ii

i ”I.“

'l ‘ . '. . ‘ r2

' at} “-8“ ‘3

ii 12$"?- «96%.:
3' 1' ‘4”
.

. L
. «q ..
' > ,4 .

‘-
u

. V
I h" M” -“ ~..
v - ‘ é! iii!” fl‘k‘kn .
. . ‘7' f .- '1‘ ’1‘? 'C. §¢§§zgr> 5‘2““?
L‘ 9 - J IV r - .-
. ‘xfiaou - '1? 1. .' V M
u.“ \_ ,.. c ._ ‘ t J. W a.
Yaw-5L ‘ b-
u . 2‘
A a __ ,5; A h
2 ,- “.1: ‘ . ’ ‘ .
‘ ,- .5113; ‘ . <
rt‘ «EL
.fiéfi..:n . ‘ ‘ ‘ l
‘ "73': a Q ~ *5
._,.,

1.,
42 Q .5
Ma;- ‘33
”‘1‘? 3?}? e

’ ‘2

Z '33“le

k . _ .

‘W‘E'fiaéfiw, .
' '.. K

3:." a;

r

(1.275
r

 

.;.,
"11:24 w
"1' vn

I 1 w
r’ ,
4w
w
‘1 ,
\‘w ‘
“$41th

.I _ ‘
.3 3 . .

 

,. “J‘w
fin “‘ 93:13.1; .r, 1
'11," 3 55:1“ » 5?.
xxx-«men . a .5

*5 v

Hg... 'f 3‘6»
35;;{w'r' V

£- A

THEScS

This is to certify that the

dissertation entitled

A Tandem Time-of-Flight Mass Spectrometer:
MS/MS on the Chromatographic Time Scale

presented by

Mary Ann Seeterlin

has been accepted towards fulfillment
of the requirements for

Ph. D . degree in Chemis tT'Y

[% Zé/e

Major professor

 

 

Date 1/24/94

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

‘ nit‘tiiititfiiiiu

 

MBRARY
Michigan 32am
University

 

 

 

PLACE N RETURN BOX to remove thin checkout from your record.

TO AVOID FINES return on or bdore dole due.

E

DATE DUE DATE DUE DATE DUE

L___J

-L_j
--

1
I
i
\

--*
r?

 

 

 

 

 

 

 

 

 

 

 

%

 

 

 

 

 

     

 

 

 

   

 

 

 

 

 

 

 

 

r-

MSU leAn Affirmative Action/Equal Opportunity lnetltuion
W

1
1

m1

 

 

 

A TANDEM TlME-OF—FLIGHT MASS SPECTROMETEFI:
MS/MS ON THE CHROMATOGRAPHIC TIME SCALE

BY

Mary Ann Seeterlin

Advisor: Christie G. Enke

A DISSERTATION

Submitted'to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

.1993

ABSTRACT

A TANDEM TlME-OF-FLIGI'IT MASS SPECTROMETER:
MS/MS ON THE CHROMATOGRAPHIC TIME SCALE

BY

Mary A. Seeterlin

The tandem time-of-tlight (T OF/T OF) instrument constructed has been
designed to obtain complete MS/MS spectra from compounds eluting from a
capillary gas chromatographic column, while maintaining unit mass resolution for
both precursor ion selection and product ion analysis. To meet this goal, the
instrument needs to acquire at least 10 product spectra per second. Such
spectral generation rates require maximum utilization of sample molecules and
ions, efficient fragmentation of precursor ions, and array detection of the product
ions produced. The use of continuous electron impact ionization with ion
storage between pulsed extractions, TOF separation and reflectron focusing of
precursor ions, pulsed laser photo-induced dissociation (PID), TOF separation
and reflectron focusing of product ions, and array detection of all products

formed, meets each of these criteria.

The electron impact source designed for this instrument utilizes two
opposing filaments which continuously emit electrons. Reduced ion loss rates
achieved by this source yield a high density of ions in each parent ion packet.

An extraction pulse applied to the rear electrode accelerates the ions out of the

ionization region. The spatial distribution of the ions in the source is focused at

a space-focus plane just outside of the source [1].

Precursor ion packets are separated due to their mass dependent
velocities. A grid-free reflectron [2] focuses the space-focus plane image at the
interaction region. The timing of a 15 ns excimer laser pulse selects the desired
precursor ion packet with unit mass resolution to 1000. The high degree of
spatial and temporal overlap obtained by this instrument provides optimal

photodissociation efficiency.

After fragmentation, acceleration of the ions just following the interaction
region gives the product ions mass dependent velocities and reduces the
relative range of the product ion kinetic energies. A second reflectron will focus

the interaction region image at a detector.

This system is capable of generating 500 transients per second (limited
by the laser). lf approximately 50 transients are summed for each product
spectrum, 10 product spectra per second will be produced. With this rate of
acquisition a substantial amount of MS/MS information can be obtained for a

chromatographically eluted component.
References

1 B. A. Mamyrin, V. I. Krataev, K. V. Shmikk, and V. A. Zagulin, Sov. Phys.
JETP, 1973, 37, 45.

2 R. Grix, U. Gruner, G. Li, H. Stroh, H. Wollnik, Int J. Mass Spectrom. Ion
Proc., 1989, 93, 323.

flight - of - time

ACKNOWLEDGMENTS

Often, as I explained to friends and family my thesis work, they would say
that it sounded much like a job. I must admit that some days it felt more like a
job than school, but looking back at how I've grown over the last five years, I
realize that the project itself was the means through which I was taught, like
homework in a course. The product of this thesis is not simply an analytical
instrument but rather my personal growth - in confidence, independence and
critical thinking. I am indebted to many for my education. Foremost, I
acknowledge and thank Dr. Enke. Always there at an impasse, and always far
away when I struggled through the problems which I was capable of solving on
my own. His guidance, enthusiasm and trust have been cherished. For my
degree in philosophy, I couldn't have made a better choice of advisor.

Studying with others has always made learning not only much more
enjoyable, but also much more efficient and insightful. ln research, I found this
to be equally true. Working with George Yefchak, Raimond Grix, Dave McLane,
Ben Gardner, Paul Vlasak, Doug Beussman, Sherri Delbert and Qinchung Li has
been great. Throwing ideas back and forth, no matter how undeveloped,
between such receptive minds has been very stimulating, and has helped me to
gain confidence with my ability to be creative.

This thesis is also the product of many people who enjoy learning so
much that they want to share it. I have been lucky to have had Mr. Levigne, Mr.
Oldham, Mr. Male, Mr. Burgett, Mr. Rule, Mr Stallcup, Dr. Blecker, Dr.

Richardson, Dr. Kren, Dr. Salari and Dr. Cope, Dr. Crouch, Dr. Allison, Dr

Reusch and Dr. Leroi. Every step in my career as a student, I have always had
an inspiring guide. Luckier still am I to have learned with great friends. Jim
Ridge, my study buddy, has taught me more than any other teacher. Learning
together has been fun. Mark Cole, and his contagious enthusiasm for
knowledge has taught me about science, politics, life and myself. I will always
be thankful for this.

Last, but most important, I would like to acknowledge my family. They
have and continue to give me encouragement and guidance in life and I could
not have gotten here without them. My dad and mom showed me how to see my
goal, see it finished, and make my work something to be proud of. They also
showed me to enjoy life along the way. With my brothers I learned the secrets
and joys of working with a team, and as I have been recently reminded, that a
team members success is often twice as sweet. John, my husband and closest
friend has shared and shares the joys, excitement and frustrations of life, making
everything that much more worth doing.

Wonderful family, teachers, friends, health . . for these I thank God and

trust him with my future.

Table of Contents

 

 

Chapter 1: Introduction .................................................................. 1
1.1 Tandem Mass Spectrometry (MS/MS) ................................................... 1
1.2 Mass Spectrometers .............................................................................. 4
1.3 Existing Instrumentation for MS/MS ...................................................... 7
1.4 MS/MS on the Chromatographic Time Scale ........................................ 8
1.5 A Tandem Time-of—Flight Mass Spectrometer ....................................... 9
Chapter 2: Time-of-Flight Mass Spectrometry ............................. 16
2.1 Advantages of Time-of—Flight Analysis .................................................. 16
2.2 TOF Resolution - Problems ................................................................... 20
2.3 TOF Resolution - Breakthroughs .......................................................... 20
2.4 Time-Array Detection ...................................................... '.24
Chapter 3: An Electron Impact Ionization Storage Source ......... 28
3.1 lon Storage for TOFMS ......................................................................... 28
3.2 The Storage Well .................................................................................. 29
3.3 A Mathematical Model of Ion Storage ................................................... 35
3.4 Effects of Helium & High Frequency Field Modulation .......................... 41
3.5 Effects of Source Design ....................................................................... 51
3.6 The TOF/TOF Source ........................................................................... 54
Chapter 4: TOF/TOF Instrumentation ............................................ 59
4.1 Instrument Overview .............................................................................. 59
4.2 Description ............................................................................................ 62
4.2.1 Vacuum Chamber ........................................................................................... 62
4.2.2 Vacuum System ............................................................................................. 64
4.2.3 Sample Introduction ........................................................................................ 64
4.2.4 First TOF Analyzer .......................................................................................... 65
4.2.4.1 Source Extraction --
4.2.4.2 High Voltage Shield 65
4.2.4.3 Einzel Lens and Steering Plates --

 

vii

4.2.4.4 First Reflectron 68

 

 

 

 

 

 

 

 

4.2.4.5 First Detector 68
4.2.5 Interaction Region ........................................................................................... 71
4.2.5.1 Optical Rail 73
4.2.5.2 Gate 73
4.2.5.3 Light Tube 74
4.2.6 Second TOF Analyzer ..................................................................................... 76
4.2.6.1 Post Acceleration 76
4.2.6.2 Reflectron 2 76
4.2.6.3 Detector 2 76
4.3 Performance .......................................................................................... 77
4.3.1 Resolution ...................................................................................................... 77
I 4.3.2 Beam Image ................................................................................................... 77
4.3.3 Preliminary Gate Studies ................................................................................ 81
Chapter 5: Photo-Induced Dissociation ........................................ 84
5.1 Introduction to Photo-Induced Dissociation ........................................... 84
5.2 PID in the TOF/T OF Instrument ............................................................ 86
5.3 Results and Discussion ......................................................................... 87
5.3.1 Precursor Selection ......................................................................................... 87
5.3.2 PID Efficiency ................................................................................................. 89
5.3.3 Photoionization ................................................................................................ 94
Chapter 6: Product Ion Analysis .................................................... 99
6.1 Historical Use of TOF for Product Ion Analysis ..................................... 99
6.2 TOF Product Ion Analysis in the TOF/T OF Instrument ......................... 100
6.3 Theoretically Modeled Resolution ......................................................... 101
6.4 Performance .......................................................................................... 103
6.5 The Second Reflectron .......................................................................... 105
Chapter 7: Conclusions .................................................................. 113

1.1

1.2

1 .3
2.1

2.2

2.3

3.1

3.2

3.3

3.4

LIST OF FIGURES

Schematic representation of the collection of an MS/MS product

spectrum. ..................................................................................................... 3

Diagram of a time-of-flight mass spectrometer. Ions formed in the
ion source are accelerated out periodically. Isomass ion packets
separate in time due to their mass dependent velocities and are

detected. ...................................................................................................... 6

Schematic diagram of the tandem time-of-flight mass instrument ................ 12

Diagram of a quadrupole mass spectrometer, demonstrating the

filtering nature of mass analysis ................................................................... 19

Depiction of the ion spatial and energy effects on TOF resolution.
Ions in the source have a spatial and energy distribution which

account for a distribution of arrival times at the detector. ............................ 21

Schematic representation of the energy focusing effect of the
reflectron. Ions at the space-focus plane have a distribution of
energies. An ion with more energy spends more time on the
potential hill created by the reflectron than a slower ion. If the
reflectron is tuned properly this extra time spent in the reflectron
matches the difference in time the ions spend in the field free

region, and they come to a temporal focus at the detector. ......................... 23

A 10,000 fold increase in peak detector current would be achieved
in a TOF instrument capable of ion storage over an instrument

which could detect all of the ions continuously extracted. ........................... 30

Diagram of the storage source developed in Professor Wollnik's
laboratory. This source'is rotationally symmetric about the ion
optical axis. G4-G1 are‘grided electrodes, for acceleration, and
storage. Pu are "pusher" electrodes responsible for focusing the

electrons generated by (K) the filament into the storage area. .................... 31

Graph of peak intensity (m/z = 18) versus the voltage applied to the

center grid (G3) ............................................................................................ 33

Voltages applied to the source electrodes in both theoretical and

experimental operation. ............................................................................... 34

ix

3.5
3.6
3.7

3.8

3.9
3.10

3.11

3.12
3.13

3.14

3.15

3.16

3.17

3.18
3.1 9
4.1
4.2

Theoretical graph of ion peak intensity versus the ionization time. ............. 37
Graph of peak intensity (m/z=44) versus ionization time. ........................... 38

Predicted ion population growth curves for different analyte partial

pressures. .................................................................................................... 39
Effect of analyte pressure on the ion population growth curves of

CHBr3. ......................................................................................................... 40
Normalized ion loss curves for ion 5 of different m/z. .................................. 42

Ion growth curves for a background analyte at different helium

pressures. .................................................................................................... 43
Visual model of the process of ion storage and the effect of the

tremendous number of helium ions on the loss of analyte ions. .................. 44
Ion storage curve for m/z = 4 (He+). ............................................................ 46

Effect of source pressure (due to He flow) on analyte peak
(m/z= 58) intensity ....................................................................................... 47

Simulated effect of a 2MHz, 30Vp-p ac voltage on the trajectory of
ions with m/z=4,28,52, and 76. Ions, starting with 5 eV of kinetic
energy and angle = 90 degrees, are oscillated from side to side in
the source, causing ions of lower mass to be accelerated out ..................... 48

Ion population growth curves with and without the application of an
ac voltage ..................................................................................................... 49

Background spectra with and without the application of an ac
voltage. Note the increase in intensity for the higher massed ions

when the ac voltage is applied. .................................................................... 50
Two different source designs depicting the use of a circular cage,

and a square cage. ...................................................................................... 52
Schematic diagram of the TOF/T OF source. ............................................... 55
lon population growth curve for the TOF/T OF source. ................................ 56
Schematic diagram of the TOF/T OF instrument and control system. .......... 60
Blueprint of the TOF/TOF vacuum chamber. ............................................... 63

4.3 SIMION model of an Einzel lens system. Collimated ions of equal
energy are focused to a point using an Einzel lens created by the
voltages applied to three "washer" electrodes. The outer two
electrodes are held at the same potential while the center electrode
is held at a either a negative potential (in this case) or a positive
potential with respect to the other two. ........................................................ 67

4.4 SIMION model of Wollnik's grid-free reflectron. The reflectron is
made up of large "washer" electrodes which have voltages applied
to them. Ions which enter the reflectron will travel up the potential
field, stop, and be accelerated out. As depicted in this model ions
of greater energy travel further into the reflectron than ions of lower
energy. ......................................................................................................... 69

4.5 SIMION model of the modified grid-free reflectron used in the first
TOF analyzer. This reflectron does not have the "lips" in the
Wollnik design, yet the contours are not significantly different. ................... 70

4.6 The top (A) and side (B) views of the ion trajectory through the
TOF/T OF instrument. ................................................................................... 72

4.7 SIMION model of the ion gate. This gate is made of fine wires
(0.003" diameter). Alternate wires are held at +300V and -300V.
The potential field generated by these alternating electrodes
extends only a few millimeters. This short field allows for the
selective gating of closely bunched ion packets and thus high
resolution gating. ......................................................................................... 75

4.8 Mass spectrum of bromobenzene collected with a detector at the
interaction region, showing resolution of 1500 (FWHM) for the
molecular ion (m/z = 156 and 158). ............................................................. 78

4.9 Mass spectrum for the monoisotopic molecular ion (m/z = 158) of
bromobenzene collected with a detector placed at the end of the
second TOF analyzer, showing resolution greater than 1700
(FWHM). ...................................................................................................... 79

4.10 Plot of the intensity of m/z = 18 as a function of the iris diaphragm
aperture area. .............................................................................................. 80

4.11 Plot of ion intensity versus the potential difference applied between
alternate wires of the deflection grid. ........................................................... 82

5.1 Spectra of the molecular ion region of bromobenzene.
a. Spectrum with the laser off. b-d. Spectra with the laser delay
set at 86.40 us, 86.70 us and 86.97 us, respectively ................................... 88

xi

Diagram of the TOF/TOF instrument showing the location of the
temporary detector approximately 12 inches after the interaction
region. .......................................................................................................... 91

A product spectrum for the molecular ion (m/z = 156) of
bromobenzene. ............................................................................................ 92

The spectrum resulting form the photoionization of residual gas
molecules in the interaction region. ............................................................. 96

Simulated resolution of product ions for precursor ions with m/z =
500, 1000, and 2000. All fragment ions of precursor ions with m/z
s 1000 will have at least unit mass resolution .............................................. 104

a. A spectrum of the molecular ion region of toluene obtained at
the final detector. b. The spectrum of the molecular ion region of
toluene when the laser is fired to interact with the molecular ion
(m/z = 92). The peak at 109.4 us is m/z = 91 product ion. .......................... 106

The bromobenzene product ion (m/z= 77) obtained after the
second reflectron was retuned for lower kinetic energy ions. ...................... 107

SIMION model of a hyperbolic reflectron. The two electrodes used
to form this reflectron include a grid electrode and a hyperbolic
electrode. In one model both of these electrodes extend
perpendicular to this page and are of infinite length, in the other
model these electrodes are symmetric about the x axis. Ions enter
these mirrors at the origin of the grid electrode and travel towards
the hyperbolic electrode with no y velocity component. ............................... 109

Graph of the location an ion along the x axis versus time for ions
with a large range of energies. The origin is located at x = 75.
Ions which have a wide range of different energies, and which are
temporally focused at the origin when they enter the reflectron, exit
the reflectron at the origin and are again in temporal focus. This
simulation was performed on an extended monopole... .............................. 110

Graph of the location an ion along the x axis versus time for ions
with a large range of energies. The origin is located at x = 75.
This simulation was performed on a conical reflectron. lons which
have a wide range of different energies, and which are temporally
focused at the origin when they enter the reflectron, are not
temporally focused with this reflectron. ........................................................ 111

xii

LIST OF TABLES

3.1 lon storage time constants (1/kL) and ion formation rate constants
(kg) for six different source designs ............................................................. 53

5.1 PID results for several precursor ion m/z values. ........................................ 93

xiii

Chapter 1: Introduction

you can imagine my surprise when I arrived
for work on my first da , to find on my desk a paper
that read "Remit: Buil a mass spectrometer.“

J. H. Beynon

1.1 Tandem Mass Spectrometry (MS/MS)

"I feel sure that there are many problems in Chemistry which could
be solved with far greater ease by this than any other method. The
method is surprisingly sensitive - more so even than that of
spectrum analysis - requires an infinitesimal amount of material
and does not require this to be specially purified:..."[1]

As the advantages of mass spectrometers impressed J. J. Thomson in
1913, the advantages of tandem mass spectrometers impress researchers
today.

Since the introduction of tandem mass spectrometry in the 1970's, this .'
technique has proven to be a valuable tool for an ever increasing number of
analytical applications as well as for studies of fundamental phenomena. The
analytical applications range from the analysis of relatively small organics such
as environmental pollutants, industrial products, or food components to larger

organic and bioorganic structures [2].

This multitude of applications stems from the variety of experiments
possible with the many different MS/MS instruments available. Despite the
differences, all MS/MS experiments involve the same fundamental elements:
Ionization of the sample is the first step and is most often accomplished using
electron impact ionization. Bombarded by 70 eV electrons a neutral molecule in
the vapor state obtains sufficient energy to eject one of its electrons thus
becoming an ion. This ion typically has additional excess energy which results
in fragmentation and the production of product ions of different m/z. A precursor
m/z value is selected by the first mass analyzer in the MS/MS instrument. The
selected precursor ions undergo a mass changing reaction in an interaction *
region, and lastly the ions formed in the interaction region are mass analyzed in
the second stage of mass analysis. Thesesteps are illustrated in Figure 1.1.

In contrast to a typical mass spectrum in which precursor ion/product ion
relationships are unclear, MS/MS clearly establishes these relationships and is
thus a valuable tool for structural elucidation of complex unknowns. A complete
MS/MS data set contains a product spectrum for each ion present in the normal
mass spectrum. Such data sets comprise all precursor-product reactions and
are extremely useful for elucidating their structure. Although not currently
available, libraries of MS/MS spectra would dramatically mprove the accuracy of
spectral searching techniques.

The widespread use of tandem mass spectrometry is due largely to its
high sensitivity and selectivity [3]. Compared to a single mass analyzer, two
analyzers in tandem eliminate a large percentage of chemical noise and
dramatically lower the detection limits. In one type of experiment, the first mass
analyzer performs a separation step while the second mass analyzer provides
the data for compound identification. MS/MS also allows for trace analysis of

target compounds in complex mixtures with minimal sample cleanup [4].

141
Mane Analysis

191'

93 112 133"
l l 'I
I I II I .

80 100 120 140 160 180 '200

 

 

 

\ /
Interaction Region " Dissociation -
/ \
I | |
‘19-;
141 ‘
Mace Analysis

I
I I I I I I I I II I I I
80 100 120 140 I 80 180 200

 

 

 

Figure 1.1 Schematic representation of the collection of an MS/MS product
spectrum.

1.2 Mass Spectrometers

Mass analyzers measure the mass to charge ratio of ions. The flexibility
and ease with which ions can be maneuvered with both electric and magnetic
fields have led to many different mass analyzers. The magnetic sector, electric
sector, quadrupole mass filter, Fourier transform-ion cyclotron resonance, ion
trap and time-of-flight mass analyzers all are based on different principles of
mass to charge separation or detection.

A magnetic sector uses a magnetic field to disperse ions according to
their momentum [5]. In the presence of a magnetic field an ion travels in a
circular path. The radius of this circular path is dependent on the momentum to
. charge ratio and the magnetic field strength. With a thin slit in front of a
detector, a mass spectrum is obtained by scanning either the acceleration
voltage of the source or the magnetic field.

In an electric sector [6] an ion travels in a circular path also. The radius
of travel is dependent on the kinetic energy and the electric field strength. Used
in conjunction with a magnetic sector and slits, the electric sector filters out all
ions which fall outside a narrow spread of energies. This combination of devices
provides very high resolution mass analysis. If,,however, an ion fragments in the
field-free region preceding the electric sector, it can act as a product ion mass
analyzer by dispersing them according to their kinetic energy [7].

The quadrupole mass analyzer [8] consists of four symmetrically arranged
parallel rods. Diagonally opposite rods are electrically connected and a
combination of dc and rf potentials are applied to the rods. With the correct
application of voltages, only ions with a narrow range of mass to charge ratio
have stable trajectories through the quadrupole. Scanning these voltages allows

sequential mass to charge values to pass, thus creating a mass spectrum.

Fourier transform-ion cyclotron resonance (FT-ICR) [9] and ion trap mass
spectrometers (ITMS) differ from the previous instruments in that they are not
beam instruments. Beam instruments form ions in one region, accelerate them
into the scanning mass analyzer, and detect the ions allowed through. Each of
these steps is performed in the same region of the ICR cell. This cell is located
in a magnetic field. The ions formed by a pulse of electrons travel in a circular
orbit with a frequency dependent upon the ion's mass-to-charge ratio and the
magnetic field strength. A very fast rf sweep is made across a range of
frequencies that corresponds to the desired mass range, causing the ions to
travel together coherently. This coherent ion motion induces an image current
in a set of receiver plates. The composite signal detected is due to the motion of
all of the excited ions. The Fourier transform of this time domain spectrum yields
a frequency domain spectrum which is can be relabeled as a mass spectrum
after calibration.

The ITMS [10] is actually a three dimensional quadrupole, and the
equations explaining the ion stability in the trap are very similar to the equations
for stable ion trajectories in the quadrupole mass analyzer. In the most popular
mode cf operation, termed mass-selective instability, no dc potential is applied to
the endcap electrode, producing an rf only quadrupole field. After an ionization
period, all ions with m/z above a threshold value have stable trajectories within
the trap. As the rf amplitude is increased, ions of increasing m/z value become
sequentially unstable. These ions pass through holes in the endcaps and are
detected.

Time-of-flight mass analysis [11], as depicted in Figure 1.2, is
conceptually and mechanically very simple. Ions produced in the source are
extracted with a high voltage pulse applied to the source backplate. All of the

ions receive approximately the same kinetic energy. Ions of different mass to

.38an Ba new $502? Emachmn $9: :9: 2 one 9:: 5 293mm
39.03 :2 mmmEom_ 5:862ch So $65.88 2m 858 :2
9: E nmELoL mac. .cmamEozooqm mmmE EQEoéE: m 3 E9930 «é 2:9“.

:oSoobxm .8
:ofioehom G3 @315

 

A2; 25..

Q

 

 

 

 

 

 

 

 

 

 

macaw

+ M... .f .
u a .. .. w 2 is...
. . coaooaoa :3me «3.5 cochlgor— 8.50m

charge ratio have different velocities and isomass packets thus become
separated in time as they travel down a field-free flight tube. These ion packets
arrive at the detector at different times. A mass spectrum is created with each

source extraction.

1.3 Existing Instrumentation for MS/MS

To obtain MS/MS spectra, two mass analyzers are typically used. As
combinatorial mathematics would imply, this has led to a plethora of different
instrument designs.

Sector based MS/MS instruments include single magnetic sector
instruments which were used in the early stages of MS/MS work [12], two-sector
instruments of the BE geometry [7] and the EB or "reverse" geometry [13] (where
B and E refer to magnetic and electric sectors respectively) as well as three and
four [14] sectored instruments. These instruments typically employ high energy
collisional activation to induce fragmentation.

The first triple quadrupole instrument for MS/MS studies was introduced
in 1978 [15]. This instrument has become very popular due to straightforward
nature of the scanning procedures for product, precursor and neutral loss
spectra, and the simplicity of operation especially under computer control. This
instrument utilizes low energy collisions for precursor ion fragmentation.

Hybrid instruments, consisting of both sector and quadrupole instruments
have also been developed [16]. These instruments take advantage of the
different qualities of the analyzers and offer tremendous experimental versatility.
The BEQQ instrument [17], for example, which incorporates a deceleration step
before the first quadrupole, can yield both high and low energy fragmentation, as

well as high resolution selection of precursor ions.

The two mass analysis stages of MS/MS can both be performed in the
trapping FT-ICR [18] and ITMS [19] instruments. The ions produced in the
ionization period are stored, as discussed earlier, in either a combination of
magnetic and electric fields in a FT -ICR or with electric fields in an ITMS. Next,
all ions except the desired precursor ions are scanned or ejected from the cell.
The precursor ions of interest, left in the cell, are now excited enough to cause
collisionally induced dissociation (CID) with background gas but not enough to
be ejected from the cell. Product ions produced are then detected with the
conventional methods. Multiple stages of MS (i.e., MS") are possible with both

of these instruments.

1.4 MS/MS on the Chromatographic Time Scale

Chromatographic introduction of sample into an MS/MS instrument is
routine. Experiments, however, typically involve monitoring a single reaction
(SRM) or a few reactions [20] while the continuous chromatographic eluent flows
into the source. Increasing the number of reactions monitored requires
sacrificing ion statistics for each reaction due to the scanning nature of the
second analyzer employed in most commercially available instrument.

To obtain a complete MS/MS data set from any commercially available
instrument takes at least one minute. The speed of this acquisition is limited by
the time it takes the second mass analyzer to scan the entire mass range while
still producing product spectra with useful ion statistics. Modern
chromatographic methods including gas chromatography (GC), high
performance liquid chromatography (HPLC), and capillary zone electrophoresis
(CZE), typically yield separated mixture components with peaks which are on the

order of 1-5 seconds wide. Clearly, available MS/MS instrumentation is not able

to obtain complete MS/MS data set from such components because the
components are not present in the source long enough. The design and
construction of an instrument capable of obtaining complete MS/MS data set
from compounds introduced chromatographically is the subject of this

dissertation.

1.5 A Tandem Time-of-Flight Mass Spectrometer

In two sector MS/MS instruments and in triple quadrupole mass
spectrometers, the acquisition of the product spectrum is the rate limiting step in
the collection of a complete MS/MS data set. The choice of product ion mass
analyzer is thus critical. The 100+ fold increase in acquisition speed needed
from an MS/MS instrument compatible with chromatographic sample introduction
requires that the second mass analyzer utilize array detection. Array detection
has been demonstrated using a magnetic sector analyzer with multielement
electrooptic ion array detection [21], a Fourier transform MS, and a time-of—flight
mass spectrometer using array detection [22]. Using array detection the
magnetic sector possesses only limited mass resolution when a large range of
masses is studied. The FT MS requires successive, alternating introduction of
the sample and the collision gas for each product spectrum and is therefore not
a high speed technique. The time-of-flight mass analyzer, however, capable of
producing thousands of transients per second, does not have these limitations
and is thus a suitable candidate for array detection in an MS/MS instrument.
Time-of-flight mass spectrometry will be examined more carefully in Chapter 2.

Time-of—flight mass analysis requires that ions start as discrete packets.
Techniques to produce such ion packets involve either deflecting a slice from a

continuous beam or pulsing the source. For a time-of—flight analyzer to be used

for product ion analysis, the first mass analyzer for the new MS/MS instrument
could therefore be a beam instrument (e.g., magnetic sector, quadrupole etc.)
utilizing beam deflection to produce a discrete ion packet of precursor ions, or a
time-of-flight instrument which produces temporally separated isomass ion
packets. An instrument based upon a magnetic sector and time-of-flight
instrument in tandem, and utilizing beam deflection was built in Dr. Enke's
laboratory. Using a technique termed time-resolved-ion-momentum mass
spectrometry (T RIMS) [23] a complete MS/MS spectrum could be obtained in
10 5. Although this is among the fastest speeds achieved to date, it is barely
fast enough for the wide peaks obtained in packed column gas chromatography
[24]. This instrument is limited by the inherently low scan rates of magnetic
sectors since acquisition of the MS/MS spectra requires the scanning of the
magnetic field. In addition, the poor duty cycle for the deflection technique
wastes a large percentage of the ions.

We have chosen to use a time-of-flight mass analyzer for the first stage of
mass analysis. By its nature a TOF analyzer yields discrete packets of
separated isomass ions. In addition, the development of an electron impact
source for TOF which stores the continuously generated ions between pulsed
extraction, eliminates the waste of precursor ions. The characterization and
improvement of this storage source for time-of-flight is detailed in Chapter 3.

The choice of fragmentation technique was made on the basis of
efficiency, ease, and minimum disruption of the ion kinetic energy distribution.
Cornish and Cotter [25] have incorporated CID in their tandem time-of-flight
mass spectrometer using a pulsed valve. This technique is a novel way to
introduce collision gas into the ion path with minimal increase in the overall
chamber gas pressure. Cooks [26] has used surface induced dissociation in his

tandem time-of-flight instrument. Both of these methods yield useful

10

 

Irr-

fragmentation but due to the collisional processes the resulting product ions
possess a large range of kinetic energies which is detrimental to good resolution
in TOF mass analysis. In addition, both rely upon a gate to select the desired
precursor. This gating process results in limited mass resolution for precursor
selection.

We have selected photo-induced dissociation (PID) as a means of
fragmentation. PID has historically been plagued by low efficiency. This low
efficiency has resulted from a very low interaction cross section and low ion
densities. In the TOF/T OF instrument, however, the storage source together
with time-of—flight produces a very high density of precursor ions at the
interaction region. Combined with the high photon density of a high power laser,
efficiency over 100% is theoretically possible. A more detailed description of the
photo-induced dissociation process will be covered in Chapter 5.

Photo-induced dissociation with a pulsed laser also provides an excellent
means for precursor m/z selection. The laser selected for this instrument
provides pulses which are approximately 15-20 ns in duration. Focused to the
width of the ion packets (approximately 1 mm), the dimensions of the laser pulse
almost perfectly match those of the precursor ion packet. This ideal overlap
results in precursor selection with resolution of approximately 1200. Laser
selection of precursors, as well as demonstration of this efficient photo-induced
[dissociation will also be covered in Chapter 5.

The tandem time-of-flight instrument, depicted in Figure 1.3, was
constructed in a stepwise manner. Chapter 4 contains a description of the
instrumentation. The vacuum chamber and vacuum system, the TOF focusing

elements necessary for both analyzers, the elements in the interaction region

11

 

 

.EwEEfiE mmmE 39:99:: Emucm. 9: so Emfimfi oszmzom n; 239”.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Loam... .o c
.2389 EEme Sofie” N c2623:
L .1 .._. ,
m . m 2. EEE eaoMcMmem
.. e
- D
= -
,e \

 

 

_. cohfiocom 530». 5.6285 850m

12

will be as well as the resolving power and beam image produced by this
instrument will be found in Chapter 4.

The second time-of-flight analyzer will be discussed in Chapter 6. A
theoretical model of the expected resolution predicts unit mass resolution for
precursors up to m/z = 1000. Preliminary experimental results of the second

mass analyzer are also presented in Chapter 6.

References

 

 

 

1 Thomson, J. J., Rays of Positive Electricity and Their Applications to

Chemical Analysis, Long mans: London, 1913.

2 K. L. Busch, G. L. Glish and S. A. McLuckey, Mass Spectrometry/Mass
Spectrometry: Techniques and Applications of Tandem Mass

Spectrometry, VCH Publishers Inc., New York, 1 988
3 J. V. Johnson, R. A Yost, Anal. Chem, 57, 758A, 1985

4 G. L. Glish, V. M. Shaddock, K. Harmon, R. G. Cooks, Anal. Chem, 52,
165, 1980

5 A. O. Nier, Rev Sci. Instrum, 11, 212, 1940

6 H. E. Duckworth, S. N. Goshal, Mass Spectrometry, 1963, McGraw-Hill,
New York, 201 .

7 M. Barber, R. M. Elliot, Presented at the ASTM E-14 Conference on Mass
Spectrometry, Montreal, June 1964.

8 P. H. Dawson, Quadrupole Mass Spectrometry and Its Applications, 1976,

Elsevier, New York.

13

 

 

20

21

M. B. Comisarow, A. G. Marshall, Chem. Phys. Lett. 25, 282, 1974.

G. C. Stafford, P. E. Kelley, J. E. P. Syka, W. E. Reynolds, J. F. J. Todd,
Int. J. Mass Spec. Ion Proc., 60, 85, 1984.

A. E. Cameron, D. F. Eggers Jr., Rev. Sci. Instrum, 19, 605, 1948.

J. H. Beynon, J. W. Amy, W. E. Baitinger, J. Chem. Soc Chem. Commun.,
1969, 723.

J. H. Beynon, R. M. Caprioli, T. Ast, Org. Mass Spectrom. 1971, 5, 229.

F. W. McLafferty, P. J. Todd, D. C. McGilvery, M. A. Baldwin, J. Am
Chem. Soc, 1980, 102, 3360.

R. A. Yost, C. G. Enke, J. Am. Chem. 800., 1978, 100, 2274.

G. L. Glish, S. A. McLuckey, T. Y. Ridley, R. G. Cooks, Int. J. Mass
Spectrom. Ion Phys, 1982, 41, 157.

A. E. Schoen, J. W. Amy , J. D. Ciupek, R. G. Cooks, P. Dobberstein, G.
Jung, Int. J. Mass Spectrom. Ion Proc. 1985, 65, 125.

R. B. Cody, B. S. Freiser, Int. J. Mass Spectrom. Ion Phys, 1982, 41,
199.

J. N. Louris, R. G. Cooks, J. E. P. Syka, P. E. Kelley, G. C. Stafford Jr., J.
F. .1. Todd, Anal. Chem. 1937, 59, 1677.

R. W. Kondrat, G. A. McCIusky, R. G. Cooks, Anal. Chem. 1978, 50,
2017.

P. A. Leclercq, H. M. J. Snijders, C. A. Cramers, K. H. Maurer, U. Rapp. J.
High Res. Chrom., 1989, 12, 652.

 

 

2!

 

 

22 E. o. Erickson, c. c. Enke, .1. F. Holland, .1. T. Watson, Anal. Chem.

1 990, 62, 1079.

23 J. T. Stults, C. G. Enke, J. F. Holland, Anal. Chem. 1983, 55, 1323.

24 B. A. Eckenrode, J. T. Watson, 0. c. Enke, Anal. Chem. 1990, 62, 1362.

25 T. J. Cornish, R. J. Cotter, Anal. Chem. 1993, 65, 1043.

25 K. Schey, R. G. Cooks, R. Grix, H. Wollnik, Int. J. Mass Spectrom. Ion

Proc., 1 987, 77, 29.

 

Chapter 2: Time-of-Flight Mass Spectrometry

"Perhaps nowhere in chemistry is the
necessity for collaboration more clear than in creative
work in instrumentation."

R. G. Cooks

2.1 Advantages of Time-of-Flight Analysis

"The mass spectrometer, in which ions are separated according to mass
by passing the beam through a magnetic field, does not lend itself readily to the
instantaneous display of the entire spectrum. Observation of the individual ion
peaks is too slow when it is desirable to monitor a system in which the
composition is changing rapidly as a function of time...(The velocitron) does,
however, lend itself to the display of a mass spectrum which can be readily
observed even though it changes rapidly as a function of time."[1]

First introduced in 1948 by Cameron and Eggers, the advantages of high
acquisition speed possessed by the "velocitron" had been understood 11 years
before a chromatograph and mass spectrometer were linked together. The
spectral generation speed of time-of-flight spectrometers is limited by the transit

time of the ion with highest m/z through the flight tube. Mass spectral generation

16

 

rates can be as high as 25 kHz [2]. As reviewed by MSU researchers, the mass
spectral generation rates of other available mass analyzers are limited to 1-
10 Hz[3]. The scan rate of the quadrupole is limited by the transit time of ions
through the quadrupole. For efficient ion transmission the rf and dc potentials
should not vary dramatically during the passage of an ion through the
quadrupole. The magnetic sector mass spectrometers spectral acquisition rate
is limited by the rate at which the magnetic field can be swept. Limited to a scan
rate of approximately 0.1 s/decade and requiring 0.2 s to reset the magnetic
field, the magnetic sector can produce only 3-4 spectra/second. Both of these
instruments are also limited in scan rate by the ion statistics necessary for useful
spectra. The commercially available Saturn ll ion trap scans at a rate of 5600
u/second. It can thus yield 9 scans/second when scanning from 30 to 650, but
usually averages 3 ”micro-scans" resulting in 3 spectra/second.

Interestingly, a time-of-flight mass analyzer was the first to be coupled
with gas chromatography [2]. Today, however, quadrupoles are the most
popular GC/MS analyzer. With the improvements in chromatographic
separations yielding temporally narrower and narrower peaks [4], researchers
are once again looking at time-of-flight M8 for its high spectral generation rates.

The evolution of gas chromatography into a high resolution separation
technique has been achieved mostly through the modification of the GC column.
The original packed columns have given way to capillary columns of narrower
and narrower width. Today chromatographic researchers are performing
separations using capillary columns which have inner diameters as narrow as
10 pm [5]. As the columns become narrower the sample capacity of the columns
drops dramatically. The amount of sample to be analyzed is increasingly less
and the peak widths are increasingly narrower, placing a demand on both the

detection limits of the mass spectrometer and its spectral generation rate. The

17

 

detection limits of time-of-flight mass analyzers are much lower than the
scanning quadrupole and magnetic sector instruments. In scanning instruments
over 99% of the ions produced in the source are not detected because ions of
only one m/z value are allowed to reach the detector at any given time (see
Figure 2.1). In magnetic sector instruments this inefficient use of ions is even
worse due to the narrow slits needed to yield good resolution. Time-of-flight due
to its non-scanning mode of analysis, and high ion transmission, yields very low
detection limits, quoted to 5 picograms [6]. Ion trap mass spectrometers, also
very sensitive, detect a large percentage of the ions produced. Ion traps,
however, are limited in ionization duty cycle because the ionization and analysis
steps occur in the same region.

High mass desorptionfionization techniques have also rekindled an
interest in time-of-flight mass analyzers. The 01-252 plasma desorption
ionization technique introduced in 1974 [7] allows desorption ionization of large
biological molecules, and was optimally suited for TOFMS. The fission-triggered
ionization source not only allowed easy TOF measurement but, the ion formation
occurred on a plane perpendicular to ion acceleration yielding good resolving
power. Other high mass ionization techniques including laser desorption [8],
secondary ion mass spectrometry (SIMS) [9] and supersonic molecular beam
laser ionization also found TOF well suited for mass analysis due to its unlimited
mass range, compatibility with event driven ion sources, and mechanical and

electrical simplicity.

The numerous and significant advantages of time-of-flight analysis have
driven researchers to overcome the limitations of the technique. They included
very poor resolving power, low sample utilization duty cycle and inadequate data

acquisition electronics.

18

 

 

.m_m>_mcm $9: .0 93m: 9525
m5 mczmzmcoemu cmEEESoam $9: 28283 m B 8935 flu 2:9...

Beach—52c
vasom
coaccaoc IIIIIII\IIIIIIIIIIIIII.IIIII-III. .
. -
m 6‘leva A

o . cu.
. l5 03395

{a seam

19

2.2 TOF Resolution - Problems

The time-of-flight instrument introduced by Cameron and Eggers in
1948 [1] had a flight path of 3 m, used beam deflection to form ion packets,
photographs of the oscilloscope screen for data collection and had a resolution
of about 2. Wolff and Stephens [10] improved the resolving power to 20 in their
instrument which had a 1 m flight path and used a pulsed electron beam to
create discrete ion packets. Katzenstein and Friedland [11] introduced a gated
ion detector to allow integration of ions from successive cycles in 1955.
Resolution was improved to 100 in their instrument which had a flight path of 1
m.

These early TOF instruments were limited in resolving power due to a
combination of effects (see Figure 2.2). Ideally, all ions in a TOF source would
start at the same point and have no initial kinetic energy. Unfortunately ions
produced from gaseous samples start from a finite range of positions in the
source, and have different initial kinetic energies. These effects result in a
distribution of arrival times for isomass ions. In addition, those ions with initial
trajectories away from the source must first decelerate, and turn around before
being accelerated from the source. The delay between these ions and those
which have an equal initial kinetic energy towards the detector is referred to as
the turn-around time. This effect results in an additional spread of isomass ion

packets.

2.3 TOF Resolution - Breakthroughs

In 1955, Wiley and McLaren [12] introduced two techniques for improving

the resolving power of time-of-flight analysis. Using a two stage ion source,

20

Source field - Free

 

 

Region
. H 3...
Ideal Behavior
0 O—D I» [
a {saw M
Spatial Efl’ect. *i i
’ i 1:» 0+
Space - Focus
.9. . a...
e—> e—> e—>
Energy Efi'ects e ’ e e
«ls—e W ewe»
Detector

Figure 2.2 Depiction of the ion spatial and energy effects on TOF resolution.
Ions in the source have a spatial and energy distribution which account
for a distribution of arrival times at the detector.

21

Wiley and McLaren showed both mathematically and experimentally that it is
possible to accelerate the ions in the source with a pulse such that ions in the
back of the source catch up to ions in the front of the source at a space-focus
located at the detector. This technique thus minimizes the spatial contribution to
the temporal spread of isomass ions.

Wiley and McLaren also devised a technique termed time-lag focusing
which helped to minimize the effects of the turn-around time. A variable delay
(time-lag) between the electron ionization pulse and the ion acceleration pulse is
used. During this time-lag, ions produced in the electron beam move to new
positions in the source. The distance traveled depends upon the initial velocity
of each ion. When the acceleration pulse is applied, those ions nearer the exit
grid, which are already moving forward, experience a smaller acceleration
potential than those moving backward and thus are farther from the exit grid. If
the time-lag is correctly adjusted the ions in the back of the source will overtake
the ions in the front at the detector and the effects of the initial kinetic energy
distribution can be minimized. Incorporated into the commercially available
Bendix TOF mass spectrometer, resolutions of almost 600 have been achieved.
Despite this improvement in resolution, the time-lag technique is mass
dependent and only a small fraction of the total spectrum is in focus.

In 1973 Mamyrin [13] published results of his time-of-flight mass
spectrometer capable of resolution to 3500 (FWHM), using a electron impact
source. Mamyrin noted that, using the Space-time focusing techniques of Wiley
and McLaren, an ion packet has a thickness which is proportional to the focal
length. If the ions were focused to a detector just outside of the source, ion
packet thicknesses of a fraction of a millimeter were possible. He invented an
ion-optical system (the mass-reflectron) to reflect the well-focused ion bunch to a

detector without allowing expansion of the bunch due to the spread in the initial

22

228.2.
2: 8 £52 .9852 a 2 9:8 .65 new .839 om: 2o: 9: :. 9.QO
mac. 9.. on... c. 8.552% m... mmcofiE c9826. 9.. 5 €QO as... Sea
25 25qu no.5. m. 5.52:: 9: 2. do. 5266 m :9: c9628. 9..
.5 3520 ..E 32:23 9.. co we: 29: mucmam .655 99: 5.; :0. :<
@9925 20 5.59.256 m 99. 983 302-8me m5 6 «do. 6282.9
9: .o 823 @538. .655 9.. 3 8.8.5852 ozmemcom ad 959“.

0:29

 

23

ion energies. The reflectron's operation is depicted in Figure 2.3. The
variations in flight time through the field-free regions due to the variations in
kinetic energy are compensated by corresponding variations in transit time in the
mass-reflectron. With a mass reflectron system the resolution increases
proportionally with the length of the drift region. Using this scheme Wollnik [14]
achieved a resolution of 20,000 with a 4 meter instrument. In addition, Wollnik
developed a grid-free mirror which yields similar results in resolution, but has
improved transmission and radial focusing. Coupled with an ion-storage source,
detailed in Chapter 3, Wollnik's time-of-flight instrument eliminated the other

problem plaguing TOF; low sample utilization duty cycle.

Oddly enough, not until Wollnik resurrected the idea of using a reflectron
to reflect a space-focus image onto a detector did this technique for improving
TOF resolution for electron impact sources receive any notable attention. This
may be due to the fact that the reflectron was introduced just as high mass
ionization techniques were being developed. These techniques produced ions
on a plane, and thus did not have the spatial distribution and turn-around
problems of the electron impact source. The reflectron was ideal for resolving
the energy distribution resultant from these surface ionization techniques.
Resolving powers of 3500 for linear instruments could be improved to about

10,000 with the addition of a reflectron [15].

2.4 Time-Array Detection

Although time-of-flight mass spectrometers could produce 25,000 spectra
per second, limitations in the acquisition electronics did not allow entire spectra
to be acquired at this rate. Researchers had to use a detection technique

termed time-slice detection. This technique only allows the integration of ion

24

current over one narrow (0.1 us) time window (slice). To collect a full mass
spectrum this window starts at 0 us (relative to the extraction pulse) and each
acquisition thereafter is successively incremented 0.1 us up to 100 us. To
collect the entire spectrum requires 1000 source extractions. This technique
limits the TOF analyzer acquisition to speeds similar to those of scanning
instruments.

Prompted by the desire to produce a mass spectral system capable of
producing mass spectra at a rate compatible with modern chromatographic
methods, researchers at MSU have developed an integrating transient recorder
(ITR) [16]. The ITR is capable of recording 80 us segment of information
(transient) with a time resolution of 5 ns. Each transient is read, summed and
stored in on of two summing registers. As the contents of one register is
processed and stored, incoming transients are summed in the other. The
unmanageable volume of data which would be produced in a GC/T OF
instmment experiment is thus dramatically reduced by integrating successive
transients.

In a TOF review article [17] Campana suggested that the popularity of
GC/quadrupole MS was due to the development and availability of computerized
data systems to collect and manipulate the GC/MS data. Given the significant
advantages of time-of—flight mass spectrometers over other mass analyzers for
GO analysis, the commercial availability of a such a system for TOF may be all

that is needed for GC/T OF to become the analytical instrument of choice.

References

1 A. E. Cameron, D. F. Eggers, Jr., Rev. Sci. lnstrum., 1948, 19,605.

25

 

 

 

R. S. Gohlke, Anal. Chem, 1959, 31, 535.

J. F. Holland, C. G. Enke, J. Allison, J. t. Stults, J. D. Pinkston, B.
Newcome, J. T. Watson, Anal. Chem, 1983, 55, 997A.

C. A. Cramers, P. A. Leclercq, CRC Critical Reviews in Anal. Chem, 1988,
20, 117.

P. A. Leclercq, C. A. Cramers, HRC & CC, 1988, 11, 845.

R. E. Tecklenburg, Jr., R. D. McLane, R. Grix, C. C. Sweeley, J. Allison, J.
T. Watson, J. F. Holland, C. G. Enke, U Gruner, H. Gotz, H. Wollnik,
Proceedings of the 38th ASMS Conference on Mass Spectrometry and
Allied Topics, Tucson, Arizona, June 3-8, 1990.

D. F. Togerson, R. P. Skowronski, R. D. Macfarlane, Biochem. Biophys.
Res. Commun., 1974, 60, 616, .

M. A. Posthumus, P. G. Kistemaker, H. L. C. Meuzelaar, M. C. Ten
Noever de Brauw, Anal. Chem, 1978, 50,985.

B. T. Chait, K. G. Standing, Int. J. Mass Spectrom. Ion Phys, 1981, 40,
185.

M. M. Wolff, W. E. Stephens, Rev. Sci. lnstrum., 1953, 24, 616.
H. S. Katzenstein, S. S. Friedland, Rev. Sci. lnstrum., 1955, 26, 324.
W. C. Wiley, l. H. McLaren, Rev. Sci. lnstrum., 1955, 26, 1150.

B. A. Mamyrin, V. l. Karataev, D. V. Shmikk, V. A. Zagulin, Sov. Phys.
JETP, 1973, 37,45.

R. Grix, R. Kutscher, G. Li, U. Gruner, H. Wollnik, Rapid Commun. Mass
Spectrom., 1988, 2, 83.

26

15 X. Tang, R. Beavis, W. Ens, F. Lafortune, B. Schueler, K. G. Standing,
Int. J. Mass Spectrom. Ion Proc. 1 988, 85, 43.

16 J. F. Holland, B. Newcombe, R. E. Tecklenburg, Jr., M. Davenport,
J. Allison, J. T. Watson, C. G. Enke, Rev. Sci. lnstrum., 1991, 62, 69.

17 J. E. Campana, Anal. lnstrum., 1987, 16, 1.

27

Chapter 3: An Electron Impact Ionization Storage
Source

we all share each others ideas and this is
the way that science develops."

Ron D. Macfarlane

3.1 Ion Storage for TOFMS

The Bendix TOF mass spectrometer, due to its commercial availability
which began in 1957, is commonly used as a benchmark for performance
comparison. The Bendix source uses pulsed ionization. Using a 0.25 us pulse
of electrons followed by an extraction pulse, the duty cycle for ion production in
this instrument, operated at 10 kHz, is only 0.25%. If a source could be made
which stored 100% of the ions produced from a continuous ionization process
the sensitivity of TOF could be increased by a factor of 400! In addition, to the
sensitivity improvement by ion storage, the ion bunching caused by
simultaneous extraction and focusing results in a great improvement in signal to
noise ratio compared to continuously detecting filtering instruments. All the ions

of each m/z value accumulated in the source over a 2 ms period (extraction

28

frequency = 500 Hz) arrive at the detector within a 20 ns interval. It 100% of the
ions formed were stored, the peak detector current would be about 100,000
times greater than the average ion current (see Figure 3.1).

In 1989 Professor Wollnik et al. [1] published the development of an
unique electron impact storage source capable of storing a large fraction
(“perhaps 20%") of the ions which are continuously produced. Interest in this
tremendous increase in sensitivity, in addition to the high resolution which
Wollnik accomplished with electron impact sources led to a collaboration
between MSU researchers and Professor Wollnik. I was graciously invited to his
Giessen laboratories to undertake fundamental studies of this source. With
guidance from George Yefchak, I performed many experiments to probe both the
ion storage process, and the resolving powerof the instrument. In this thesis, I
will discuss the ion storage experiments which began in Giessen and continued
on an instrument very similar to the one in Giessen which was built at MSU. A
thorough review of the resolution experiments and conclusions can be found in

Yefchak's thesis [2].

. 3.2 The Storage Well

The ion storage source developed in the Giessen laboratories is depicted
in Figure 3.2. This source is rotationally symmetric with respect to the ion
optical axis. Ionizing electrons are produced continuously by a ring filament (K).
Two additional ring electrodes (Pu), are used to direct, or "push", the emitted
electrons through the slit created by two grids (G3 and G2). G2 and G4, are
held at slightly positive potentials with respect to G3. This design was intended
to create a potential well about G3 in which ions are stored between pulsed

extractions, and to allow independent variation of the depth of this well. A pulse

29

.cmfimzxo >.m:o::=:8 2:. :5 .o ._m .88: 2:8
:02; EoEEfiE :m 85 mamas...“ :o. .o 2:38 59:26:. mob m :_
35.5: on 2:95 E95: 882:: 3.me :. ommeoc. v.2 80.9 < fin 2:9".

“2
258 Howl .

 

  

31A:

 

4/

moon—5350

   

mares 2:2 as, :2.

\

macaw sons: 5.3 :2.

 

OH

 

 

30

G4 G3 G2 G

I'M—i I

Pu Pu

Figure 3.2 Diagram of the storage source developed in Professor Wollnik's
laboratory. This source is rotationally symmetric about the ion optical
axrs. G4-G1 are grided electrodes, for acceleration, and storage. Pu are
"pusher" electrodes responsible for focusing the electrons generated by
(K) the filament into the storage area.

31

is applied to G3 to extract the stored ions from the source through G1, the final
grid, and into the field-free flight tube.

One of the first experiments undertaken in Giessen was aimed at
measuring the depth of the potential well in which the ions are stored. If Wollnik
et al.'s theory that the ion storage capability of the new TOF source is due to a
potential well created by the voltages applied to the three source grids: G4, G3,
and G2, varying the voltage on the center grid (G3), even slightly, should cause
a significant effect on the ion storage capabilities.

To perform this experiment all of the instrument control voltages were first
tuned for optimal sensitivity and resolution. The voltages on G4 and G2 were
- 462.0 V and 451.1V respectively after tuning. The voltage on G3 was then
varied between 451.0 - 455.2 V. The peak intensity for ions of m/z = 18 was
measured from the oscilloscope screen as the voltage on G3 was varied. Figure
3.3 is a graph of peak intensity versus G3 voltage. The optimal voltage for G3
was found to be approximately 453.0 V. Instead of forming a potential well
between G4 and G2 the voltage applied to G3 was higher than the voltage
applied to G2 thereby forming a potential hill (see Figure 3.4). Another point of
interest is that the intensity was at a maximum for a relatively wide range of
voltages (452.6 V - 453.6 V). Both of these results contradicted the original
hypothesis that the potential well was formed by the potentials applied to the
grids. In fact, modified source configurations which do not even incorporate the
back-most electrode G4 have demonstrated similar storage capability.

Another possible explanation for the significant ion storage observed is
that the high density of electrons focused into the center of the source from the
ring filament forms a space-charge potential well. A strong dependence of the
storage effect upon the tuning of the pusher electrodes and filament alignment

supports this hypothesis. Support of this theory also came from a thorough,

32

.8. 25 9:8

:5 2 3.3:: $9.? on. m:m.m> 8. u NE. 3.2.9:. xmoa .o 390 Rm 2:9“.

.3 33.2, no

mm:

.0:

 

 

_ _

 

 

ON

0:

ow

om

(71111) 41181194111

33

 

 

 

 

potential well

 

.;.f. __‘
"" e s' .

' " I I o . n
55' '.'. .' .
II. ' ’

Voltage

 

 

 

.........

Experimental Operation

 
 

Voltage

 

 

Distance

Figure 3.4 Voltages applied to the source electrodes in both theoretical and
experimental operation.

34

albeit dilatory, literature search which uncovered a paper by Studier [3], who in
1963 developed a modified source for the commercially available Bendix time-of-
flight instrument. With continuous ionization Studier achieved a factor of 300
increase in sensitivity over that of the pulsed mode of operation by using a
focused electron beam to store ions. Curiously, Studier made mention of an
alternate source for the Bendix TOF, which was made available for instruments
using a continuous source of ions and which utilized three grids. This assembly
incorporated a center grid, which is biased 0.2 V lower than the electrodes on
either side of it, to "trap" positively charged ions. Despite an ionization duty
cycle of approximately 100%, Studier found that this three grid assembly
demonstrated little increase in sensitivity over the pulsed ionization source [4].
Unfortunately, the hapless designers of this Bendix source did not incorporate

electrodes to focus the electrons into the center region.

3.3 A Mathematical Model of Ion Storage

To characterize the performance of the storage source, a mathematical
model of ion behavior was developed by Yefchak, and. myself [4]. A model
would help us to quantitatively analyze the effects of pressure, mass,
instrumental parameters and source designs on storage time in order to fully
understand and optimize the storage process.

We hypothesized that ions are formed and lost-according to the rate

equation:

—d—(-t-5- = HF - RLn. (3'1)

35

where n is the number of ions in the source, tis time, and RF and RL are the
formation and loss rates, respectively. Integrating equation 3-1, we get the

following equation.
W) = 51(1 _ e-RLt) (3-2)
FIL

If this model accurately describes the storage process in the source, one would
expect that, after the previous ion extraction, as the ionization time tincreases
the number of ions in the source and consequently the number and ion trap
mass spectrometers which can actually trap or store an ion for a long,
measurable period of time, the "storage" of ions cannot be detected in this
source, but only the steady-state level of ions attained. By trying to improving
the ”storage" effect of this source we actually are trying to minimize the rate of
ionloss.

With this mathematical model in hand, we could begin to postulate about
the influence of analyte partial pressure, m/z, and carrier gas (in the case of
GC/MS) on ion storage. Figure 3.7 demonstrates the expected dependence of
storage on analyte partial pressure. We predicted that as the analyte partial
pressure is increased the rate of ion formation RF should increase, while the
rate of ion loss RL should remain constant. Thus the growth curves are
expected to rise with similar time constants (1/RL) but to different population
values. Figure 3.8 is a graph of the growth curves for CHBr3 obtained at 10

mTorr, 20 mTorr and 40 mTorr. As expected the curves reach different plateaus

36

.oE. :o.
. ..m~.:o. m5 m:m:m> 3.39:. xmma :o. .o :98: 30:98: .. md 23
. E

25H. come—www.3—

HI

fitment] uoI

 

37

.oE: :o..m~.:o. m:9m> .3195 3.9.9:. xmmn. .o .390 ad 9:9“.

6.5 .259 come—«3:3

 

 

 

 

 

 

can m.“ c.« m.—
u u u c
r cm I
u
1
- 2. m
S
n”
t 8 K
m,
.. an m
t 2:
C
n n n n - AVNH

38

 

39:39:
3:9. 922: 599:: :9 32:0 539: 3.8.2.0: :o. :90.:9n. fin 9:9...—

oEmE :omenumasm

 

9:mm9n. .m.:mn.
93:5. 950..

 

9:mm9n. Eran.
9>_m:< 9.9:

thsuestul )[BGJ uoI

 

39

@510 3 $38 5390 8338 :2 9: cc 9885 9298 8 .85 m6 939....

33 25a. «83333

 

 

 

 

 

 

8. o: 02 8. 9 8 8 8 on o
v w 9 “ u v w w w. o
0
4
r. 8’
M
F 8N u
4 d
a
an m
U
m
has: 3 . 8.. m.
m
5.? om ILI 8m, m
59: 3. Ill /./
/

 

§

but the storage time constants did not change significantly. We also postulated
that ion loss rates would be a function of ion velocity. Thus, lighter, faster ions
are expected to be lost more rapidly than heavier, slower ions. To directly
compare ion loss rates, ionization was terminated at a variable time before
extraction. Figure 3.9 is a graph of the normalized peak intensity as a function
of loss time for m/z = 58, 43, 18 and 4. As expected, the loss rates were greater
for lower masses, and were approximately proportional to 1/ J5.

Preliminary data collected by Tecklenburg in Giessen from samples
introduced via a gas chromatographic interface indicated that the presence of
the helium carrier gas attenuated the intensity of the background ions compared
to lower pressure conditions. This observation indicated that the source
pressure also effects the steady state concentration of analyte ions in the
source. The extent of this effect is demonstrated in Figure 3.10. This is a graph
of ion intensity (rn/z = 58) versus ionization time for five different source
pressures. Clearly analyte ion storage efficiency is inversely related to source
pressure. Under GC conditions, the presence of helium is therefore a detriment

to optimal source performance.

3.4 Effects of Helium & High Frequency Field Modulation

To understand the effect of helium on ion storage, a visual model was
found to be useful (see Figure 3.11). The storage source can be viewed as
similar to a bucket with holes in it. The rate at which the bucket is filled depends
upon the ionization rate of the molecules in the source. These rates are
dependent upon many factors including the partial pressures of the different
analytes, their ionization cross sections, and the electron current. The loss rates

are dependent upon ion mass, and also upon the volume of the bucket. Given

41

.NxE E99=n E 9.2 8. 82:0 mmo. :2 33.9502 ad 939“.

33 9.59 mood

o m v . n m _
- b w ’ fl

.0-

vnQEIT
SudEIIT
nvu~>cli|l

mmufiElloll

 

l I I I U I r

N. 0. m. V. n. N. —. o

O O O O O O O
figment} aread pazgeunoN

I l
0: “2
o o

 

O
F

42

 

.m93mm9q EEE; 2.99% 6 2298 33.30% m .2 $28 £39m co. ofim 9:9“.

8:: 9.59 55333
N 0; o; v... N— p .md 0.0 v.0 Nd o

 

 

 

cop

 

0
L0

0
O
N

O
t!)
N
(Am) thsuaqul speed 1101

con

0
I!)
F)

O
O
V

 

0
ID
V

43

Analyte 2

 

 

 

Figure 3.11 Visual model of the process of ion_storage and the effect of the
tremendous number of helium ions on the loss of analyte ions.

this model, it is clear that if the bucket is filled very quickly with the numerous
helium ions produced, the rate of loss for all other analyte ions is very high.

This model would tend to imply that due to the high partial pressure of
helium present in the source under GC/MS conditions, the number of helium
ions stored in the source would rise very sharply to a maximum indicating that
the "bucket" is full. Further storage time, however should yield an increase in
the number of heavier ions stored due to their lower velocities and thus lower
loss rates. As the source fills with heavier ions the intensity of helium ions
should drop. Figure 3.12 demonstrates that this is what is observed under
GC/TOF conditions.

Another experiment supporting this model involved lowering the electron
energy below the ionization threshold for helium, thereby eliminating the
formation of helium ions. If in fact the helium ions, and not the helium neutrals,
cause the attenuation of the analyte intensity, an increase in helium pressure
should, in such an experiment, not effect the analyte intensity. Figure 3.13
shows that the m/z= 58 peak (background hydrocarbon) drops as previously
observed as the pressure of helium is increased when the electron ionization
energy is at 70 eV. At an electron ionization energy of 20 eV, however, an
increase in helium pressure has no significant effect on analyte intensity.

In order to optimize the storage of higher massed analyte ions, a scheme
was developed to enhance the preferential loss of He+ ions using an ac voltage
applied to one of the source grids (C52). As modeled by SIMION [5] (see Figure
3.14), a 2 MHz, 30 Vp-p ac voltage applied to one of the two source grids
causes the He+ ions to be quickly accelerated from the storage region. The
heavier ions, on the other hand, are only slightly disturbed. Experimentally, this

removal of helium ions provides more efficient storage and higher sensitivity for

45

 

 

(£295 8m u NE: xmma 922m :0 36: m: 2 2.2 9385 858 3 695 m; 9:9”.

2 3. NF

<l

>995 8:83 >0 on IT

>995 co:oo_o >o 05 ill

op

4|

$393 c.5305

m

‘I

.1.-

4|

‘I

d)-

4|

 

ON

on

00

0w

8—

ON—

03

00—

(Am) Kutsumul 3mad “01

46

 

f9: v u 95 .2 9,50 mamafi :2 «mm 9:9”.

A95 959 95333

C!—
qr-
d
-h

 

C)
L!)

O
O

O
LO

O
O
0]

0mm

(Am) 4161191111 1188.1

47

source grid source grid

 

r l W
/VV’N/“f\fld\/
J

/\

3
II

.p.
N
do

52 76

 

 

2 MHz, 30Vp_p applied held at ground

Figure 3.14 Simulated effect of a 2MHz, 30Vp—p ac voltage on the trajectory of
ions with m/z=4,28,52, and 76. Ions, starting with 5 eV of kinetic energy
and angle = 90 degrees, are oscillated from side to side in the source,
causing ions of lower mass to be accelerated out.

48

.mmm=o> om cm 8 cozmgaam 9: 52:? new 5;, $23 5365 c2538 :2. mfim 920E

a.”

 

r mmm=o> cm 502:; - I- -

3:: 9:5. :3?an—

WN cfl m4 c.—

- _ — _ _

 

 

mmm=o> om 5? lol

             

 

g 53

§
(Am) WI“! JD 4118an

§§§

49

Intensity (ITR counts)

 

w/o AC Voltage

 

)-
.;:___._
L44“

 

 

 

‘

 

 

30001

2503'1

 

1500 -_

Intensity (ITR counts)

1000 ~-

w/ AC Voltage

-e-‘_‘-L.L_L-A- ..

.L.‘

-a 4 art.

 

 

Ll l-l.lll LLlllllllJlllmmei’luiL. - . ._

 

 

A .L-.L_L

3) 40 50 6)
Flight Time (us)

Figure 3.16 Background spectra with and without the application of an ac
voltage. Note the increase in intensity for the higher massed ions when

the ac voltage is applied.

50

 

analyte ions (see Figures 3.15 and 3.16), and is thus very useful under GC/MS

conditions.

3.5 Effects of Source Design

Together with Ben Gardner and George Yefchak, I performed a
systematic study of source geometry to determine the parameters which are
critical for ion storage. Modifications included adding different "cages" between
G2 and G3 (see Figure 3.17). The independent voltage applied to the cage(s) is
optimized for sensitivity and is typically very similar to G2 and G3. In addition,
different filament geometries were tested. For each source tested, the ion
storage efficiency was evaluated in terms of the ion storage time constant (1/kL)
and ion formation rate constant kF. These time constants were obtained by
fitting the theoretically derived equation for ion population (Eqn. 3.2) to the
experimental ion population growth data. Results from these experiments are
tabulated in Table 3.1. These data along with daily observations indicate that
the most favorable conditions for ion storage are those which produce source
volumes that are essentially field free. Sources A-D, which have symmetric
cages were designed to produce such a field free region. The production of a
dense electron cloud also was determined to be critical to ion storage. Source
B, which has a circular filament, and sources A and D, which have two opposing
linear filaments were designed to optimize the electron density. Other significant
parameters for source performance are the alignment of the filament with
respect to G2, G3, and to the pushers, and the design of the pushers. Modeling
of the source with SIMION has indicated that with correct design and voltage

adjustment, the emitted electrons have multi-pass trajectories through the source

51

 

-. “V

circular filament

 

 

 

circular cage

 

‘32 9&—

 

 

 

 

—— -
pushers \l . .
G3 \ ‘5! :1 ('— lmear filament
7‘ m

square cage

Figure 3.17 Two different source designs depicting the use of a circular cage,
and a square cage.

52

 

53

 

 

 

 

 

 

 

.69.: N mac: u.
904 no.0 59.: P 59:95 w
omm mod 59.: w 996m 0
84a Cd 59: F 935 o
85 mmd 952”. team m
8Q: 96 69.: N 996w w 950m <
my. 5.2. BEBE—2.". Emmao 350m

 

 

 

 

 

 

 

2996 858 9:99.? 5m
.6. to: 9:928 99 5.6.59 :2 new 3%: 95928 9:: 392m :0. fin 9an

region which should yield very high electron densities in the center of the

source.

3.6 The TOF/TOF Source

The source for the tandem time-of-flight instrument was designed in light
of the information learned about ion storage. Advantageous for GC/T OF, ion
storage is critical for GC/TOF/TOF. To obtain 10 product spectra per second
from a chromatographically eluting compound, a maximum of 50 (limited by laser
frequency of 500 Hz) product transients can be summed. To obtain spectra with
a statistically sufficient number of ions it is thus critical that for each laser pulse
a significant number of ions be present in the interaction region.

The TOF/TOF source is depicted in Figure 3.18. Two linear filaments
were used instead of a circular filament. Experience proved that the shaped
circular filaments were terribly irreproducible, and often became misaligned.
More importantly, the two-filament design performed just as well. A "cage" was
incorporated into the source by using lips on G2 and G3. An ion population
growth curve for this instrument (Figure 3.19) demonstrates that this source
does, in fact, have excellent storage capability. The ion storage constant (1/kL)
was determined to be 1.4 ms, and the ion formation rate constant RF was 212
(residual water peak, m/z = 18). The approximate number of ions in an ion
packet arriving at a detector placed at the interaction region was also
experimentally determined. By measuring the peak area of several single ion
events at the detector as recorded by the transient recorder, a mean value for
peak area resulting from a single ion event was determined. This peak area was
divided into the area produced as the result of the arrival of a single ion packet

at the detector. The ion packet measured was m/z = 18 (residual water) which

54

 

.858 motmoh 9: B Emamfi o:mEm;om 3d 9.6:.

 

 

 

 

 

 

 

 

 

9U Illv
NU llv
33:82.: .8823
s - . w _ IAlA
o _ _ ufiobomfim
96 w . _ . All Manager:
muzgflm

8.59553 $332.5

55

69:8 motmoh 9: :9 :28 5:59: 5:330: :2 mtm 9:9“.

3:: :85. 55323

G m.m m m.N N m._. _. m.o o
_ _ _ _ _ _ .l. _ _ _.L.-_ : ._._..-_.-_Ll»l_.Ll_::!~ -_l._--:::l_7_._ _ _,_ _ _ _. ..+ O

I
1..om
1 m.2:
\- .omp
\ .com
-\ \ H.93

.l.i..--l. Loam

 

jrrr.

omm

(Aux) thsuenq )199d

56

represented approximately 25% of the total residual gas spectrum for a source
pressure of 10'6 Torr. It was determined that approximately 2000 ions comprise
each ion packet.

Although this instrument is designed for GC/MS/MS, the unlimited mass
range of time-of-flight, and the high mass potential of laser photo-induced
dissociation indicate that it should be ideally suited for techniques such as
CZE/MS/MS and LC/MS/MS. In the present source, however, the electrons
necessary for ion storage also limit this source to electron impact ionization. A
TOF source design which uses an ion trap for storage [6] should not have this
limitation, and will be capable of very accurate low mass destabilization to rid the

storage volume of unwanted eluent ions.

References

 

1 R. Grix, U. Gn'iner, G. Li, H. Stroh, H. Wollnik, Int. J. Mass Spectrom. Ion
Proc., 1989, 93, 323.

2 G. E. Yefchak, Doctoral Dissertation, Michigan State University, 1985,
Ch. 5, 107.

3 M. H. Studier, Rev. Sci. lnstrum., 1963, 34, 1367.

4 G. E. Yefchak, M. A. Puzycki, J. Allison, C. G. Enke, R. Grix, J. F.
Holland, G. Li, Y. Wang, H. Wollnik, Proceedings of the 38th ASMS
Conference on Mass Spectrometry and Allied Topics, Tucson, Arizona,

June 3-8, 1990, 540.

5 D. A. Dahl, J. E. Delmore, SIMION PC/PSZ Version 4.0, EGG-CS-7233
Rev. 2, April 1988.

57

6

S. M. Michael, M. Chien, D. M. Lubman, Proceedings of the 40th ASMS
Conference on Mass Spectrometry and Allied Topics, Washington, DC,

May 31-June 5, 1992, 1765.

58

 

1".“-
s“.

 

Chapter 4: TOF/TOF Instrumentation

"Perhaps creative people are those who
believe that the impossible can e done and that they
are the ones to do it."

Elizabeth Monroe Drews

4.1 Instrument Overview

The tandem time-of—flight mass spectrometer which has been constructed
was designed to be capable of obtaining complete MS/MS spectra from
compounds eluting from a capillary gas chromatographic column, while
maintaining at least unit mass resolution for both precursor ion selection and
product ion analysis. The goal was therefore to develop a system capable of
acquiring at least 10 product spectra per second. Figure 4.1 is a diagram of the
tandem time-of—flight instrument and the system control devices.

The electron impact storage source, detailed in Chapter 3, uses
continuous ionization and ion storage between pulsed extractions. A high
extraction pulse applied to the back electrode causes a space-focus plane to be

formed just beyond the source. Parent ion packets are separated due to their

59

 

 

. Emmi 65:8 ucm EmEgms mete. on. B 69920 29:28 3 9:9“.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

v 83950 owe v 3.9.280 A
>28
2 x.
.358: 3.2280
E239.» .58.. .mEme 5:25“.
.2088 coaom 5:222... . N 5.80:0:
_ / H
/ . I . can?“
3? s 11H. ill . 308-com m
. li.lw..§rll:l\l&ll ...... .. lillrl ‘\ /

 

 

 

 

 

 

 

« .
w 5.80:2. mmfiE mczmgm mam; .mNEm 3.58

mass-dependent velocities. The large distribution of energies the ions acquire
in the source are focused by a reflectron. This reflectron is tuned such that the
spatial and temporal image at the space-focus plane is focused at the interaction
region. A gate located prior to the interaction region can be used to deflect
away ions which have m/z values lower than the precursor so that they do not
interfere with the product spectra. The arrival time of a 17 ns excimer laser
pulse focused at the interaction region selects the desired precursor ion packet.
After fragmentation, the product ions formed have the same velocity as their
precursor ions, but only a fraction of the precursor ion kinetic energy.
Acceleration of the ions just following the interaction region, gives these ions
mass-dependent velocities. Product ion packets, as well as the unfragmented
precursor ion packet are separated due to their mass-dependent velocities. A
second reflectron is used to focus the energy distribution of the isomass ions at
the detector. Summing successive time-array detected transients using a
LeCroy Transient recorder, and in the future an MSU Integrating Transient
Recorder yields the product spectra.

Many functions in this instrument require accurate and precise control.
These functions include the timing of the source extraction pulse (:2 ns), the
firing of the laser (:3 ns), and the triggering of the transient recorder (11 ns). To
meet these needs a LeCroy 4222 delay generator (:2ns) and a Questek 2580 v8
excimer laser (jitter and drift i2ns) are utilized and controlled remotely via GPIB
interfaces using National Instruments LabWindows software on a 486 computer.

This system will generate 500 product transients per second (limited by
the laser). If approximately 50 transients are summed for each product
spectrum, 10 product spectra will be produced per second. With this rate of
acquisition a substantial amount of MS/MS information can be obtained for a

chromatographically eluted component.

61

4.2 Description

The design and implementation of the complete TOF/TOF system was an
effort shared between David McLane and myself, and more recently also shared
with both Paul Vlasak and Doug Beussman. Working in parallel, David focused
his efforts on all aspects of the laser, gas chromatograph, and system control.
Further details on the GC sample introduction, criteria for selection and
operation of the laser, transient recorder, delay generator, and documentation of
the control software can be found in his thesis [1].

My focus on the other hand was to design and construct the vacuum

chamber, source, mass analyzers, and the necessary ion optical elements for

the TOF/TOF instrument. These components, their function, and their

performance are detailed in this chapter.

4.2.1 Vacuum Chamber

The vacuum chamber for the TOF/TOF instrument was designed to be as
versatile and accessible as possible. Figure 4.2 is a blueprint showing the major
features. This "coffin" design was patterned after the Finnigan TSQ 700's
vacuum chamber in that it is a welded stainless steel box with removable lids
providing access to the entire ion optical path. The internal dimensions of this
chamber are 60" x 9" x 11" and the walls are 3/4" thick. The entire chamber was
electropolished‘after welding to minimize the surface area upon which water and
other molecules can deposit. Such treatment reduces the outgassing rate of
unbaked metals by a factor of approximately 20 [2]. All electrical and
mechanical feedthroughs are made through flanges allowing for non disruptive

modifications. In addition, many tapped holes were included in the interior of the

62

..mnEmco E::om> doing. on. 6 .5535 we 9:9”.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

63

chamber at various defined locations allowing for accurate positioning of ion
optical elements along the flight path.

The source is housed in a separate chamber which is interfaced to the
main chamber. This independent, multiport chamber was designed to allow
access from multiple directions. This has allowed for different sample
introduction systems to be added without major modifications or removal of the

GC interface.

4.2.2 Vacuum System

The vacuum system for the instrument consists of two Balzers
turbomechanical pumps (TPH270 and TPH510) each backed by a Leybold-
Heraeus (T rivac D 16 A) rotary vane rough pump, two standard Bayard-Alpert
ionization gauges to measure the source and flight chamber pressures, two
thermocouple gauges to measure the foreline pressures, and the necessary
controllers, tubing, connectors, and valves. The larger of the two turbo pumps
(T PH510) is used on the flight chamber, while the TPH270 is used to pump the
source. A 0.5" diameter aperture between these chambers allows for differential
pumping so that under GC/MS/MS conditions the majority of the flight path will
be under high vacuum conditions.

With this system, the measured source and flight path chamber pressures
are typically 5 x 10'7 Torr and 1 x 10'6 Torr, respectively after about 20 minutes
of pumping without the GC connected. Still lower pressures are expected if the

chambers were to be baked overnight.

4.2.3 Sample Introduction

Currently the tandem time-of-flight instrument is capable of sample
introduction from three sources. A direct leak system, which was designed by

E. Hemenway, allows for continuous introduction of a pure compound for

extended periods. A Finnigan TSQ 700 heated transfer line has been modified
by D. McLane and D. Beussman to interface the GC oven to the ion source. In
addition, a Hewlett-Packard direct insertion probe has been modified for TOF
use by B. Gardner and has been interfaced to this instrument for nonvolatile

compound introduction.

4.2.4 First TOF Analyzer

42.4.1 Source Extraction

A nearly linear acceleration field is established during the source
extraction pulse. The extraction pulse applied to the back electrode (G3) is
200 V. An additional 500 V potential drop is established between the second
grid (G2) and the field-free region (G0)- The average energy of the ions,
assuming that the storage region is located midway between G3 and 62 is
600 V. Due to the spatial distribution of ions in the source, a kinetic energy
distribution is created as ions at the back of the source receive more energy
than ions of the same mass near the front of the source. A few inches down the
flight tube, a space-focus plane is created where ions originating near the back
of the source overtake ions created near the front. The finite widths of isomass
ion packets at the space-focus plane (~20-40 ns) are due to the initial kinetic
energy spread of the ions in the source, and imperfections in the space-focus

conditions.
4.2.4.2 High Voltage Shield
The MTOF instruments built in Geissen and at the MSU Mass
Spectrometry Facility, both float the ion source at high voltage (~1000 V), while

the flight tube is grounded. In the TOF/TOF instrument, however, the entire TOF
flight path is floated. The first TOF region is floated at -500 V and the second

65

 

TOF region is floated at -1500 V. This design allows the source electrodes to be
at or near ground, and has alleviated the problem of arcing under GC/MS
conditions due to the fairly high pressures of helium. The high voltage shields,
which establish the field-free flight region were constructed of stainless steel
mesh. These are segmented and have easily detachable lids allowing for easy

modification and access to the optical elements within them.

4.2.4.3 Einzel Lens and Steering Plates

An Einzel lens and steering plate assembly has been incorporated into
the instrument a few inches from the source. An Einzel lens is analogous to a
glass lens were the light has the same speed on either side of the lens. This
lens has three elements. The outer two elements are held at the same potential
and the middle element at another potential. Originally concerned about the
effects of spherical aberrations on the temporal resolution, this effect was
modeled using SIMION. These studies indicated that for the lens system
modeled (Figure 4.3), temporal spreads of up to 9 ns were observed at the focal
point (m/z = 500, KE =500 eV). Another inescapable feature of ion optical
lenses is chromatic aberration. In ion optics this means that ions with different
energies will be focused at different points. Despite these features, Einzel lens
systems have proven to dramatically improve sensitivity. Resolution, which is
routinely 1500 (FWHM), is clearly not overly impaired.

Two sets of orthogonally oriented steering plates are used to steer the ion
beam. These steering plates consist of two parallel plates. One plate is set at
the shield voltage (500 V) and the other has an adjustable voltage. If each
optical element were perfectly aligned, these steering plates would not be
necessary. This would be ideal, because the magnitude of deflection in such an

electrostatic field is energy dependent. Careful alignment, and motion

 

 

.92 850 m5 2 6932

55, 3:58“. czzmoa m .o 6me £5 Ev .3:an czfiomc m 853
m um 2m; .2 32.86 5:8 9: 223 .35qu 9:8 9: “m Em: 9m
monozoofi 92 .230 9:. .8828? 12mm? 8.5 2 3:an 33:?
m5 3 623.6 mam. 3.25 :m use: Eon m 2 8302 9m 35cm
.mzcm B 22 umfiE___oo ESQ? 22 ESE cm ..o .39: 20:25 mé 2:9“.

>2: 3 0255905
$86 1 >83 932:8 3.8.858

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

cop-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

> OS. > 09:. > OS.

67

feedthroughs on the reflectrons (Section 4.2.8) has minimized the need for high

voltages on these elements.

4.2.4.4 First Reflectron

The ion reflectron used in the first stage of mass analysis, is grid-free,
and is similar to the reflectron designed by Wollnik et al [3]. (See Figure 4.4)
The more energetic and thus faster ions travel further into the reflectron. The
extra time spent in the reflectron compensates for their shorter time these higher
velocity ions spend in the other regions, and thus the ions of different energy are
focused at the interaction region. As modeled by SIMION, (Figure 4.5) the grid-
free reflectron constructed for the TOFfT OF instrument does not incorporate
"lips" on the first two washers as in Wollnik's design. Considerably more difficult
to construct, these were said to establish a better lensing field, considered
helpful for radial focusing. Their absence, however, does not significantly
change the field produced, or the observed performance.

The reflectron is constructed from a series of large stainless steel
washers separated by ceramic insulators. Four of these washers have voltages
directly applied to them, while the others, linked by a series of 5 MO resistors,
have voltages which are dependent upon the others. The angle of the reflectron
'with respect to the ion beam is controlled via two linear motion feedthroughs
connected at three locations using universal joints. Optimization of the

reflectron's angle minimizes the necessity for steering plates.

42.4.5 First Detector

Two detectors are used in the TOF/TOF instrument. One of these is
permanently mounted on a flange located at the end of the second TOF
analyzer. The other, mounted on a stand, is capable of being located

throughout the instrument and is helpful for troubleshooting. This detector,

68

 

.385 832 B was :9: c2823. 9: 25 855. .622. >995
.236 B was .39: £5 E 3838 m< .26 8.9288 on new .36
.22. .3298 9: a: .99; as c9825. 9: 82m :02; go. .69:
2 6239.... 393.? 9m... 22;; 33:86 Locum}. 3:2 6 a: mnwE
m_ 5:023. och .cozomzm. mgficm m.x_c=o>> B .89: zo__2_m .3 2:9...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

’

>2: 3 62:25.8.
$8: I. 28.3 232:8 326898..

69

.EmBEn 22.853? 6: 9m 832:8 05 .9» 692%
xE=o>> m5 5 ..ma__.. 9: 9m; 8: meet 5.823. 25. ..m~>_mcm “.9.
“we 9: 5 new: 5:023. $5-25 umEuoE 9: B .39: 20:25 mé 65m."—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

98: +7 28.3

 

p.505

 

 

’

>2: .3 62:25.8.
2358 3.5328

70

eventually, will be permanently mounted below the interaction region. Deflection
of the ion beam in the direction of this detector will allow for very simple
measurement of arrival times and focusing of precursor ion packets at the
interaction region.

The detectors constructed incorporate two Galileo non-optical grade
microchannel plate's (MCP's). These MCP's are thin glass wafers with a
honeycomb structure of millions of tiny channels. With a bias voltage applied
across these wafers, each channel functions as a single electron multiplier. Two

MCP's in series yield a gain of up to 107.

4.2.5 Interaction Region

An isomass ion packet, focused temporally and radially, is selected for
photo-induced dissociation by timing a 17 ns laser pulse to intersect it as it
enters the interaction region. The interaction region is thus defined by the
cross-section of the ion packet and laser pulse.

The top and side views of the ion beam trajectory through the entire
instrument is diagrammed in Figures 4.6 A and B respectively. The laser beam
is aligned to enter and exit the vacuum chamber through windows in the sides of
the instrument. To eliminate the possibility of interaction with the ion beam
except at the interaction region a slight angle downward was incorporated into
the ion trajectory.

Ideally intended for photo-induced dissociation, the TOF/T OF instrument
was also designed to allow for surface induced dissociation (SID) studies, and
possibly collisionally induced dissociation (CID) studies. This region thus needs
to be very flexible, and many different elements must be accurately positioned in
the ion flight path. Together with Paul Vlasak, the necessary elements of this

region were designed and constructed.

71

.Emeams ”BEE 2: 3:25 539. c2 9: .o 32> av 6% 2m E as a: 3. 2:9“.

 

llllllllllllllllllllllll'IIIIIIIIIII' ‘-"‘l

'll'I' II'IIIIIIIIIIIIIIlllllllllll'l'

 

 

comomm 5308(6ch \fovumfidgh com

 

72

m

 

coamm cofiuomecH / llllllllllllllllllll

Y“nllll'lI'llllll'l'lll|||'lll"lll|ll'll||l"l-lll'llll'l'I'l'll' 'l'll'llll

/

>LowumfioLH com

 

 

 

4.2.5.1 Optical Rail

A vacuum compatible ion optical rail from Newport Corp. is used in the
interaction region to provide a base for aligning the multiple elements in this
area. The gate, detector, acceleration grids, apertures, and other elements are
all connected to Newport carriers which can be positioned along this rail or

easily removed.

4.2.5.2 Gate

When using photo-induced dissociation the implementation of the ion
gate will serve to deflect ions of m/z lower than that of the selected precursor.
These ions, if not deflected, continue along the flight path and create a
background spectrum which is convoluted with the desired product spectrum.
Only one precursor ion packet is affected by the laser pulse and thus a "light on"
minus "light off" background subtraction algorithm can be used to produce the
product spectrum.

If SID or CID is utilized, however, all precursor ions passing through the
interaction region may interact. The result would be a convolution of all product
spectra of all precursor ions for the compound(s) ionized. Such a convoluted
spectrum would be nearly impossible to decipher. For these experiments, a gate
which allows exactly one isomass precursor ion packet to the interaction region
and rejects all of the others is necessary.

The gate designed, was patterned after a gate which was only briefly
described in a paper by Schlag et al. [4]. This gate uses a plane of thin (.003"
thick), parallel wires placed orthogonal to the ion beam about 6 inches prior to
the interaction region. With 500 V and -500 V potentials (relative to the shield
voltage) applied to alternate wires, ions are deflected to the right or left,

depending upon the polarity of the field through which the ions pass. When the

73

voltage applied to the wires is set at the shield voltage the gate is ”open" and
ions pass through unaffected. Modeled by SIMION (Figure 4.7), the advantage
of this gate design is that the deflection field extends only a short distance down
the flight path compared'to a parallel plate deflection system, which is typically
employed for deflecting ion beams. With this new gate, modeling suggests that
unit resolution for precursor ion selection to m/z = 1000 can be achieved if
pulsers having very short rise times (~10 ns) are employed (assuming mass

resolution of 1500 FW HM).
4.2.5.3 Light Tube

An obsthction free path through the vacuum chamber for the high power
excimer laser beam is necessary to eliminate the ionization and volatilization of
molecules on any surface. A preliminary experiment in which the laser beam
was simply passed through the stainless steel mesh shields caused the ion
beam to be reduced to 10% of its original intensity. Although not studied
thoroughly we believe that this attenuation in ion signal was caused by
ion/molecule, ion/electron, and/or ion/ion collisions with species photo-desorbed
off of the mesh shield. A stainless steel tube was designed which extends
through both the first and second TOF regions, and has an aperture at the
interaction region through which the ions travel. This tube, held at -500 V in the
first TOF analyzer and -1500 V in the second mass analyzer allows the laser to
pass unobstructed (except for background gas molecules and analyte ions) and

maintains a field free drift path in both analyzers.

74

 

....
. . r!

. I 3... ’

u 4. w.... ..1. i ..v u. I 1:. .. ... f .. Hum-i! —

 

6.56: 5:282 :9: ma... 6:6 22.23 :2 86:2.
>686 .o o:_6m m>=om6m 6:. 6. 6326 :6: :96 6...; .266E____E 26.
a 2:0 66:26 862666 8.6526 68:. E 8.26:2. Em: 6::28
2: Soon- 23 >89. .6 22. em 8%, 29:26 66.65% .80....
62.; 6:: 6 26E 6. 2mm 2.... .23 :o. 6:. 6 6.6:. zo_.2_m he 2:2".

 

....» D b h»
EIE.Eit.bODO‘-‘O‘O..OI‘DVODOOODOOaid I
q .u l a

[II
D

 

 

 

 

 

75

4.2.6 Second TOF Analyzer

42.6.1 Post Acceleration

Product ions produced in the interaction region will retain, approximately
the velocity of their precursor ion, but will only have a fraction of the kinetic
energy of the precursor. To impart mass dependent velocities, the ions are
accelerated immediately following the interaction region. The acceleration field,

established by two parallel grids, is mounted on the optical rail.

42.62 Reflectron 2

The requirements of the second reflectron are different than the first
reflectron, because the ions to be focused have a much larger range of
energies. The product ions have a mass dependent fraction of energy obtained
from the precursor. Two types of reflectrons have been constructed for this
second stage of analysis. The first grid-free reflectron is, except for the voltages
applied, identical to the grid-free reflectron used for the first TOF analyzer. The
second is a gridded, two-stage mirror. To construct this gridded reflectron, a
very thin, electrodeposited sheet of grid material was secured to the first and
second washers, and 5 M0 resistors are used between the back eight washers.
A more detailed analysis of the requirements and operation of these reflectrons

is found in Chapter 6.
4.2.6.3 Detector 2‘

The final detector, flange mounted at the end of the second TOF
analyzer, detects the unfragmented precursor ions and product ions produced.
This detector was designed to incorporate three multichannel plate amplifiers if
necessary for low current signals. This third MCP has not been necessary due

to the ample sensitivity observed.

76

4.3 Performance

This instrument was built in a stepwise manner. The source, first TOF
analyzer, detector, and necessary ion optical elements were built first, and their
performance evaluated. Next the second TOF analyzer, and its components
were installed and tested. The results of these experiments are reviewed in this

section.

4.3.1 Resolution

With a detector placed at the interaction region of the TOF/TOF
instrument, mass resolution is measured to be approximately 1500 (FWHM).
Figure 4.8 is a mass spectrum of the molecular ion region of bromobenzene,
demonstrating the resolution achieved.

. Figure 4.9 is a spectrum of bromobenzene (m/z = 158) collected at the
final detector. Calculated resolution for this spectrum is over 1700. The ions
which produced this spectrum were created in the source, reflected by the first
ion reflectron to form a space-focus plate at the interaction region, accelerated
by the acceleration grids, reflected and focused by the second reflectron and
detected at the final detector position. Unfortunately, the resolution achieved at
the final detector is not twice the resolution achieved at the interaction region as
would be expected with the extended flight path. The reason for this anomaly is

not yet determined.
4.3.2 Beam Image
, In order to maximize the photo-induced dissociation efficiency, the ion
beam must be both temporally and radially focused at the interaction region. In

order to measure the radial distribution of the ion beam, an iris diaphragm which

can be controlled via a rotational motion feedthrough was constructed and

77

.89 Em R: u #5 :2
5.8206 9: :2 22:3”: 83 B 5:289 @5265 6059 5:855
9: :m .286: a 5:5 8826: 23:89:05 .0 E28QO «.922 ad 2:9”.

as 25:.
98 :8 v.8 «.8 8 mg

q d -

di-

mna u a):

:3 u a):

wdw

antenatal

 

78

.2213“: 005 :05 :2020 :0:282 0:568 03208 ”:0:
0:008 0:: :0 0:0 0:: :0 0000.0 5820: 0 :22, 0202.00 082000805
:0 89 n ~>5 :0_ :200205 2028658 0:: :0: E2808 8.22 0.: 0:09“.

3.: 25:.

mud: ”NE 2N3 9N3 mAvH wAE

_ p —
q _

-o-
..
.0—

p
«

fireman}

 

79

.020 22:30 5002020 0:: 0:: :0 :0:0::: 0 mm m: n NE: :0 38:22 2: :0 SE 0 iv 0:09“:

A :18: 002 83.83:
mi v m.» m ed N :4 a :6 o

— : p b p : _ : _
u i

_ _ — a _ P c.

(Am) MESUSWI

 

 

8O

placed in front of the detector at the interaction region. Figure 4.10 is a plot of
ion intensity (m/z = 18) versus the area of the aperture opening. The intensity
increased linearly until the aperture area was approximately 0.5 cm2, and was
still increasing, but less rapidly, when the maximum area for the detector was
reached (~ 4.1 cmz). This indicates that although the ion beam is more intense
in the center, a substantial percentage of the ions at the interaction region are in
a less dense outer radius. In addition, a large number of ions are probably not
being detected due to their radial energies. Improvements in the source design,
to minimize non-linear extraction fields, and in the Einzel lens design, to
minimize the aberration effects, may help to tighten the radial focus of the ions

and may help reduce the ion loss.

4.3.3 Preliminary Gate Studies

A study of the gate voltage necessary to effect ion deflection was done to
establish conformity with the SIMION modeling of this process. Since the dual
pulsers needed for this experiment are not yet completed, static voltages were
used on the alternate gate wires. The gate was placed 6" in front of a detector
located at the interaction region. An 0.5" aperture placed in front of the gate was
used to define the diameter of the ion beam. Figure 4.11 is a graph of the
relative ion intensity versus the potential difference applied between alternate
wires. In agreement with modeling, potentials of 130 V and -130 V, were
necessary to deflect the entire beam from the detector. Much larger voltages will
actually be used in order to cause deflection of ion packets which only
experience a fraction of this deflecting field due to their close proximity to the
packet being selected. The amplitude and rise time of this deflecting pulse will

ultimately limit the resolution of the gate selection.

81

_ _
FEE.

 

0:0 5:00:00 0:: :0 02.3 2052.0 :00500 00.300 00:22:.0 .0::200 0:: 08:0> 2.22:. :0. :0 :20 1.0 0:00.“.

CC 00:? 0:00.820 000253 00:20:50 35:89:

com com com o3 o3 on o
. . . . o

p
a
O

l

.2.

O
In
51391191111 eAtqueH

O
O
O}

 

o o o o :. ca

82

References

D. Ft. McLane, Doctoral Dissertation, Michigan State University, 1993.

J. F. O'Hanlon, A User's Guide to Vacuum Technology. John Wiley &
Sons, Inc., 1989, 445. 7

H. Wollnik, M Przewloka, Int. J. Mass Spectrom. Ion Proc., 1990, 96, 267.

Ft. Weinkauf, K. Walter, C. Weickhardt, U. Boesl, E. W. Schlag, Z.
Naturforsh., 1989, 44a, 1219.

83

Chapter 5: Photo-Induced Dissociation

Truth as subject
Beauty as means
Utility as purpose

Alessandro Manzoni

5.1 Introduction to Photo-Induced Dissociation

Ion photo-induced dissociation is the process:

AB++hv-)A++B

The pioneering work in ion photo-induced dissociation (PID) began in the
1960's [1]. These early studies were performed by physicists interested in
determining the interaction cross section of very simple species (e.g., H2+). In
the 1970's, Dunbar [2] began using the ion cyclotron resonance (ICR) mass
spectrometer to study photo-induced dissociation of more complex species of
chemical interest. PID was shown to provide information about the optical
spectroscopy and the electronic and vibrational levels of gas phase ions.
Capable of trapping ions for minutes, the ICR was well suited for early PID
studies due to the very low PID cross sections and low photon sources available.

PID cross-sections have been determined to be between 10'17 and 10'19 cm2

34

[2] for ions such as N201: CH4+, and the molecular ions of trimethylbenzene,
butadiene, and hexatriene. In ICR instruments, sufficient photofragment-ion
yields were obtained using conventional arc lamp sources. Unfortunately, these
experiments are difficult to interpret due to the formation of metastable, CID, and
ion-molecule product ions formed during the relatively long trapping time in the
ICR [3]. Background subtraction techniques do not eliminate these interferences
because the neutral molecules involved in these processes are often desorbed
from the ICR electrodes by photon excitation.

In the 1980's Beynon [4] took advantage of the introduction of high-
powered continuous-wave lasers and performed PID in a beam instrument. To
effect sufficient fragmentation yields, it was necessary to bring the ion and laser
beams into axial coincidence. .

Morrison [5] used a flashlamp-pumped dye laser with relatively long flash
duration (1 us) oriented perpendicular with respect to the low density ion beam
in the interaction region of a triple quadrupole mass spectrometer. Again the low
sensitivity with orthogonal excitation observed was improved by the use of a
coaxial geometry which exposed a greater number of ions to the laser flash.

With the development of lasers producing short-duration, high-energy,
pulses, came the implementation of time-of-flight mass spectrometers in
conjunction with PID. Russell has used a hybrid EB/TOF mass spectrometer
with PID occurring between the magnetic sector and the TOF mass
spectrometer [6]. Using pulsed Cs+ i0n liquid SIMS ionization and a pulsed
excimer laser, Russel demonstrated an increase in sensitivity of 103-104 over
experiments with a CW laser. Colby and Reilly have performed TOF MS/MS
analysis on benzene and aniline [7]. These studies relied on the ability to
selectively photodissociate a chosen ion over all others in the source of a TOF

mass spectrometer. Duncan [8] has demonstrated the utility of PID in a

85

reflectron TOF mass spectrometer by exciting gate-selected ions in the turn-
around region of the reflectron. Using the reflectron solely to decelerate the ions
for interaction was necessary due to the difficulty he experienced in timing the
laser pulse. Schlag et al. [9] have demonstrated MS/MS using linear TOF
separation of gate-selected precursor ions, PID at the space-focus plane, and
sequential focusing of the product ions with a reflectron voltage scanning
technique.

With the availability of very high-powered, short duration, pulsed lasers
which can be operated at frequencies to 300 Hz, photo-induced dissociation is
uniquely suited to the fragmentation needs of our tandem time-of-flight mass
spectrometer. In this instrument, isomass ion packets are well focused to
. temporal and spatial dimensions that are very similar to those of our focused
laser pulse. The resulting high degree of spatial and temporal overlap not only
provides optimal PID efficiency, but high resolution selection of the precursor ion
packet of interest is also achieved. In addition, the ability to operate the laser at
high frequency allows many product spectra to be produced per second. In this
chapter the achievement of high-efficiency PID consistent with the overall

instrumentation goals is presented.

5.2 PID in the TOF/T OF Instrument

To maximize the photo-induced dissociation fragmentation efficiency, the
maximum number of the desired precursor ions produced in the source must be
exposed to as dense a population of photons as possible. In addition, a high
percentage of the product ions formed must be detected. Using a storage

source and reflectron focusing of precursor ions, the tandem time-of-flight

86

instrument provides this high overlap of ions and photons and with TOF analysis
of product ions detection efficiency is maximized.

For precursor ion selection we rely on the laser pulse interacting with
exactly one precursor m/z value. The laser pulse is focused to a width of
approximately 1 mm. The isomass ion packets are also focused to 1 mm in
length (resolution = 1200 FWHM). At m/z 1000, the trailing edge (FWHM) of the
m/z1000 ion is separated from the leading edge of the m/z 1001 ion by 0.2 mm.
During the 17 ns temporal length of the laser pulse, the m/21000 ion packet
travels 0.2 mm. At higher m/z values, the distance between ion packets
becomes smaller than the distance traveled in 17 ns. and unit resolution for
precursor selection is not achieved. If greater resolution were required, focusing
the laser beam to less than 1 mm or improvements in the precursor resolution
(above 1200) could be made.

The laser used in this system is a Questek 2580vl3 excimer. Operating
with ArF (193 nm) at 30 Hz, this laser was measured to yield between 4-5 W, or
approximately 150 mJ per pulse. When focused by a cylindrical plano-convex
lens down to the dimensions of the ion packet (1 cm x 1 mm), the laser pulse
photon density is 1.46 x 1018 photons/cmz. Given that PID cross-sections have
been determined to be between 10'17 and 10'19 cm2- [2} the PID efficiency
should theoretically be between 100% and 15%.

5.3 Results and Discussion

5.3.1 Precursor Selection

Figure 5.1a is a spectrum of the molecular ion region of bromobenzene.

Ions in this spectrum have traveled through the entire instrument. With the laser

87

m/z 156 m/z 158

0.01 ._ m/z 157

 

A A A A
r v v v v v7

 

0.03 'r b

0.01 "' A
A ..
O . c . : v : t ~‘Aw

0.0. I

0.03 '- c
0.02 "
0.01 ‘-

0 : : -.—- m... - ~ ¢ ¢ - .../~44...
0.04 I.
0.03 -- d
0.02 “r .
0.01 “r A

o - A— . ...

140.9 141.3 141.7 142.1 ' 142.5

Intensity (V)

 

Flight Time (#3)

Figure 5.1 Spectra of the molecular ion region of bromobenzene. a. Spectrum
with the laser off. b-d. Spectra with the laser delay set at 86.40 us, 86.70 us and
86.97 us, respectively.

88

on and the delay time between source extraction and laser firing set such that
the laser pulse arrives at the interaction region at the same time as the precursor
ions, interaction between the ions and photons occurs. Figures 5.1b-d are the
spectra obtained when the laser delay times are set at 86.40 us, 86.70 us and
86.97 us to coincide with the ion packet arrival times for m/z values 156, 157
and 158 successively. The ion depletion associated with each of these times is
limited to the selected ion and demonstrates the effectiveness of precursor
selection by laser timing adjustment. Calibration of delay times is easily
accomplished due to the linear dependence of flight time on the square root of
m/z. The ion arrival time jitter and laser thyratron discharge jitter were both
measured to be less than 2 ns. Once the instrument is calibrated, selection of a
new precursor ion simply involves changing the laser delay time to the
calculated value for the desired precursor m/z value In addition, the day to day
arrival time drift is less than 100 ns. Recalibration requires only 3-4 attempted
laser delays. Precursor ion areas have been depleted by 55.2%, 56.3%, and
51.8% for m/z=156, 157, and 158, respectively. This substantial depletion
demonstrates that considerable photon-ion interaction is achieved. To prove
that the observed fragmentation is due to PID and not CID with laser desorbed
molecules, the laser was triggered so that the laser pulse arrives at the
interaction region 50 ns before the m/z=156 ion packet. No depletion was
observed. In addition, the fact that we have selectively excited ion packets
separated by 300 ns conclusively indicates that only PID is affecting the

observed ion intensity.

5.3.2 PID Efficiency

The efficiency of the PID process is determined by the efficiency of

fragmentation and the efficiency of collection of the ions. Letting P and P0

89

represent the precursor ion current with and without laser excitation,
respectively, and 21‘} the total current of all fragment ions, the fragmentation

efficiency is E, = 21‘} /(P+2‘,F;), the collection efficiency is
EC -.- (P+2F})/P,, and the overall PID efficiency is the product
Em = Er'Ec = Eli/RUG]-

To perform efficiency studies, a detector was placed approximately 20 cm
beyond the interaction region as depicted in Figure 5.2. This arrangement
allowed us to study the PID process independent of the effects of the second
reflectron. Figure 5.3a-b are product spectra for the molecular ion (m/z= 156)
of bromobenzene and for rer = 91 of toluene. In these cases, the precursor
ions have been separated by their m/z-dependent velocities and focused at the
detector. These product spectra were obtained by subtracting the Laser Off
spectra from the Laser On spectra. In the bromobenzene product spectmm, ion
peaks at t = 93.60 us, 93.39 us, 92.95 us, 92.50 us and 91.50 us correspond to
product ions with 6, 4, 3, 2 and 1 carbon atoms, respectively. The calculated
fragmentation efficiency, collection efficiency and overall PID efficiency are 56%,
116%, and 65%, respectively. Compared to the El spectrum, the relative
intensities of the lower m/z fragment ions are enhanced in this PID product
spectrum, indicating that either multiple single photon or multiphoton events are
likely to be occurring. '

Product spectra from precursor ions for bromobenzene, nitrobenzene,
acetophenone, triethylamine, N,N-diethylformamide, N-methylacetamide, and
cyclohexene were also collected. Efficiencies for these compounds are given in
Table 5.1. The compounds and precursor ions chosen demonstrate PID
fragmentation of odd electron, even electron, aromatic and nonaromatic
moieties. In most of these studies, the ions were space-focused at the detector

rather than at the interaction region to improve product ion resolution. This

90

 

.569 58925 9: 5:0 $55 up zmfiéxoaam
582% ESSEE 05 B 8:86. 05 9:265 595.55 ”5.55% m5 8 E9905 Wm 9:9“.

 

.893
.88me

uSoSoQ

cBofloQ 380808 N sebum—mom

 

 

 

      

98E
300% 3QO

e

H sebum—mom 550m :050335 occzom

91

 

05288.05 8 amp n NE: :2 5.3029: 9: .2 83:0QO 6:090 < 0m 059...

 

 

 

ea eea. Emma
mafia mmém mmdm mmdm mmAm
«I w n 1 mad-
0030300
080303 .- Hod-
r mood-
- o
. mood
- Hod

 

 

 

-- £06

(A) KutsumUI

92

.NOU‘¢-‘> N\..\ I..\. ..I."Irlilll!1.lfl I.

.0069me 05 «m *0 U699: scam. €269.25 9.: am Ummsog 0.53 9.2 009300;...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Z 5 mi oi No ecexecege
0262000

B so om omi on 1566-2
00_E0E.2

em oo em ooi 00 45292.2

00 o: R o: 00 0550352:

om Bi em oi moi 2820280

5 k: 8 oi mi 08282::

00 moi mm 0: im 0:022

om . o: 3. o? .R

eoi mm: on o? .09

ow no on of R

mo o: oo oi of 0520er92

st Goo 3.0 35

>0:0_0_=w >0c0_0_=m 5:005“... >90cm
OE 00:00:00 020900;.QO 00.30 0000.. NE :0. 00050000 0:309:00
00:_0> NS: :0_ 005000 .9000 ..2 0:000. en. mm 0.03

93

resulted in reduced PID efficiencies. To determine the effect of improved
focusing on PID efficiency, two precursor ions (indicated by asterisks in Table
5.1) were studied with the space focus at the interaction region. As
hypothesized, PID efficiencies were increased. Collection efficiencies of greater
than 100% were sometimes observed. This could be the result of greater
detector gain for ions with higher velocity (lower m/z) [11].

The overall PID efficiency of 104% achieved for bromobenzene is much
higher than the 15% achieved by Mclver ef al. [12] in an ICR instrument.
Furthermore, it is higher even than the efficiency achievable in the CID process
in a triple quadrupole instrument (65% for n-hexane) [10]. These results give
evidence that PID can be an analytically useful fragmentation technique when

implemented in a system with high ion/photon overlap.

5.3.3 Photoionization

In addition to the high efficiency photo-induced dissociation observed,
photoionization of residual background molecules present in the interaction
region was also observed (See Figure 5.4). The high frequency signal at 67 us
in this spectrum is due to the thyratron discharge of the laser. The three broad
peaks located at 71-80 us are the peaks due to the multiphoton ionization of
background molecules. The ions produced in this process must drift from the
field free interaction region into the acceleration region. After acceleration they
are separated by their mass-dependent velocities and detected after a 12 inch
flight. The predominant molecules typically present in the background matrix
include H20, N2, and 02. These molecules have ionization energies of 12.6
eV, 15.6 eV, and 12.1 eV, respectively [13]. The ionization must therefore be a
multiphoton process since a 193 nm photon has 6.44 eV of energy. To eliminate

the detection of these photoionization ions when collecting product spectra a

94

repelling grid with a low voltage (10V) applied to it was placed immediately after

the interaction region to deflect the multiphoton ionization products.

95

.0060. 00000.02.
05 0. 00500.9: 000 03200. 00 0800022020 05 Eo: 00230. 5.50000 9:. 0.0 050E

3:5 055.

no

 

 

 

 

96

10

11

References

 

 

G. H. Dunn, Atomic Collision Processes, M. R. C. McDowell, ed., North

Holland Publ., Amsterdam, 1964.

Dunbar, R. C. in Gas Phase Ion Chemistry; Vol. 2, Bowers, M. T., Ed.
Academic: New York, 1979; p 181.

R. E. Tecklenburg, Jr., D. H. Russel, Mass Spec. Rev., 1990, 9, 405.
F. M. Harris, J. H. Beynon, Gas Phase Ion Chemistry, 1984, 3, 99.

D.C.McGiIvery, J. D.Morrison, Int. J. Mass Spectrom. Ion Physics 1978,
28, 81.

R. E. Tecklenburg, Jr., D. H. Russell, Presented at the Pittsburgh
Conference; Atlanta, GA, March 1989; p 488.

S. M. Colby, M. Yang, J. P. Reilly, Presented at the Pittsburgh
Conference; Atlanta, GA, March 1989; p 482.

K. LaiHing, P. Y. Cheng, T. G. Taylor, K. F. Willey, M. Peschke,
M. A. Duncan, Anal. Chem. 1989, 61, 1458.

R. Weinkauf, K. Walter, C. Weickhardt, U. Boesl, E. W. Schlag, Z.
Naturforsch 1 989, 44a, 1219.

R. A. Yost, C. G. Enke, D. C. McGilvery, D. Smith, J. D. Morrison, int. J.
Mass Spectrom. Ion Physics 1 979, 30, 127.

R. Meier, P. Eberhardt, Int. J. Mass Spectrom. Ion Processes 1993, 123,
19.

97

M_—_—=
1 2 W. D. Bowers, S. S. Delbert, R. T. Mclver, Jr., Anal. Chem. 1 986, 58, 969.

13 F. W. McLafferty, 1980, Interpretation of Mass Spectra, University

Science Books, 278.

98

Chapter 6: Product Ion Analysis

. "The easiest way to lose success is to become
convrnced that you are successful."

American Express Commercial

6.1 Historical Use of TOF for Product Ion Analysis

The high sensitivity, high speed, low cost and inherent simplicity of time-
of-flight analysis have made it an attractive alternative for the analysis of product
ions. In 1976 Johnsen et al. [1 ] built a tandem magnetic sector/T OF instrument
to study surface induced dissociation (SID). In 1985 LeBeyec and Della-Negra
[2 ] developed a 252Cf TOF mass spectrometer which utilized a reflectron, and a
neutral detector placed behind the reflectron. This instrument was developed to
study metastable ions. The decay of a metastable ion in the field free region
produced a neutral and a product ion. The time-of-flight of the neutral ion,
detected at the neutral detector allows the determination of the precursor ion m/z
value, and the reflectron gives the product ion a mass dependent velocity. A
similar instrument using Cs+ SIMS was developed by Standing et al. [3 ]. Glish

etal. [4], interested in developing a low cost MS/MS instrument built a

99

quadrupole! TOF instrument which used CID in an intermediate collision cell,
and beam deflection to create a discrete ion packet for TOF mass analysis.
Cooks etal. [5] built a tandem TOF/T OF instrument which incorporated SID.
The theoretically unlimited mass range and high sensitivity made TOF
advantageous for the SID technique which is particularly useful for the
fragmentation of large molecules requiring considerable amounts of internal
energy to dissociate. More recently, Russel [6] has built a tandem magnetic
sector (EB)/T OF instrument, and Schlag [7] has built a tandem TOFfI’ OF both
using laser photo-induced dissociation. Although the instruments built to date
have demonstrated some of the advantages of time-of-flight analysis of product
ions, none has achieved unit mass resolution of the product spectrum over the

full GC/MS mass range.

6.2 TOF Product Ion Analysis in the TOF/T OF Instrument

The tandem TOF/T OF was designed to obtain unit mass resolution for
both precursor ion selection and product ion analysis. Achieving unit mass
resolution for the production ions, however, is complicated by the fact that the
kinetic energy of the product ions is mass dependent.

After photo-induced dissociation in the interaction region has occurred,
the product ions produced move with the same velocity as their precursor ion.
The product ions also have a fraction of the kinetic energy spread which was
obtained by the precursors in the source extraction process as well as the
additional kinetic energy acquired in the photofragmenation process. The
product ions are not accelerated after photo-induced dissociation, they will all

reach the second reflectron at approximately the same time, and the field in this

100

reflectron will serve to mass separate the ions. Serving as a mass analyzer,
however, the reflectron's energy focusing ability is lost.

In order to focus the kinetic energy distribution of the product ions, the
products are accelerated just following the interaction region. This acceleration
causes the ions to acquire mass dependent velocities, and isomass ion packets
separate prior to the second reflectron. The total kinetic energy of the product
ion is mass dependent. If the m/z value of a product ion is rm and the m/z value
of the precursor ion is mp, the kinetic energy of the product ion (Kd) after
fragmentation is md/mp x Kp where Kp is the kinetic energy of the precursor ion.
With a precursor ion energy of 500 eV and a post fragmentation acceleration
potential of 1000 eV, the range of energies for a given selected precursor and its
products is 1000 eV to 1500 eV with the unfragmented precursor ion having the
greatest energy.

Unfortunately, for a reflectron to achieve perfect energy focusing, all of
the ions should have relatively the same kinetic energy. The second reflectron,
if optimized for the energy of the precursor ions, will not achieve perfect focusing

of the product ions.

6.3 Theoretically Modeled Resolution

Concerned about the extent to which the resolving power of the second
reflectron is dependent upon the energy spread, this reflectron was
mathematically modeled.

Using an equation for the resolution of a linear mirror developed by
D. loanoviciu (Equation 6-1), the effect of the energy spread was theoretically

estimated.

101

 

 

 

Ff: 6-1
2,1, ( I
where: 4d1
u1cose
=h+§tan9(1+20tan9+-a—f-)
L 2
2
+-1-20ttan9+—f-a+B _Tl___
L 2 2 Tlcor
1 2

+—or
’- 2 floor

2
+— L-l-oaftan9[1+ Tl ——)+ L[3- 11 )
1lcor 8 1lcor

 

tan29+l(1-—3—)

 

 

IO = the ion packet length at the interaction region

S = the ion packet width at the interaction region

9 = the incidence angle at the second reflectron's entry

or = the radial half angular aperture of the beam (packet's divergence)
[3 = the maximum axial angle normal to the deflection plane

af- — the relative ion energy spread in the second TOF region
n = reference energy in the second TOF analyzer. Products of ratio p (not
energy focused)

= Uop + AU
U0 = the energy of the precursor before interaction

AU = the additional energy obtained by acceleration after interaction

Tlcor = reference energy in the second TOF analyzer. Products of ratio p
(correctly energy focused)

= UoPcor 1’ AU
p = mass product (md)/mass precursor (mp)
80 = the initial relative energy spread of the precursors

102

Given the TOF resolving power achieved in Giessen (2000 FWHM),
values for the parameters necessary for this calculation were estimated, and the
resolution for the second TOF analyzer was calculated for product ions of
precursors with m/z values of 500, 1000, and 2000. As seen in Figure 6.1, the
product ions formed from precursor ions with m/z values to 1000 were calculated
to have at least unit mass resolution. In addition it was determined that
improvements in resolution are expected if the range of energy was minimized.
For example, resolution should improve if the ions receive an additional 2000 eV
after the interaction region giving the product ions an energy spread from 2000
to 2500 eV rather than 1000 to 1500 eV. Another possible method for reducing

the relative energy spread would be to decelerate the ions prior to interaction.

6.4 Performance

The reflectron tested in the second analyzer of the TOF/T OF instrument
was a grid-free reflectron. This grid-free reflectron was indicated to be capable
of resolving a larger range of energies than a gridded reflectron [8].

Initial photo-induced dissociation experiments of the bromobenzene
molecular ions (m/z .= 156, 157 and 158) showed a dramatic depletion of the
precursor ion peaks (Chapter 5, Section 5.3.1) but no product ions were
observed. In was anticipated that the predominant product peak would be from
the loss of Br- (m/z = 77). As observed in Figure 6.1 the resolution for product
ions which retain half of the kinetic energy of their precursors is dramatically less
than that of the precursor. It was hypothesized that no product ions were
detected in this experiment due to their very poor resolution. Another possible

explanation of the absence of product ions is that upon dissociation a significant

103

 

Resolvmg' Power

3WD -

N

i

1000..

200011 ’///

 

 

Figure 6.1

500,

 

5000 1000 u
i /x/
5 x”
s r \
5 // Unit Mass
1 Resolution
.5
i
l.
’1: I". ’,
“.y. .' ....... "
.’/. ...--
//
0 1000 2000 3300
hagment Mass

Simulated resolution of product ions for precursor ions with m/z =
1000, and 2000. All fragment ions of precursor ions with m/z .<. 1000

will have at least unit mass resolution.

104

amount of kinetic energy is released causing a displacement of the products
from the precursor ion axis [9].

To test these hypothesis, the molecular ion of another compound,
toluene, was photodissociated. Toluene was chosen because one of the
dominant product ions from the molecular ion (m/z = 92) results from the loss of
H- (m/z = 91). A spectrum of the molecular ion region of toluene is shown in
Figure 6.2 a. Figure 6.2 b is a spectrum of the same region with the laser timed
to interact with the molecular ion. The peak at 109.4 us corresponds to the
m/z = 91 product ion peak demonstrating that product ions with energies similar
to their precursors are focused by the reflectron.

When the second reflectron's voltages were lowered to optimally focus a
product ion with 1250 eV (rer = 77) rather than the 1500 eV of the precursor ion
(an = 156), a product ion peak was observed (See Figure 6.3). It is thus
possible to obtain good resolution for the complete product ion spectra by
scanning the reflectron's voltages but the advantages of simultaneous product
ion detection are lost. This technique was used by Schlag [7] in his tandem

TOF/T OF instrument.

6.5 The Second Reflectron

Ideally, the second reflectron would be capable of resolving a very large
range of ion kinetic energies. Mamyrin [10] has shown that second-order time-
of-flight focusing for the ions with respect to energy is possible using a two stage
reflectron. In addition, Rockwood [11] has developed a method which gives
infinite order correction against energy variations. In his "perfectron" the
potential field is of the form ¢=ka/2. Ions start at x=0, travel into the

perfectron, are reflected by the parabolic field and are focused at the origin. A

105

m/z = 91
0.07 "

0.06 "

99
00
#0!

m/z=92

Intensity (V)
P
8

P
o
to

     

 

 

0 4. W. m : i
108.9 109.1 109.3 109.5 109.7 109.9
Flight Time (us)

0.07 ‘-
0.06 ~- 0
0.05 «-

0.04 «-

Intensity (V)

0.03 4»

0.02 1r-

O.01 d. M
o c . .

108.9 109.1 109.3 109.5 109.7 109.9
Flight Time (us)

 

Figure 6.2 a. A spectrum of the molecular ion region of toluene obtained at
the final detector. b. The spectrum of the molecular ion region of toluene
when the laser is fired to interact with the molecular ion (m/z = 92). The
peak at 109.4 us is m/z = 91 product ion.

106

 

.22 30:0 000:0. .030. .0. 00:30. 00.5 5.80:0. 0:800
05 .000 005050 AR u~>5 :0_ 8000.0 052000505 0.: 0.0 0.39".

 

3308.0.
0.02 vfi mama 2 mama «2

i 800.0

f 30.0 m...
1-
m
“.0

-. 22:. x

I «00.0

1. good

 

107

detector must be placed at the origin. Knowing that the monopole field is
hyperbolic (See Figure 6.4), this focusing ability was modeled with SIMION. As
shown in Figure 6.5 the focal point for ions with 100-150 eV as well as for ions
with 800 - 875 eV were indeed focused at the origin (x = 75) of this reflectron.
Unfortunately, constructing such an extended monopole would be quite difficult.
A conical monopole reflectron was modeled to determine if it could achieve the
same resolution as an extended monopole system. A conical reflectron has
electrodes which are symmetric about the x axis. As seen in Figure 6.6, this
reflectron did not achieve energy independent focusing.

The challenge of developing a reflectron capable of focusing a large
range of kinetic energies has recently been undertaken by both Oinchung Ji,
and Paul Vlasak. . Using SIMION, mathematical modeling and a numerical

method for achieving a nearly perfect reflectron, they have already achieved

very promising results.

108

equipotential contours ($2 —> 82 + 2000V)
incremented by 100V

     

grid electrode
held at the shield potential (82) \gu'

hyperbolic electrode
3 held at $2 + 2000V

Figure 6.4 SIMION model of a hyperbolic reflectron. The two electrodes used
to form this reflectron include a grid electrode and a hyperbolic electrode.
In one model both of these electrodes extend perpendicular to this page
and are of infinite length, in the other model these electrodes are
symmetric about the x axis. Ions enter these mirrors at the origin of the

grid electrode and travel towards the hyperbolic electrode with no y
velocity component.

0.00058 0005.5
:0 :0 006.050 003 500.050 0.: 0. .0000. .0.00:.0. :. :_000 0.0 0:0
505 0:. .0 550.5. 0:. .5 .5505. 0:. 5.5 5:. 5:; E05 0:. .0
000000. 2.0.00.5. 0.0 :03; 0:0 00.0.05 .5.0...0 .0 00:0. 00.; 0 0.0:
:0.:.s 0:0. .00 u x .0 00.000. 0. £05 0: h 00.0.05 .0 00:0. 00.0. 0 :._.s
05. .0. 0E: 000.0> 0.5 x 0:. 05.0 5. :0 50000. 0:. .o 50.0 0.0 0.00....

 

 

00.08....
m. o. m o m-
- ., o
- om
-- co.
.1 cm.
- _ w . I com

 

 

 

stxv 3: eq; 30018 uotqeooq

110

4. ... ...vv

$.47

 

...—0302.9.

0.:. :...s 000000. 2.0.0060. .o: 0.0 .5505. 0:. 5.5 .0... 5:3
505 0:. .0 000000. 2.0.00.5. 0.0 :0.:.s 0:0 00.0.05 .5.0...0 .0 00:0.
00.3 0 05: 5...? 0:0. .5..00_.0. .00.:00 0 5 008.050 00; 55.250
0.: ... .00. u x .0 00.000. 0. 50.5 0:... 00.0.05 .0 00:0. 00.0. 0 5.?

0:0. .0. 0:... 000.0> 0.5 x 0:. 0:0.0 5. :0 50000. 0:. .0 50.0 0.0 0.00.“.

om

o.

03 as...

m

_
_

 

loo.

o0.

 

.- oom

 

 

(tum) stxv x 30019 uogeooq

111

10

11

References

R. M. Gandy, R. Ampulski, J. Prusaczyk, R. H. Johnsen, Int. J. Mass
Spectrom. Ion Physics, 1977, 24, 363.

S. Della-Negra, Y. L. SeBeyec, Anal. Chem, 1985,57, 2035.

K. G. Standing, R. Beavis, G. BoIbach, W. Ens, F. Lafortune, D. Main, B.
Schueler, X. Tang, J. B. Westmore, Anal. lnstru, 1987, 16, 173.

G. L. Glish, S. A. McLuckey, H. S. McKown, Anal. lnstr., 1987, 16, 191.

K. Schey, R. G. Cooks, R. Grix, H. Wollnik, Int. J. Mass Spectrom. Ion
Proc., 1987,77, 49. '

D. H. Russell, Presented at the 37th ASMS Conference on Mass
Spectrometry and Allied Topics, Miami Beach, FL, May 21 -26, 1989.

R. Weinkauf, K. Walter, C. Weickhardt, U. Boesl, E. W. Schlag, Z
Naturforsch, 1989, 44a, 1219.

Personal communication with R. Grix, 1989.

K. LaiHing, P. Y. Cheng, T. G. Taylor, K. F. Willey, M. Peschke, M. A.
Duncan, Anal. Chem. 1989, 61, 1460.

V. I. Karataev, B. A. Mamyrin, D. V. Shmikk, IFT Physics-Technical
Physics, 1973, 37, 1177.

A. L. Rockwood, Presented at the 34th ASMS Conference on Mass
Spectrometry and Allied Topics, Cincinnati, OH, June 8-13, 1986, 173.

112

1:1

Chapter 7: Conclusions

"The reward for a thing well done is to have
doneit"

Ralph Waldo Emerson

Throughout the design and construction of the tandem time-of-flight
instrument there were numerous instrumental requirements each of which were
critical to the goals of achieving MS/MS on the chromatographic time scale with
this instrument. Although this instrument was significantly based on existing
technology, fulfillment of these requirements was still uncertain. They included:
0 The ability to produce a source which could store continuously produced

ions and perform reproducibly from one day to the next.

0 The ability to focus the ions enough radially such that a significant
fraction would interact with the laser pulse.

0 The ability to accurately and precisely time the laser pulse to interact with
only one m/z value.

c The ability to achieve significant fragmentation with photodissociation.

. The ability to guide ions through two reflectrons without significant ion

loss.

113

o The ability to focus the fragment ions despite the large range of kinetic

energies.

One steady step at a time, a new instrument has been constructed, and
these instrumentation goals have been met [112]. With the instrumentation
requirements of the TOF/T OF demonstrated, actually achieving MS/MS
acquisition speeds compatible with chromatographically eluting compounds is
imminent. There is still, however, the need for an instrument control system, a
time-array detection system, and a data control system which are compatible
with this instrument. Complete implementation of the GC/MS/MS will require
real-time calculation of interaction region arrival times of precursors and real-
time laser timing control. Achieving these capabilities, will undoubtedly require
significant effort and creativity.

The future of this TOF/T OF instrument is very promising and exciting.
The capability of obtaining product spectra at least 100 times faster than current
instrumentation will allow researchers to obtain complete MS/MS spectra from
chromatographically eluting compounds. In addition, researchers currently
limited to single or multiple reaction monitoring in GC/MS/MS applications will be
able to monitor all possible reactions of a precursor m/z value with no loss in
sensitivity.

Although this instrument was designed for GC/MS/MS, its use with other
separation techniques and other forms of ionization including HPLC, CZE,
MALDI and electrospray are equally provoking. Some of the most significant
advances in mass spectrometry in the last 10 years have been in the ionization
of molecules of higher and higher mass. These methods, however, do not
provide significant fragmentation for structural interpretation. The unlimited

mass range of the TOF/T OF instrument, as well as the high mass fragmentation

114

capability of photodissociation, may allow the TOF/T OF instrument to provide
structural information from such large molecules. Equally significant, these data
may be produced from small aliquots of unpurified real samples allowing direct
analysis of biological molecules from blood, cell tissue and other samples of

limited quantity.

References

1 Paul Vlasak, Doctoral Disseration, Michigan State University (in progress)

2 Qinchung Ji, Doctoral Dissertation, Michigan State University (in progress)

115