulsed glow discharge offers potential advantages for atomic emission measurements. Figure 3-7 shows a representation of the temporal responses for various elements that are produced with a pulsed glow discharge, and 38
illustrates the temporal resolution possible with an atomic emission spectrometer. The initial emission response corresponds to the discharge gas, argon, which is detected early due to its photon release immediately after the plasma voltage is applied and the gas is broken down electrically. Emission from the sputtered sample







58










0
N










U) E
=L %*MOO E
2.
E 0

E
w .Y

m
c
m

L









Aipuaiui







59

atoms (e.g., copper and iron) appear later due to the time required for sputtering and excitation of these sample atoms. Emission from metal ions (e.g., copper ion) is detected last, since these species require additional time to undergo ionization prior to detection.

The temporal response for pulsed discharges have been employed for obtaining enhancements in the signal-to-background for selected elements in iron samples.8 In these studies by Bengtson and Yang, nitrogen emission originating from the samples was preferentially detected, while background nitrogen emission from the discharge gas was discriminated. This temporal detection was achieved by positioning a collection window (i.e., data gate), at an extended delay in which emission responses originate from the sample atoms and ions, as shown in Figure 3-7. The high instantaneous powers of the pulsed discharge also yielded a two-fold signal-to-background enhancement.2



Time-of-Flight Mass Spectrometry

Operating a glow discharge in the pulsed regime offers the biggest rewards for mass spectrometry. Temporal advantages have been shown for the a, millisecond pulsed discharge source coupled to a quadrupole mass spectrometer, in which calcium was detected in the presence of argon through time-gated detection.10 However, the largest benefit of a pulsed source can be obtained by coupling a microsecond pulsed GID to a time-of-flight mass spectrometer (TOFMVS). A TOEMS requires a pulsed introduction of ions and forms individual mass spectra based on a duty cycle of these pulsed events. The ion packets produced by a







60

microsecond pulsed GID are well-suited for the duty cycle of a time-of-flight, and the time-dependent events offer special opportunities.'14,38

Figure 3-8 shows the time-resolving principle that is possible with the combination of a microsecond pulsed GID and TOEMS. Argon ions, shown as red spheres in Figure 3-8, are formed first in a discharge, followed by the production of sample ions. After being sampled and skimmed, the ions are focused by ion lenses and enter a repelling region. This region redirects ions into the time-of-flight by applying a high positive voltage (typically 1 kV) to the back plate of the repeller. Extraction is unselective any ions in the repelling region during the voltage on-time may be sampled and detected. Controlling the repeller pulse timing allows discrimination of certain ions based on their temporal characteristics. Typically, argon ions are measured between delay times of 10 and 100 l~s, while sample ions are detected between 80 and 250 Pts. The magnitude of the delay time depends on the operating conditions and configuration of the GID source. Temporal resolution is achieved by operating at conditions that maximize the separation between the argon ions and sample ions as they enter the repelling region.

Figure 3-9 represents the time resolved spectra that can be obtained by varying the delay between the application of the repeller voltage and the plasma voltage.8 Within the first 100 jlts, ions extracted are representative of the background gas ions (e.g., argon, water, etc.). Analyte ions have yet to reach the sampling orifice due to the time required for diffusion after the sample is sputtered. As the delay time is increased beyond 100 pis, analyte ions








61























14

m E






0
.000 -000


E co E E





CL
m



06
(975 IU) ---T
CL U- OL
E
c







62







op


CN LO LM
m
LO + to
C3 9



co CL




+ R




loulhs Lo E

tun) U)

m
WE E
V
E






JOU61S Lo m
Irl leuBiS







63

(e.g., copper ions) begin to arise while discharge gas ions disappear. Typically, mass spectra are collected at optimum delay times to preferentially detect sample ions while potentially interfering gas species are discriminated. This ability is particularly important for low-mass elements, as shown in the subsets of Figure 3-9, which are prone to interferences from background contaminants. 37,38
















CHAPTER 4
QUANTITATIVE DEPTH ANALYSIS USING MICROSECOND PULSED GLOW
DISCHARGE ATOMIC EMISSION SPECTROMETRY Introduction

Glow discharge atomic emission spectrometry (GD-AES) is a mature spectroscopic technique, dating back approximately 100 years. However, modem glow discharge spectrometry is generally associated with the development of the Grimm source design in 1968 .80 A common application of this technique, surface and depth profile analysis, first done in 1973 by Belle and Johnson, 107 employed the Grimm-type glow discharge, the most commonly used GD emission source today.

Successful glow discharge depth profile analysis has been shown using both direct current (dc) and radio frequency (rf) sources, including quantitative analysis comparisons. 108,109 While rf sources have the advantage of being responsive to conducting and nonconducting samples, the majority of depth profile work has been performed using de sources. A third type of source is the pulsed glow discharge source, which has certain potential analytical advantages over a steady-state dc source, including enhanced control of sputtering rate, increased emission intensities, and possible temporal factors.

Because the sputtering process can be altered by controlling the pulse frequency and pulse width, extension of the pulsed technique to depth profiling affords more flexibility in thin layer analysis. Layers of a few atomic distances can be removed over an extended time scale by using a lower duty cycle that provides a


64







65

slower sputtering process. This research shows the application of a microsecond pulsed glow discharge for quantitative depth profile analysis of thin foils. The quantification method is based on a similar method done by Bengtson and colleagues with a dc GD source, a technique applied here to the pulsed GD source.



Experimental

Instrumentation

A standard Grimm GD source (LECO Corp., St. Joseph, Michigan, USA) was used for all experiments. The tubular-shaped anode (4 mm W.) of the source extends to approximately 0.2 mm from the sample. The spectrometer was a slightly modified LECO SA-2000 direct reader, which uses a 0.4 m direct-reading spectrometer with a 2400-groove holographic grating. A picture and schematic of the microsecond pulsed Grimm GD-AES is shown in Figure 4-1 and Figure 4-2, respectively.

The Grimm glow discharge was operated in the pulsed mode using a highvoltage pulsed power supply (Model: M3k-20-N, IRCO, Columbia, Maryland, USA). This power supply is capable of internal pulsed operation, but for greater flexibility, a separate pulse generator (Model: 8003A, Hewlett-Packard, Palo Alto, California, USA) was used to trigger both the pulse power supply and the boxcar integrators to synchronize the firing of the discharge and resultant signal collection. A pulse counter (Model: 5740, Data Precision Corp., Wakefield, Massachusetts, USA) was added to monitor the number of pulses from the power supply. The signal readout and data system of the LECO spectrometer was altered to permit incorporation of






66






























AWW IN


Figure +1. Photo of the microsecond pulsed Grimm glow discharge atomic emission spectrometer.







67














x



21 7FL





r

7ij

E CL
C v E
"5 =
CL

E o





7r-;







68

the microsecond pulsed signal. The output currents of the photomultiplier detectors were converted to voltages by dropping the signal across a 66.3 k() resistor. The signals were processed in two ways. The first system consisted of a 2 Gsals, 400 MHz, four-channel digital oscilloscope (Hewlett Packard, Palo Alto, California, USA). The output signal, discharge voltage and discharge current were displayed simultaneously on the oscilloscope, which afforded plasma stability monitoring. The voltage was attenuated through a 1000:1 voltage probe, while the discharge current was determined by calculation from the peak voltage dropped across a 5.1 0) carbon resistor.

The second readout system consisted of gated boxcars and a computer. The signals from the photomultiplier tubes were temporally gated and processed with a boxcar integrator (Model: SR250, Stanford Research Systems, Sunnyvale, California, USA), and then sent to a computer via an analog-to-digital converter (Model: SR245, Stanford Research Systems, Sunnyvale, California, USA). The timing diagram of the system involves a waveform generator to control both the pulsed power supply and the boxcar integrators.



Sample Material

Sputter rate protocol. Samples used for obtaining sputter rates were high purity elemental foils approximately 0.1 mm thick (Puratronic, Alfa AESAR, Ward Hill, Massachusetts, USA). Samples were cut into small discs just larger in diameter than the 0-ring surrounding the anode, approximately 15 mm, and weighed using an analytical microbalance (Model M2P, Sartorius Corp., Edgewood, New York, USA)







69

prior to analysis. The thin nature of the films necessitated sample cooling, which was achieved using a cooling puck option that was available on the LECO spectrometer. The foils of lead and tin required modification prior to sputtering. These samples were so soft that, once put under vacuum, contact was made with the anode, thereby shorting out the discharge. This was overcome by adhering the sample disc to a less malleable sample, e.g., a copper disc of equal size, prior to analysis.

The samples were sputtered for 40 minutes and then re-weighed to determine the weight difference due to sputtering. Although sputtering rates could be determined using the net "sputter time," it was typically more precise and more accurate to measure the number of pulses from the power supply via a pulse counter. Each sample was analyzed three times, using a new sample for each run.



Additional samples. Additional samples were employed in the quantification study. Bulk materials used in generating the calibration curve were iron standard samples (spectrometric reference materials, cast iron C18.8, CKD232-239, Research Institute CKD, Prague, Czech Republic) as well as NIST 1107, 1110, and 1113 brass standard samples (National Institute of Standards and Technology, Gaithersburg, Maryland, USA). Samples used to show the quantification method were pure copper films deposited on polished iron substrates and were prepared in the Department of Materials Science and Engineering, University of Florida. Copper and iron were the only elements of significant mass fraction in the sample; the purities are 99.5% copper and 99% iron, respectively.







70

Results and Discussion

Optimization of Conditions

Glow discharge quantitative depth profiling depends on the sample material being removed uniformly layer by layer across the area exposed to sputtering by argon ions and fast atoms. Modeling of excitation processes have shown that both operating voltage and pressure are responsible for erosion uniformity, while the discharge current determines the number density of bombarding particles and the sputtering rate. Sputter induced effects, such as surface roughening and sample mixing, are common to all surface techniques employing sample erosion for depth analysis, including glow discharge atomic emission spectrometry and secondary ion mass spectrometry.

The formation of crater curvature by non-uniform erosion due to poorly chosen parameters tends to degrade measured intensity-time curves and consequently decreases the accuracy of the quantitative measurement of film thickness. The depth resolution is limited largely by the curvature of the sputtered crater and its roughness. In the case of secondary ion mass spectrometry, the surface roughness can be improved by('rastering,,the ion beam across the sample surface; however, mastering is not possible with glow discharge analyses. Therefore, it is important to use optimum experimental conditions that sharpen the intensitytime profiles, thereby enhancing the depth resolution. The literature shows reports regarding the influence of operating parameters on crater shape for both the dc GD' 10-113 and the rf GD. 114 More recently, our group has studied this influence using the microsecond pulsed Grimm glow discharge,90 which shows some significant







71

deviations from the characteristics of a dc GD. For example, the depth resolution of micron sample layers analyzed via dc GD sources is not significantly affected by surface roughness. Conversely, the analysis of sub-micron sample layers, such as via our pulsed glow discharge, requires that the crater curvature and surface roughness be carefully controlled.




Effect of voltage and pressure. For the case of the microsecond pulsed glow discharge, we see that high operating pulse voltages and pressures yield an increased sputtering rate, while low voltages and pressures yield decreased sputtering rates. Therefore, controlling the discharge power is critical to the crater shape and depth resolution. Figure 4-3 shows the effects of three different operating powers on crater profiles.90 The respective peak discharge settings were: (a) a low power regime of 100 W (800 V, 0.12 A, 3 torr); (b) a medium power regime of 270 W (1200 V, 0.22 A, 4 torr); and (c) a high power regime of 680 W (2000 V, 0.35 A, 3 torr). The sputtering time of the three trials was slightly varied to normalize the depth of sputtering.

As shown in Figure 4-3 (a), concave crater shapes, those that have an inherent dip in the middle of the profile, result from sputtering at low power (voltage). Conversely, convex crater shapes, which have a bulge in the middle, are the result of employing high power (voltage), as shown in Figure 4-3 (c). Only a mid-range pulse voltage yields a relatively flat crater bottom, which is indicated in Figure 4-3

(b). For example, the optimum pulse voltage is between 1200 V and 1400 V for a 3 torr argon gas discharge. The same trend between power and crater shape was







72




800 a
a

600

400 200


0
800


600











400

200


0
-3 -2 -I 0 I 2 3
Radius/mm Figure 4-3. Crater profiles for different operating conditions: (a) 100 W; (b) 270 W; (c) 680 W.90







73

observed with a dc discharge,110 except the range of voltages is higher in the pulsed mode. A possible explanation for this trend is that the crater shape results from the electric field focusing or defocusing the bombarding ions responsible for sputtering.90

Figure 4-4 shows the effect operating voltage can have on crater profiles.90 Figure 4-4 (a) is a magnified digital image of a sputter pattern of a copper layer on steel at low voltage (800 V, 3 torr, 10 [ts pulse width, 200 Hz). The image has been digitally altered to enhance the contrast between the copper and steel layers. The dark circular region in Figure 4-4 (a) marks the edge of the deep sputter erosion area. The lighter surface represents the copper top layer, whereas the gray center region illustrates the exposed steel layer. The low power conditions result in a concave crater, as shown in Figure 4-3 (a), and also a small emission intensity due to a lower sputtering rate.

At high pulse voltages (e.g., 2000 V), the sputter rate and emission intensity are increased. However, the crater shape shows the underlying layer to be exposed asymmetrically at the edge of the crater first. Figure 4-4 (b) is a digital image of a sputter region under the same operating conditions as in Figure 4-4 (a), except at a higher pulse voltage (i.e., 1800 V). A contribution from the underlying layer is observed early, while significant amounts of the top layer are still present in the center of the discharge area, as shown in Figure 4-3 (c). This awkward crater shape extends the signal overlap of the two layers, rather than the desired sharp transition. Observing the sample crater under a simple optical microscope, the contrast between the copper and iron layers can be clearly followed as a function of discharge conditions. Since both high and low voltage discharges produce higher






74






































Figure 4-4. images of burn spots for a copper layer on a steel substrate: (a) low voltage at 800 V; (b) high voltage at 1800 V-90







75

sputtering in one localized region and reduced sputtering in another, the signals from each are correspondingly unsatisfactory. The cause of poor depth resolution is the non-uniform erosion under very high or very low operating power.



Effect of pulse width and pulse frequency. Pulse widths and pulse frequencies had little influence on crater shapes, as long as the pulse voltage and gas pressure conditions remained constant. The effect of pulse width and pulse frequency on the sputter rate of a brass sample is shown in Figure 4-5 and Figure 46, respectively. The results show that depth resolutions at different pulse widths and frequencies are essentially constant under the conditions used.

The main advantage of these two pulse parameters lies in their ability to increase or decrease the sputtering rate without altering the depth resolution. Because of the reduced duty cycle of the pulsed source, less material can be removed per unit time. Thus, a wide range of layers from several nanometers to tens of micrometers can be analyzed. The optimum pulse width and pulse frequency conditions for microsecond pulsed depth profiling of the samples involved in this research were 10 gs and 400 Hz, respectively. These conditions were chosen because they allow a significant amount of sample material to be sputtered for the given analysis time.



Effect of sputtering time. The final parameter to be studied was the time allowed for the sample to sputter. In theory, time should have no influence on sputtering rates of pure samples since sputtering will simply penetrate deeper into a








76














3.50


3.00

2.50

CL
a.
CO 2.00 U
a
.4-I
1.50 aL
a.
c/ 1.00


0.50


0.00
0 20 40 60 80 100 120
Pulse Width (s)


Figure 4-5. Effect of pulse width on sputter rate (brass sample,
1.4 kV, 3 torr Ar, 400 Hz pulse frequency, 20min.).









77














1.60


1.40 1.20 c. 1.00 .~0.80 0.60


0.40 0.20 0.00
0 100 200 300 400 500 600 700 800 900
Frequency (Hz) Figure 4-6. Effect of pulse frequency on sputter rate (brass
sample, 1.4 kV, 3 torr Ar, 10 jis pulse width, 20 min.).







78

sample with longer analysis times. However, if the increasing depth should affect net diffusion of atoms from the surface, then calculated sputter rates could be affected. Figure 4-7 shows a plot of the sputtering rate dependence on time using a brass sample consisting of 70% copper and 30% zinc (Alfa AESAR, Ward Hill, Massachusetts, USA). No significant change in sputtering rate with increased analysis times is observed, as shown in this figure. Uncertainty measurements were obtained through triplicate analysis for each sputter time and plotting the standard deviations. It is possible that longer analysis times could degrade the quality of the crater profiles due to differential sputtering of the crater sides with respect to the bottom of the crater. For the current study a fixed time of 40 minutes has been used for all sputter data studies. Also, it should be noted that other sample types, such as pure copper and pure zinc foils, have been shown to yield similar parametric trends as those reported above for a brass thin film.90



Sputtering Rates and Penetration Rates

The net sample sputter rate q (ng/pulse), as described in Chapter 2, can be calculated for a GD using:

q = AW1 (t AWIPN (Eqn. 4-1)

where AW (ng) is the amount of sample lost due to net sputtering, t (s) is the sputter time, f (Hz) is the pulse frequency. If sputtering time is measured, the first part of Equation 4-1 is used in which the number of pulses must be calculated. However, a pulse counter can be used to directly monitor the number of pulses (PN pulses) from the power supply, making the second part of Equation 4-1 applicable. Furthermore,








79













1.4 1.2



0.

.O 0.8


0.6


0.
yn 0.4


0.2


0
0 10 20 30 40 50 60 70
Sputter Time (mins) Figure 4-7. Sputtering rate dependence as a function of total sputtering time (brass sample, 1.4 kV, 3 torr Ar, 10 gs pulse width, 400 Hz pulse frequency).
Error analysis (standard deviation) obtained through triplicate measurements.







80

using a pulse counter should yield a more accurate and precise sputtering rate due to limitations of measuring sputter time, since some fractional time in each pulse is required for the plasma to stabilize, making it difficult to measure the "true" sputter time.

To obtain a quantitative depth profile it is necessary to determine concentration as a function of depth. This can be accomplished by converting the sputtering rate of the sample to its penetration rate (w nm), which is the amount of sample eroded per unit time and is dependent on the density of the sample. As described by Boumnans,6 the penetration rate has units of distance per time (limfsec). However, in the present study, the units are changed to distance per pulse (nm/pulse) to account for the use of a pulsed glow discharge source. The penetration rate can be shown quantitatively by: w=q/(A *p) (Eqn. 4-2)

where p (g/cm2) is the density of the target material, and A (cm2) is the sputtered area, 7tr2, where r (cm) is the radius of the Grimm anode (typically 2 mm). Sputtering rate values and corresponding penetration rates for the various pure elements studied in this project can be found in Table 4-1. The values in this table are based on three trial runs with the standard deviation included to show uncertainty.

As seen in Equation 4-2, the penetration rate has a direct dependence on the density of the material being analyzed. This dependence does not present a problem for determining the penetration rate of pure elements. However, determining the sputtering depth of a multi-layered sample presents a concern. As one layer of the sample is sputtered away and the next layer is exposed, the







81







Table 4-1. Sputtering rates and corresponding penetration rates for thin films of brass and various pure elements. Based on three sample runs with the standard deviation included to show uncertainty. Sputtering conditions: 1400 V, 3 torr, 10 gs pulse width, 400 Hz frequency, and a sputtering time of 40 min.


Element Sputtering rate (q) Penetration rate
ng per pulse (nj / nm per pulse

Brass 0.83 0.01 0.0078 0.0001


Iron 0.12 0.02 0.0012 0.0002


Nickel 0.268 0.006 0.00239 0.00005


Copper 0.67 0.01 0.0059 0.0001


Zinc 1.74 0.01 0.0194 0.0001


Silver 2.06 0.03 0.0157 0.0002







82

interface involves a change in density of the pure element comprising the first layer to that of the density of a second layer, along with changes in sputtering rate and concentration of the two elements. If the transition from one layer to the next were instantaneous, this concern would not exist. However, in the interface between layers, sample material from both layers is being removed, and the sample will take on multi-element characteristics, rather than acting as separate pure layers. Hence, a quantification method must take into account this interface concern.



Calibration Curve Correction

A variety of bulk reference standards is used to quantify a depth profile because large concentration ranges in different matrices are typically required. For example, when analyzing a layered sample, the element of interest must be calibrated over the full concentration (0 to 100%), because of the concentration range created during the transition. However, sputter rates are unique due to matrix differences, producing a different calibration curve for each family of materials. To correct for this, the sputter rate for each standard can be used to correct for inherent differences, permitting a single calibration curve. The method of quantification described in this dissertation requires accounting for these sputtering rate differences, which makes this correction step crucial to the overall quantification method.

An intensity-concentration diagram for copper (copper 11 219.227 nm emission) is shown in Figure 4-8. The copper 11 219.227 nm emission line was chosen due to the self-absorbing nature of the copper 1327.396 atom line. Figure 4-







83


8 is a calibration curve for 12 copper samples, 8 in an iron matrix, 3 in a zinc matrix, and one a pure copper sample. A non-linear response of copper intensities is shown, resulting from matrix-dependent sputtering rates. The resultant calibration slope (Figure 4-8) would yield erroneous copper concentrations based on the uncorrected intensities. The quantification method in this dissertation relies on correction of these intensity differences.

The procedure used for correcting intensity differences due to varying sputtering rates has been previously reported,' 15 and requires the use of a variety of calibration samples, each having different sputtering rates. In this previous publication, which is based on the Swedish Institute of Metals Research (SIMR) calibration model (described below), the intensity correction relies on determining the sputtering rates of an iron alloy. In the current research, intensity correction employs the sputtering rate of pure iron, which can be directly determined through weight loss measurements. As long as the calibration for each element is normalized to the same reference sputtering rate, the reference sample (i.e., pure iron or an iron alloy) will not result in a discrepancy. Intensity correction in the current research is done by multiplying the measured intensities of each calibration sample by a factor equal to the ratio of the sputtering rate of the reference material to that of the sample. Mathematically this relationship is given by: Inorm = / (qr,,f / q,) (Eqn. 4-3)

where /,,,, is the normalized intensity, I is the "raw" intensity, and qef and q, refer to the sputtering rates of the reference material (e.g., iron) and sample, respectively.







84

Figure 4-9 shows a sputter rate-corrected calibration plot, using the results of Equation 4-3, for the non-linear copper calibration curve shown previously in Figure 4-8. Using this corrected calibration curve affords better estimations of unknown copper concentrations based on intensity measurements. A similar study was done using nickel alloys, with comparable results being obtained, as shown in the uncorrected calibration curve of Figure 4-10 and sputter-rate corrected curve of Figure 4-11, respectively. A marked advantage of this calibration procedure lies in the fact that bulk standard materials can be used, not requiring high-cost, less readily available layered samples. Also, the analytical procedure is simplified by only requiring one reference sputtering rate to be determined, rather than multiple reference material sputtering rates. After a linear calibration curve is obtained, the information can be used for the subsequent quantification procedure.



Quantification Overview

When performing depth analysis, the data obtained typically represents intensity-time profiles. However, the information desired is in the form of depthconcentration profiles since more detailed information can be extracted from such a plot. An example of this added information is that, for a given sample, the concentration of a thin layer in the sample can be monitored at varying depths of the sample. The time to depth conversion procedure involves knowledge of the sputtering rates for each element and calibration of the instrument from a series of standard samples. This qualitative to quantitative conversion gives a better estimation of the true depth profile.










85



10

9 Cu:Zn
(3 samples)

8
70


Pure Cu E (I sample)
5
N
o 4 y : 0.0849x + 0.1038
3 -32 R2 0.9626 Cu:F&
(I samples)
2




0 10 20 30 40 50 60 70 80 90 100
Copper Concentration (%)


Figure 4-8. Plot of uncorrected calibrated data for various copper
samples (Cu 219.227 nm emission) in different matrices.



1.4

1.2 Cu:Zn
(3 samples) Pure Cu 1( 0amPie) 'A
c
S 0.8
E
c
NY = 0.0127x + 0.0329 N0.6
R' = 0.9997
N
0 0.4 Cu:Fe
(8 semples)

0.2


0
0 10 20 30 40 50 60 70 80 90 100
Copper Concentration (%)

Figure 4-9. Plot of sputter rate-corrected copper calibration
data of Figure 4-8.









86



3.5 3


~2.5

0
2
E
S1.5 2=087
CM'




0.5

0
0 2 4 6 a 10 12 14
Nickel Concentration(% Figure 4-10. Plot of uncorrected calibrated data for various
nickel samples (Ni 225.35 nm emission) in different matrices.


14


12



0


E

N) y =0.8583x + 0.2848
N% R 2 = 0.9972







0J
0 2 4 6 8 10 12 14
Nickel Concentration(% Figure 4-11. Plot of sputter rate-corrected nickel
calibration data of Figure 4-10.







87

Two prominent quantification techniques have been described in the literature; the IRSID (Institute de la Recherch6 Siderurgie) method and the SIMR (Swedish Institute for Metals Research) method. The IRSID method, first introduced by Pons-Corbeau in France, 116,117 uses the emission yield concept as a basis for quantification of depth profiles. This method relies on the assumption that the integrated intensity from one element is proportional to the sputtered mass of that element, which in turn implies that the emission yield, defined as the emitted light per unit sputtered mass of an element, is independent of the sample matrix provided 86
that the excitation conditions remain at least nearly constant. Most researchers now believe that this assumption is valid at least to a first approximation.

Another quantification method was developed by SIMR.' 15 Unlike the IRSID method, the SIMR quantification method is not directly based on the emission yield concept. Rather, it is based on a sputtering rate-intensity proportionality. Essentially, the difference between the quantification techniques lies in the fact that in the SIMR method, the concentration, instead of the sputtered mass of each element, appears in the calibration function, making the SIMR method similar to standard techniques for bulk analysis. Measurements are based on a calibration at pre-determined levels of voltage and current, referred to as the reference excitation condition. One of the main goals of the SIMR quantification method is a calibration procedure that resembles a standard bulk analytical method.

In both quantification methods a steady-state dc glow discharge was employed. However, the pulsed glow discharge in our work adds two additional parameters, pulse width and pulse frequency, that offer additional control over the








88

sputtering process. As described previously, extension of the pulsed technique to depth profiling should increase performance by allowing thin sample layers of a few atomic distances to be measured more easily by selecting a slower sample removal rate. Furthermore, reduced sample heating by the pulsed discharge compared to dc operation allows certain thin layer samples to be analyzed. Since these two additional parameters are inherent in the pulsed glow discharge, the quantification method must be modified to incorporate them. The quantification method chosen for this research is based on the SIMIR method, though it should be noted that slight modifications to this model have been made, such as the use of different reference sputtering rates.



Principle of Quantification

The first step in the quantification method, after acquiring sputtering rates and calibrating the instrument, is to collect a raw intensity plot from the layered sample. An intensity-time plot of a 50 nm layer of copper deposited on a steel substrate can be seen in Figure 4-12. This plot shows the intensity of the copper 11219.227 nm ion line, as well as the Fe 1371.993 atom line. The data in this plot only show intensity as a function of time. Therefore, quantitative results, such as depth, cannot be extracted from the plot as it is shown.

The next step in the quantification method involves determining the selected elements' concentrations. This is done by converting the intensity measurements shown in Figure 4-12 to concentrations using the sputter rate-corrected calibration







89














g 1.4
.Cu219
7 219 Fe371 1.2


6
- 0
C C
0

-0.8
E E
c 4-'C
N
3-3 '
N
2 0.4 .
0 2


-10.2


0 0
0 50 100 150 200 250 300
Time (s)


Figure 4-12. Intensity-time plot of a 50 nm layer of
copper deposited on a steel substrate.







90

curve of the bulk standards shown in Figure 4-9. These concentrations are then normalized to 100%, as shown in Figure 4-13.

The third step in the method is to determine the composite sputter rate and composite density as a function of time from the normalized concentrations. The composite sputter rate is simply a summation of each element's percent contribution of the reference material's sputtering rate over all the elements in the sample. Similarly, the composite density is a summation of each element's percent contribution of the reference material's density over all elements in the sample.

After the composite sputtering rates and composite densities have been determined, they can be used in collaboration with the previously determined sputter area and sputter time to obtain the penetration depth (d). The sputtered depth increment (Ad) at a given point in time can be obtained by multiplying the penetration rate (Equation 4-2) at this point by the number of pulses per sampling point: Adt = Wt *PN (Eqn. 4-4)

where, Adt (nm) is the sputtered depth increment, wt (nm/pulse) is the penetration rate at that point, and PN (pulses) is the number of pulses per sampling point. The overall penetration depth (d) can be obtained by summing all sputtered depth increments (Eqn. 4-4) for each sampling point over the entire sampling range:

n
d=EAd (Eqn. 4-5)
1=0

Based on Equations 4-4 and 4-5, the time axis (s) of Figure 4-13 can be converted to depth (nm) for the entire sampling depth (150 nm). Therefore, the final