Analysis of lightweight gases by quadrupole ion trap mass spectrometry for the safety of the American Space Shuttle program

MISSING IMAGE

Material Information

Title:
Analysis of lightweight gases by quadrupole ion trap mass spectrometry for the safety of the American Space Shuttle program
Physical Description:
Book
Creator:
Ottens, Andrew Keith, 1976-
Publication Date:

Record Information

Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 30311373
oclc - 52619409
System ID:
AA00022322:00001

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
    List of Tables
        Page ix
        Page x
    List of Figures
        Page xi
        Page xii
        Page xiii
        Page xiv
        Page xv
    Abstract
        Page xvi
        Page xvii
    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
        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
    Chapter 2. Monitoring lightweight gases for space shuttle safety
        Page 32
        Page 33
        Page 34
        Page 35
        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
        Page 64
    Chapter 3. Lightweight gas monitoring by QITMS
        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
        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
    Chapter 4. Ion-molecule reactions: A limitation of QITMS
        Page 128
        Page 129
        Page 130
        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
        Page 162
        Page 163
    Chapter 5. Optimizing QITMS for lightweight gas analysis
        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
        Page 198
        Page 199
        Page 200
        Page 201
    Chapter 6. Performance evaluation of QITMS
        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
    Chapter 7. Conclusions and future directions
        Page 220
        Page 221
        Page 222
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
    Appendix A. Code for mass calibration
        Page 229
        Page 230
        Page 231
    Appendix B. Code for real-time data plot
        Page 232
        Page 233
        Page 234
    Appendix C. Code for polarisq AHGD scan
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
        Page 242
        Page 243
    Appendix D. Code for AHGD optimized UP-IT scan matrix
        Page 244
        Page 245
        Page 246
        Page 247
        Page 248
        Page 249
    Appendix E. List of acronyms and abbreviations
        Page 250
        Page 251
        Page 252
    List of references
        Page 253
        Page 254
        Page 255
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
        Page 262
        Page 263
    Biographical sketch
        Page 264
        Page 265
        Page 266
        Page 267
        Page 268
Full Text









ANALYSIS OF LIGHTWEIGHT GASES BY
QUADRUPOLE ION TRAP MASS SPECTROMETERY
FOR THE SAFETY OF THE AMERICAN
SPACE SHUTTLE PROGRAM













By

ANDREW KEITH OTTENS


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


2003
































To Tamara, to my parents, and to my brother for always providing love and support.














ACKNOWLEDGMENTS

I express my utmost thanks to my parents, Paul and Ann Ottens, for supporting me

through all that I have done always remember their advise to stay in school as long as I

can, and to stand back-up whenever life knocks me down. I thank my brother Richard

whose dedicated interest in NASA inspired my research. I thank my entire family for

their continued love and support.

I sincerely thank my fiancee Tamara Blagojevi6 for her dedicated support and help in

writing this document. Her unconditional love brightens my life. For all the little things

that seemingly went unnoticed, I am forever grateful.

Special thanks go to my good friends Kevin McHale and Josh Coon. Kevin's open

kindness to me as a stranger helped bring me to the University of Florida. He was always

available to lend a hand or to debate a topic; I too will remember the pint-o-Killians talks.

Josh's friendship made the laboratory bearable, despite the awful noise of his ever-

expanding mass spectrometer. From countless lunch runs to exhaustive conversations,

his comradeship kept me on track. I also acknowledge my past and present group

members for their support over the years.

I express my gratitude to Drs. Richard A. Yost, David H. Powell, and Jodie V.

Johnson who have been helpful in their understanding of mass spectrometry throughout

the years. I also thank the staff of the business office, the electronics shop, the IT shop,

and the machine shop. These people make the chemistry department run. I give special








thanks to Steve Miles who on many occasions dropped what he was doing to help me

resolve complex electronics issues

At the Kennedy Space Center, I sincerely thank Drs. Timothy P. Griffin and C.

Richard Arkin for their professional support and friendship. Tim was my mentor through

much of my graduate school career to him I am utterly grateful. Richard as a fellow

chemist brought a sense of sanity to the engineering world of the space center. Tim and

Richard are truly great people.

I would also like to thank the people of Dynacs Engineering for their support of my

research: specifically, Guy Naylor, Bill Haskell, Charles Curly, and David Floyd of the

HGDL lab were an immense help during my time at the space center. I wish to

acknowledge those from NASA (William R. Helms, Fredric W. Adams, and Duke W.

Follistein) who helped provide project support and make my KSC experience enjoyable.

Many thanks go to Scott Quarmby and Brody Guckenberger of ThermoFinnigan who

have been immensely helpful with technical issues and in contributing hardware,

software, and documentation for the QITMS instrumentation used in this research.

Finally, I extend my sincere gratitude to my research advisor Dr. W. W. Harrison,

who despite our late introduction was an immense help in the successful completion of

my doctoral studies. He has served as a steady guide, and as a role model with his

professional prowess.














TABLE OF CONTENTS
Raeg

A CKN O W LED G M EN TS ................................................................................................. iii

LIST O F TA BLES ........................................................................................................ ix

LIST O F FIG U RE S ....................................................................................................... xi

A B STRA CT ..................................................................................................................... xvi

CHAPTER

1 IN TRO D U CTION ........................................................................................................... 1

Theory of Ion Motion within Quadrupole Devices .................................................... 2
Q uadrupole Electric Field ................................................................................... 2
Ion M otion in Q uadrupole D evices ...................................................................... 3
D eriving the M athieu Equation ............................................................................. 7
Stability D iagram ............................................................................................... 10
Pseudopotential W ell Theory ............................................................................. 18
M odes of Quadrupole Ion Trap O peration ............................................................... 22
M ass-Selective D etection .................................................................................... 23
M ass-Selective Storage ..................................................................................... 25
M ass-Selective Ejection ...................................................................................... 27
Collision G as ...................................................................................................... 28
M ulti-Purpose A uxiliary W aveform s ................................................................. 28
O verview of D issertation .......................................................................................... 31

2 MONITORING LIGHTWEIGHT GASES FOR SPACE SHUTTLE SAFETY .......... 32

Introduction ................................................................................................................... 32
Present D ay G as D etection .................................................................................. 35
A dvanced H azardous G as D etection Project ..................................................... 38
AHGD Leak Detection System Requirements ................................................. 38
Experim ental M ethodology and Equipm ent ............................................................. 40
A H G D Test Procedure ........................................................................................ 40
A nalytical Procedures ........................................................................................ 44
Tw o-point calibration ................................................................................. 44
Calculated detection lim it .......................................................................... 44
A ccuracy of quantification ........................................................................... 44
Precision of quantification .......................................................................... 45








Response tim e ............................................................................................ 45
Recovery tim e ............................................................................................ 45
Gas Delivery Assem bly ...................................................................................... 45
Evaluated M ass Analyzers ................................................................................. 47
Linear quadrupole ........................................................................................ 47
Quadrupole array ....................................................................................... 48
Tim e-of-flight ............................................................................................ 48
Cycloidal focus .......................................................................................... 49
Results of the Exam ination ...................................................................................... 49
Detection Lim its ................................................................................................. 50
Accuracy ................................................................................................................ 55
Precision ................................................................................................................. 57
Response and Recovery Tim e ............................................................................. 57
Conclusions and Consideration for Future W ork ..................................................... 57

3. LIGHTWEIGHT GAS MONITORING BY QITMS ............................................... 65

Introduction ................................................................................................................... 65
History of QIT with Hydrogen and Helium Ions ............................................... 65
M odem QITM S ................................................................................................. 66
Instrumentation and Equipment for Lightweight Mass Analysis ............................ 69
University of Florida Custom QITM S (UF-IT) ................................................. 69
M ass analyzer, source, and optics ............................................................... 69
Vacuum system .......................................................................................... 71
Control, source, optics, detector, and acquisition electronics ..................... 76
Drive circuitry ............................................................................................. 80
Chassis and power supplies ....................................................................... 84
Custom ized software .................................................................................... 84
Comm ercial QITM S (PolarisQ) ........................................................................ 85
Hardware differences between QITMS systems ........................................ 85
Drive circuitry ............................................................................................. 89
Custom Software ........................................................................................ 94
Tuning the RF Circuit ........................................................................................ 96
Gas Dilution System (GDS) ............................................................................... 99
Instrum ent Developm ent Discussion .......................................................................... 101
GD S Evaluation ................................................................................................... 101
Results of UF-IT A ssem bly ................................................................................. 106
Changing the drive frequency ........................................................................ 109
Pum p difficulty with hydrogen ...................................................................... 114
Exploring the PolarisQ ......................................................................................... 116
Collision gas and external ionization ............................................................. 116
Injection waveform s ....................................................................................... 118
Resonant ejection ........................................................................................... 121
PolarisQ m odified for lightweight gas analysis ............................................. 123
Detecting hydrogen ions with the PolarisQ ................................................... 123
Conclusions ................................................................................................................. 125








4 ION-MOLECULE REACTIONS: A LIMITATION OF QITMS ............................... 128

Introduction ................................................................................................................. 128
Types of Ion-M olecule Reactions ........................................................................ 129
Charge-exchange reactions ............................................................................ 130
Proton transfer reactions ................................................................................ 130
H ydrogen-atom transfer ................................................................................. 132
K inetics of Ion-M olecule Reactions ..................................................................... 132
Therm odynam ics of Ion-M olecule Reactions ...................................................... 135
Equipm ent and M ethods ............................................................................................. 136
Gas Delivery Setup .............................................................................................. 136
N um ber Density D eterm ination ........................................................................... 137
Custom Scan Functions ........................................................................................ 139
Ion-M olecule Reaction Results ................................................................................... 143
H e+ Reactions ...................................................................................................... 143
H 2+ Self-Protonation and Other Reactions .......................................................... 151
N 2+' Reacting w ith H20 ........................................................................................ 155
Ar+ Reactions ....................................................................................................... 157
Therm odynam ics of O2+ and H30 ....................................................................... 158
Conclusions ................................................................................................................. 161

5 OPTIMIZING QITMS FOR LIGHTWEIGHT GAS ANALYSIS ............................. 164

Introduction ................................................................................................................. 164
Conditions for Ionization ..................................................................................... 164
Electron energy .............................................................................................. 164
Tim e and pressure .......................................................................................... 165
Low -m ass cutoff ............................................................................................ 165
Space charge effects ....................................................................................... 167
Ion Storage in the QIT .......................................................................................... 168
Ion Ejection from the QIT .................................................................................... 168
Ejection by nonlinear resonance .................................................................... 169
M ass resolution controlled by ejection rate ................................................... 174
Specialized scan routines ............................................................................... 174
Experim ental Details ................................................................................................... 175
Sam ple Gas .......................................................................................................... 175
Instrum entation .................................................................................................... 176
Softw are ............................................................................................................... 176
Results and Discussion of QITM S Optim ization ........................................................ 177
Ionization of the QITM S ...................................................................................... 177
Electron optics ............................................................................................... 177
Ion trap pressure versus ionization tim e ........................................................ 178
The optim al q, for ionization ......................................................................... 183
Interm ediary Scan Function Activity ................................................................... 186
Ion Ejection and A cquistion ................................................................................. 188
Post scan activity ............................................................................................ 190
N onlinear resonance ....................................................................................... 190








High-resolution .............................................................................................. 194
Conclusions ................................................................................................................. 198

6 PERFORM ANCE EVALUATION OF QITM S ......................................................... 202

Introduction ................................................................................................................. 202
Experimental ............................................................................................................... 203
Analytical M ethodology ....................................................................................... 203
Sample Delivery ................................................................................................... 204
Instrumentation .................................................................................................... 206
Results and Discussion of QITM S Perform ance ........................................................ 206
Custom Built QITM S ........................................................................................... 207
M odified Comm ercial QITM S ............................................................................. 208
Perform ance Comparison ..................................................................................... 211
Conclusions ................................................................................................................. 218

7 CONCLUSION S AND FUTURE DIRECTIONS ...................................................... 220

APPENDIX

A CODE FOR M ASS CALIBRATION ......................................................................... 229

B CODE FOR REAL-TIM E DATA PLOT .................................................................... 232

C CODE FOR POLARISQ AHGD SCAN .................................................................... 235

D CODE FOR AHGD OPTIMIZED UF-IT SCAN MATRIX ...................................... 244

E LIST OF ACRONYM S AND ABBREVIATION S .................................................... 250

LIST OF REFERENCES ................................................................................................. 253

BIOGRAPHICAL SKETCH ........................................................................................... 264















LIST OF TABLES


Table page

2-1. The AHGD project analytical performance requirements ................................ 39

2-2. The AHGD project system specifications ........................................................ 39

2-3. Linear quadrupole analytical performance ....................................................... 50

2-4. Quadrupole array analytical performance ........................................................ 50

2-5. Time-of-flight analytical performance ............................................................. 50

2.6. Cycloidal focus analytical performance .......................................................... 50

3-1. Change in peak width with and without using resonant ejection ......................... 123

4-1. Recombination energies (RE) for ions of interest ................................................ 131

4-2. Ionization energies (IE) for atoms and molecules of interest .............................. 131

4-3. Proton affinities for molecules of interest ............................................................ 131

4-4. Correction factors for partial pressure measurements ......................................... 139

4-5. Repetitive ion-molecule reaction experiments between He+' and N2 .................. 146

4-6. Ion-molecule reactions at different He pressures between He+' and N2 .............. 150

4-7. Ion-molecule reaction at different N2 pressures between He+' and N2 ................ 150

4-8. Ion-molecule reaction with reversed gas lines between He+, and N2 ................... 151

4-9. Self-protonation reaction results for hydrogen .................................................... 154

4-10. Hydrogen ion-molecule reaction at different N2 pressures .................................. 155

4-11. Determination of water vapor pressure via reaction with N2 .................... 157

4-12. Ion-molecule reaction between argon ions and water .......................................... 158

5-1. Observed nonlinear resonance ejection using a slow scanrate ............................ 195








6-1. Custom buit (UF-IT) QITMS analytical performance .................. 208

6-2. Modified commercial (PolarisQ) QITMS analytical performance .......... 210















LIST OF FIGURES


Figure pMage

1-1. Electrode arrangement for a linear quadrupole ................................................. 4

1-2. Cross-sectional view of a QIT ............................................................................ 5

1-3. Mathieu stability diagram in one dimension .......................................................... 11

1-4. Linear quadrupole stability diagrams ............................................................... 13

1-5. Quadrupole ion trap stability diagrams ............................................................. 14

1-6. Ion motion within quadrupole ion trap depicted in two and three dimensions ...... 17

1-7. Parabolic pseudopotential wells in the r and z dimensions ............................... 20

1-8. Plot of ion motion relative to the RF phase angle ............................................. 21

1-9. Mass-selective detection methods used for in-situ detection ............................ 24

1-10. Plot of a basic QITMS scan function ............................................................... 29

2-1. The fully assembled Space Shuttle (Atlantis) ................................................... 33

2-2. Space Shuttle main engines (SSMEs) ............................................................... 34

2-3. Hazardous gas detection system (HGDS) ........................................................ 36

2-4. Map of transport lines used for delivering sampled gas to the HGDS ............. 37

2-5. Stanford Research System model RGA-100 linear quadrupole mass analyzer ..... 41

2-6. Ferran Scientific model POD-01 quadrupole array mass spectrometer ............ 41

2-7. IonWerks time-of-flight mass spectrometer ..................................................... 42

2-8. Monitor Group MG2100 cycloidal focus mass spectrometer ........................... 43

2-9. The AHGD gas delivery system ........................................................................ 46

2-10. Detection limits for the evaluated mass spectrometers ..................................... 52








2-11. Update Rates for four evaluated mass spectrometers ....................................... 53

2-12. Quadrupole array data ...................................................................................... 54

2-13. Accuracy of the four evaluated mass spectrometers ........................................ 56

2-14. Precision of the four evaluated mass spectrometers ......................................... 58

2-15. Response time of the four evaluated mass spectrometers ................................. 59

2-16. Recovery time of the four evaluated mass spectrometers ................................. 60

2-17. T O F data ................................................................................................................ 6 1

2-18. Ferran Scientific miniature quadrupole array ................................................... 64

3-1. Location of hydrogen and helium ions on the stability diagram ..................... 67

3-2. The ion trap assembly of a Finnigan ITS-40 used in the UF-IT ....................... 70

3-3. Electron optics of an ITS-40 used on the UF-IT ............................................... 72

3-4. Open-configuration on the UF-IT ion trap ........................................................ 73

3-5. The vacuum manifold, turbo-drag pump and chassis of the UF-IT .................. 74

3-6. The U F-IT transfer line ..................................................................................... 75

3-7. Backing pumps used with the UF-IT ............................................................... 77

3-8. The main system board (MSB) of the UF-IT .................................................... 78

3-9. The electrometer board, RF control board, and RF amplifier of the UF-IT .......... 79

3-10. The RF drive circuit diagram of the UF-IT ..................................................... 81

3-11. RF coils used on the UF-IT ............................................................................... 82

3-12. Power supplies of the UF-IT ............................................................................ 83

3-13. UF-IT mass calibration software ..................................................................... 86

3-14. New UF-IT real-time graphical display ............................................................. 87

3-15. PolarisQ external ion source and ion trap assembly ........................................ 88

3-16. The new RF coil of the PolarisQ ............................................................................ 90

3-17. The low-pass filter board (LPF) board of the PolarisQ ................................... 92








3-18. The RF amplifier of the PolarisQ ...................................................................... 93

3-19. The AHGD scan software written for the PolarisQ .......................................... 95

3-20. Method for determining the resonant frequency of an RLC network ............... 98

3-21. Devised gas mixing and delivery system (GDS) ................................................. 100

3-22. Gas Dilution System (GDS) control software ..................................................... 102

3-23. The experiment schedule function of the GDS software ..................................... 103

3-24. Performance comparison of the NASA discrete bottle method and the Gas
D ilution System (G D S) ........................................................................................ 105

3-25. First UF-IT mass spectrum taken of room air ...................................................... 107

3-26. The use of a collision gas with the UF-IT ............................................................ 108

3-27. UF-IT ion signal improvement after regulating the emission current ................. 111

3-28. Mass Spectrum of 1.25% each analyte in nitrogen acquired on the UF-IT
at a drive frequency of 2.76 M Hz ........................................................................ 113

3-29. Difficulty with pumping hydrogen ...................................................................... 115

3-30. Removing the collision gas from the PolarisQ .................................................... 117

3-31. Oscilloscope plots of electrometer signal (red) as the RF amplitude was
ramped (blue) during mass analysis ..................................................................... 119

3-32. Effect of injection waveforms on analyte ion signals .......................................... 120

3-33. Mass spectra illustrating the effects of using resonant ejection ........................... 122

3-34. Mass spectra with the PolarisQ tuned to 2.82 MHz ............................................. 124

3-35. Hydrogen as the collision gas in the PolarisQ at 2.82MHz ................................. 126

3-36. Use of a hydrogen collision gas when monitoring the AHGD-analyte ions
while changing from pure nitrogen to 5000-ppm of each in nitrogen ................. 127

4-1. Diagram of the UF-IT vacuum chamber .............................................................. 138

4-2. N itrogen correction factor plot ............................................................................. 140

4-3. Hydrogen correction factor plots ......................................................................... 141

4-4. A rgon correction factor plot ................................................................................. 142








4-5. QITM S scan functions for low-mass ions ........................................................... 144

4-6. QITMS scan function for higher mass ions ......................................................... 145

4-7. Ion intensities versus reaction time for reaction of He+* and background gases. 147

4-8. Plot of the change in helium ion signal versus reaction time .............................. 148

4-9. Ion intensity versus reaction time for self-protonation reaction of H2 ................. 152

4-10. Plot of the change in hydrogen ion signal versus reaction time .......................... 153

4-11. Five plots of the change in hydrogen ion signal versus reaction time ................. 156

4-12. Ion intensities versus reaction time for reaction of Ar+ with background gas .... 159

4-13. Plot of the change in argon ion signal versus reaction time ................................ 160

4-14. Monitored ion signals of background gases ......................................................... 162

5-1. Distribution of electron energies of an internal ionization QITMS ..................... 166

5-2. Lines of nonlinear resonance within the stability diagram .................................. 170

5-3. Three-dimensional representation of black canyons ............................................ 172

5-4. Distortion of the pseudo-potential well caused by a +4% octopole field ............ 173

5-5. Hydrogen ion signal response with changing ionization conditions ................... 179

5-6. Helium ion signal response with changing ionization conditions ....................... 180

5-7. Oxygen ion signal response with changing ionization conditions ....................... 181

5-8. Argon ion signal response with changing ionization conditions ......................... 182

5-9. Plot of H2+" and He+* ion signals versus RF amplitude during ionization ........... 184

5-10. Plot of O2+" and Ar+" ion signals versus RF amplitude during ionization ............ 185

5-11. Oscilloscope display of post-ionization activity .................................................. 187

5-12. Oscilloscope display of H2+" and He+* ion peaks using different scan rates ........ 189

5-13. Oscilloscope display of delayed acquisition ........................................................ 191

5-14. Mass spectrum taken of pure nitrogen at a scan rate of 0.06 ms/Da ................... 192

5-15. Mass spectrum taken while sampling pure nitrogen at a scan rate of 1 ms/Da ... 193








5-16. Mass spectra of the isobars, nitrogen and ethylene ...................................... 197

5-17. Optim ized UF-IT scan function .......................................................................... 199

5-18. Optimized PolarisQ scan function ..................................................................201

6-1. Revised diagram of gas dilution system (GDS) ..................................................205

6-2. Plot of UF-IT response over the concentration range from 0 to 25,000 ppm
(2.5%) for each AHGD analyte in nitrogen .........................................................209

6-3. Plot of detection limits for six mass analyzers...................................................212

6-4. Plot of quantification error for six mass analyzers .................... 213

6-5. Plot of precision for six mass analyzers........................................................ 214

6-6. Plot of response times for six mass analyzers ......................................................215

6-7. Plot of recovery times for six mass analyzers .................................................216

6-8. Plot of the update periods for six mass analyzers .................................... 217

6-9. Plot of high-mass cutoffs for six mass analyzers ..................... 219

7-1. Plot of system volume for six mass spectrometer systems ............... 224

7-2. Plot of system weight for six mass spectrometer systems ......................... 225

7-3. Sectioned image of the UF-IT QITMS instrument .................... 227














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

ANALYSIS OF LIGHTWEIGHT GASES BY
QUADRUPOLE ION TRAP MASS SPECTROMETRY
FOR THE SAFETY OF THE AMERICAN
SPACE SHUTTLE PROGRAM

By

Andrew Keith Ottens

May 2003

Chair: Dr. W. W. Harrison
Department: Chemistry

The quadrupole ion trap mass spectrometer (QITMS) was patented nearly 50 years

ago, when it was proposed for trace analysis of lightweight gas mixtures. Though a

commercial success, QITMS has been used with analytes of ever-increasing size. We

evaluated QITMS for quantifying lightweight gas mixtures with the performance

compared to other mass analyzer technologies.

The National Aeronautics and Space Administration (NASA) uses mass

spectrometers to monitor the amount of hydrogen, helium, oxygen, and argon in the

nitrogen-purged Space Shuttle. The explosive hazard of the cryogenic hydrogen and

oxygen used to propel the Space Shuttle makes leak detection imperative. The two

present-day leak detectors are remotely located because of their large size and sensitivity

to vibration. Analysis is delayed by up to 45 s, and only two samples can be monitored

simultaneously. In 2000, NASA initiated the Advanced Hazardous Gas Detection project








to develop a compact, rugged, and fast mass spectrometer to be placed in multiple

locations next to the Space Shuttle to provide real-time analysis with increased

redundancy.

The QITMS instrumentation was modified specifically for this application. The RF

drive frequency was increased to 2.5 MHz to adequately trap lightweight hydrogen and

helium ions. Internal ionization was preferred for use without a collision gas, along with

an open source configuration that provided rapid sample replacement. The modem

electronics incorporated were controlled by customized software.

Analytes were found to react rapidly with abundant background gases. The QITMS

operating conditions were optimized to minimize negative effects of ion-molecule

reactions while maximizing analytical performance. A custom segmented scan function

was developed with a total scan time of 14 ins, averaging 70 scans per data point at the

required 1 Hz update rate.

The QITMS met requirements for detection limits, accuracy, precision, response

time, and recovery time. The linear quadrupole was the only other instrument to perform

similarly, but was six times slower than the QITMS. Thus, QITMS was the preferred

analyzer technology for the application, which NASA will further develop to improve the

safety of Space Shuttle program for years to come.


xvii













CHAPTER 1
INTRODUCTION

Wolfgang Paul introduced the quadrupole ion trap (QIT) in 1953,' but he did not

know then that the device, later to be named after him, would become a great

achievement in mass spectrometry.2,3 In the same patent, Paul introduced the linear

quadrupole, which also became a leading choice of mass spectrometrists.4 Both are

quadrupole devices; essentially, the QIT is a 3600 rotation of a linear quadrupole forming

a central ring flanked by two endcaps.5 Both devices operate by manipulating ion motion

in a periodic electric field. Despite their similarities, the two devices have evolved along

entirely different paths, which has led to a divergence in the applications they are used

for.3-7 The linear quadrupole was quickly adopted as a compact, simple, mass analyzer

used for trace analysis of gas mixtures a residual gas analyzer (RGA).4 In contrast, for

three decades the QIT remained a relatively unknown device because of its poor

performance as a mass spectrometer.6'7

This changed in 1983 with the introduction of the Ion Trap Detector (ITD) by

Finnigan MAT Corporation (San Jose, CA). Commercial success was made possible

with the use of a new mass analysis method (mass-selective ejection) that was further

enhanced using a helium collision gas.8'9 Finnigan marketed the ITD as a simple,

inexpensive detector for gas chromatographs (GC), beginning nearly 20 years of

innovation by numerous manufacturers.

Today's quadrupole ion trap mass spectrometer (QITMS), coupled with

chromatography, can analyze complex mixtures, with femtomole detection limits and a








106 dynamic range. It has become widely used in biological, pharmaceutical, and

environmental industries,2,3 with a focus on high-mass analytes. With the success in

these areas, little attention has been paid to Paul's intended use of the QIT as a tool for

trace analysis of lightweight gases' the focus of our study.

This chapter provides a background of QITMS fundamentals, operation, and history.

The theory of ion motion in a quadrupole field is discussed in some detail, because of its

importance for understanding the instrumental modification and results of this work.

Most equations are provided in a general context to explain behavior inside a quadrupole

device, thus units are provided only for essential terms used in calculations, and the

reader is referred to the literature for further explanation. The three modes of QITMS

operation are reviewed in historical order. The advantages of the mode selected for this

research are emphasized in contrast to the other two modes. Mentioned next are recent

advances in QITMS technology, some of which are evaluated in later chapters for

possible performance enhancements. The chapter concludes with an overview of the

dissertation.

Theory of Ion Motion within Quadrupole Devices

Quadrupole Electric Field

A mass spectrometer uses electric and/or magnetic fields to influence an ion's

trajectory in such a way as to distinguish its mass-to-charge (m/z) ratio.1' The electric

field in an ideal quadrupole device will have a potential, (p, that is a quadratic function

relative to the Cartesian coordinates x, y, z as in the equation:

(p(x,v. ) + 2 (1-1)








where a, 13, and y are multiplicative constants.7'10'12"14 The potential is applied to the

electrodes of the device, four rods in a two dimensional array for the linear quadrupole

(Figure 1-1), and a three dimensional configuration of a ring with two endcaps for the

QIT (Figure 1-2). The emanating electric fields are geometrically related by the

constants a, 13, and y. For both devices Laplace's equation must be satisfied, V2(P = 0.

This condition requires that a + /1+ y = 0. For the linear quadrupole y = 0, since there is

no field in the z direction. This requires then for a = 1 that 13 = to give a + 13 = 0.

Thus, the potential p is applied to the x-set of rods, while is applied to the y-set of

rods (Figure 1-1). Ions entering the device are confined in the x and y directions by

opposing fields and travel lengthwise along the unconfined axial dimension. This differs

from the QIT that has an effective field along all three axes. The fields in the x and y

directions are identical in the 3D structure of the QIT, thus for a = 1 then 13 = 1, and to

satisfy Laplace's equation, y = -2. This was originally accomplished by applying +(p to

the ring and --p to the two endcaps (a factor of-2 when taken together). More recently it

was found to be easier to ground the endcaps and apply the potential p to the ring. This

has the effect of halving the field strength, thus an ion at the center of the trap

experiences +p/2 from the ring while each endcaps appears as --/2 (the one-half factor is

later accounted for in the equations of ion motion). In the QIT, fields in three dimensions

confine ion motion, trapping ions for as long as necessary.

Ion Motion in Quadrupole Devices

The influence of a quadrupole electric field on an ion's motion can be expressed

mathematically, derived from the applied potential equation:


(PpOO) = A P (cos 0) (1-2)

















































Figure 1-1. Electrode arrangement for a linear quadrupole. Four hyperbolic rods are
arrayed in the x-y plane. An inscribed circle of radius r0 describes the rod
spacing. Potentials +p and -9 are applied to alternate rod pairs. (Adapted
from Ghosh, P. K. Ion Traps; Oxford University: New York, 1995; Figure
2-1, p. 8.)
















































Cross-sectional view of a QIT: a) the exit endcap allows ions to exit and
strike a detector; b) the ring electrode together with the endcaps, it defines
the volume and field within which ions are trapped; c) the entrance endcap -
allows either electrons or ions to enter the QIT. (Adapted from March, R. E.
in Encyclopedia of Analytical Chemistry; Meyers, R. A., Ed.; John Wiley &
Sons: Chichester, UK, 2000; Vol 13; Figure 2-9, p. 11850.)


Figure 1-2.








which is expressed differentially here using Laplace's equation in spherical polar

coordinates, where A are arbitrary coefficients and P,,(cosO) are Legendre

polynomials. 14 Equation 1-2 can be expanded and rewritten in Cartesian coordinates as
(P(r,z) = A' + A lz + A 2 ( r 2 z_) + Az(r z2)+... (1-3)



where r = + y2 The parts n = 0, 1, 2, 3, ... represent discrete components of the

field (monopole, dipole, quadrupole, hexapole, etc.), but in the ideal quadrupole device,

only monopole and quadrupole components are present, which is assumed hereafter.

Assigning the coefficients A = B and A = 2A, Equation 1-3 becomes


9(r,) = A(r2 2z2) + B (1-4)

which can be used to determine A and B. This is accomplished using the boundary

conditions qoL = 0 at the grounded endcaps, and ?O = po at the ring electrode. The

radial boundary potential at the edge of the ring electrode (ro, 0) is

O( ,.) = A(r2))+B=VO (1-5)


and the axial boundary potential at the edge of the endcaps (0, z,) is

A(-Zz)+ (1-6)

thus the values for A and B are


r2+ 2ZoB= ro2 + 2zo) (1-7)


which provides the potential equation:
qo0(r2 -2z2) VO 2z'0

(, (t") +2z2) r + 2z0) (1-8)

This expresses the potential experienced by an ion at any point (r, z) inside the ion trap.









Deriving the Mathieu Equation

The potential described in Equation 1-8 exerts a force on an ion in each direction

controlling its motion. The force on an ion increases linearly as it is displaced from the

center of the ion trap,'5 which can be expressed independently in the r and z directions as

d2r dqp
dt2 dr (1-9)


d2z dVo
F = dt2 _e dz (1-10)


where ddr and / are the derivatives of Equation 1-8, which for each dimension


are

dq_ _(_o 2r
dr (r"2 + 2z2) (1-11)


dVp 2)2 4z
dz (ro2+2z) (1-12)

and when substituted into Equations 1-9 and 1-10 with the relationship

2 thus j d2 (1-13)



produces the following equations of ion motion:

d2r 8rerp =0
d_2 m(r2 + 2z20 (1-14)

d2z 16ze (o. 0
d2 m(r2+ 2z'))2 (1-15)

of a form similar to the equation of motion of a simple harmonic oscillator:





8


d u k 0 (1-16)
dt2 m

where the force constant, k, relates to the driving quadratic field, with each of three axes

of motion being treated as separate harmonic oscillations with respect to time.

The applied potential can be expressed as a combination direct current (DC)

component U and radio frequency (RF) component Vcos(f t) as in qp0 = U V cos(Ot),

which when substituted into Equations 1-14 or 1-15 forms the second-order differential

equation:

d2uW +[a. 2q. cos(2 )u =0
(1-17)

which is the conical form of the Mathieu equation, used originally to describe the motion

of vibrating membranes,16 and here illustrating the motion of ions within a quadrupole

device. This general form can be used for any quadrupole device, and in any direction

where u can represents x, y, r, or z. The dimensionless variables, au and qu, are dependent

on instrumental parameters as expressed below specifically for a QIT with grounded

endcaps and the potential qTo applied to the ring electrode:

8eU
a m(r02 +2z0) 2 (1-18)


-16eU
m(r2 +2z)2 (1-19)


-4eV
q, m(r + 2z02)Q2 (1-20)


8eV
= m(r0 +2z 2 (1-21)








where m is the mass of the ion in kg, ro and zo are the dimensions of the ion trap in m, U

and V are the respective DC and RF potentials in V, Q is the angular frequency in rads/s,

and e is a unit of charge in C.

The movement of an ion can be determined as a combination of separate harmonic

motions in each direction. With u representing any dimension in Equation 1-17, the

function u( ) specifies the position (displacement from the trap center) of an ion in each

direction. The derivative u'(4) provides the velocity of the ion at a given A complete

solution to u(4) is composed of two independent parts, which using Floquest's theorem is

stated as

u() = AeU"t'() + Be-"Vt(-) (1-22)

where g is a complex constant, pu = a + if, and the function y is periodic. From

Fourier's theorem a periodic function can be written as

v/( ) = .C2,,e 2n"
n=-c (1-23)

Thus Equation 1-22 can be expressed in the form:

u(4) = Ae' Z C2ne2"i + Be-u C 2,
,,=_,, ,=_ o (1-24)

It is only under specific conditions of the constant g. that the solution of u(4) will

remain finite as the phase (4) goes to infinity. The ion is considered stable within the ion

trap when it has a finite position and velocity. This occurs when g. is purely imaginary,

thus a always equals zero, leaving pu = +#/ that must also remain imaginary. In the

specific case where 03 is an integer (i.e., 0, 1, 2, ...), g. is still imaginary, but ion motion is

unstable; these are the boundary points between stable and unstable trajectories.








Therefore, it is of interest to determine the 13 values of the trapped ions. Often 13 is better

expressed in terms of the dimensionless values a, and q,, since Equations 1-18 through

1-21 relate them to known ion trap parameters. The values are interconnected using the

continued fraction:

2
)6,2 = a, + q.
2
(fl. + 2)2 -au qu
(flu, + 4)2 u
(flu +6)2 -a.


2
+ 3
(f8.-2)2 -au q3
(,f,, 4)2 _a. qu
(,8. -6)2 -au (1-25)

which requires an initial 3 estimation derived from the Dehmelt approximation:

2
2 = a, + q126
,8 U 2 (1-26)

that is reasonably accurate for values of q, -< 0.4.

Stability Diagram

To better visualize solutions of the Mathieu equation, the function u( ) can be plotted

in a, qu values, referred to as the stability diagram. 19,2 This is done by first writing

Equation 1-24 as

u( ) = (A + B) C2,, cos(2n /8) + (A B) ,C2, sin(2n f )
n=--o n=-0 (1-27)

using the relationship e" = cos(x) + isin(x) to convert into trigonometric terms. The sine

type functions are then plotted in a,,, q,, coordinates as shown in Figure 1-3. The formed

pattern is identical in all directions u (e.g., x, y, r, z). To determine an ion's complete

trajectory, motion in all directions must be considered at the same time. It happens that
































Mathieu stability diagram in one dimension. Ion trajectories that map within
the shaded regions are stable. The lines marked a, and bn are respectively
derived from cosine and sine functions of order n. These lines represent the
boundary points between stability and instability at integral values of P3.
(March, R. E.; Londry, F. A. Chapter 2 in PracticalAspects of Ion Trap Mass
Spectrometry; March, R. E., Todd, J. F. J., Eds.; CRC Press: Boca Raton, FL,
1995; Vol 1, Figure 3, p. 38.)


Figure 1-3.








for both devices, only two-dimensional coordinates need be considered. For the linear

quadrupole, only the x and y dimensions are plotted, since no field is exerted in the z

direction. For the QIT, the x and y fields are identical because of rotational symmetry,

thus motion is plotted in dimensions r and z.

To create a 2-D stability diagram, two plots (each of a different dimension) are

superimposed using a common coordinate system. For the linear quadrupole, the x and y

dimensional plots are inverted relative to each other (in satisfying LaPlace's equation,

a = I and 3 = -1). This is illustrated in Figure 1-4a where two identical plots are reflected

across the q-axis; therefore, the plot looks the same whether given in a, q, or a, qy

values (only the right half of the diagram is usually shown, because +qu and -q, values

are indistinguishable). Shaded areas indicate regions of stability in one dimension.

Areas of overlap are where ions have stable trajectories in both dimensions. Linear

quadrupoles are operated in the region marked as A in Figure 1-4a, because ions can be

manipulated in this region with the lowest electric potentials. This is known as the first

stability region, shown in an expanded view in Figure 1-4b.

The QIT stability diagram is formed by superimposing plots of the r and z

dimensions. Recalling the QIT solution for LaPlace's equation, a = P = 1 and y = -2, the

r and z stability regions will differ in magnitude by a factor of-2. Figure 1-5a shows the

QIT stability diagram in a,, qz coordinates, with the first stability region expanded in

Figure 1-5b. The r stability region is inverted and is twice that of the z, resulting in a lack

of symmetry across the q-axis, which differs from that of the linear quadrupole. Ions that

have a, q, coordinates that coincide with the first stability region shown in Figure 1-5b

will be trapped inside the QIT.











(a)


B


















(b)
0.8 0.2
0.2
0.4
0.6

0.1 0.6
0.4

0.2 0.8
0.0 q


Figure 1-4. Linear quadrupole stability diagrams: a) stability diagrams in two
dimensions of a,, q, space. Regions of simultaneous stability are marked
A through D; b) Enlarged plot of the first stability region, shown for only
positive a, values (symmetrical across the q, axis). (Adapted from March,
R. E.; Hughes, R. J.; Todd, J. F. J. Quadrupole Storage Mass
Spectrometry; Wiley-Interscience: New York, NY, 1989; Figures 2-7 and
2-8, pp. 46-47.)















a.

0

A7

-5



-10


1.6 q.


Quadrupole ion trap stability diagrams: a) overlapping plot of the r and z
stability diagrams; simultaneous regions are marked A and B; b) enlarged
diagram of the first stability region, section A. (Adapted from March, R.
E.; Londry, F. A. Chapter 2 in Practical Aspects of Ion Trap Mass
Spectrometry; March, R. E., Todd, J. F. J., Eds.; CRC Press: Boca Raton,
FL, 1995; Vol 1, Figures 5 and 6, pp. 40-41.)


Figure 1-5.





15


Ion trajectories are oscillatory; having angular frequency components related to the

13u values where the ions are stored and the RF drive frequency (fl) by Equation 1-28.

0)' = (2n/fl)-n
2 (1-28)

The primary component, known as the fundamental secular frequency,7,14 is of order

n = 0, expressed as


2 (1-29)

which at the stability boundaries of 13u = 0 and 13u = 1 is ou,O = 0 and (ou,O = K2/2

respectively. At these points, the secular frequency of an ion comes into resonance with

K, suddenly increasing the amplitude of the ion motion by forced oscillation of the

harmonic motion. A forced harmonic oscillator must take into account both the natural

oscillation forces and the applied force. This is expressed by adapting Equation 1-16 into

the form of Equation 1-30 where Fmcosf2t is the applied field. The general solution to

Equation 1-30 is shown in Equation 1-31 where the displacement value u increases as the

natural resonance frequency (o) approaches the applied frequency (Co") and is in

resonance as G goes to 0.

d2u k F
S -- cos"t = 0 (1-30)
dt2 m m
u = Fsin(ot 0)
G

G= ,m2(o"2- 2)2 (1-31)

Superimposed on this motion are the higher-order frequency components of n > 0,

otherwise known as micromotions. This ion motion at specific Or, 0z values is illustrated








in Figure 1-6. The amplitude of the ion motion is determined by the C2n coefficient of

Equation 1-27, which is based on the a,, q, values and the order n. Only the secular

frequency and the higher-order frequencies of n = 1, 2 are of significance, because C2n

diminishes rapidly as n increases.4 In summary, the terms C2n and 3, respectively

determine the amplitude and frequency of an ion's oscillation at specific au, q,, values that

are based on the m/z of an ion and the other instrumental parameters used in Equations

1-18 through 1-21.

Until now the above theory has dealt with the motion of a single ion within an ideal

quadrupole field. This of course is not an exact measure of true ion motion in a real QIT

where more than one ion is trapped. Once two or more ions are confined within the

quadrupole field, the repulsion of the like charges will affect the field experienced for

each ion. Such a condition is known as space charge and must be considered when

describing true trajectories.

Space charge is of particular importance when discussing ion behavior within a QIT,

since the ions are confined in a finite volume for extended periods (ions have been

trapped for days7). Various mathematical methods have been deduced to explain the

motion of multiple ions within the trap,2126 which can then be simulated on computers.21

One such method known as the pseudopotential well approach estimates ion motion

based on secular oscillation. The approach was first described by Major and Dehmelt22'23

and later developed by Todd et al.24'25 to quickly determine average ion velocity, energy,

and ion cloud density at the space charge limit (the maximum ion capacity).
















25



0.0 3- Trajectory

C

N

1.21.25



Y (Ring) / nrn 1.251.25



Figure 1-6. Ion motion within quadrupole ion trap depicted in two and three dimensions.
(Forbes, M. W.; Sharifi, M.; Croley, T.; Lausevic, Z.; March, R. E. J. Mass
Spectrom. Vol. 34, 1999; Plate 1, p. 1221.)








Pseudopotential Well Theory

In deriving the pseudopotential well model, ion motion is first considered as a

combination of low frequency macromotion and high frequency micromotion

components, U and 6, respectively where

u = U + 4 (1-32)

Then with the assumptions that 8 << U and that d8/dt >> dU/dt, the Mathieu equation

(Equation 1-17) becomes

d%5
d -9 =-[a 2qu cos(2 )]U
d 2 (1-33)

Integration over one RF period results in an equation for 6, and using Equation 1-32, the

6 term can be removed altogether from the ion motion function (converted into units of

time) giving the equation:

d2U n 2 q U
dt2 a.+- U
2) 4 (1-34)

which describes ion secular motion as simple harmonic motion, which can also be written

as

d 2U 2 u
dt2 1 O,' (1-35)

using the Dehmelt approximation (Equation 1-26) and Equation 1-29.

For the condition a, = 0 (no applied DC potential), and by substituting for qu using

Equations 1-20 or 1-21, Equation 1-34 becomes

d2U e2V,
dt2 2m2u 2 )2 U
_YP 0(1-36)


Using the force equation relation:









_eU dD'
t dU (1-37)

the estimated potential becomes
dD ( e V
-- 2 ou--2 U
dU 21uQ (1-38)

which when integrated from U=0 to U=uo results in
eV'
4muf2 (1-39)


Equation 1-39 is the pseudopotential equation for any dimension u, with the same units as

used in Equations 1-18 through 1-21. Equation 1-39 can also be expressed in terms of qu

values using Equation 1-20 or 1-21.

The model describes the motion of ions as harmonic oscillators, precessing at their

secular frequency within the pseudopotential well, which being parabolic (as illustrated in

Figure 1-7) acts to confine the ions. A deeper well will confine more ions, thus increases

in m, 03, and Q will result in greater storing capacity. This runs contrary to a decrease in

storage volume during ionization as either 03 or m is increased.24

This model is only one approximation of the complete solution to the Mathieu

equation. The estimation is fairly accurate for q, < 0.4, but deviates significantly at

higher values because of the increased significance of the ignored micromotion. Figure

1-8 compares the pseudopotential well approximation with the complete solution and the

smoothed general solution (SGS) that tracks the mean of the macro and micro motions

together, accounting for some displacement from the micromotion.

The pseudopotential well model was devised to determine the space charge limit of

an ion trap given as


















p' JJ ?'

i I ,*





a, n

I I









Figure 1-7. Parabolic pseudopotential wells in the r and z dimensions. The relationship
of D. = 2Dr is based on the assumption that r0 = 2zX- (Ghosh, P. K. Ion
Traps; Oxford University: New York, NY, 1995; Figure 2-9, p. 23.)












1.0

0.8

0.6

0.4

0.2

0.0

-0.2

-0.4

-0.6

-0.8

-1.0


0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0



RF phase
(arbitrary units)


Plot of ion motion relative to the RF phase angle. The relationship between
the complete solution to the Mathieu equation and two approximations,
Dehmelt's pseudopotential well model and the smoothed general solution, is
illustrated. The plot is shown for 13=0. 1, where the micromotion amplitude
is minimal. (March, R. E.; Hughes, R. J.; Todd, J. F. J. Quadrupole Storage
Mass Spectrometry; Wiley-Interscience: New York, NY, 1989; Figure 2-16,
p. 75.)


C)
,,,i.
D'- --
E .


Figure 1-8.









3D,, 3V2
0 4n00 16nu2 (1-40)


Other important approximations using this model are the average velocity and kinetic

energy:

(u-2u060.,0
2r (1-41)


S O0U1O 4 -eD
2(7 2 /T2 (1-42)

which are derived using the equation for harmonic motion u = u0 sin co',0t, with an

amplitude of uo and a frequency oO.

The theory presented in this chapter provides a means of explaining the behavior of

ions within the QITMS, and serves as a starting point for understanding basic principles

behind this work. The next section reviews how the QIT has been used over the course

of QITMS development, which divides into three periods each characterized by a

different mode of mass analysis.

Modes of Quadrupole Ion Trap Operation

The quadrupole ion trap, like other mass spectrometers, separates ions based on their

mass-to-charge ratios. This is accomplished by varying an electrical field to affect an

ion's trajectory. Three distinct mass analysis methods have been employed with the QIT:

mass-selective ion detection, mass-selective storage, and mass-selective ejection.4-6'2728

Though the first two have been around longer, it wasn't until the introduction of mass-

selective ejection in the early 1980's that the QIT became commercially available.

Presented is a summary of each method.








Mass-Selective Detection

Mass-selective detection was the analysis method first presented by Paul.1'1 Two

versions of this method were further developed by Fischer29 and Rettinghaus,30 each

based on sensing ions inside the ion trap without having to remove them. This is

accomplished by applying a low amplitude auxiliary field across the endcap electrodes at

a frequency corresponding to a selected 3, value (a set fraction of the drive frequency).

Ion motion is manipulated in a,, q, space by ramping either the amplitude of the DC

(Fischer's method) or the RF (Rettinghaus' method) potential applied to the ring

electrode, as illustrated in Figures 1-9a and 1-9b, respectively. In either case, as the

secular frequency of ions at a specific m/z comes into resonance, the applied field will

become attenuated and phase shifted, the degree of which is proportional to the number

of ions at that particular m/z. Ions of different m/z values are sequentially brought into

resonance with the applied auxiliary field to produce a mass spectrum.

Both the Fischer and the Rettinghaus methods suffered from the same problems.

Large ion populations or long detection periods were required because of the low

sensitivity with tuned circuit detection. Space charge effects were a prominent problem

with the high ion densities of the QIT, causing the secular frequency distribution of the

ion packet to spread out, broadening the detected signal. Space charge further

complicated matters, because ions of differing m/z values experienced different ionic

environments. An ion at resonance is influenced by ions of higher m/z values since both

are stable within the trap at the same time, but this is not the case vice-versa, leading to

nonlinear mass spectra.















U
(volts)


V (volts)


(a)






amplitude of
sowtooth wave










(b)


Mass-selective detection methods used for in-situ detection: a) the Fischer
method using a DC ramp to shift ions of different m/z values sequentially
into resonance at a specified 3, value; b) the Rettinghaus method using an
RF ramp to shift the q, value of ions along the a, = 0 line and into resonance
at a specific I3, value. (March, R. E.; Hughes, R. J.; Todd, J. F. J.
Quadrupole Storage Mass Spectrometry; Wiley-Interscience: New York,
NY, 1989; Figures 3-1 and 3-2, pp. 113-114.)


Figure 1-9.








To minimize space charge interference, an ion signal can be averaged over longer

periods with a lower ion density; however, signal intensity may be affected (enhanced or

diminished) by ion-molecule reactions. Maintaining a low ion trap pressure reduces the

number of collisions, thus lessening such reactions, as demonstrated by Rettinghaus, who

reports detecting a minimum of four ions with the QIT. This suggests that, under specific

conditions, the sensitivity of this mode is quite good, but complications from space

charge and ion molecule reactions have caused the technique to remain in obscurity.

More recently, Fourier transform spectrum analysis was used to improve detection, but

unit mass resolution still was not achieved because of problems with space charge.31

Mass-Selective Storage

To avoid the complications of mass-selective detection, Dawson et al. opted for a

mass analysis method akin to the operation of a linear quadrupoles.32'33 Called mass-

selective storage, this mode operates by ramping both DC and RF potentials at a fixed

ratio (UN) such that ions of sequential m/z are placed at either apex of the QIT stability

diagram. By selecting an appropriate U/V ratio, ions of a single m/z value are stable at

any given time, while all other ions fall beyond the 3u = 0 or 1 stability boundaries. Ions

are ejected through holes in the endcaps to impinge on an external detector. This method

of mass analysis has the benefit of avoiding ion-ion interactions between different

species, and of providing improved mass resolution because of the selective storage

approach. However, this method, like mass analysis with a linear quadrupole, has a much

lower duty cycle. Ions of differing m/z values are stored separately, requiring more time

to produce a full mass spectrum. Under conditions where the gas sample is available

only for a short period of time (such as with gas chromatography), the prolonged duty

cycle becomes a critical limitation.








Mass-selective storage also requires more electronics to control the ramping of DC

and RF potentials at a fixed ratio. Though such electronics are common (used on linear

quadrupoles), they add cost, complexity, and size/weight. The addition of endcap holes

for ion ejection is another area of concern. Their presence disturbs the quadrupole field,

causing further deviation from ideal conditions (truncation of electrodes and machining

tolerances are other imperfections), resulting in nonlinear behavior of the QIT. These

effects will be considered in more detail in Chapter 5.

Mass-selective storage for QIT mass analysis turned out to be unpopular.6'7 The

technique provided little improvement over the well developed linear quadrupole

technology of the day. The QIT did achieve some popularity at this time after Todd et al.

demonstrated that it could be used as a chemical ionization source. Called a quadrupole

ion store, or QUISTOR,34 a QIT would replace the normal ionization source of another

mass spectrometer (e.g., linear quadrupole, sector, time-of-flight). The QUISTOR was

particularly useful for studying ion-molecule reactions, having been developed to

examine the decay rates of metastable ions.35

The QIT as a QUISTOR now stood out as a gas-phase ion test-tube.6'7'28 Ions could

be confined at low pressure and monitored over specified time intervals. This provided

better control of gas-phase chemistry where pressure and time are key factors. The low-

pressure environment allowed reactions to reach equilibrium slowly, so that reactant loss

and product formation could be monitored for kinetics studies. QIT devices were used to

determine rate constants, proton affinities, and recombination energies for charge-transfer

processes. Reactions at thermodynamic equilibrium could also be examined by simply

extending the reaction period. Ions have been reacted with neutral molecules, other ions,








and photons inside the QIT, with complementary experimental results to other

techniques,36 most notably those of ion cyclotron resonance, another ion trapping

technology that grew in popularity at the same time.37-39

Mass-Selective Ejection

Until the late 1970's QIT instruments were rare and all home-built. This was about

to change with the advent of another mass analysis method, mass-selective ejection.8,'9

This method allowed ions of differing m/z values to be trapped simultaneously and then

ejected one m/z at a time to an external detector. This provided a greater duty cycle for

most applications than with the use of mass-selective storage for the QIT or with a linear

quadrupole. Additionally, mass-selective ejection was performed with only an RF ramp,

thus hardware was simpler and cheaper. These two advantages made the QIT more

competitive with the linear quadrupole, thus commercially viable.4

The first mass-selective ejection instruments were simple to operate. A single RF

potential would be applied to the ring electrode with the endcaps grounded. Without a

DC potential a_ = 0, and only the q component of ion motion changes. All ions with q,

values lower than 0.90805 would have stabile trajectories, thus the widest mass range is

collected at a low RF amplitude. Electrons are injected into the QIT through holes in the

endcaps for internal electron impact (El) ionization. The electron beam is stopped once

enough ions have been collected, allowing ions with stable trajectories to settle inside the

ion trap. Mass analysis occurs by ramping the RF potential rendering ions of increasing

m/z values axially unstable at the 3, = 1 boundary, and ejecting them through the endcaps

into a detector. These steps taken together form a scan function. Figure 1-10 depicts a

typical scan function showing the changes in the RF, electron gate lens, and electron

multiplier voltages for each step, all of which can be optimized for a specific application.








Collision Gas

Commercialization spurred many advances in QITMS technology. One of the

earliest was the incorporation of a lightweight gas inside the ion trap at a high pressure of

1 mtorr, which significantly improved both mass resolution and sensitivity. 6-9,12,40 These

effects were noticed early on;24 a background gas of considerably lower mass than an

analyte ion would dampen motion by viscous drag (essentially a dampened harmonic

oscillator). This buffering effect collapsed the trapped ion cloud into the center of the

device, lowering the displacement and the energy of the ions. The collisionally cooled

ions of the same m/z value absorb energy uniformly, and eject together forming narrow

peaks. Improvements are most pronounced with the collision gas (B) being considerably

less massive than the analyte (A ) (i.e., B << A ). For the opposite case, B > A+, a

significant amount of energy is transferred, resulting in the heating of the ion cloud and

elastic scattering of the ions, which are lost as their displacement and velocity increase.

Helium was preferred as the collision gas because of its low mass, and high

ionization energy (IE = 24.1 eV; most analytes are well below 20 eV). It provided a large

mass difference with most analytes, and promoted charge-exchange reaction to produce -

not neutralize analyte ions. A collision gas was also important for use with external

ionization techniques. Ions enter the ion trap with a large displacement and are subjected

to a high RF field strength. Collisional cooling reduces the energy of the ions so that they

do not collide with the electrodes.6'41'42

Multi-Purpose Auxiliary Waveforms

Supplemental waveforms applied across the endcap electrodes are used with the

advanced functions of today's QITMS instruments. These functions operate on the








Ionization


Time (ms)


Figure 1-10.


Plot of a basic QITMS scan function. The scan is subdivided into discrete
steps for ionization, post-ionization, mass analysis, and post-scan activities,
with the electric fields changing as needed between each step.


Post
Ioniz-
ation


Mass
Scan


Post
Scan


RF
(V)

Gate
Lens

Elec.
Mult.


2000


I_








principle that an ion in resonance with an auxiliary field will quickly absorb energy, to

increase its velocity and displacement. Broadband waveforms are used during ionization

to minimize the storage of unwanted ions. Frequencies of desired ions are left out of the

broadband waveform, which keeps their energy low while all other ions are excited and

removed.

A single frequency can be used to isolate desired ions. The RF drive amplitude is

first ramped to eject all lower m/z ions. Then the higher m/z ions are sequentially ejected

as they come into resonance with the auxiliary field, which has a frequency just below the

secular frequency of the desired ions. Once isolated, the ions of interest can be

collisionally fragmented by application of another auxiliary waveform. This process can

be repeated, allowing the QITMS to perform multiple stages of mass spectrometry (MS')

for improved selectivity.

Supplemental waveforms can also be used to enhance detection. An auxiliary field

is applied with a frequency corresponding to a q, value that is typically just inside the

N = 1 boundary (i.e., q, 0.9). The main RF is ramped in the normal manner, but ions

are ejected by resonating with the auxiliary field before they reach the 3z = 1 stability

boundary. The process is known as resonant ejection in which ions absorb energy more

uniformly to produce tighter ion packets, increasing mass resolution and sensitivity. The

resonant ejection frequency can also be decreased to eject ions at lower q2 values for

extending the mass range of the instrument.

The enhanced functionality provided on today's commercial QITMS make it a

popular choice for many applications.43 The QITMS has found wide acceptance in

biological, pharmaceutical, and environmental industries as a versatile, sensitive,








instrument with the power of multiple-stage mass spectrometry an economical

alternative to comparable instrumentation. In this work, however, Paul's intended use of

the QIT is explored to perform trace analysis of lightweight permanent gases an

application typically performed by linear quadrupole RGA mass spectrometers.

Overview of Dissertation

The next chapter discusses the needs of the National Aeronautics and Space

Administration (NASA) for a small rugged mass analyzer for monitoring lightweight

permanent gases. Four mass analyzer technologies, other than QITMS, are evaluated

with a discussion of their performance and suitability for the NASA application. The

results provide a comparison basis for evaluating the performance of the developed

QITMS instrumentation. Chapter 3 begins with a historical perspective on the use of QIT

with lightweight gases. Next, the details of the QITMS instrumentation developed for

this application are presented. One QITMS was custom built, incorporating specific

features that were found ideal for lightweight gas monitoring. Another QITMS was a

modified commercial instrument that offered advanced functionality. In Chapter 4 we

examine ion-molecule reactions found to be a significant limitation of QITMS for the

analysis of permanent gases. Chapter 5 presents a parametric study of the QITMS, and

the development of a custom scan function to minimize ion-molecule reaction effects

while maximizing quantitative performance. In Chapter 6 QITMS performance is

examined and compared with other analyzer technologies evaluated in Chapter 2.

Chapter 7 draws together final determinations from the presented work with

consideration for future development of QITMS instrumentation for this application.













CHAPTER 2
MONITORING LIGHTWEIGHT GASES FOR SPACE SHUTTLE SAFETY

Introduction

Since 1981, the Space Shuttle has been the only NASA operated vehicle capable of

carrying man into space. As the primary transport to the International Space Station,44

the Shuttle is expected to continue operation well into the 21st century. A fully assembled

Space Shuttle consists of three major components (Figure 2-1): 1) two solid rocket

boosters (SRBs), used in the first 2 minutes of flight; 2) the external tank (ET) containing

the cryogenic fuel and oxidizer needed for the Space Shuttle main engines (SSMEs); and

3) the Orbiter with crew compartments, a payload bay, and the SSMEs (Figure 2-2) used

during the entire eight minute flight into space. Together the Space Shuttle weighs 2

million kilograms, nearly half of which is amassed by the 1.8 million liters of liquid

hydrogen and 0.7 million liters of liquid oxygen held in the ET. 44,45 Because of the

immense volume of explosive cryogens, leak monitoring is required and in light of the

long-term future for Space Shuttle operation, NASA is exploring new technology for

improved leak detection.

Mass spectrometers have been the analytical tool of choice for cryogenic fuel leak

monitoring since the start of the Space Shuttle program.46 They provided low detection

limits, stable readings, and fast analyses. The effectiveness of leak detection was proven

in 1985 with STS-6 and later in 1989 with STS-35 and STS-38 when during the

cryogenic fueling process leaks were discovered, stopping the fueling process and



















































Figure 2-1. The fully assembled Space Shuttle (Atlantis). At launch the Space Shuttle
comprises three major parts: the solid rocket boosters (SRBs); the external
tank (ET); and the orbiter.


















































Space Shuttle main engines (SSMEs). Three SSMEs are located in the aft
end of the orbiter. Cryogenic hydrogen and oxygen are combusted to
produce 2 million Newtons of thrust from each SSME. The SSMEs together
with the solid rocket boosters place the Space Shuttle into orbit in only eight
minutes.


Figure 2-2.








avoiding a potential accident.46'47 Recently with STS-93, limitations of these systems

became apparent when the rupture of three coolant lines on a SSME was inconclusively

detected because it could not be confirmed because of a lack of multiple detectors around

the space shuttle.485

Present Day Gas Detection

Upon inquiry, NASA determined that the remote location and size of the detection

systems were limiting factors. The antiquated rack-sized systems (Figure 2-3) were

placed beneath the Space Shuttle in an environmentally controlled room of the mobile

launch platform (MLP) (Figure 2-4) where air conditioning and shock absorbing floors

helped minimize the impact on the sensitive and costly equipment of the violent launch.51

Gas sampled from the Space Shuttle therefore was transported over long distances to the

remotely located mass spectrometers, causing long delays from sample extraction to

analysis. Another limitation was that only two mass spectrometer systems could fit into

the MLP because of their large size. Just two of the five available gas lines could be

monitored simultaneously, though often both mass spectrometers were used to monitor a

single line for redundancy. Switching between gas lines imposed additional delays as the

gas load inside the line was replaced.

NASA begins monitoring the Space Shuttle two days prior to liftoff. First the

cryogen plumbing is pressurized with helium to trace for leaks. While sampling the

nitrogen purged compartments of the Space Shuttle, launch controllers look for helium

elevations, but too few sampling lines are available and so pin-pointing a leak is done by

hand, a time consuming process. The following day at 8 hours prior to liftoff mass

spectrometers are used again while the Space Shuttle is filled with liquid hydrogen and


































Hazardous gas detection system (HGDS). Large rack-sized systems are
used to detect leaks within the Space Shuttle. The original HGDS was built
in the late 1970s and was upgraded recently (2001) with new technology.
The HGDS' large size and sensitivity to launch vibration required it to be
placed remotely in an environmentally controlled room with shock
absorbing floors. The HGDS used two mass spectrometers (A and B) to
provide redundancy.


Figure 2-3.


















Service
Structure
(FSS)


226' (69 m)
227' (69 m)
200' (61 m)


Mobile
Launch
Platform
(MLP)
I I /. Platf.orm


Map of transport lines used for delivering sampled gas to the HGDS. With
the HGDS remotely located inside the mobile launch platform (MLP), gas
from the Space Shuttle had to be transported through transfer lines for
analysis. The long length of the transfer lines, between 61 m and 117 m,
caused a time delay from 25 s to 50 s between sampling and analysis.


Figure 2-4.








liquid oxygen. Monitoring continues until liftoff, but transport delays prevent analysis of

the final critical seconds when a lot of events occur such as checks of the gimble

actuators and SSME startup.

Advanced Hazardous Gas Detection Project

Noting the rapid advancement in mass spectrometer technology, NASA was poised

to develop a new leak detection system that would resolve present day limitations. In

2000 the aptly named Advanced Hazardous Gas Detection (AHGD) project was

contracted to Dynacs Engineering Company at Kennedy Space Center. The recent flurry

of work published on miniature mass spectrometers52 substantiated the possibility of

developing a miniaturized, rugged mass spectrometer unit for leak detection. The unit,

about the size of a toaster oven, would be placed up next to the Space Shuttle, thereby

reducing transport delay. The small footprint would allow many of these units to be

placed around and possibly within the Space Shuttle, providing launch controllers with a

continuous stream of data from an entire network of detectors. The added redundancy

would also minimize launch cancellations currently caused when an anomalous reading is

observed on one mass spectrometer, but cannot be confirmed with the other. The AHGD

unit will be of simple design, which lowers costs and allows quick hot swapping of

defective units to minimize downtime.

AHGD Leak Detection System Requirements

The analytical performance required of the AHGD prototype, detailed in Table 2-1,

is much the same as that required of the original mass spectrometer detection systems,5354

but the speed and size of the new system are significantly different as outlined in Table

2-2." The AHGD system is required to report ion signals for all four analytes every

second. This is considerably faster than the 30-s update period required of the current








system. The new 35,000 cm3, 10 kg sized AHGD system must also perform all the same

functions as the current 1.25 million cm3, 450 kg system.

Table 2-1. The AHGD project analytical performance requirements
Requirement Hydrogen Helium Oxygen Argon
Detection limit (ppm) 25 100 25 10
Accuracy (% error) <10 < 10 <10 < 10
Precision (% deviation) <5 <5 < 5 <5
Response time (s) <10 <10 <10 < 10
Recovery time (s) < 30 < 30 < 30 < 30
Mid-level 2-hour drift (%) <10 < 10 <10 < 10
Upper limit, dynamic range (%) 10 100 25 1

Table 2-2. The AHGD project system specifications
Characteristic Specification
Total system volume < 35,000 cm3
Total system weight < 10 kg
System power (from 1 15V 60Hz line) 230 VA
Vibration and shock 18 G, 8 Hz along 3 axes
Final production cost < $20,000
Data update rate 1 Hz

Such miniaturization is not trivial; a typical backing pump alone is half the target

volume of 35,000 cm3. Miniaturization requires use of the latest in microelectronics,

vacuum components, and small mass analyzers. Recently, many types of mass analyzers

have been miniaturized including linear quadrupole (Quad),56-62 quadrupole array

(QuadArray),63-66 QITMS,67-73 time-of-flight (TOF),74-81 and sector/cycloidal.82"90 All

were considered for the AHGD project, with representative prototypes supplied from

industrial manufacturers when available, and from academia when not. Commercial

QITMS instruments have been available for some time, but none were able to mass

analyze necessary hydrogen or helium ions. NASA turned to the University of Florida to

develop QITMS for the AHGD application

Before beginning QITMS development, the author evaluated four other mass

analyzers to provide a comparison basis: 1) a Quad, Figure 2-5; 2) a QuadArray, Figure








2-6; 3) a TOF, Figure 2-7; and 4) a cycloidal focus, Figure 2-8. The analytical

performance of each was evaluated using standardized test equipment and methodology

and their strengths or weaknesses were determined.91

Experimental Methodology and Equipment

Four analyte species are monitored during launch preparations. The molecular ion of

hydrogen, the cryogenic fuel, is monitored at 2 Da (the atomic ion signal at 1 Da is less

intense under standard electron impact (El) ionization conditions). Helium (4 Da) is

monitored, because it is used to trace plumbing leaks and is the pneumatic gas of the

SSME actuators (a leaking actuator could disable the associated SSME). Oxygen (32

Da), the cryogenic oxidizer, presents a significant fire hazard at high concentrations, and

because of the abundance of oxygen in air (20.9%), argon (40 Da) is monitored to

distinguish an air leak from a cryogen leak (argon is the third most abundant constant

component of air at 0.93%).

NASA has outlined the following evaluation methodology to be used with all mass

spectrometers for determining detection limits, accuracy, and precision, as well as the

response and recovery of the system to changing analyte concentrations. The AHGD test

procedure below is used to determine all of these criteria. The two-hour drift and

dynamic range requirements (Table 2-1) were not evaluated in this preliminary study,

because time and materials were not available for the required additional testing.

AHGD Test Procedure

The instrument is exposed to three gas streams: 1) "zero gas" (ZG), which is pure

nitrogen; 2) "test gas" (TG), which is a mixture of each analyte (500 ppm H2, He, 02, and

100 ppm Ar) in a balance of nitrogen; and 3) "span gas" (SG), which is a mixture of




















Stanford Research System model RGA-100 linear quadrupole mass
analyzer. An Alcatel ATH-30+ turbo drag pump was attached to a custom
vacuum chamber. A 0.002 in. inlet orifice admits the sample gas, while a
Granville Phillips Microlon gauge measures the pressure of the chamber.
To the right is the quadrupole assembly, which attaches to the control
electronics.


Ferran Scientific model POD-01 quadrupole array mass spectrometer.
Ferran assembled the shown system specifically for the AHGD project.
Inside the case, the system comprises a Ferran Scientific Symphony
quadrupole array mass analyzer (45 Da MPA and CNI-06 control
electronics), and a custom vacuum chamber evacuated by a Varian V70LP
turbo drag pump backed by a KNF diaphragm pump. Gas was sampled
through a long capillary, to allow sampling directly from atmospheric
pressure.


Figure 2-5.


Figure 2-6.











































lonWerks time-of-flight mass spectrometer. The flight chamber of the
lonWorks TOF (a) was compact when compared to other TOF instruments.
On the far left, a capillary tube enters an inlet cavity through a 2.75 in.
Conflat flange. A mechanical pump evacuates this region via the 0.5 in.
pipe (seen extending downward). An orifice admits gas from the inlet
cavity into the El ion source. An orthogonal extraction pulse samples from
a constant ion stream, sending ions into a "V" shaped reflector flight path
and impinging on an MCP detector. The TOF's fast speed required
sophisticated electronics that were contained in a large rack (b) that
dwarfed the mass analyzer.


Figure 2-7.


































Monitor Group MG2100 cycloidal focus mass spectrometer. The crossing
magnetic and electric fields of the MG2100 produced a double focusing
effect similar to that of a double focusing sector, but with a single analyzer.
A gas sample was drawn through a narrow bore capillary into the El ion
source. The ions traveled in cycloidal paths, focused through an aperture to
strike a Faraday cup. Ions of different m/z ratios traveling in different paths
were sequentially focused through the aperture by scanning the electric
field.


Figure 2-8.








each analyte at 10 times the concentration in TG. The following steps are made:

1. Run each gas line for a minimum of 5 minutes to remove air
2. Run ZG for a minimum of 15 minutes to stabilize the system
3. Begin data recording for 2, 4, 32, and 40 Da
4. Run ZG for 3 minutes
5. Run TG for 3 minutes
6. Run SG for 3 minutes
7. Twice, repeat steps 4 through 6
8. Run TG for 3 minutes
9. Run ZG for 3 minutes
10. Stop data recording

Analytical Procedures

Two-point calibration

Data acquired while running ZG and SG were used to generate a two-point

calibration. Data points between 2 and 2.5 minutes are averaged together for ZG and SG

(a 1 s update rate would average thirty points). Known concentration values certified by

the gas mixture manufacturer are used to generate an ion-signal-to-concentration plot of

each analyte. This is repeated for each ZG and SG set (total of three).

Calculated detection limit

From the three-minute ZG period of the test procedure, ten subsequent data points

after the two-minute mark are converted to values of concentration using the calibration

plots. The calculated detection limit is reported as three times the standard deviation of

the ten values.

Accuracy of quantification

Data acquired from 2 to 2.5 minutes period while running TG are averaged. This is

then converted to a concentration value ([test]meas.) using the calibration plots, and then

compared with the manufacturer certified value ([test]ce.). Accuracy is reported as the

percent error between the two values (Equation 2-1).








[test],e, [test]ce,. x 100 = %Error (2-1)
[test]ce,.


Precision of quantification

In the test procedure, the sequence ZG-TG-SG is repeated three times from which

three separate TG concentration values are measured; the average (PttestI) and standard

deviation (attestl) are calculated. Precision is reported as the percent deviation determined

using Equation 2-2.

'Iesx x 100 = %Deviation (2-2)
Al testj

Response time

Response time is defined as the time in seconds needed for transition from ZG to

within 95% of TG.

Recovery time

The recovery time is defined as the time in seconds needed for transition from SG to

within 5% of TG.

Gas Delivery Assembly

A differential-pressure, flow-by gas delivery system shown in Figure 2-9 was built

for the mass spectrometer evaluation. Modifications were made only when necessary to

work with a specific mass spectrometer. The delivery system included a gas selection

section and an inlet section. Gas cylinders of ZG, TG, and SG, each regulated to an

outlet pressure of 10 PSI, were plumbed through three computer-controlled solenoid

valves for fast switching. Once selected, a gas passed into a 4 in. i.d. plenum that was

regulated to a constant pressure with a mass flow controller (MFC) (MKS Andover,









MKS 146 Gauge Measurement Control System
MKS Flow Controller (1000 SCCM)
MKS Baratron Gauge (1000 Torr)


Granville-Phillips
Convectron Gauge


Computf Controlled
- Sample Valves
Zero Gas


Test Gas


Span Gas


To Orifice or Capillary
Inlet of Mass Spectrometer


To Vacuum Pump


Figure 2-9. The AHGD gas delivery system. Computer controlled solenoid valves were
used to select one of three gas streams: 1) pure nitrogen (ZG); 2) 500-ppm
H2, He, 02, 100-ppm Ar mixture in nitrogen (TG); and 3) 5000-ppm H2, He,
02, 1000-ppm Ar in nitrogen (SG). The gas stream was regulated to a
constant pressure (900 torr) via a feedback controlled mass flow controller
and adjusted to a 500 sccm flow by a regulator. A mechanical pump drew a
side stream into an inlet block at a pressure of a few torr. A sampling orifice
or capillary then delivers gas into the mass spectrometer.








MA model 1479A) operated by a MKS 146 control box in a feedback loop with a MKS

626A Baratron pressure gauge. A target pressure of 900 Torr was selected, and a flow of

approximately 500 SCCM was achieved by adjustment of a regulating valve situated

upstream of the MFC. A side stream was drawn out of the plenum (from between the

regulating valve and the MFC) by a mechanical pump. This ran past a sampling orifice

or capillary, which passed a small amount of gas into the mass spectrometer. The design

provided these benefits: 1) the high flow rate of the plenum provided rapid replacement

of the gas volume when switching gases; 2) the above-ambient pressure of the plenum

kept air from diffusing into the gas system; and 3) a low pressure (2 to 5 torr) side stream

allowed sampling into the high vacuum of the mass spectrometer, which was measured

with a Granville-Phillips (Hudson, NH) convectron gauge. Short length and small

diameter tubes were used where possible for speeding up gas switching.

Evaluated Mass Analyzers

Linear quadrupole

A Stanford Research Systems RGA-100 (Sunnyvale, CA) linear quadrupole mass

spectrometer was evaluated. Marketed as a residual gas analyzer (RGA), it was designed

specifically to monitor lightweight gases with a mass range from 1 to 100 Da. The

system came without a vacuum system, so a small vacuum chamber was constructed in-

house (Figure 2-5) to attach an Alcatel (Annecy Cedex, France) ATH-30+ turbo-drag

pump, backed by a Vacuubrand (Wertheim, Germany) MZ / 2D diaphragm pump. The

system performed best with a chamber pressure in the mid 10-5 torr range. The

quadrupole measured 11 cm long, with an ro = 4.5 mm and a drive frequency of 2.7648








MHz. The ion optics and quadrupole settings were tuned as outlined by the

manufacturer. Gas samples were admitted through a 0.002 in. orifice directly into the

vacuum chamber. Ions were generated in an El ionization source (-70 eV, 1 mA). A

Faraday cup was used for detection, though an electron multiplier was also available.

Quadrupole array

The QuadArray mass analyzer from Ferran Scientific (San Diego, CA) used sixteen

rods arrayed four by four, forming nine quadrupoles. Each rod was 1 cm in length with

an inscribed r0 0.5 mm. The drive frequency applied to all rods was 16 MHz, with a mass

range from 2 to 45 Da. The evaluated unit was custom assembled for the AHGD

application. A model Symphony mass analyzer (45 Da MPA, and a CNI-06 electronics

box) was attached to a small chamber evacuated by a Varian V70LP (Lexington, MA)

turbo-drag pump that was backed by a KNF (Munzingen, Germany) type UN84.4

diaphragm pump. The capillary inlet was designed to sample directly from a pressure of

1 atmosphere. Because of this, the capillary was attached directly to the high-pressure

plenum of the gas delivery system. The instrument was factory tuned; additional in-

house tuning was not possible. Ions were generated by El (-70 eV, 0.3 mA), and were

detected by nine Faraday cups.

Time-of-flight

The IonWerks system used an orthogonal-extraction, reflector-TOF packaged into a

10 in. x 10 in. x 2.5 in. vacuum chamber (Figure 2-7). A Varian V70LP turbo drag

pump backed by a BOC Edwards (Wilmington, MA) model RV-0.5 rotary vane pump

provided vacuum. Like the Ferran Scientific instrument, a capillary was used to transport

gas directly from the plenum (900 torr) to the instrument; however, unlike the Ferran a

flow-by inlet was built into the IonWerks system. A gas stream was drawn into a








chamber (roughed by the backing pump) and sampled through a 0.005 in. orifice into the

El ionization source (-70 eV, 0.015 mA). The detector was a custom "two-zone" micro

channel plate that was designed to increase the response range.92 IonWerks optimized the

TOF on location for monitoring lightweight gases over a mass range of 1 to 100 Da for

the AHGD project.

Cycloidal focus

The Monitor Instruments MG2100 cycloidal focus mass spectrometer used a crossed

magnetic and electric double focusing mass analyzer. The complete system was packed

into a small 9 in. x 13 in. x 23 in. case. A pair of permanent magnets surrounded the

vacuum chamber, which was evacuated by a Varian V70LP turbo drag pump backed by

two KNF 84.3 diaphragm pumps (used in series). The inlet capillary of the MG2100 was

connected to the in-house inlet block on one end and on the other passed into the EI

ionization source (-70 eV, 0.02 mA). The mass range was from 2 to 100 Da, achieved by

scanning the electric field of the mass analyzer, and a Faraday cup was used for a

detector with a variable gain electrometer (unique to this instrument).

Results of the Examination

The results of the examined analytical criteria are reported for each instrument in

Tables 2-3 through 2-6, and are detailed in the following sections. A significant

observation is that none of the instruments met all of the AHGD project requirements

(Table 2-1), which demonstrates the challenge of the AHGD project. The chapter closes

with concluding remarks on each of the evaluated instruments and with general

observations pertinent to QITMS development.








Table 2-3. Linear quadrupole analytical performance
Analytical criteria Hydrogen Helium Oxygen Argon
Detection limit (ppm) 10 12 7 3
Accuracy (% error) 15.7 3.3 16.1 2.2
Precision (% deviation) 1.4 1.8 0.8 0.3
Response time (s) 6 6 6 6
Recovery time (s) 6 6 36 6

Table 2-4. Quadrupole array analytical performance
Analytical criteria Hydrogen Helium Oxygen Argon
Detection limit (ppm) 58 6 55 5
Accuracy (% error) 44.7 58.4 11.5 4.6
Precision (% deviation) 27.9 10.8 5.2 5.9
Response time (s) 15 15 15 15
Recovery time (s) 15 15 15 15

Table 2-5. Time-of-flight analytical performance
Analytical criteria Hydrogen Helium Oxygen Argon
Detection limit (ppm) 56 47 80 16
Accuracy (% error) 11.2 60.8 20.3 5.4
Precision (% deviation) 1.9 8.3 4.4 7.5
Response time (s) 15 14 15 16
Recovery time (s) 31 33 32 30

Table 2.6. Cycloidal focus analytical performance
Analytical criteria Hydrogen Helium Oxygen Argon
Detection limit (ppm) 27 16 105 4
Accuracy (% error) 11.6 35.1 15.2 13.3
Precision (% deviation) 3.7 7.3 7.9 5.9
Response time (s) 22 22 22 22
Recovery time (s) 22 22 44 22

Detection Limits

A calculated detection limit is based on the deviation about the mean of an ion signal

acquired when sampling ZG. Ideally the ion signal would remain steady over time, but in

reality this does not happen. The background concentration of oxygen and argon may

drift because of virtual leaks in the vacuum chamber or trapped gas in the sample

delivery system. This is not a problem with hydrogen and helium, found at low

concentrations in air (nominally 5ppm and 0.5 ppm respectively); however, these gases








are difficult to remove with turbo-drag pumps once added (poor compression of

lightweight gases is discussed in Chapter 3), which can result in signal drifts. Low

frequency noise is cause by the signal drifts mentioned above. Higher frequency noise is

due mainly to variation in the emission current, the El ionization process, and the

dynamic gas composition within the ion volume at high vacuum (10-6 torr).

The calculated detection limits from Tables 2-3 to 2-6 are presented together

graphically in Figure 2-10. The Quad results appear better than the other systems. This

is somewhat misleading, since the Quad's update time was six times as long as that of the

TOF (update times for the four systems are shown in Figure 2-11). The TOF was the

only system to meet the 1 Hz AHGD requirement, and it may perform better than the

Quad if a longer update time were used, since more data points could be averaged.

The QuadArray had poor detection limits for hydrogen and oxygen, with a large

reported error (error bars indicate the 95% confidence interval). This was because of

large ion signal fluctuation, shown for oxygen in Figure 2-12 where the signal fluctuates

by more than 25% (the similar looking hydrogen signal is off scale). The detection limit

for helium was the lowest, but artificially so. The low background concentration of

helium produced a very low ion current when sampling ZG. Nearly all of the data points

were below the system's detection threshold, which caused them to be reported as zero

current. The data set of almost all zeros misleadingly showed low signal deviation.

The worst detection limit was reported for oxygen on the cycloidal focus system,

caused by a downward drift of the oxygen signal because of a virtual leak. This was

exacerbated because the detection limits were calculated from deviation of ten data









120



100 Hydrogen
U Helium
0 Oxygen
9 80 -0Argon


-60


40- T



20


~4Th


EmE E


0 1 1
Quad Quad Array TOF Cycloid


Figure 2-10. Detection limits for the evaluated mass spectrometers. The limits of
detection (based on S/N = 3) for hydrogen, helium, oxygen, and argon are
provided for the four mass spectrometers evaluated. The linear
quadrupole mass spectrometer (Quad) was the only system to meet the
AHGD requirements (Table 2-1). (Error is reported at the 95%
confidence interval).














20 1


15




0



5


0


Quad


Quad Array


Figure 2-11.


Update Rates for four evaluated mass spectrometers. The IonWorks TOF
met the AHGD requirement of a 1 Hz update rate, while the other systems
were considerably slower. The TOF analytical performance could be
improved if data were averaged from 6 to 22 times the measurement
period, as was the case with the other instruments.


TOF


Cycloid









4.OOE-07
TG
3.50E-07
--Helium

3.00E-07 Oxygen
I- Argon

2.50E-07

S2.00E-07

I .50E-07 __

1.00E-07

5.00E-08

O.OOE+O0 ,
0 100 200 300 400 500
Time (s)


Figure 2-12. Quadrupole array data collected while sampling pure nitrogen (ZG) and
then a 500 ppm mixture of hydrogen, helium, oxygen, and 100 ppm of
argon (TG). The oxygen detection limit was poor because of the large
amount of signal fluctuation (ZG data). The ZG helium ion signal was
below the mass spectrometer's detection limit, and the data points were
mostly zeros.








points, but with an update period of 22 s, this meant that the standard deviation was taken

over the entire three-minute period (compared to a 30-s for the TOF).

Accuracy

An accurate measurement would require a linear signal response, minimal signal

drift, and accurate gas cylinder certifications. Accuracy results for the four systems are

shown together in Figure 2-13. None of the instruments meet the AHGD requirement of

less then 10% error for all four gases. This was surprising for the Quad, which although

it was the closest with values for hydrogen and oxygen only slightly above 10%, was

expected to perform better. The SRS-100 Quad was considered to be the benchmark

instrument for the AHGD project. It had already been extensively evaluated with the

same analytical requirements in an earlier NASA project. Previous users of the Quad

were able to get all gases below 10%, which suggested that the problem was not caused

by the mass spectrometer but by the experimental setup.

To rule out inaccurate gas cylinder certifications as the cause, the experiment was

repeated with new gas cylinders. The new accuracy results were: hydrogen, 10.6%;

helium, 49.8%; oxygen, 7.4%; argon, 9.2%, showing improved oxygen results but worse

helium accuracy. These new gas cylinders were used with the other instruments.

Between all three instruments the measured helium concentration ranged from 660ppm to

785ppm, the average of which was within 1.3% of the latest Quad value. This agreement

substantiated that inaccurate certifications were the cause for the inaccurate helium

measurements, and possibly caused the high oxygen inaccuracy found in the first Quad

experiment.









100

90

80

70


30

20

10-

0-



Figure 2-13.


Quad Quad Array TOF Cycloid


Accuracy of the four evaluated mass spectrometers. The % error in
calculating a median concentration using a two-point calibration is
reported for the mass spectrometer systems evaluated. An erroneous gas
cylinder certification resulted in inaccurate helium readings for three
systems (a different gas cylinder was used for the reported Quad data).
None of the mass spectrometers met the < 10% NASA requirement for all
four gases. (Error is reported at the 95% confidence interval).








Precision

The AHGD project required less than 5% deviation throughout the three runs. Only

the Quad met this requirement as illustrated in Figure 2-14. The TOF and cycloidal focus

systems were within 10%, and the QuadArray showed a significant deviation only for

hydrogen which stemmed from large fluctuations in the TG data, caused by unstable ion

transmission at a low RF amplitude (Vp 0C m/z).

Response and Recovery Time

In addition to a long update period, slow response or recovery to a changing gas

concentration would impact the system's effectiveness for the AHGD application. The

10 s and 30 s requirements for response and recovery, respectively, are generous; NASA

would prefer to have the time further reduced. Yet even at these values, only the Quad

met the response time requirement (Figure 2-15). The other instruments fared better with

the longer recovery time requirement (Figure 2-16). An anomalous reading caused the

Quad oxygen recovery time to be above 30 ss.

The TOF results were above the requirements, because of its built-in flow-by inlet.

The design was similar to the in-house inlet shown in Figure 2-9; a subtle difference

resulted in poor response and recovery times. The capillary diameter was much smaller

than that used on the in-house inlet, which significantly reduced flow into the inlet cavity

and increased the time needed to replenish the inlet's volume and attached vacuum tubes.

This caused the slow changing signal shown in Figure 2-17.

Conclusions and Consideration for Future Work

The Quad provided the best analytical performance of the four systems. Still, the

Quad was slow with an update period of six times the AHGD requirement. The update













n Hydrogen
M Helium
o Oxygen
0 Argon


Quad


Quad Array


Figure 2-14.


Precision of the four evaluated mass spectrometers. The % deviation of
three repetitive runs is reported for the four mass spectrometer systems.
The linear quadrupole (Quad) was the only system to meet the < 5 %
NASA requirement.


30



25-



20


15-


10



5



0-


TOF


Cycloid









25-



20-
















0




Figure 2-15.


Quad Quad Array TOF Cycloid


Response time of the four evaluated mass spectrometers. The linear
quadrupole (Quad) was the only mass spectrometer to meet the < 10-s
NASA requirement. Both the quadrupole array and the cycloid focus
instrument have slow update rates that prevent a precise determination of
the actual response time. The TOF instrument responds slowly because
of a poor built-in inlet design.










50-

45

40

35

030

~25-
.
0--
0
Fu 20

15

10









Figure 2-16.


Quad Quad Array TOF Cydoid


Recovery time of the four evaluated mass spectrometers. The NASA
requirement of < 30 s was met only by the quadrupole array (Quad
Array) for all four gases. Both the linear quadrupole (Quad) and the
cycloid focus (Cycloid) had anomalously high recovery times for oxygen.
The oxygen ion signal of the Quad began to reduce at the same time as
the other three gases, but stopped for a few data points before continuing
to decrease to the final level. The Cycloid had a 22 s update rate, which
imprecisely defined the response time with only one or two data points.









4000


3500


3000


2500

C
@2000
C

1500


1000


500


0


150


Hydrogen
* Helium
Oxygen
. Argon


TG




ZG










....... .... ...................f
.S SS*USU.333535.SSS.SS*UWSU.SU3SomSS


170


190


210


230


250


Time (s)

Figure 2-17. TOF data acquired while transitioning from pure nitrogen (TG) to a gas
mixture (TG: 500 ppm each of hydrogen, helium, oxygen, and 100 ppm
of argon in nitrogen). The transition occurs gradually over approximately
15 s because of the built-in inlet design. A narrow bore capillary
provides a minimal flow causing slow replacement of the gas volume
inside the inlet.








rate can be increased; however, a significant degradation of performance results. This

was a tradeoff common to all of the mass analyzers evaluated. The Quad did have the

advantage of a higher emission current, which was from 3.3 to 67 times the emission

current used on the other systems. More electrons meant more ions, resulting in

improved performance up to the space charge limit of the Quad.

Quadrupole array technology was intended to provide performance comparable to a

standard linear quadrupole in a small package. The QuadArray was the smallest mass

analyzer (Figure 2-18), but contrary to this intention it performed worse than the Quad.

The QuadArray was able to perform about the same as the TOF and the cycloidal focus

with all gases except hydrogen. The potential well (Chapter 1) at 2 Da is shallow, which

caused poor transmission efficiency, resulting in signal fluctuations. The QuadArray was

also designed to operate at a higher pressure than the other analyzers. This was meant to

reduce pumping requirements, but the operating pressure of 104 torr did not provide any

benefit since a turbo-drag pump was still required.

The IonWerks' TOF had the largest mass analyzer and vacuum chamber of the

group. This was a significant disadvantage, made worse by the large rack of electronics

needed to run the instrument. The instrument was fast, but had the shortest update

period.

The cycloidal focus instrument was packaged nicely into a portable case that met the

AHGD size requirement. The system was also the slowest with an update period of 22 s.

The manufacturer indicated that the electrometer was the limiting factor. It had a

variable gain, but even at the fastest setting it remained slow. This was because a 1012








ohm feedback resistor that prevented quick current changes, which required the electric

field to be scanned slowly.

The preliminary evaluation of these mass spectrometer systems revealed the

importance of a well-designed inlet, one that had a high flow with minimal dead volume

and used a sampling orifice instead of a capillary. It was also discovered that gas

cylinder certification values were not accurate. This would lead to the development of an

alternative gas preparation system (Chapter 3). The study also demonstrated the

difficulty in achieving the analytical requirements of the AHGD project at the required

update rate of 1 Hz. This was later achieved with QITMS technology (Chapter 6), the

development of which is presented in the next chapter.
















0.5" 1.33.3


I I





Figure 2-18. Ferran Scientific miniature quadrupole array. The ion source, mass
analyzer, and detector were combined into a small package less than 1.5
in. long. A four-by-four array of lcm long rods formed nine
quadrupoles.













CHAPTER 3
LIGHTWEIGHT GAS MONITORING BY QITMS

Introduction

When Wolfgang Paul patented the QITMS in 1960, he realized its applicability to

analyzing trace components in a sampled atmosphere.'0 However, commercial

development of the QITMS occurred decades later with the advent of mass-selective

ejection used with a helium collision gas inside the ion trap.8'9'12 The use of this collision

gas precluded QITMS use for trace analysis of hydrogen and helium, and previously little

effort was made to fill this void. This chapter presents the development of modem

QITMS instrumentation for the analysis of lightweight gases.

History of QIT with Hydrogen and Helium Ions

In its early history the QIT was used to store hydrogen and helium ions. In the

1960's, Dehmelt utilized ion traps to contain hydrogen93 and helium94 ions in

spectroscopy studies of physical and chemical processes. Later, Fulford and March used

a QUISTOR for studying ion-molecule reactions of helium ions with molecules in air.95

Alheit et al. employed atomic and molecular hydrogen ions to characterize instabilities

associated with higher-order anharmonicities within the ion trap.96"98 But, in all the

above cases, the ion trap was used exclusively as a storage device; the mass spectrometry

was performed by ejecting ions into another mass analyzer, typically a TOF or linear

quadrupole. Dawson was the first and until now the last to report on mass-analyzed

hydrogen ions with a QITMS.24 '99 Dawson used the mass-selective storage mode, which

involved trapping and detecting ions one mass-to-charge (m/z) value at a time. At this








time, Dawson suggested that the QITMS, when operated in this mode, would be useful

for performing leak detection.100 Compared with mass-selective ejection, Dawson's

method was less practical for performing trace analysis of multiple permanent gases,

because it was more complex and time-intensive.9

Modern QITMS

Mass-selective ejection is the mass analysis method used on all commercial'0' and

most research QITMS instruments, because of numerous benefits. It is more time-

efficient with a high duty cycle, large dynamic range, high sensitivity, and a lower

cost.9,2 Operating in this mode, mass analysis depends on the ion trap dimensions (ro

and zo in in), the angular frequency (Q in rads/s), and the RF amplitude (Vp in Vp) as

related through the Mathieu equation (discussed in Chapter 1) shown in its axial

transformed state in Equation 3-1. To extend the QITMS mass range up to thousands of

Daltons, commercial QITMS instruments are set to ion trap parameters that sacrifice their

ability to analyze hydrogen and helium. This is best pictured with the stability diagram

shown in Figure 3-1. At the lowest stable Vp of a commercial instrument, helium ions

(He+) and hydrogen ions (H2+) are beyond the stability boundary at q, = 0.908.

q- 8eVP (3-1)
m(ro2 + 2z 031

QITMS performance is improved with the use of a collision gas to dampen ion

motion.'9 This process is most efficient with a large mass difference between the analyte

and the collision gas species.7'99 Therefore, lightweight helium is typically used as the

collision gas, with improved ion signal and mass resolution most noticed at m/z values

greater than 100 Da. Consequently, permanent gas analysis benefits little from the use of







az
0.2-
% =O0.908
0.12

0.0 --q
-0.1

-02 H;

-0.3


-0.4

-0.5

-0.6

-0.7
Figure 3-1. Location of hydrogen and helium ions on the stability diagram. Available
commercial QITMS instrumentation offers a minimum mass cutoff
(q,=0.908) of 6Da. This places hydrogen ions at q,=2.724 and helium ions
at q,=1.362, well beyond the stability boundary.








a collision gas. Mass resolution is unimproved, and sensitivity is, at most, doubled.102

Gases of the same or lower mass than the collision species are difficult to analyze,

because scattering occurs and ionized collision gas presents a large amount of chemical

noise.

A collision gas was also used for ion injection from an external ionization source.

Ion injection was first suggested by Dawson et al. as a means to adapt QITMS to other

ionization techniques such as plasma discharge.100 Since then many ionization types

have been adapted for QITMS including electron impact ionization (El), chemical

ionization (CI), electrospray ionization (ESI), atmospheric pressure chemical ionization

(APCI), and matrix assisted laser desorption ionization (MALDI). External ionization

sources are used in cases when high pressures are needed, or when neutrals from the

ionization region would interfere with ion trap operation.3 Modem GC/QITMS

instruments use combination El/CL external ion sources. Though both El and CI can be

performed on an internal ionization ion trap,7 an external ionization source would permit

higher reagent gas pressures for faster analysis, and also allow analysis of negative ions

by separating them from positive ions, which is not possible with internal ionization

source.7,42

An injected ion has a large initial displacement at the surface of the entrance endcap;

therefore, the ion oscillates with a large amplitude of motion, and without any opposing

dampening force, the ion will quickly strike an electrode and be neutralized.41' 5 A

means must be provided to relax the motion of ions to make ion injection possible. The

collision gas, used earlier to improve sensitivity and mass resolution, now serves to

reduce the displacement of injected ions. Other alternatives included electrostatic








deceleration and pulsed ion trap operation, but are less efficient and more elaborate.

Even with a collision gas, the trapping efficiency for externally injected ions is < 5%,

depending on the injection angle and the RF phase angle at injection.'014'5 With internal

ionization only the RF phase angle limits trapping efficiency, with values reported

between 15% and 40% efficient.27

Instrumentation and Equipment for Lightweight Mass Analysis

This section describes two QITMS systems developed for the AHGD application

(described in Chapter 2); each used mass-selective ejection. The first used internal

ionization with no helium gas, in an open trap configuration. The second used all the

latest QITMS innovations from Finnigan. With these two instruments modem QITMS

was explored for low-mass analysis, an application envisioned since the early days of the

ion trap.1'10'

University of Florida Custom QITMS (UF-IT)

A custom QITMS was assembled at the University of Florida to be more compact

and simpler than commercially available QITMS, while achieving the analytical

requirements of the AHGD project. Components were recycled from available

commercial hardware when possible, and were engineered to work with low-mass ions

between 2 Da and 40 Da. Custom software was written to operate the instrument, and

acquire data in a format suitable to the AHGD project.

Mass analyzer, source, and optics

Instrument assembly began by selecting an appropriate mass analyzer. The ion trap

chosen for the UF-IT was that of a Finnigan (Austin, TX) ITS-40 GC/QITMS (circa

1990) shown in Figure 3-2. It had already been designed for internal ionization, and no










2.22"


Figure 3-2.


Exit
Endcap
Ring
Electrode
Entrance
Endcap


The ion trap assembly of a Finnigan ITS-40 used in the UF-IT custom
instrument. Three electrodes are stacked together with two quartz spacer
rings that serve to electrically isolate and close off the ion trap. The
external dimensions are as shown; the internal dimensions are ro = 1.000
cm, zo = 0.785 cm.








collision gas was required, plus the optics were simple and compact. The ion trap was

machined with hyperbolic surfaces that better approximated an ideal quadrupole field,

and would have been difficult to reproduce on a home-built ion trap.

The ion source components of the ITS-40 were also used (shown in Figure 3-3),

since they were compact and were already designed to work with the selected ion trap. A

bias voltage set between the rhenium filament and the tube lens produced an electron

beam turned on and off by the gate lens. The optics sat directly inside the entrance

endcap, taking up little space and only the filament assembly was outside, protruding

from the Conflat flange.

The solid ring spacers were replaced with three ceramic standoffs as shown in Figure

3-4 to increase conductance of gas through the ion trap (an "open" configuration). Both

endcap electrodes were electrically grounded by connection to the Conflat flange, which

also held the assembly together on three rods.

Vacuum system

The ITS-40 vacuum manifold (Figure 3-5) was selected for the UF-IT having been

designed to mate with the ion trap assembly, and already being compact in size. The

transfer line of the ITS-40 was replaced with a shorter 3.5 in.-long and '8 in.-inner

diameter (i.d.) tube (Figure 3-6) designed to easily connect with a 0.001 in. inlet orifice

(O'Keefe Controls, Trumbull, CT). Gas traveled faster by molecular flow with the

shorter, larger i.d. tube, which ended inside the ion trap. A 1-inch NW-16 fitting added

to the vacuum chamber allowed connection of a Granville-Phillips model 354 Micro-Ion

high-vacuum gauge, placed 900 to the chamber wall to prevent gauge-generated ions

from striking the electron multiplier.







Conflat
Flange


Tube
Filament Lens
Assembly


Entrance
Ceramic Endcap
Spacers Eda


Gate
Lens


Figure 3-3. Electron optics of an ITS-40 used on the UF-IT. An electron beam from the
rhenium filament is focused through the tube lens, and gated on and off
from entering the ion trap by the gate lens. The electron optics are recessed
into the entrance endcap, taking up minimal space. The filament assembly
and the ion trap are bolted on opposite sides to the Conflat flange. (Adapted
from Finnigan ITS-40 Schematics Manual, p. 2-15.)







Ceramic
Standoff


/Conflat


Flange
Figure 3-4. Open-configuration on the UF-IT ion trap. The ring spacers of the ITS-40
ion trap were replaced by ceramic standoffs to increase gas conductance
through the ion trap. The assembly is bolted together onto a vacuum
Conflat flange.









Conflat Seat
for Ion Trap 0


RF Feed-
SThrough

Electron-
Agdooo Multiplier


Figure 3-5.


The vacuum manifold, turbo-drag pump and chassis of the UF-IT. The
vacuum manifold of an ITS-40 was selected for its compatibility with the
ITS-40 ion trap assembly and its compact size. A Pfeiffer TPH-065 turbo-
drag pump was selected for its high compression ratios for lightweight
gases. Components were secured to the shown chassis constructed of in.
aluminum plate.




























The UF-IT transfer line. A 3.5 in. long, 'A in. i.d. transfer line was used to
transport gas samples from the 0.001 in. inlet orifice directly into the ion
trap. This configuration performed better than fused silica for rapid
response to changing gas concentrations.


Figure 3-6.








The high-vacuum was provided by a Pfeiffer TPH-065 turbo-drag pump that offered

high compression ratios. Two backing pumps, shown in Figure 3-7, were evaluated

independently: 1) a Vacuubrand (Wertheim, Germany) MZ/2D diaphragm pump that

featured oil-free operation for minimal contamination (low hydrogen background) in a

compact size, and with a base pressure of 2 torr; and 2) a BOC Edwards (Wilmington,

MA) rotary vane pump used with an oil mist trap to minimize contamination, which had a

low base pressure of 5 mtorr, but was three times the size of the diaphragm pump.

Control, source, optics, detector, and acquisition electronics

The UF-IT was built with electronics from a Finnigan GCQ GC/QITMS (circa

1995.). The main system board (MSB), shown in Figure 3-8, contained circuits for

digital control, system monitoring, serial communication, the lens voltage power

supplies, and a filament power supply designed for an external El ionization source. The

filament circuit was modified to run the UF-IT filament, keeping it in regulation. One of

the lens voltage supplies on the MSB (lens 1 of the GCQ) was connected to the tube lens

of the UF-IT, which was originally grounded on the ITS-40. This provided an additional

means for tuning the electron beam. The switching voltage supply designed for the gate

lens of the GCQ was applied to the UF-IT gate lens. The on/off polarities, configured

originally to positive ions, were reversed to gate electrons.

The GCQ detector electronics were also used, which included the DeTech model

2312 electron multiplier (105 gain), the high voltage power supply for the electron

multiplier (variable from 500 to 3000 V), the electrometer (shown in Figure 3-9), and the

16-bit data acquisition card. A holder for the electron multiplier was designed to

accommodate the vacuum manifold with its feed-through connectors, and was positioned

directly below










































Figure 3-7. Backing pumps used with the UF-IT. Two mechanical pumps were tried for
backing the turbo-drag pump: A) a BOC Edwards RV3 rotary vane pump
with oil mist trap, base pressure 5 mtorr; B) an oil-free Vacuubrand MZ/2D
diaphragm pump, base pressure 2 torr. The compact size and oil-free
operation of the diaphragm pump was more in line with the small size
requirement of the AHGD project, but the high base pressure was not
adequate for backing the TPH-065 turbo-drag pump. Combined with the
rotary vane pump, the TPH-065 was able to pump hydrogen faster from the
vacuum chamber.





















,nte-









Electronics



Figure 3-8. The main system board (MSB) of the UF-IT. The MSB contained integral
electronic circuits, including the high-speed serial computer interface, the
digital and analog control circuitry, and the electron filament and optics
power supplies. The board was the bulkiest component of the system, but
contained circuitry beyond that used on this system, including three more
lens power supplies, and two heater supplies. Eliminating the unnecessary
circuits and switching to surface-mount components would reduce the board
size significantly.






















i7










Figure 3-9. The electrometer board, RF control board, and RF amplifier of the UF-IT. In
addition to the MSB, three other GCQ circuit boards were incorporated into
the UF-IT system: the electrometer board; the RF amplifier; and the RF
control board. The electrometer was cut away from a larger GCQ circuit
board, and was made to fit with the ITS-40 hardware. The RF amplifier
with the RF control board and the RF load circuit (RF coil and ion trap)
formed a feedback loop to control the RF amplitude applied to the ring-
electrode. A 24 in. ruler suggests the size of the UF-IT.








the exit endcap; no dynode was used. The GCQ electron multiplier did not require

additional impedance matching with the electrometer circuit, which was connected to a

16-bit analog-to-digital converter (ADC) that provided more precise acquisition than the

ITS-40 12-bit counterpart.

Drive circuitry

The RF drive circuitry of a GCQ was adapted to work with the ITS-40 ion trap (a

drawing of the circuit is shown in Figure 3-10). The waveform generator, located on an

expansion card inside the control computer, put out a 5 Vpp high-impedance signal at a

user-selected frequency between 1 and 5000 kHz. This signal passed to an RF amplifier

(Figure 3-9) where the waveform was multiplied with a variable DC voltage generated by

a 16-bit digital-to-analog (DAC) located on the RF control board. The variable DAC

output modulated the amplitude of the RF signal as needed for mass analysis. Power

transistors increased the output-power of the amplifier. The signal passed through an

impedance matching capacitor and onto an RF coil transformer, which increased the

signal amplitude by a factor of 200. A detect capacitor (0.25 pF) sampled the final

amplitude, which was compared with the RF DAC setting for feedback control. A high

voltage vacuum feed-through transmitted the signal to the ring electrode.

Three RF coils with different diameters were fabricated (shown in Figure 3-11), each

with multiple tapping points. This allowed the RF frequency to be varied between 1 and

4 MHz in small increments. The RF coil was placed inside an aluminum box which was

used for electro-magnetic shielding (Figure 3-12).









Detect
Capacitor


Figure 3-10. The RF drive circuit diagram of the UF-IT. An RF waveform is DC
modulated and power amplified by the RF amplifier (AMP). The RF coil
than steps up the amplitude to be applied to the ring electrode of the ion
trap. A fraction of the final voltage is passed through the detect capacitor
for feedback regulation of the output. The primary and secondary sides
of the RF coil are RLC networks, which must resonate together for
optimal power transmission. The matching capacitor is used to adjust the
primary resonant frequency to coincide with the secondary resonant
frequency.


































Figure 3-11.


RF coils used on the UF-IT. The three RF coils were fabricated to change
the frequency of the RF signal applied to the ion trap. A range from 1 to 4
MHz was achieved with the coils, each with multiple tapping points: A)
1.0 to 2.3 MHz, o.d. = 3.48 in.; B) 1.9 to 3.1 MHz, o.d. = 1.85 in.; C) 2.8
to 3.9 MHz, o.d. = 1.49 in..












































Figure 3-12. Power supplies of the UF-IT. AC and DC power supplies from the GCQ
were used on the UF-IT. The DC power supplies were installed into a
duct-like portion of the chassis where fans provided forced air cooling.
An AC line voltage switch allowed use of either 110 V or 220 V 60 Hz
power. An AC transformer converted the line power to multiple
amplitudes as needed to run the UF-IT.