fl- 87V

1“

 

 

 

 

 

     

   

 

 

 

 

 

7 Quorum L'

\M lilijmijlilll

 

 

 

 

 

 

 

 

3| 1

This is to certify that the

dissertation entitled

Applications of Ion—Molecule Reactions for
Distinguishing Organic Isomers in a Tandem
Quadrupole Mass Spectrometer

presented by

Siu H. Stephen Chan

has been accepted towards fulfillment
of the requirements for

Ph .D degree in Chemistry

Major professor

Mini-7,, W/

MSU is an Affirmative Action/Equal Opportunity Institution O~ 12771

 

newer ,3
Michigan Stag-‘9
L University

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i
i

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l i

MSU Is An Affirmative Action/Equal Opportunity Institution
czwmmma-DJ

 

 

 

 

 

 

 

APPLICATIONS OF ION-MOLECULE REACTIONS FOR
DISTINGUISHING ORGANIC ISOMERS IN A TANDEM QUADRUPOLE
MASS SPECTROMETER

By

Siu H. Stephen Chan

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1991

 

DIS

my
a t
seli

moi

 

630-5437

ABSTRACT

APPLICATIONS OF ION-MOLECULE REACTIONS FOR
DISTINGUISHING ORGANIC ISOMERS IN A TANDEM QUADRUPOLE
MASS SPECTROMETER

By

Siu H. Stephen Chan

Many chemical reactions are isomer-specific whereas simple ionic
fragmentation, upon which the specificity of normal mass spectrometry is
based, is much less so. The unique location of the collision chamber (Q2) of
a tandem quadrupole mass spectrometer (TQMS) provides enhanced
selectivity for mechanistic studies and analytical applications of ion-
molecule reactions. This can be exploited to develop new analytical

techniques for differentiating organic isomers.

Low-energy collisions between the [M-1]' ions of dichlorobenzene
isomers and D20 or ND3 are found to produce distinct hydrogen/deuterium
(H/D) exchange product patterns that allow the isomers to be clearly
differentiated when the traditional mass spectrometric techniques fail. The
predominant products observed for 0—, m- and p-dichlorobenzene are anions
substituted with 3, 2, and 1 deuteriums, respectively. The reactivity of D20
is about three times that of ND3 for these reactions. A mechanism
involving the formation of a five-membered-ring intermediate is proposed
for the exchange reaction and is found to fully explain all the experimental
results. The results of computer simulations based on the proposed

mechanism are consistent with the observed results.

 

tetr

1,3,
proc

i931

dist

PW

bet

crez
usii
use
90h
deb

Ion-molecule reactions are also explored for analyzing isomers of
tetrachlorobenzo-p-dioxin (TCDD). For reactions between D20 and the
[M-1]' ions of TCDD isomers, 2,3,7,8-TCDD is distinguished from 1,2,3,4-,
1,3,6,8-, 1,2,7,8- and 1,3,7,8-TCDD by its unique formation of a single
product having only one H/D exchange. When alcohol molecules are
reacted with the [M-1]' ions from isomeric TCDDs, 2,3,7,8-TCDD is
distinguished from the other less toxic isomers by its lower reactivity for the

formation of the [M-1+alcohol]' adduct products.

While ions formed by low-energy ion-molecule reactions in Q2
provide useful information for isomeric differentiation, these ions will not
be detected if their mobilities are too low to reach the detector. The detection
of these ions can be improved if a gentle longitudinal potential gradient is
created along the ion path of Q2. Theoretically, this can be accomplished by
using Q2 made of BeO rods on which are winded high-resistance wires
used for the generation of the transverse field. The nearly constant

potential gradient will provide the ions with a continuous force towards the

detector.

 

 

 

 

 

to people who strive to make a better world for all human races

iv

 

 

helj
his
alse

esp
rea

QXF
sati
tea.
clas

trei
3001
to a

tea:
con

for

The

live

 

 

 

ACKNOWLEDGEIVIENTS

I would like to thank my research advisor, Dr. Christie Enke, for his
help and guidance that make the completion of my Ph.D a reality. Through
his scientific insight, I have learned a lot about being a better scientist. I
also would like to thank the other members of my guidance committee,
especially Dr. William Reusch who agreed to take on the duty of a second
reader for my dissertation just two weeks before my oral defense.

During my graduate school career at Michigan State University, the
experience of teaching has provided me a great deal of personal
satisfaction. Thanks to all my students who allowed me to enjoy myself as a
teacher. Especially during the rough years of my research, seeing my
classes was something that I was looking forward to every week.

The most rewarding experience of my life comes from the year when
I was a secondary school teacher many years ago. Thanks to the
tremendous friendship and help offered by all of my colleagues who
accepted me just the way I was. I would also like to extend a special thank
to all the kids of my classes who gave me so much fun for being a teacher. I
will never forget the hiking trips with all the kids. It certainly was the
toughest job I have ever loved.

I also want to express my gratitude to two of my undergraduate
teachers, Drs. Paul Merritt and Nick Zevos for their kindness, help and
confidence in me that stirred me into the field of chemistry. My
undergraduate career at Potsdam, New York, would be less enjoyable if not
for the warm hospitality of Mr. & Mrs Sellers during my final year there.
Thanks for all the gifts, camping trips and the party in honor of my
graduation. You guys have made the cold winter of Potsdam so warm to
live in.

I would like to thank my former roommates, David Bell, Andy Teng,
Joe Foley and Mark Meyer for their sincere friendship and putting up with

 

 

 

my different moods. Thanks also go to the members of Enke's group for
their help. Also I would like to thank Susan Cady for her persistent long-
distance friendship and listening to some of my problems.

Over the past many years as a graduate student, I am grateful to
have learned many things beside chemistry. Now I understand more about
the importance of humanity for the development a mature decent human
being. Through appropriate education starting from very young, perhaps,
one day we may live in the "dream world" of Dr. Martin L. King J r..

Although the road to the completion my Ph.D have been a long and
tough one, I would like to express my greatest thank to my parents, sisters
and brother for their love, moral support and understanding. Finally, I
can say with enthusiasm that I am still alive.

 

 

LIST

LIST

INTR(

Introdu
Collisio

Distingi

 

 

TABLE OF CONTENTS
LIST OF TABLES ............................................................................... xi
LIST OF FIGURES ........................................................................... xii
CHAPTERI
INTRODUCTION ................................................................................ 1
Introduction.................................. ..................................................... 1
Collision Processes in Ion-Molecule Reactions ........................................ 2
Distinguishing Organic Isomers in Tandem Mass Spectrometry .............. 8

A. MS/MS in Sector Instruments for Isomeric Differentiation... 9

B. Isomeric Differentiation in MS/MS Instruments Suited for

Ion-Molecule Reactions .................................................... 11
a. Tandem Flowing Afterglow-SIFT-Drift ..................... 12
b. Quadrupole Ion Trap .............................................. 13
c. Fourier Transform Ion Cyclotron Resonance ............ 14
d. Triple Quadrupole Mass Spectrometry ...................... 15

.- u-n_.._ ,_ .

 

 

 

Con

Cone

 

 

CHAPTER2

REACTIVITIES OF DEUTERATED REAGENTS TOWARDS THE
[M-l]' IONS OF CHLORINATED BENZENES FOR H/D EXCHANGE

 

 

 

 

REACTIONS ................................ 27
Intrndm‘tinn . .............................. 27
Experimental .................................................................................... 28
Result and Discussion .................................... 29
A. Characterization and Optimization of H/D Exchange
Reactions Involving the [M-1]' of o-Dichlorobenzene ............ 29
B. H/D Exchange Reactions between Deuterated Reagents and
the [M-1]' of o-Dichlornbenzene 36
C. H/D Exchange Reactions for the [M-1]' of Isomers of
Substituted Chlorobenzenes with D20 ................................ 39
Conclusions ..................................................................................... 5O
CHAPTER3

MECHANISTIC STUDY OF H/D EXCHANGE BETWEEN [M-1]' IONS
OF CHLORINATED BENZENES AND DEUTERATED WATER OR

 

DEUTERATED AMMONIA .................................. 52
Introduction ..................................................................................... 52
Experimental ................................................................................... 54
Results and Discussion ...................................................................... 56
A. Mechanistic Analysis ....................................................... 56

B. Collision Pressure Dependence on Products of H/D
Exchanges ..................................................................... 65
C. Reactions with Other Aromatic Isomers ............................ 86
Conclusions ..................................................................................... 93

viii

 

 

 

mRATIO
MASS SPECI‘I
POLYCHLOF

Introduction...
Experimental.
Results and l

A. Ge

B Re
C. Re.
D Re:

Conclusions .....

IMPROVEMEN
UA]
CONSIDlllIA'.

Introduction .....

DeSigDS for the E
Instrument.

Conclusions...

 

CHAPTER4

EXPLORATION OF ION-MOLECULE REACTIONS IN TANDEM
MASS SPECTROMETRY FOR DISTINGUISHING ISOMERS OF

 

 

 

POLYCHLORO-DIBENZO-P-DIOXINS 99
Introduction ..................................................................................... 99
Experimental ............................................... 101
Results and Discussion ..................................................................... 102
A. Generation of Reactant Ions 102
B. Reactions with D20 ........................................................ 103
C. Reactions with Alcohols .................................................. 112
D. Reactions with 02 ........................................................... 115
Conclusions .................................................................................... 116
CHAPTER5

IMPROVEMENT TO THE DETECTION OF LOW-ENERGY IONS IN
A TRIPLE QUADRUPOLE MASS SPECTROMETER - A THEORETICAL

CONSIDERATION ........................................................................... 120
Introduction ................................................................................... 120
Designs for the Second Quadrupole Collision Chamber of a TQMS

Instrument ..................................................................................... 122
Conclusions ..................................................................................... 140

 

CONCLUSIOI

APPENDIX.

 

 

CONCLUSIONS AND SUGGESTED FUTURE EXTENSIONS ............... 142

APPENDIX ................................................. 145

 

 

Table 2-1

Table 31

Table 41

Table 42

F: rth-la—s.‘

H "3‘"

WP!

 

 

Table 2-1

Table 3-1

Table 4-1

Table 4-2

LIST OF TABLES

Total ion intensity of all products and ion intensity for
the 3-D substituted product at different collision
pressure settings. The optimum collision offset

energy is shown right next to each pressure .....................

Maximum number of H/D exchanges for the [M— 1]‘
ions of isomers of chlorinated benzenes (in cases of
no isomerization) ..........................................................

Ion intensity ratios for m/z 319 and m/z 320 derived
from the ammonia NCI mass spectra of the selected
TCDD isomers ..............................................................

Percentages of the [M-1+alcohol]' adducts based on
100% of the [M-1]' reactants

xi

35

92

106

113

 

 

Figure 1-1

Figure 1-2
Figure 2.1

Figure 2-2

Figure 2-3

Flsure 2-4

ll8111‘s 2.5

llgllre 2.6

H“ H- “.14 h—1

 

Figure 1-1

Figure 1-2

Figure 2-1

Figure 2-2

 

LIST OF FIGURES

Center-of—mass axial kinetic energy spectrum of
protonated or-hydroxyisobutyrate ethyl ester. The
yield of adduct ions, m/z 150 (O), and the most
intense fragment, m/z 115 (I), were recorded against
the experimental center-of-mass energy of the
protonated molecular ion (m/z 133). (Adapted from
Ref. 18) .......
Schematic of a triple quadrupole mass spectrometer
based on the Finnigan TSQ-70B model ...........................

Ion intensity as a function of collision energy offset at Q2
for the [M-1]‘ of o-dichlorobemene

A plot of optimum collision energy as a function of
collision pressure of D20. At a specific pressure value,
an optimum collision energy is obtained by varying
the collision offset voltage of the instrument until the
D3-substituted product is optimiwd

Percent of detected ions that have exchanged with
three deuteriums at different collision pressure
settings. Offset energy is the optimum for each
pressure. The percent detected D3-substituted product
is based on 100% for the total ion intensity of all
detected inns

Product mass spectra obtained from H/D exchange
reactions between the [M-1]' of o-dichlorobenzene

(m/z 145) and different deuterated reagents. The
reagents that do not produce any observable H/D
exchange peaks are not shown in the figure ...................

Selected isomers for H/D exchange reactions with
D20 ............................................................................

Product mass spectra for the [M-1]' ion of the isomers
of chloroaniline upon H/D exchange reactions
with D20 .....................................................................

17

31

33

34

37

4O

41

 

Figure 2-7

Figure 2-8

Figure 2-9

Figure 2-10

Figure 2-11

Figure 2.12

Figure 3-1

Figure 3-2

Pitore 3-3

llillre 3.4

 

Figure 2-7

Figure 2—8

Figure 2-9

Figure 2-10

Figure 2-11

Figure 2-12

Figure 3-1

Figure 3-2

Figure 3-3

Figure 3-4

 

Product mass spectra for the [M-1]' ion of the isomers
of chlorophenol upon H/D exchange reactions with D20... 42

Product mass spectra for the [M-1]' ion of the isomers
of chlorobenzaldehyde upon H/D exchange reactions
with D20 ...................................................................... 43

Product mass spectra for the [M-1]' ion of the isomers
of chlorotoluene upon H/D exchange reactions
with D20 ....................... 44

 

Product mass spectra for the [M-1]' ion of the isomers
of chloroanisole upon H/D exchange reactions
with D20 ...................................................................... 45

Product mass spectra for the [M-1]‘ ion of the isomers
of chlorobenzoic acid upon H/D exchange reactions
with D20 ...................................................................... 46

Product mass spectra for the [M-1]' ion of the isomers
of chloronitrobenzene upon H/D exchange reactions
with D20 ...................................................................... 47

Product spectra obtained from the H/D exchange

reactions between the [M-1]' ions of the individual

isomers of dichlorobenzene and deuterated water

and deuterated ammonia. The reactant ions are at

m/z 145. Peaks at higher mass unit are due to one

or more H/D exchange reactions 57

 

The possible charge sites for the [M-1]' ions

of dichlorobenzene isomers and the consequent

location and number of exchangeable hydrogens

according to the proposed reaction mechanism ............... 62

Plots for normalized relative product intensity vs

collision pressure for H/D exchange reactions between

the [M-1]' ions of dichlorobenzene isomers and

D20 mnlpr‘nlpq .......... 66

 

Plots for normalized relative product intensity vs

collision pressure for H/D exchange reactions between

the [M-1]' ions of Chlorobenzene isomers and ND3

molecules .................................................................... 70

xiii

 

 

 

 

Figure 3.5

Figure 3-6

Figure 3-7

Figure 3-8

Figure 3-9

innit-10

“are an

O H to n M N

 

 

Figure 3-5

Figure 3-6

Figure 3-7

Figure 3-8

Figure 3-9

Figure 3-10

Figure 3-11

 

A simulated collision-dependence plot that represents
a composite of individual collision-product plots for
different percentages of reactive collision. The
simulation is based on the proposed H/D exchange
reaction mechanism between the structure II of

o-dichlorobenzene and D20 or ND3 molecules ..................

The reaction pathway for the structure II anions of
o-dichlorobenzene upon reactive collisions with D20

or ND3 molecules .........................................................

A simulated collision-dependence plot that
represents a composite of individual collision-product
plots for the different percentages of reactive collision.
The simulation is based on the proposed H/D
exchange reaction mechanism between the

structure I anions of o-dichlorobenzene and D20

or ND3 molecules .........................................................

The reaction pathway for the structure I anion
of o-dichlorobenzene upon reactive collision with D20

or ND3 molecules .........................................................

The reaction pathway for the structure III (top) and
structure IV (bottom) anions of o-dichlorobenzene

upon reactive collisions with D20 or ND3 molecules .........

A simulated collision-dependence plot that

represents a composite of individual collision-

product plots for different percentages of reactive
collision. The simulation is based on the proposed

H/D exchange reaction mechanism between a

mixture of different m-dichlorobenzene anions and
D20 or ND3 molecules. The mixture of anions is
assumed to contain 50% structure IV, 25% structure V
and 25% structure II ....................................................

A simulated collision-dependence plot that
represents a composite of individual collision-
product plots for different percentages of reactive
collision. The simulation is based on the proposed
H/D exchange reaction mechanism between
p-dichlorobenzene anions (structure VI) and D20 or

ND 3 molecules ............................................................

xiv

 

73

75

77

80

82

84

87

 

 

 

Figure 3-12

Figure 3-13

Figure 4-1

Figure 4-2

Figure 4-3

Figure 5-1

douse
Pisure 5-3

llElite 5.4

llEuro 5.5

Htitre 5-6

SPQHW

Figure 3-12

Figure 3-13

Figure 4-1

Figure 4-2

Figure 4-3

 

Product spectra for the tetra and trichlorobenzene
isomers that produce H/D exchange ion products. The
H/D exchange products are due to reactions between the
[M- 1]' ions of the isomers indicated and D20 molecules.
No H/D exchange products are observed for the other
tetra and trichlorobenzene isomers ................................

A product spectrum for H/D exchange reactions
between the [M-1]‘ ions of 1-chloronaphthalene and
D20 molecules. Reactant ions having five hydrogens
substituted with deuteriums are the predominant
products ........... ' ..........................................................

Mass spectra of TCDD isomers derived from negative
chemical ionization (NCI) using ammonia as a
reagent........... ................. . .........................................

Mass spectra of 1,2,4,7,8- and 1,2,3,7,8—penta-
chlorodibenzo-p-dioxin derived from negative
chemical ionization (NCI) using ammonia as
a reagent .................................................................

Product mass spectra of [M-1]' ions derived from
TCDD isomers upon hydrogen/deuterium (H/D)
exchange reactions with D20. .The exchange products
are indicated by peaks at one or more m/z units above
the reactant ions .......................................................

A potential contour diagram for the extraction lens
design ......................................................................

A design of segmenting the second quadrupole rods ......

The configuration of a potential contour diagram for
Q2 made of 6-segmented rods with a 5-V potential
applied .....................................................................

The configuration of a potential contour diagram for
Q2 made of 11- segmented rods with a 5—V potential
applied .....................................................................

The configuration of a potential contour diagram for
Q2 made of 21-segmented rods with a 5-V potential
applied................. ....................................................

The configuration of apotential contour diagram for
Q2 made of 26-segmented rods with a 5-V potential
applied .....................................................................

 

90

94

104

107

110

123

125

126

127

128

129

 

 

Figure 5-7

Figure 5-10

Figure 5-11

Figure A-1

Figure A-2

Figure A-3

Figure A-4

Fibre A—5

“ere A-6

Ifigure A.7

 

gure 5-7

gure 5-8
gure 5-9
gure 5-10

gure 5-11

gure A-1
gure A-2
re A-3
re A-4
re A-5
re A-6

re A7

The configuration of a potential contour diagram for
Q2 made of 35-segmented rods with a 5-V

potential applied

Potential along the ion path for the designs of Q2
made of different numbers of segments when a

 

 

potential difference of 5 V is applied

The 1st derivative plots representing potential

gradient vs spatial distance into the Q2 ion path ............

A design of modifying the second quadrupole using
the non-circular zig-zag wire-winding technique.
Dash lines represents wires winded on the back side
of the rod

 

Room temperature thermal conductivity of BeO
compared to various materials. (Adapted from
literature data compiled by Ceradyne, Inc., Costa
Mesa, CA) ' ......

 

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 5% reactive collision

 

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange

 

reactivity 0f 10% reactive mlliqinn

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange

 

reactivity of 20% reactive collision

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 25% reactive collision

 

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 30% reactive collision

 

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 35% reactive collision

 

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 40% reactive collision

 

xvi

 

130

139

146

147

148

149

150

151

152

 

 

Figure A-20

Figure A-21 .

Figure A-22 .

Figure A-23 1

Figure A24 1

Figure A-25 A

Pierre A-26 A

Figure A-27

evenings»

“We A-2e

593-32289.»

BlEllie A.29

5.19235 EL>

 

 

gure A-8

gure A-9

gure A—10

gure A-11

gure A-12

gure A-13

gure A-14

re A-15

re A-16

re A-17

re A-18

re A-19

A simulated collision-product plot for the structure I
of o-dichlorobenz‘ene based on an H/D exchange
reactivity of 45% reactive collision ................................

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 50% reactive collision ................................

A simulated collision-product plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 55% reactive collision ................................

A simulated collision-product plot for the structure I
of o-dichl'orobenzene based on an H/D exchange
reactivity of 60% reactive collision ................................

A simulated collision-produCt plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 65% reactive collision ................................

A simulated collision—product plot for the structure I
of o-dichlorobenzene based on an H/D exchange
reactivity of 70% reactive collision ................................

A simulated collision-product plot for the structure 11
of o-dichlorobenzene based on an H/D exchange
reactivity of 5% reactive collision .................................

A simulated collision-product plot for the structure II
of o-dichlorobenzene based on an H/D exchange
reactivity of 10% reactive collision ................................

A simulated collision-product plot for the structure II
of o-dichlorobenzene based on an H/D exchange
reactivity of 20% reactive collision ................................

A simulated collision-product plot for the structure II
of o-dichlorobenzene based on an H/D exchange
reactivity of 25% reactive collision ................................

A simulated collision-product plot for the structure 11
of o-dichlorobenzene based on an H/D exchange
reactivity of 30% reactive collision ................................

A simulated collision-product plot for the structure II
of o-dichlorobenzene based on an H/D exchange
reactivity of 35% reactive collision ................................

xvii

 

 

 

153

154

155

156

157

158

159

160

161

162

163

164

 

 

 

 

 

figmAfiT

humA38

Figure A-39

Fiaxre A-40

PiEMA-41

Figure A_42

llEllie A4 3

d

l
l
l
r
i

Ln.‘il\qI—IAL_

ADP—39321032)- mambox k‘)h—JO§HA~.

eve—392:9.»

 

 

|‘igure A-30 A simulated collision-product plot for a mixture
of dichlorobenzene anions originated from
m-dichlorobenzene. The calculations are based on
an H/D exchange reactivity of 25% reactive collision.
The mixture contains anions of 50% structure VI,
25% structure V and 25% structure II .......................... 175

I‘igure A-31 A simulated collision-product plot for a mixture
of dichlorobenzene anions Originated from
m-dichloroben'zene. The calCulations are based on
an H/D exchange reactivity of 30% reactive collision.
The mixture contains anions of 55% structure VI,
25% structure V and 20% structure II .......................... 176

I‘igure A-32 A simulated collision-product plot for a mixture
of dichlorobenzene anions originated from
m-dichlorobenzene. The calculations are based on
an H/D exchange reactivity of 30% reactive collision.
The mixture contains anions of 50% structure VI,
25% structure V and 25% structure II .......................... 177

I‘igure A-33 A simulated collision-product plot for a mixture
of dichlorobenzene anions originated from
m-dichlorobenzene. The calculations are based on
an H/D exchange reactivity of 35% reactive collision.
The mixture contains anions of 55% structure VI,
25% structure V and 20% structure 11.. ........................ 178

igure A-34 A simulated collision-product plot for a mixture
of dichlorobenzene anions originated from
m-dichlorobenzene. Thecalculations are based on
an H/D exchange reactivity of 35% reactive collision.
The mixture contains anions of 50% structure VI,
25% structure V and 25% structure II .......................... 179

igure A-35 A simulated collision-product plot for a mixture
of dichlorobenzene anions originated from
III-dichlorobenzene. The calculations are based on
an H/D exchange reactivity of 40% reactive collision.
The mixture contains anions of 55% structure VI,
25% structure V and 20% structure II .......................... 180

'gure A-36 A simulated collision-product plot for a mixture
of dichlorobenzene anions originated from
m-dichlorobenzene. The calculations are based on
an H/D exchange reactivity of 40% reactive collision.
The mixture contains anions of 50% structure VI,
25% structure V and 25% structure II...........' ............... 181

 

 

 

 

 

Figure A44

Figure A45

Figure A-46

Figure A-47

Figure A-48 .

FiEuro A49 .

iigure A-50

Figure A51

Figure A-52

Flgm‘e A_ 53

igure A-54

 

re A-44

re A-45
re A-46
re A-47
re A—48
re A-49
re A-5O
e A-51
e A-52
e A-53

e A-54

A simulated collision-product plot for a mixture

of dichlorobenzene anions originated from
m-dichlorobenzene. The calculations are based on
an H/D exchange reactivity of 20% reactive collision.
The mixture contains anions of 50% structure VI,
25% structure V and 25% structure II ..........................

A simulated collision-product plot for the structure VI
of p-dichlorobenzene based onan H/D exchange
reactivity of 5% reactive collision .................................

A simulated collision-product plot for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 10% reactive collision .................................

A simulated collision-product plot for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 20% reactive collision .................................

A simulated collision-product plot for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 25% reactive collision .................................

A simulated collision-product plot for the structure VI
of p—dichlorobenzene based on an H/D exchange
reactivity of 30% reactive collision .................................

A simulated collision-product plot for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 35% reactive collision .................................

A simulated collision-product plot for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 40% reactive collision .................................

A simulated collision-product plot for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 45% reactive collision .................................

A simulated collision-product plot for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 50% reactive collision .................................

A simulated collision-product pIOt for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 55% reactive collision .................................

xxi

 

189

190

191

192

193

194

195

196

197

198

199

Figure A-55 A

Figure A-56 A

Figure A-57 r

 

ure A-55 A simulated collision-product plot for the structure VI
of p-dichlorobenzene based on an H/D exchange
reactivity of 60% reactive collision ................................. 200

um A-56 A simulated collision-product plot for the structure VI
of p-dichlorobenZene based on an H/D exchange
reactivity of 65% reactive collision ................................. 201

:ure A—57 A simulated collision-product plot for the structure VI
of p—dichlorobenzene based on an H/D exchange
reactivity of 70% reactive collision ................................. 202

 

 

 

 

 

Introduction

Solvent
occurring in :
reactions are
complete labo
itCllllique req

recluired for 51'

J. J. Th
1913(1). The
bimdaY’s star
Hummus ion.
EEIlerally trea
mass sDectro
reactions Was
Interest in in:
1960s. Overt;
usefm pm Can

bilefly desmm

 

 

CHAPTER 1

INTRODUCTION

oduction

Solvent molecules often play an important role in ionic reactions
irring in solution. The effects of solvation can be eliminated if the
:tions are carried out in the gas phase. A mass spectrometer can be a
iplete laboratory for studying ion—molecule reactions. The gas phase
mique requires only a small fraction of the sample amount normally

uired for similar studies in the condensed phase.

J. J. Thomson first reported the products of ion-molecule reactions in
3 (1). The vacuum technology in Thomson’s day is considered primitive
:oday’s standard. As a result of poor vacuum in the mass spectrometer,
1erous ion-molecule reactions often took place; and the reactions were
erally treated as a nuisance. With the advent of better vacuum and
iS spectrometric technology, a systematic study of ion-molecule
:tions was begun in 1952 with the work of Tal’roze and Ljubimova (2).
rest in ion-molecule reactions has been widespread since the early
is. Over the past three decades, numerous techniques have been found
ii] for carrying out ion-molecule reactions. These techniques will be

ly described in the latter part of this chapter.

 

The ma
molecule re:
compounds.
selection of a
selected reagu
many MS/Mf
quadrupole m;
ion-molecule r
of other tech]

reactions as a1

 

 

2

The main objective of this dissertation is the application of ion-
ecule reactions for distinguishing among isomers of organic
upounds. A tandem mass spectrometric system (MS/MS) allows
action of a specific mass ion from the ion source for reaction with
acted reagent molecules in an isolated reaction chamber. Although
ny MS/MS systems exist today, the unique features of a triple
1drupole mass spectrometer are well suited for analytical applications of
,-molecule reactions. In this thesis, these features are compared to those
other techniques with emphasis on the applicability of ion-molecule

Lotions as analytical tools.

llision Processes in Ion-Molecule Reactions

The first step in an ion-molecule reaction is a collision between an
I. and a neutral molecule. The collision causes energy or momentum to
:hange between the interacting species. The results of the collision
)end very much on the collision energy involved. Although an exact
atment of collision processes is quantum mechanical (3), most collision
ults can be sufficiently explained by models based on classical
chanics. The classical ion-molecule collision theory has been discussed
detail by Su and Bower (4). In general, a fraction of the relative
islational energy prior to collision is converted into internal energy of
collision partners. The internal energy gained may promote both the
and molecule to some new excited electronic, vibrational and rotational
es. Depending on the energy within the interacting systems, the

tations can cause fragmentation, ion-molecule reactions and even

 

photon emisr

energy can alu

Most st
energy regime
and magnetic
collisions at h
typically mom
MIXES, for st
retime, mome
induced by th.
the short inte
are often not
ions resulting

Charge exchan

In contr,
energy COHisio
1°“ energy “ bi
momentum tr
those in the .
Processes, the a
extremely high
Because the efl
With increaser
beans to Plum.
l, the intGFacti.

3

oton emission. For vibrationally excited ions, collisions at thermal

ergy can also result in the reduction of internal energy.

Most studies of ion-molecule collisions have been performed in high-
zrgy regimes (over 1 KeV) or low-energy regimes (below 100 eV). Electric
1 magnetic sector instruments have been used almost exclusively for the
lisions at high energy ( > 1 KeV ). The products of these collisions are
ically monitored in mass-analyzed ion kinetic energy spectrometry, or
KES, for structural elucidation of ionic species (5). In this high-energy
ime, momentum transfer is minimal. The direct electronic excitation
uced by the collisions often causes unimolecular decay (6). Because of
short interacting time for the collision species, ion-molecule reactions

often not observed during high energy collisions. However, product
3 resulting from dissociation, charge stripping, charge inversion and

rge exchange may be formed.

In contrast to the high-energy collisions, the energy involved in low-
‘rgy collisions can not efficiently promote electronic excitation. These
energy “ billiard ball ” collisions involve mostly vibrational excitation by

entum transfer, and produce large scattering angles compared to
e in the high collision energy regime. During these low energy
esses, the energy conversion efficiency from translational to internal is
mely high. In many cases, efficiencies close to 100% are observed (7).
use the efficiency of collisionally activated dissociation (CAD) increases

increased internal energy, low energy collisions are an efficient
s to promote CAD. When collisions occur at very low energies ( < 3 eV

interaction time between ions and neutral molecules can be several

 

 

 

vibrational pr
reactions ins
Collisions at
studies of ion
studied in sec
use of a com
devices can 8
these reaction

and ion cyclot

In mos
designated fr
FHStrument, n
thenna] energ
the kinetic em
or energy in 12
between the i
Energy and m(
available for e
efiective collis
Center of mass
limes the ratio
11183393 of the

the equation as

 

4

'ational periods long. During this long interaction time, ion-molecule
:tions involving the formation of new chemical bonds may occur.
isions at thermal or near thermal energies are typically used for
lies of ion-molecule reactions; thus, these reactions are not easily
'ed in sector instruments. Although hybrid instruments involving the
of a combination of sector (magnetic and electric) and quadrupole
es can also be used for low-energy ion-molecule reactions, most of
reactions have been studied in quadrupole, ion trap, flowing afterglow

ion cyclotron resonance instruments.

In most experiments, a well-defined collision cell or region is
gnated for ion-molecule reactions. In the case of a quadrupole
'ument, neutral molecules are injected into the collision cell under
nal energy. The collision energy for ion—molecule reactions is due to
Linetic energy of the ions entering the collision cell. The kinetic energy
uergy in laboratory frame (Elab) is derived from the potential difference
een the ion source and the collision cell. However, conservation of
ry and momentum in a collision requires that only a fraction of E131, is
able for excitation of the interacting species (8). This fraction is the
.ive collision energy (Ecm) in a coordinate system moving with the
r of mass of the collision partners. As such, Ecm is defined as Elab
the ratio of the mass of the neutral molecule (Mn) and the sum of the

as of the neutral molecule and the reactant ion (Mi). It is described by

nation as:

Ecm = Elaban /(Mn+Mi)l

x-H‘.

 

This equation
energy can b.
relative to the
heavier neutr
into internal 1

aneutral mole

The av
pressure of 11
multiple colli
increases over
internal energ
molecule coll
fiequency of i.
yield from su
multinle collis
intermediates
products, 01;

order Sequenti

Most i0
eneri’ies. Alt
thermoneutral
Studied (12.15:
differs fol. end
orientation (Al

 

5

This equation implies that for a collision with a specific ion, the effective
energy can be varied by either changing the potential of the collision cell
relative to the ion source or changing the mass of the neutral molecules. A

heavier neutral molecule provides a more efficient means to convert Elab

into internal energy. In the case of electron impact, the large mass ratio of

aneutral molecule to an electron makes Elab nearly equal to Ecm.

The average rate of collisions can be increased by raising the
pressure of neutral collision gas in the collision cell. Generally, under
multiple collision conditions, the average internal energy of ions often
.ncreases over that of single collision conditions (9,10). This increase in
.ntemal energy may improve the yield of CAD products. For reactive ion-
nolecule collisions, multiple collision conditions result in a higher
‘requency of ion-molecule interactions, which in turn increase the product
rield from such reactions. When collisions occur at thermal energies,
nultiple collision conditions may also have a stabilizing effect on reaction
ntermediates which are often needed for the formation of final ion
iroducts. Clearly, for reaction products arising from second or higher

rder sequential reactions, multiple collisions are necessary.

Most ion-molecule reactions occur at thermal or near thermal
nergies. Although the majority of these reactions are exothermic or
nermoneutral, some endothermic ion-molecule reactions have also been
:udied (12-15). The dependence of product yield on translational energy
iffers for endothermic and exothermic reactions (16). Average dipole
.ientation (ADO) theory predicts that ion-molecule reactions proceed via

rmation of a long-lived collision complex (17). Accordingly, product ion

 

 

 

yield of an
approaching
maybe expla
complex lives
complex becor
adduct ions is
energy for ad
available for
However, the
exothermic re:
cases, the ex.

Chemica] bone

lon fray
Valuable struc
311d efficiency
vibrational ene
0f the low.ene
incFeases 801m
increase levels
For a large In
due to the larg
fora large ior

b

muse each .

Internal energ

fragmentation,

6

eld of an exothermic reaction is greatest for ion kinetic energy
)proaching zero and decreases with increasing ion kinetic energy. This
ay be explained assumming that products are formed only if the collision
mplex lives long enough. As kinetic energy increases, the life-time of the
mplex becomes too short to form products. The formation of ion-molecule

duct ions is particularly energy sensitive (18,19). The optimum kinetic

 

ergy for adduct ion formation is often barely positive; the energy range
ailable for these reactions is extremely narrow ( see Figure 1-1 ).
wever, there are many instances where adduct ions formed through
thermic reactions may not be observed as final products at all. In such
es, the excess energy released is large enough to cause rupture of

mical bonds.

Ion fragmentation derived from ion-molecule collisions provides
uable structural information regarding the original molecule. The yield
1 efficiency of the fragmentation depend largely on the amount of excess
rational energy in the chemical bonds. For collisions at the very low end
:he low-energy regime, the total internal energy deposition to an ion
feases somewhat linearly with an increase in collision energy. This
'ease levels off as the collision energy increases up to a certain level.
a large molecule, the internal degrees-of—freedom is a large number
to the large number of chemical bonds. The fragmentation efficiency
a large ion is extremely low, even at very high collision energies,
ruse each chemical bond shares only a small fraction of the total
rnal energy gained (20,21). As an alternative, to collision—induced

mentation, ion—molecule reactions may be explored for structural

 

 

 

 

1 00
80
60
3
i
40
20
0
~ (It
Figllre 1.1 0,
col
In)
re
thv

 

 

    

L A A

6.64 ebcfoba 0:1 av

 

 

O

-063 6 0.62

center-of-mass collision energy

ligure 1-1 Center-of—mass axial kinetic energy spectrum of protonated
oc-hydroxyisobutyrateethyl ester. The yield of adduct ions,
m/z 150 (O), and the most intense fragment, m/z 115 (l), were
recorded against the experimental center-of-mass energy of
the protonated molecular ion (m/z 133) (Adapted from Ref. 18)

information t

simple ionic f1

N '1'

Mass sp
samples that
Many example
equipped with
produce suflici
certain organic
VFW Similar mg
Particular, the
conmounds are
notorious comp
POlFClflorodjben
0f different app
sFectral differe
frameWOYk of It
deIlension to all
i1ltlude metast,
(CAD), photons,
demOnstrated i
Farting”, the s
thefragmenifitio:

8

:brmation that is often much more structure-specific than the result of

nple ionic fragmentation (22,23).

stinguishing Organic Isomers in Tandem Mass Spectrometry

Mass spectrometry (MS) has been the method of choice for analyzing
nples that demand low detection limits and structural information.
Lny examples may be found where a single stage mass spectrometer
ripped with electron ionization (E1) or chemical ionization (CI) is used to
tduce sufficiently distinct mass spectra to allow the differentiation of
tain organic isomers. Still many isomers are found to have identical or
y similar mass spectra when the traditional MS techniques are used. In
'ticular, the differentiation of many positional isomers of aromatic
ipounds are known to be extremely difficult. These isomers include

rious compound types such as polychlorinated biphenyls (PCBs) and
chlorodibenzo-p-dioxins (PCDDs). Over the past few decades, a number
'fferent approaches have been adapted in order to increase the mass
tral differences of some isomers. These are primarily within the
ework of MS/MS, which provides an additional mass spectrometric
ension to allow extra processes for the ions. These extra processes may
de metastable decomposition, collisionally activated dissociation
), photodissociation and ion—molecule reactions. All have been
onstrated to be useful in differentiating organic isomers. In

icular, the specificity of ion-molecule reactions has been useful when

agmentation techniques failed (24).

 

 

am

The ap
described by S
A two-sector in
decompositiom
the extent ol
differentiate th
MS/MS for dis
instruments.
different inst]
anallitical appl

next section.

The apph
Mel‘afle’ty et a

the same instru
38 stereoisome
differentiated 0:
metaStable ions

The yield
to have C011ision
imminent. Sn
cause CAD to 0

en .
ergetic collisic

9

ISZMS in Sector Instruments for Isomeric Differentiation

The application of MS/MS for isomeric differentiation was first
ribed by Shannon and McLafferty (25) for analyzing isomers of 02H5O+.
o-sector instrument was used to monitor ions derived from metastable
npositions in the field-free region between sectors. The differences in
extent of the competing decomposition pathways were used to
rentiate the isomers of 02H5O+. Since this beginning, the application of
/IS for distinguishing isomers has been extended to different types of
uments. Different processes have also been applied to each of the
rent instruments for differentiating isomers. The features and
rtical applications of non—sector instruments will be discussed in the

section.

The application of metastable decomposition was further extended by
,fferty et a1. (26) to differentiate isomers of 02H6N+ and C3H8N+ using
ime instrumental settings as described before. Larger molecules such
ereoisomers of 2-acetamido-2-deoxyhexose were also successfully
entiated on the basis of differences in the relative abundances of the
:table ions (27). Despite these successes, metastable decompositions

do not produce useful information for isomeric differentiation.

The yield of ionic fragmentation can be improved if ions are allowed
'e collisions with neutral molecules in the field-free region of a sector
ment. Such collisions increase the internal energy of the ions and
CAD to occur. In order to reduce the effect of scattering due to

tie collisions, neutral molecules with low molecular weight such as

 

He are introd
have used 4
differentiatior
a]. have emplo
differentiate (
ion source ur
molecule reac
three different
C4H7N+adduc
four-sector ins
isomers to be t

are almost ind

The inte
With Photons.
region of a r.
differentiate isc
Wiene, the diff
funciion of phc
isomers of 6th}
Photons. Diffe:
[M'OH]+'

Ions Vt

Although
l0fl~m01ecu1 e re
huge Stu'Dpin
urge“ et a1.

10

.re introduced into the field-free region to induce CAD. Kiremire et al.
2 used CAD to produce diagnostic fragments that allow the
rentiation of pyranococumarin isomers (28). Recently, McLafferty et
ave employed a combination of both ion-molecule reactions and CAD to
rentiate C4H4+ isomers (29). In this case, NH3 was introduced into the
source under EI conditions to produce distinctive products of ion-
:cule reactions for three different C4H4+ structures originated from
a different hydrocarbons. The ion products at m/z 69, presumably the
7N+ adduct, were subjected to CAD with He in the field-free region of a
sector instrument. The distinctive spectra thus obtained allow the
ers to be distinguished. The CAD spectra of unreacted C4H4+ isomers
llmost indistinguishable.

The internal energy of an ion can also be increased by interaction
photons. The process of photodissociation was applied in the field-free
n of a reversed geometry double focusing mass spectrometer to
entiate isomers of xylene and ethylnitrobenzene (30,31). In the case of
e, the difference in translational energy released by the [M]+ ions as a
ion of photon energy was used to differentiate the isomers. For the
rs of ethylnitrobenzene, [M-OH]+ ions were selected to interact with
as. Differences in the relative photoabsorption cross-sections of the

{]+ ions were used for isomeric differentiation.

Although the energetic processes in sector instruments do not allow
)lecule reactions involving the formation of new chemical bonds,
stripping (32,33) and charge exchange reactions (34,35) do occur.

'8 et a1. (36) have used the charge stripping reactions to analyze

 

isomers of C2
collision with
spectm permi
charge excha

differentiate

exchange a cl
singly chained

[(

The isomers <

abundances of

Remains
Bantams

IoII-mole
illn‘mOlecule on
is well Sujted f
frafluentationr
collision energj
favored With
take Place in
reactions. Wit]
ionmolecule re:

b .

 

11

ers of C2H5O+ ions. Stripping of an electron from C2H5O+ by energetic
sion with He generates a doubly charged ion. The resulting mass
tra permit the clear identification of four distinct C2H5O+ species. The
ge exchange reactions were explored by Guilhaus et a1. (37) to
rentiate isomers of [C5H6]2+ ions. The doubly charged ion can
ange a charge with Xe atoms in a high-energy collision to form two

y charged ions. These processes can be represented as :

[Cd—1612+ + Xe ----- > [Csflsl’r + Xe+

isomers of [C5H6]2+ were distinguished primarily by the relative

dances of [C6H6]+ ions formed by the charge exchange.

:Qmeric Differentiation in MS/MS Instgiments Suited for Ion-Molecule
:agtigns

Ion-molecule reactions can be performed on instruments that allow
iolecule collisions at low energies (a few eV or less). A technique that
[1 suited for ion-molecule reactions can still allow CAD, because some
tentation occurs even at very low energies. Thus, by adjustment of the
'on energies, either ion-molecule reactions or CAD products can be
d. With the advent of laser technology, photodissociation can also
place in instruments that are primarily used for ion—molecule

ns. With this in mind, applications of CAD, photodissociation and

lecule reactions will be discussed in this section. The emphasis will

ion-molecule reactions and their potential in analytical applications.

 

 

 

 

Currer
commonly u:
flowing aften
(38), fourier
quadrupole 1
quadrupole us
will be descril
analytical app

isomers.
3- Tandem Fl(

Tandem
develoved in 1
aflerglow sour
111Eunselected
tube, Where no
Helium buffer
can? them to
detected by a .
Infect, this in
qualdruP0163 Ina
design and con
the FA'SIFT t

phenOmeHa 0f 1'

with Such inStr

 

Currently, there are four different MS/MS techniques that are

ammonly used to perform ion-molecule reactions. They are tandem
owing afterglow-selected ion flow tube (SIFT)-drift (or tandem FA-SIFT)
8), fourier transform ion cyclotron resonance (or FTMS) (39-41),

radrupole ion trap mass spectrometry (ITMS) (42-45) and triple

 

iadrupole mass spectrometry (TQMS) (46-48). The unique features, which
11 be described later, of TQMS are utilized in this dissertation to explore
alytical applications of ion-molecule reactions for distinguishing organic

mers.

Tandem Flowing Afterglow—SIFT-Drift

Tandem FA-SIFT is an extension of the flowing afterglow technique
veloped in 1963 by Ferguson et a1. (49). In this technique, the flowing
erglow source is used to generate the reactant ions. These ions are then
iss-selected by the selected ion flow tube (SIFT) to enter the flow—drift
)e, where neutral reagents are added to carry out ion-molecule reactions.
lium buffer gas is used in the flow-drift tube to thermalize the ions and

y them to the reaction regions. The ionic reactants and products are
ected by a quadrupole mass filter coupled with an electron multiplier.
act, this instrumental configuration is very similar to that of a triple
drupole mass spectrometer. The major differences of the two are in the
ign and configuration of the ion source and the collision region, So far,
FA-SIFT technique has been limited to fundamental studies of the
nomena of ion-molecule reactions. Only a few laboratories are equipped

1 such instruments, and they are not commercially available. Perhaps,

 

 

 

 

the popularit
and greater 9

b. Quadrupole

Althoui
described by 1
ion trap (I'I‘Mi
in 1987 (42).
cap electrode:
radiofrequenc
determines th
the field and
bad‘gmund p1
damP the 1m
ionization (CI
Pressures ordr
(51-53). Photo
56) 53) have be
hW years, the
is Particularly
the filture, ur

 

 

 

13

opularity of this technique is overshadowed by the higher versatility
eater sensitivity of the TQMS for analytical applications.

adrupole Ion Trap

Although the basic concept of a quadrupole ion trap was first
ibed by Paul and Steinwedel in the 1950s (50), the use of a quadrupole
rap (ITMS) as an MS/MS instrument was first reported by Louris et al.

87 (42). This three-dimensional quadrupole device consists of two end-

 

electrodes and a central ring electrode to which direct current and
frequency voltages can be applied. The radiofrequency voltage
mines the mass-to-charge ratios of ions which can be trapped within
eld and can be used to selectively eject ions from the traps. A high
"round pressure of helium ( 10'3 torr ) is normally used in the trap to

the motion of stored ions for optimum operation. Chemical
ltiOI‘l (CI) with a variety of CI reagents can also be carried out at
ires orders of magnitude less than those employed in traditional CI
1). Photodissociation (54), CAD (55-57) and ion-molecule reactions (53,
) have been performed in the MS/MS mode of an ITMS. Over the past
ears, the popularity of ITMS has grown tremendously. The technique
ticularly well suited for ion-molecule reactions. Its versatility will, in
lture, undoubtedly allow for the development of many analytical

tions in a wide range of areas.

 

 

 

 

c. Fourier Tr:

Fourier
It is an exten
by E. 0. laws
in ICR by Cor
for ion-moles
typical FTMS
cell at a total
due to the el
trajectory. At
reactions with
1tiger orbits t
motion of all t
receiver plats
comPllter for
Spectrum (60,!
exCPPt for a c
resolving pow
illStl"orient, '

studies of ion-:

Althougi
Processes (62,6
The feasibility
isomers has t

lSOIDEI's 111C111 (1‘

 

14

>urier Transform Ion Cyclotron Resonance

Fourier transform ion cyclotron resonance is also known as FTMS.
an extension of ion cyclotron resonance (ICR) that was first described
I. 0. Lawrence in 1930 (59). The introduction of fourier transformation
3R by Comisarow and Marshall (39) has made FTMS the best technique
ion-molecule experiments requiring ultra-high mass resolution. In a
cal FTMS experiment, ions are generated by electron impart in a cubic
at a total pressure of 1 to 100 utorr. These ions are trapped in the cell
to the effects of magnetic and electric fields and follow a circular
ectory. After a certain time, referred to as trapping time, during which
:tions with neutral molecules can take place, the ions are then excited to
'er orbits by a fast frequency-swept radiofrequency pulse. The coherent
ion of all excited ions induces image currents in a circuit shunting two
river plates. These currents are amplified and transmitted to a
puter for fourier transformation to generate an interpretable mass
:trum (60,61). The basic features of FTMS are quite similar to ITMS,
=pt for a difference in resolving power and operational pressure. A
lving power of over a million is achievable with a modern FTMS
rument. The technique is particularly well suited for fundamental

ies of ion-molecule reactions.

Although some of the first papers about ICR were related to CAD
asses (62,63), the technique was mostly used for ion-molecule reactions.
feasibility of using ion-molecule reactions to differentiate organic

ers has been demonstrated by numerous reports (64-76). These

ers include C2H50+ (68,71), Cero (69), CsHesO“ (70>, trans- and Bis-4-

 

tert-butylcycl
demlone (73).
were motivat
sufficient ma;
success of ti
specificity of
addition to io
FTMS to difi‘e
odei902+ (7
may be used f

d~ Tfiple Quac

The dev
Yost and En}
Widesoread in
involves three
(91) and the
Combination 01
aPplied. Mass
the DC and R]
behavim of ion
differential eq
Whine Wit
(Q2) is need 1
Potential is ap]
during the M3,

 

 

15

 
   
 

~butylcyclohexyl acetate (75), Cngo (72), C7H30 (72), trans- and cis-oo—
alone (73), C3H3+ (76) and tautomers of CgH3N+ (74). Most of the studies
re motivated by the fact that the majority of these isomers do not give
‘ cient mass Spectrometric information for isomeric differentiation. The

cess of these examples demonstrates the usefulness of the high
cificity of ion-molecule reactions compared to the other processes. In
. 'tion to ion-molecule reactions, photodissociation has also been used in
I S to differentiate isomers. Under irradiation by a laser beam, isomers

4H902+ (77,78) and C5H6+ produce characteristic fragment patterns that

s be used for their differentiation.

[‘riple Quadrupole Mass Spectrometry

The development of a triple quadrupole mass spectrometer (TQMS) by
it and Enke (46,47) for general MS/MS applications has stimulated
espread interest in MS/MS techniques. The instrumentation of TQMS
)lves three quadrupole devices positioned in series. The first quadrupole

and the third quadrupole (Q3) are used as mass-filters to which a
bination of direct-current (DC) and radiofrequency (RF) potentials are
ied. Mass selection and mass scanning are accomplished by sweeping
DC and RF potentials applied to the four poles of Q1 and and Q3. The
vior of ions traveling through a quadrupole is described by the Mathieu
rential equations (79,80) that describe the stable trajectory of ions

acting with the fields inside the quadrupoles. The second quadrupole
is used as a non-mass-selective collision chamber where only RF
tial is applied. The processes of ion-molecule interactions occur in Q2

g the MS/MS operation. The main function of the RF potential applied

 

 

 

 

nun-1.4..

omens
pmperties as
mass filter h:
a]. (82). T]
operational .
information.
discussed in c
this (83). 0

reactions will

All the
Performed on
instrument.
exPel'iments v
adaPl the (leg
troditionrr] 1111.
due to the for]
the curved 01
difference in tl
fact, “Clinical
quadeOIe sy
chamber, The
“Mid by th
(85). The Oper
Figure 1-2, B]

enter the 0011is

16

3 is to focus ions scattered by the collision processes. The transmission
erties and the ion containment efficiency of an RF-only quadrupole
filter have been discussed in detail by Miller et al. (81) and Dawson et
32). The versatility of a TQMS instrument allows a variety of
rtional modes, which provide different types of MS and MS/MS
nation. All the possible modes of a TQMS have been reviewed and
ssed in depth by others, including colleagues and students of Dr. C. G.
(83). Only the mode used for the experiments involving ion-molecule

.ons will be discussed.

All the experimental work described in this dissertation was
'med on either a Finnigan TSQ-70B or a Finnigan TSQ-700 TQMS
Lment. Except for some of the work described in Chapter 3, all the
.ments were performed on the TSQ-70B model. Both Finnigan models
the design of a curved collision chamber that is different from the
ional linear design. It is believed that the curved design reduces noise
the formation of neutral products that cannot be transmitted beyond
1rved chamber (84). In addition, there is also a fundamental
nce in the design of Q2 between the TSQ—7OB and TSQ-700 models. In
echm'cally the TSQ-700 model should not be considered as a triple
upole system because it uses an octopole design for the collision
er. Theoretically, the octopole design can reduce the radial energy
ed by the RF potential and allow a higher ion transmission efficiency
[‘he operational mode used for all MS/MS experiments is depicted in
1-2. Briefly, ions created in the source are mass-selected by Q1 to

the collision chamber (Q2) where neutral reagents are introduced for

 

 

 

Epoa mObémE Samara 2.: no 3wa $880.50me 39: mfiomgvaU 39.5 a mo osmfimnom NA 8sz

nmwaam wmmE
agovwmw no“ 903098”

Willi.

m5
ez<eu<mm
‘1'
529:3: ma
_
.

3
L\l‘ _
.

28.5020 “A
s “and...
flu @\ . me
96:53 sees? Fl Fll
dogmamfiuoo Engage .Hoaomfimm mmma

COU‘OGOL COM #CMuomvh

      

17

 

 

 

interaction 1

analyzed by

The vs
techniques.
bombardmer
contrast, on
Sli'l‘, ITMS
unit mass re
with high se
TQMS is our
imPOI'tant in:
greatest adv:
uniQue loca'
SimUltaneous
molecule rear
are SPah'ally.
asetting ens
collision char
Scanning all t
can be iHired
In contrast, I
and mass sca
when an ion

fiHlCtion as E

allowing sufli

18

action with the ions. Product ions emerging from Q2 are then mass-

vzed by Q3 for product mass spectra.

The versatility of this TQMS allows for very wide range of ionization
riques. They include, but are not limited to, CI, El, fast atom
sardment, atmospheric pressure ionization and electrospray. In
ast, only limited ionization techniques are allowed in Tandem-FA-
, ITMS and FTMS. The capacity of a TQMS to provide independently
mass resolution to both mass filtering stages results in an instrument
high selectivity. An even higher selectivity can be achieved when a
S is coupled with gas or liquid chromatography. This capacity is
.tant for trace analysis of environmental samples. Perhaps, the two
est advantages of a TQMS over the ion trapping techniques are the
re location of its collision chamber and the independent and
taneous of the Operation of the three quadrupoles. In a TQMS, ion-
ule reactions take place in a region where the ions and the molecules
tatially separated from the ion source by a mass filtering device. Such
ing ensures that only ions with a specific m/z value can enter the
on chamber. Since ion selection, ion—molecule reaction and mass
.ng all occur stimultaneously in different regions, analytical samples
: introduced continuously into the system without any interruptions.
trast, ITMS and FTMS require the ionization, ion-molecule reaction
ass scanning to be processed in the same region at different times.
an ion trapping instrument is in its trapping mode, it loses its
in as a mass analyzer. Finally, a TQMS has the advantage of

.g sufficient control of a wide range of axial kinetic energies. The

 

capability 0;
studies of be

Despi
disadvantag:
all, the resid
short compar
prohibit the
intermediate
undetermine
applied to th.
is believed b
Spectra obtai
Q3, they are
in the collisit
ions derived
and high col'

discussed in t

The p-
quadeOle 5
Significantly

130111913, infor

 

 

ability of controlling collision energy is important for energy-resolved

lies of both endothermic and exothermic reactions.

Despite all the advantages described above, there are some
rdvantages in using a TQMS to study ion-molecule reactions. First of
the residence time of the reactant ions in the collision chamber is very
rt compared to those of ITMS and FTMS. This short residence time may
iibit the formation of some ion products which require long-lived
rmediates and have slow kinetics. Second, an ion can also gain an
etermined amount of radial kinetic energy due to the RF potential
‘ied to the collision chamber. The octopole design of the TSQ-700 model
elieved to minimize this problem. Finally, because the product mass
:tra obtained from the TQMS represent only the ions transmitted into
they are not necessarily truly representative of the product distribution
1e collision chamber. This could be a problem when sampling product
derived from ion-molecule reactions conducted under thermal energy
high collision pressure conditions. This particular problem will be

issed in greater detail in Chapter 5.

The primary ion-molecule interaction used in most tandem
lrupole systems is CAD. The CAD efficiency in a TQMS is often
.ficantly higher than that of a multisector instrument. For some
ers, information provided by CAD in a TQMS is selective enough to
Lit their differentiation (86-88). Isomers of dimethoxyindole (87),
cal morphinan (86) and o/p bisphenol-A (88) have been successfully
iguished by CAD. However, when CAD fails to produce useful

 

 

_.-___ -.-_... ...' ,_
i “mama‘s- .. -..-____-._—_.. unsh—

information

reactions ms

The 11
Yost et al. (8
differentiate
reported sue
Bursey et al.
of 1,2-cyc10pt
(92). The sa
reactions am
1»2'°Y010pent
PM by J alor
isomers. K0
differentiate
Heath et a1
differentiate t

DeSpitr
imotions in 1
successes den
tnPie quadro
Weir ofi
TQMs, We he
distinguishin
dissertation.

 

 

rmation for isomeric differentiation, the specificity of ion-molecule

tions may be useful.

The use of the TQMS for ion-molecule reactions is relatively new.
et al. (89) first explored the use of ion-molecule reactions in a TQMS to
entiate isomeric structures. Since then, several laboratories have
ted successes in using ion-molecule reactions to differentiate isomers.
ey et al. used ammonia as the neutral reagent to differentiate isomers
-cyclopententanediol (90) C2H5O+ (91) and hexachlorinated biphenyl
The same group also explored the use of endothermic proton transfer

ions and trimethysilyl ions to differentiate isomers of C5H6+ (93) and
yclopententanediol (94), respectively. Buta-1,3-diene and benzene were
by J alonen (95) to differentiate the structures of C2H5O+ and C2H3O+
ers. Kostianinen et al. (96,97) have explored the use of oxygen to
rentiate several isomers of tetrachlorodibenzo-p-dioxin. Recently,
h et al. (98) have reported success in using dimethyl ether to

entiate C7H7+ isomers.

Despite the applications described above, the use of ion-molecule
ons in the TQMS is still limited to only a few laboratories. Their
rses demonstrate the analytical potential of ion-molecule reactions in
quadrupole mass spectrometry. By taking the advantages of the
city of ion-molecule reactions and the sensitivity and selectivity of the
i, we have explored novel applications of ion-molecule reactions for
guishing organic isomers, which will be described in this

tation.

 

 

 

References

1. Thom;
to Cht

2. Tal’ro:
209.

3. Berry,

4. Su, T.
Ed, A

5, McLaf
Am. Ct

6. Yamao
7. Michae

3- Bernst
Claren

9- Todd, 1
44, 37.

10' WYSoc]
Spectre

11- White,
22, 132.

12- Orland
AnaL C

13‘ Orland!
24, 1033

‘ Orlandt
Spectm;

‘ Orlandr
25.485,

16' Kinter,

21

ferences

Thomson, J. J ., “ Rays of Positive Electricity and Their Applications
to Chemical Analysis ”, Longmans : London, England, 1913.

Tal’roze, V. L.; Ljubimova, A. K., Dokl. Akad. Nauk SSSR. 1952, 86,
209.

Berry, M. V.; Mount, K. E., Repts. Prog. Phys. 1972, 35, 315.

Su, T.; Bower, M. T. “ Gas Phase Ion Chemistry, Vol ”, Bower, M. T.
Ed., Academic Press : New York, 1979.

McLafferty, F. W.; Bente, P. F.; Kornfeld, R.; Tsai, S.-C.; Howe, I., J.
Am. Chem. Soc. 1973, 95, 2120.

Yamaoka, H.; Pham, D.; Durup, J., J. Chem. Phys. 1969, 51, 3465.
Michael, K. T.; Bursey, M. M., J. Am. Chem. Soc. 1986, 108, 1797.

”

Bernstein, R. B.; Levine, R. D., “ Molecular Reaction Dynamics ,
Clarendon Press : Oxford, England, 1974.

Todd, P. J .; McLafferty, F. W., Int. J. Mass Spectrom. Ion Phys. 1982,
44, 37.

Wysocki, V. H.; Kenttamaa, H. 1.; Cooks, R. G., Int. J. Mass
Spectrom. Ion Processes 1987, 75, 181.

White, E. L.; Tabet, J.-C.; Bursey, M. M., Org. Mass Spectrom. 1987,
22, 132.

Orlando, R.; Murphy, C.; Fenselau, C.; Hansen, G.; Cotter, R. J.,
Anal. Chem. 1990, 62, 215.

Orlando, R.; Fenselau, C.; Cotter, R. J ., Org. Mass Spectrom. 1989,
24, 1033.

Orlando, R.; Fenselau, C.; Cotter, R. J., Rapid Commun. Mass
Spectrom. 1990, 4(7), 259.

Orlando, R.; Fenselau, C.; Cotter, R. J., Org. Mass Spectrom. 1990,
25,485.

Kinter, M. T.; Bursey, M. M., Org Mass Spectrom. 1987 , 22, 775.
Bowers, M. T.; Su, T.; Anicich, V. G., J. Chem. Phys. 1973, 58, 5175.

 

 

 

 

18.

19.

21.

3

33,23

8‘33?

$52,838

Craig,

Bente
Chem

Hudsc
1984, 1

McAd<

Kintel
1988, 1

Shaun
Lersen
Fourm'

Kirerm'

Rodigh

Wesde
Spectnc

Mukht
Mass 5

' Morga]

Mass 8

Maque:
enu, .

Bowen,

Beynon
1971, 5,

Beylmn

Burger:
1982, 17

 

22

Schmit, J. P.; Dawson, P. H.; Beaulieu, N., Org. Mass Spectrom.
1985, 20, 269.

Schmit, J. P.; Beaudet, S.; Brisson, A., Org. Mass Spectrom. 1986, 21,
493.

Craig, A. G.; Derrick, P. J., J. Am. Chem. Soc. 1985, 107, 6707.

Bente, P. F.; McLafferty, F. W.; McAdoo, D. J .; Lifshitz, C., J. Phys.
Chem. 1975, 79, 713.

Hudson, C. E.; McAdoo, D. J., Int. J. Mass Spectrom. Ion Processes
1984, 59, 325.

McAdoo, D. J.; Hudson, C. E., Org. Mass Spectrom. 1987, 22, 615.

Kinter, M. T.; Bursey, M. M., Biomed. Environ. Mass Spectrom.
1988, 15 583.

Shannon, T. W.; McLafferty, F. W., J. Am. Chem. Soc. 1966, 88, 5021.
Lersen, K.; McLafferty, F. W., J. Am. Chem. Soc. 1974, 96, 139.
Fournie, J.-J.; Puzo, G., Anal. Chem. 1985, 57, 2287.

Kiremire, B. T.; Chiarello, D.; Traldi, P.; Vettori, U.; Guiotto, A.;
Rodighiero, P., Rapid Commun. Mass Spectrom. 1990, 4, 117.

Wesdemiotis, C.; Zhang, M.-Y.; McLafferty, F. W., Org. Mass
Spectrom. 1991, 26, 671.

Mukhtar, E. S.; Griffiths, I. W.; Harris, F. M.; Beynon, J. H., Org.
Mass Spectrom. 1981, 16, 51.

Morgan, T. G.; Kingston, E. E.; Harris, F. M.; Beynon, J. H., Org.
Mass Spectrom. 1982, 17, 594.

Maquestiau, A.; Van Haverbeke, Y.; Flammang, R.; De Meyer, 0.;
Menu, A., Org. Mass Spectrom. 1977, 12, 706.

Bowen, R.; McLafferty, F. W., Org. Mass Spectrom. 1980, 15, 51.

Beynon, J. H.; Mathias, A.; Williams, A. E., Org. Mass Spectrom.
1971, 5, 303.

Beynon, J. H., Anal. Chem. 1970, 42, 97A.

Burgers, P. C.; Terlouw, J. K.; Holmes, J. L., Org. Mass Spectrom.
1982, 17, 369.

 

 

 

 

 

23

Guilhaus, M.; Kingston, R. G.; Brenton, A. G.; Beynon, J. H., Org.
Mass Spectrom. 1985, 20, 424.

Van Doren, J. M.; Barlow, S. E.; Depuy, C. H.; Bierbaum, V. M., Int.
J. Mass Spectrom. Ion Phys. 1987 , 81, 85.

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

Marshall, A. G.; Comisarow, M. B.; Parisod, G., J. Chem. Phys.
1979, 71, 4434.

McIver, R. T., Amer. Lab. 1980, 12, 18.

Louris, J. N.; Cooks, R. G.; Syka, J. E. P.; Kelley, P. E.; Stafford, G.
C., Jr.; Todd, J. F. J ., Anal. Chem. 1987, 59, 1677.

March, R. E., Hughes, R. J ., “ Quadrupole Storage Mass
Spectrometry ( Chemical Analysis Series, Vol. 102 ) ”, Wiley : New
York, 1989.

Berberich, D. W.; Hail, M. E.; Johnson, J. V.; Yost, R. A., Int. J.
Mass Spectrom. Ion Phys. 1989, 94, 115.

March, R. E., Org. Mass Spectrom. 1991, 26, 627.
Yost, R. A.; Enke, C. G., J. Am. Chem. Soc. 1978, 100, 2274.
Yost, R. A.; Enke, C. G., Anal. Chem. 1979, 51, 1251A.

  
  
   
  
   
   
   
   

Yost, R. A.; Enke, C. G., American Lab. 1981, June, 88.

Fehsenfeld, F. C.; Ferguson, E. E.; Schmeltekopf, A. L., J. Chem.
Phys. 1966, 44, 3022.

Paul, W.; Steinwedel, H., US Pat. 2939952, 1960; German pat. 944900,
1956.

Boswell, S. M.; Mather, R. E.; Todd, J. F. J ., Int. J. Mass Spectrom.
Ion Processes. 1990, 90, 139.

Johnson, J. V.; Berberich, D. W.; Pedder, R. E.; Yost, R. A., Adv.
Mass Spectrom. 1989, 11A 276.

 

Berberich, D. W.; Hail, M. E.; Johnson, J. V.; Yost, R. A., Int. J.
Mass Spectrom. Ion Processes 1989, 94, 115.

Louris, J. N.; Brodbelt, J. S.; Cooks, R. G., Int, J. Mass Spectrom.
Ion Processes 1987, 75, 345.

 

 

 

 

- Beauc
- Gross.

- Nibbe

Stale)
Am, C

MCIVe

Henio
1975, .

Tome!

 

 

24

Johnson, J. V.; Yost, R. A.; Kelley, P. E.; Bradford, D. C., Anal.
Chem. 1990, 62, 2162.

Louris, J. N .; Brodbelt, J. S.; Cooks, R. G.; Glish, G. L.; Van Berke],
G. J .; McLuckey, S. A., Int. J. Mass Spectrom. Ion Processes 1990,
96, 117 .

Evans, C.; Traldi, P.; Mele, A.; Sottani, C., Org. Mass Spectrom.
1991, 26, 347.

Brodbelt, J. S; Cooks, R. G., Analytica Chimica Acta 1988, 206, 239.
Lawrence, E. 0.; Edlefsen, N. E., Science 1930, 72, 376.
McIver, R. T., Jr.; Rev. Sci. Instrum. 1970, 41, 555.

Hunt, D. F.; Shabanowitz, J.; McIver, R. T.; Hunter, R. L.; Syka, J.
E., Anal. Chem. 1985, 57, 765.

Kaplan, F., J. Am. Chem. Soc. 1968, 90, 4483.

Henis, J. M. S., J. Am. Chem. Soc. 1968, 90, 844.

Henis, J. M. S., Anal. Chem. 1969, 41, 22A.

Bursey, M. M.; Hoffman, M. K., Can. J. Chem. 1971, 49, 3395.

: Benezra, S. A.; Bursey, M. M., J. Am. Chem. Soc. 1972, 94, 1024.

Bursey, M. M.; Kao, J .-I.; Henion, J. D.; Parker, C. E.; Huang, T.-I.,
Anal. Chem. 1974, 46, 1709.

Beauchamp, J. L.; Dunbar, R. C., J. Am. Chem. Soc. 1970, 92, 1477.
Gross, M. L.; Lin, P.-H.; Franklin, S. T., Anal. Chem. 1972, 44, 974.
Nibbering, N. M. M., Tetrahedron 1973, 29, 385.

Staley, R. H.; Corderman, R. R.; Foster, M. S.; Beauchamp, J. L., J.
Am. Chem. Soc. 1974, 96, 1260.

McIver, R. T., Org. Mass Spectrom. 1975, 10, 396.

Henion, J. D.; Kao, J.-I.; Nixon, W. B.; Bursey, M. M., Anal. Chem.
1975, 47, 689.

Tomer, K. B., Org. Mass Spectrom. 1974, 9, 686.

 

Burse

. Auslo<

. Baykt

Socl!

. A3811]!

1988, l

McLa<
Oxfort

Mathi

. Miller

1986, 1

. Dawsc

1982, 4

. VinceI

G., Or!

. Syka,

1990, g

Syka,
Matus
Analyt

- P61870111

- Emary

00mm;

' Charle

Spectnc

FGttem
19. 104

Meb’erl
169.

Kinter,

white,
18. 413,

25
Bursey, M. M.; Hass, J. R.; Stern, R. L., Anal. Chem. 1975, 47, 1452.

Ausloos, P. J.; Lias, S. G., J. Am. Chem. Soc. 1981, 103, 6505.

Baykut, G; Watson, C. H.; Weller, R. R.; Eyler, J. R., J. Am. Chem.
Soc. 1985, 107, 8036.

Asamoto, B.; Dunbar, R. C., Int. J. Mass Spectrom. Ion Processes
1988, 86, 387.

McLachlan, N. W., “ Theory and Application of Mathieu Function ”,
Oxford University Press : New York, 1947.

Mathieu, E., J. Math. Pures Appl. 1868, 13, 137.

Miller, P. E.; Denton, M. B., Int. J. Mass Spectrom. Ion Processes
1986, 72, 223.

Dawson , P. H.; Fulford, J. E., Int. J. Mass Spectrom. Ion Processes
1982, 42, 195.

Vincenti, M.; Schwartz, J. C.; Cooks, R. G.; Wade, A. P.; Enke, C.
G., Org. Mass Spectrom. 1988, 23, 579.

Syka, J. E. P.; Schoen, A. E., Int. J. Mass Spectrom. Ion Processes
1990, 96, 97.

Syka, J. E. P.; Schoen, A. E.; Stafford, G. C., Jr.; Hail, M. E.;
Matuszak, K. R, Abstract Book of Pittsburgh Conference on
Analytical Chemistry and Applied Spectroscopy, 1990, Abstract #151.
Peltonen, K.; Kostiainen, R., Org. Mass Spectrom. 1988, 23, 278.

Emary, W. B.; Allegri, G.; Costa, 0.; Frison, G.; Traldi, P., Rapid
Commun. Mass Spectrom. 1989, 3, 413.

Charles, B.; Cole, R. B,; Bruneton, J.; Tabet, J.-C., Org. Mass
Spectrom. 1989, 24, 445.

Fetterolf, D. D.; Yost, R. A.; Eyler, J. R., Org. Mass Spectrom. 1984,
19, 104.

Meyerhoffer, W. J.; Bursey, M. M., Org. Mass Spectrom. 1989, 24,
169.

Kinter, M. T.; Bursey, M. M., J. Am. Chem. Soc. 1986, 108, 1797.

White, E. L.; Bursey, M. M., Biomed. Environ. Mass Spectrom. 1989,
18, 413.

 

93.

Kim:

Meyei

Jalom
Kostiz

Kostiz
135.

Heath
1991, .

 

 

 

%

Kinter, M. T.; Bursey, M. M., Org. Mass Spectrom. 1987, 22, 775.

Meyerhoffer, W. J .; Bursey, M. M., Org. Mass Spectrom. 1989, 24,
246.

Jalonen, J., J. Chem. Soc., Chem. Commun. 1985, 872.
Kostiainen, R.; Auriola, S., Org. Mass Spectrom. 1990, 25, 255.

Kostiainen, R.; Auriola, 8., Rapid Commun. Mass Spectrom. 1988, 2,
135.

Heath, T. G.; Allison, J .; Watson, J. T., J. Am. Soc. Mass Spectrom.
1991, 2, 270.

 

REACT]

Inhoducfior

The a
reactions v.
differentiati
emPlayed ch
that would '
The extent 0
and tertiary
later reporte
compounds x
from many .
reaetions w;
reactions 1
”Women-1
analytical pc
with {11.1}
between the
reagents. 1
different de:

 

 

CHAPTER 2

REACTIVITIES OF DEUTERATED REAGENTS TOWARDS
THE [M-ll' IONS OF CHLORINATED BEN ZENES FOR H/D
EXCHANGE REACTIONS

roducfion

The analytical potential of hydrogen/deuterium (H/D) exchange
ctions was first demonstrated by Hunt and co-workers (1) for
erentiating primary, secondary and tertiary amines. This early work
)loyed chemical ionization (CI) using ND3 as a reagent to produce ND4,+
t would then undergo H/D exchange reactions with amine molecules.
extent of the degree of H/D exchange is different for primary, secondary
tertiary amines. Based on a comprehensive study, the same group
r reported H/D exchange reactions on the [M-1]' ions of several aromatic
pounds with deuterated reagents (2). Although protonated molecules
1 many aromatic compounds were also found to have H/D exchange
:tions with deuterated reagents (2,3), thus far only H/D exchange
:tions with [M-1]‘ ions were shown to produce useful mass
trometric information for differentiating aromatic isomers (4). Such
ytical potential encourages further studies on H/D exchange reactions

[M-1]' ions. This chapter will describe the H/D exchange reactions
'een the [M-1]' ions of various chlorinated benzenes and deuterated
ants . The focus of this chapter is a study of relative reactivities for

rent deuterated reagents with different types of [M-1]* ions. The

27

 

 

 

 

mechanistic
will be died

Expainmt

A trig
was used to
of the triple
the experim.
to enter the
mass resolu‘
tuned for s11,
i1110 the ion
PrOduced by
and the ma
reSpectively.
the eXDeTiIm
collision en e:
instrument).
consistency i
Moms of r
DuDlicate n
same. The

Werage of th

.. .— m .U

 

28

anistic details for the reactions of anions of dichlorobenzene isomers

>e discussed in Chapter 3.

rimental

A triple quadrupole mass spectrometer (Finnigan TSQ-7OB model)
ised to perform all the experiments. The operation and configuration
1 triple quadrupole system have been described in Chapter 1. During
:periments, only [M-1]' ions were allowed by the first quadrupole (Q 1)
:er the collision chamber for reactions with deuterated reagents. Unit
resolution was maintained in the third quadrupole whereas Q1 was
for slightly better than unit mass resolution. Samples are introduced
.he ion source by a direct insertion probe. The [M-1]' ions were
ced by CI using NH3 as a reagent. The temperature of the ion source
he manifold of the spectrometer were set at 150 °C and 70 °C,
:tively. The pressure of the CI source and the collision chamber for
:periments were maintained at 1.5 torr and 4 mtorr, respectively. A
)n energy of 2.8 eV (according to the Q2 offset voltage reading of the
ment) was used. All the experiments were performed twice to test for
tency in the results. The replicates were compared for discrepancies
us of relative peak ratios and the absence or presence of stray peaks.
:ate runs consistently gave spectra that were approximately the

The data shown come from one set of results rather than the

e of the replicates.

 

 

All t1
Company.
purchased 1
which was
chemicals w

purification

 

Z)

All the chlorinated benzenes were purchased from Aldrich Chemical
1pany, Inc., Milwaukee, WI. All the deuterated reagents were
chased from Cambridge Isotopes Laboratory, Woburn, MA except ND3
ch was obtained from MSD Isotopes, Montreal, Canada. All the
micals were the best purity available and were used without further

ification.

ults and Discussion

Sharagterizgtign ind Optimization 9_f_‘ H/D Exchange Reactions

Inyglving h M-1 ' qf'Q-Dichlorgbenzene

Many ion-molecule reactions have been found to occur only within
' narrow collision energy ranges (5,6). In a triple quadrupole system,
1 ionic products derived from successful ion-molecule reactions may not
bserved if the instrumental parameters of the system are not set at
opriate values. Usually, the standard tuning program can do a
factory job in optimizing many instrumental parameters. For a given
molecule reaction system, the two major factors that control the

ency of the reaction system are collision energy and collision pressure.

 

:e factors must be optimized in addition to the optimization by the
1g program. Sometimes, compromises have to be made between the
factors in order to achieve a maximum yield of a certain ion product.
a H/D exchange reactions are the main theme of this research work,
tection is devoted to the characterization and optimization of the effects

,lision energy and collision pressure on the exchange reactions. The

 

 

 

study is be
dichloroben

The 1
controlled h
collision cha
chamber if t
complex inst
an ion as it
can affect it
the offset p01
1. This figu
collision cha
derivative pl(
with a width
about 3.0 v.
woold be littl
the Offset pot
comm-mg at l
ionic dissocia
potentials ar
chtimber coul.
transmission ;

but it Will aim

1011.11101‘
and D20 Dmd

30

y is based on the H/D exchange reactions between the [M-1]' of o-

lorobenzene and D20.

The kinetic energy of an ion entering the collision chamber is
rolled by the offset voltage potential between the ion source and the
sion chamber. Theoretically, no ion should be able to enter the collision
iber if the offset voltage potential is set at zero or less. In reality, for a
)lex instrument such as the Finnigan TSQ-7OB, the kinetic energy of
)n as it leaves the source, as well as other instrumental parameters,
affect its transmission. The ion transmission efficiency as a function
iffset potential for the [M-1]‘ of o-dichlorobenzene is shown in Figure 2-
‘his figure demonstrates that some ions can still manage to enter the
iion chamber even at an offset potential below zero. From the
ative plot, the transmission window of the ions centers at around 1.8 V
a width of about 3 V. Transmission of the ion reaches a maximum at
; 3.0 V. This information is important because it suggests that there
i be little or no gain in ion transmission into the collision chamber if
ffset potential is increased beyond 3.0 V. For ion-molecule reactions
ring at high collision pressures, a higher collision energy would favor
dissociation and increase the chance of ion scattering. If the offset
tials are set at below 3.0 V, ion transmission into the collision
ner could be reduced significantly. Optimum is balance between ion
nission and reaction efficiency. It may be different for each reaction;

will almost always be in the low energy region.

Ion-molecule reactions between the [M-1]'ions of o—dichlorobenzene

2O produce H/D exchange ion products. All three hydrogens on the

 

1600 '

1400 ‘

1200 '

mom

0.
«3C3 -DL-wv \nu—QCGHC— C

200 '

oih

FlgUIe 2.]_ I(

 

 

31

600 '
—-El-—- normal plot
400 ‘ ——0— 1st derivative plot

  

200 "
000 ‘
BOO "
5005
too--
200 ‘

uJ-IA . .
0 " l ' l ‘ I ' I r I ' l ' l ' l ' I ' I ' l ' I

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

 

Collision energy offset at 02 (eV)

l-l Ion intensity as a function of collision energy offset at Q2 for the
[M— 1]‘ of o-dichlorobenzene.

 

 

anions of .
resulting fi‘l
a specific 1':
collision ene
potential an
At collision

no H/D exch
optimum col
as a functio
Optimum col
until the pe
product Spec
the collision
causes a big}
mtorr as Shc
DFOduct reac
increased to
reduce the 1,
increase of c,
ion tralismis:
3.0 to M Int:
increases fror
merely 6%.

between the [
of the react;
deuterillms.

SUbSh'tuted W

 

32

s of o-dichlorobenzene are exchangeable, but the ion products
ing from the exchange of all three hydrogens are detected only within
cific range of collision energy and collision pressure. The values of
on energy and collision pressure were obtained from the offset voltage
tial and the pressure gauge reading of the instrument, respectively.
Ilision energies above 5.0 eV or at collision pressure below 0.3 mtorr,
I exchange products of any kind were observed. Figure 2-2 shows the
urn collision energies for the formation of the D3-substituted product
unction of collision pressure. At a specific collision pressure, the
urn collision energy is determined by varying the collision energy
the percentage distribution of the D3-substituted product in the
ct spectrum is optimized. The optimum collision energy increases as
lHiSiOIl pressure is increased. An increase in collision pressure also
3 a higher relative yield of the D3-substituted product up to around 4.0
as shown in Figure 2—3. The relative yield of the D3-substituted
ct reaches a maximum of about 85% when the collision pressure is
sed to 4.0 mtorr. . Further increases of the collision pressure only
a the ion transmission efficiency (see Table 2-1). For example, an
se of collision pressure from 4.0 to 5.0 mtorr would reduce the total
msmission by 60%. However, for a collision pressure increase from
4.0 mtorr, the percentage distribution of the D3-substituted product
ses from 65% to 85%, but the total ion transmission is reduced by
' 6%. The results suggest that for the H/D exchange reactions
n the [M-1]' of o-dichlorobenzene and D20, a maximum of about 85%
reactant ions passing through Q2 are substituted with three
iums. The rest of the reactant ions are either nonreactive or

uted with only one or two deuteriums. The optimum collision energy

 

 

 

.Qo
EDE—
00:0

co co_n:_oo a

>050

- nun-v

A>o c

figure 2-

 

 

 

optimum offset colllslon energy (lab In eV)

 

 

1.0 . . . , . fl
6

colllslon pressure of deuterated water (mtorr)

re 2-2 A plot of Optimum collision energy as a function of collision
pressure of D20. At a specific pressure value, an optimum

collision energy is obtained by varying the collision offset
voltage of the instrument until the D3-substituted product

is optimized.

 

 

 

Ao\ov «01:00.5 80.2.5025 0A.. Unaooueu

Figure

 

100 "

ueIeCIeu u-u suosuru eo prOouct %

 

 

 

collision pressure at optimum energy (mtorr)

{ure 23 Percent of detected ions that have exchanged with three
deuteriums at different collision pressure settings. Offset

energy is the optimum for each pressure. The percent
detected D3-substituted product is based on 100% for the

total ion intensity of all detected ions.

 

 

 

.t—hT— C3c£m mm th050 gumbo Gommfiwoo 859:qu Garb .mMCBuom mhfimmmhnn GommeOU afimhmcflflu

aw QOUUOHQ UGaSumamfldmlmQ 03» how kammGQQGm QOm USN muOSUOhQ :6. M0 Kawmanmvunuw 30w TWQOE Hnmu ®~n-wrh

v.__- _

r_. 5.07-

 

 

 

000.000
000.0004
000.0005
000.00m.0
00068.3
000.00N.NH
00008.2
000.000.?
000.006...“

8500.3 085.53:th
on» .80 .35:wa c3

000.0004
000.00N.N
000.034»
000.00m.m.
00068.3
80000.3
000.000.NN
000.0093u
000.00%.3V

305on :m mo
303820 :2 033

C6 «.00 0.0
So «.8 S.
Owe 0.00 0.0
So 0.00 00
96 mg 06
So wdv 0.0
Sm NNV 0N
Se 0.: 04
So «.3 0.0

REES 5658 898503
08380 6.8395 H8658

.8383 some 8 use:

anwfl 852? m« .8."me 0300 5658 8:839 23. .mwfifimm 9:583 5658 0:8th

aw OSUOMQ Umpdfipmodmumn 33 -34 5.33931 13 313 2>>$$>ch 11 1. hi.

 

 

and collisi

mtorr, resi

B. Eflh
o-Dighlg

Althl
reactions, 4
undergo H
compound:
approxima'
difference,
deuterated
systems,
exchangs 1:
than 20 kc;
react With .
H/D excha
deuterated

The (
C2H50D, (
(Pyridine),
01338001),
The reager
BlEllie 2.4_
p°tentials e

 

36

Iision pressure for the exchange reactions are at 2.8 eV and 4.0

espectively.

mmmmwmmmrm

110 o p ---nzn

Ithough D20 has been widely used as a reagent for H/D exchange
3, a number of other deuterated reagents have also been found to
H/D exchange reactions with both anions and cations of organic

nds (2,7). An H/D exchange reaction requires reagents of
:nately equal relative acidity or proton affinity (2,7). For large
ce, the preferred reaction is proton transfer; thus, no single
bed reagent is universally effective for all kinds of H/D exchange
For negative ions reacting with a deuterated reagent, H/D

e is generally not observed if the difference in acidities is greater
kcal (7). In this section, different deuterated reagents are used to
th the [M-1]' of o-dichlorobenzene for H/D exchange reactions. The
:hange reactivities and efficiencies are compared among the

ed reagents employed.

e deuterated reagents chosen for this study are D20, ND3, CH3OD,
, CDCl3, CD3COCD3, CD3CN, CD3SOCD3, CD4, C5D5 and C5D5N
). Among these reagents, CDC13, CD3000D3, CD3CN and

D3 were found in this study to produce no H/D exchange products.

ents that have produced H/D exchange products are shown in

4. All the spectra were obtained under optimized collision offset

8 and collision pressure settings except for ND3 which could not be

 

 

 

 

~
5
2

Hm :meHHH 09.1w
NA“.

100‘

75‘

50‘

25‘

 

 

 

100'

75‘

50‘

25‘

 

 

 

 

 

I1 I

I

144 145 146 147 148 14

 

 

 

 

CH3OH

 

 

 

 

 

 

I I

I
144 145 146 147 148 149

CD4

 

 

 

 

 

 

 

 

 

I
144 145 1

 

I I I I

46 147 148 149

D

in:

J .

 

 

00" 100'
ND3
75‘ 75‘
50' 50‘
5‘ 25‘
I I
144 145 146 147 148 149
)0“ 100‘
5_ 02H5OH 75 __
0‘ 50‘
5" 25‘
l
I I I I
144 145 146 147 148 149
0‘ 100'1
D
-_ D D 75 .l
D n
)J 50d
SJ 25'
. . 1 .
144 145 146 147 148 149
m/z

 

I I I I

I
144 145 146 147 148 149

4 Product mass Spectra obtained from H/D exchange reactions

between the [M-1]' of o-dichlorobenzene (m/z 145) and different
deuterated reagents. The reagents that do not produce any
observable H/D exchange peaks are not shown in the figure.

 

 

 

optimized
expected t
instrumen‘
for the arr
for all the
varies sub:
the collisir
while the .
pressure se
5.0, 4.5 an.
setting djf
gauge from
for the djfi
one anothe
other magi
its somewh
[M'll‘ of o-
hl'drogens,
Well be the
reactivitieE

application

 

38

'zed because the Optimum collision pressure setting for the reagent is
ted to be well beyond the highest collision pressure allowable by the
ent. The results clearly show that D20 is the most effective reagent

e exchange reactions. Although the optimum collision offset setting
the reagents is at about 3.0 V, the optimum collision pressure setting
substantially depending on types Of reagents used. From optimizing
Ilision pressure settings Of the different reagents, it is found that
the Optimum pressure setting for D20 is at 4.0 mtorr, the Optimum
ire settings for CHgOD, CzH5OD, CD4, C6D5 and C5D5N are at 2.5, 1.5,
.5 and 2.0 mtorr, respectively. It is quite possible that these pressure
g differences are due to biases in pressure calibrations for the ion
from one compound to another. In other words, the real pressures

a different readings Of the different reagents may actually be closer tO
10ther than indicated. In addition tO D20 and ND3, C6D5 is the only

reagent tO allow the production Of the D3-substituted product despite
ewhat bulky structure. Since H/D exchanges between CeDe and the
Of O-dichlorobenzene involve only aromatic deuteriums and aromatic
ens, the chemical similarity between the two exchanging atoms may

the main contributing factor for the unusual reactivity Of CGDS. The

ties of the other reagents are tOO low to be useful for analytical

tions.

 

 

 

0.1110123
Chilled

Den
exchangin.
dichlorobe
differentia
in the ner
hydrogen
substituti'
functional?
functionali
deprotonat
Chlorobenz

are explore

The
were Chos
donatingI
Selected so
these hydr
artitratic r
reactent i
SultStituen
reaCtions
tblotch F
excbangles

39

mmmmmmmm m r Offiyhstituted
blnmhenmesmthm

 

Deuterated water has been found tO be a powerful reagent for
nging deuterium for the hydrogen atoms Of the [M-1]' ion from O-
robenzene. Successful applications Of the exchange reactions for
'entiating isomers Of chlorinated benzenes will be discussed in detail
.e next chapter. Because the characteristic chemistry Of a specific
)gen depends on the chemical environment Of its parent molecule,
:ituting one Of the two chlorine atoms with other chemical
ionalities may alter the reactivity Of its [M-1]' with D20. If the
ionalities also contain hydrogen atoms, [M-1]' may be formed from
rtonation Of one Of these hydrogens. In this section, the [M-1]' ions Of
)benzene isomers substituted with different chemical functionalities

:plored for H/D exchange reactions with D20.

The isomers selected for this study are listed in Figure 2-5. They
chosen originally for their electron-withdrawing and electron-
ing properties towards the benzene rings. Except for N02, all the
d substituents also contain one or more hydrogens. Because most Of
hydrogens are expected to be more acidic than the hydrogens on the
tic ring, it is conceivable that for most Of the isomers, the [M-1]'
nt ions are formed by direct or indirect deprotonation Of the
tuent instead Of the benzene ring. The results Of H/D exchange
ons between the [M-1]' ions and D20 are shown in Figures 2-6
h Figure 2-12. The product ions derived from successful H/D

ges are represented by peaks having one or more m/z units above the

 

 

 

 

 

 

 

chlgroaniling

H H H ' H H
.9... 9m 019%

H Cl Cl H H H
ghlgrgphgngl

H H H H

H C1 C1 H H H
thgrobenzaldehyde

H H H H
H) H i: H5010 Cldcio

H C1 Cl
ghlorotolueng

H
H H H
V) H‘CHB H@0H3 018cm,

HH

H Cl C1 H H H

thQrQQiSQng

’1 Hdo-CH3 HQO—CH3 ClHoHH3
H C1 Cl H

, chlorobenzoic ag’d

. H9: “992 019::
H ClO C1 H

hlrnitr

Him» 9%:

ure 2-5 Selected isomers for H/D exchange reactions with D20

 

 

 

2
4

6

E

C

0m
HWHHQQHHH 05,594
and .

10

Figure 26

 

 

41

[M-1]' O-chloroaniline
(reactant ion)

100 - \
80 e
60 -
40 -

20-

 

 

 

0 l ' l I ' W H T
124 126 128 130 132

100 - m-chloroaniline
80 -
60 -
40 -

20-

 

 

 

12'4 ' 126 ' 128 ' 1230 I 1532
100 - p-chloroaniline

sot
I 60-
’ 40-

20-

 

 

 

194 ' 126 - 129‘ 130 ' 132
m/z

26 Product mass spectra for the [M-1]' ion of the isomers
Of chloroaniline upon H/D exchange reactions with D20.

 

 

 

thawwflmvaH ®>B®H®MH

Ifigure

 

42

[M- 1] " O-chlorophenol
(reactant 10n)

100- st
80—
60-
4O-

20

 

 

0 , l l
128 130 132 ' 134 '

 

100 - I m-chlorophenol
80 -
60-
40-

20-

 

 

0 l I I I
128 130 132 134

 

100 _ p-chlorophenol

80-
60-
40-

20-

 

 

 

128 130 132 134

m/z

ire 2-7 Product mass spectra for the [M-1]‘ ion of the isomers
of chlorophenol upon H/D exchange reactions With D20.

1

PlEllre 2

NAHHMQQHHHH ®>wu~w~®m

 

 

 

43

[M-1]' O-chlorobenzaldehyde
(reactant ion)

100-
80-
60 -
40 -

20

 

 

 

o . , r, | .
138 140 142 144

'm-chlorobenzaldehyde

 

100 -

80-

60-

40-

20

 

 

 

I

l
138 140 ' 142 ' 144

p-chlorobenzaldehyde
100 -

80—
60-
40-

20

 

 

 

l
138 ' 140 p 142 144

m/z

re 2-8 Product mass spectra for the [M-1]' ion of the isomers Of
chlorobenzaldehyde upon H/D exchange reactions With D20.

18

IL

HHMIVMVHHH ®>Hnw
enamel

10(

8C

68

4C

2C

Figure 2.9

 

44

[M— 1] ' O-chlorotoluene
(reactant ion)

100- \k
80—
60-
40-

20-

 

 

., l,.;;
124 126 128 130

 

100 - m-chlorotoluene
80 -
60 -
4O -

20

 

 

0 ' II I i I
' 124 126 128 130

 

100 - p-chlorotoluene

80-
60-
40-

20~

 

 

130

 

124 126 ' 128
m/z

2-9 Product mass spectra for the [M-1]' ion of the isomers
Of chlorotoluene upon H/D exchange reactrons With D20.

0 00 6 4 2 1

\AQMWHQGHHH mwxrwummmwm

10

rrrrr

 

 

45

[M-l]: O-chloroanisole
(reactant ion)

100 - \
80 -
60 -

40-

 

20- l

 

 

o, . . l l . ;
140 142 144 146

100 - .m-chloroanisole
80 -
60 -
4O -

20

 

 

 

0 l ‘ l I l
140 142 144 146

p-chloroanisole
100 —

80—
60-
40-

20 l

 

 

 

140 142 144 146

m/z

a 2-10 Product mass spectra for the [M-1]‘ ion Of the isomers
of chloroanisole upon H/D exchange reactlons With D20.

10

(0

0
anHMHHQUHHH ®>wu9~ NH

108

80

60

40

20

Fltitre 2~1

 

46

[M-1]' O-chlorobenzoic acid
(reactant ion)

100. st
80 -
60 t
40 -

20-

 

 

of

 

154 156 ' 158 160

100 _ 'm-chlorobenzoic acid
80 -
60 -
40 -

20-

 

 

 

0. l . . .
154 156 158 160

100 1 p—chlorobenzmc acid

80—
60-
40-

20-

 

 

 

154 156 158 '7’ 160

m/z

2-11 Product mass spectra for the [M-1]' ion of the isomers of
chlorobenzoic acid upon H/D exchange reactions with D20.

 

1C

0 8 6 4 4|}-
1

unpfimHHvaHHH mirwnmeQNH

10

Ifigure 2.

 

[MJJ'

(reactant ion)

100-
80-1
60-
4OJ

20-

 

 

47

1-chlorO-2—nitrobenzene

 

154

100 1
80 -
60 -
40 -

20-

 

156

 

158 ' 160 162

' 1-chlorO-3-nitrObenzene

 

Relative Intensity

154

100.
80-
60-
407

20-

 

156

l
158 160 162

1-chlorO-4-nitrobenzene

 

 

154

156

igure 2-12 Product mass spectra for the [M-1]’ ion Of the isomers Of
chloronitrobenzene upon H/D exchange reactions with D20.

 

 

 

m/z of tht
atoms wit
the [M4]
three ison
hydrogen:
reactions
interestin

minimal.

For
are obserr
that whil
chlorophe
deuterium
afld chlorr
demonstrz
addition it
these thr
compared
in reactii
chlOroani
chloTObem
SuhStitute
Chlorobem
deuterim
chlorobem

merenfia

 

 

48

m/z Of the reactant. It is clear that substituting one Of the two chlorine
atoms with the selected substituents substantially reduces the reactivity Of
the [M-1]' ion with D20. Figure 2-6 shows that the product spectra Of all
three isomers Of chloroaniline are essentially identical. Only one Of the five
hydrogens Of [M-1]' from the chloroaniline isomers is exchangeable upon
reactions with D20. Although the results suggest that there may be some
interesting chemistry involved, the analytical value Of the reactions is

minimal.

For the other substituted Chlorobenzenes, some spectral differences
are Observed for the different isomers. In Figure 2-7, the spectra indicate
that while no H/D exchange product is Observed for either p— or m-
chlorOphenOl, O-chlorophenol gives product spectra indicative Of two
deuterium substitutions. For isomers Of chlorobenzaldehyde, chlorotoluene
and chloroanisole, the results presented in Figure 2-8 - Figure 2-10 clearly
demonstrate that hydrogens on the substituents are also exchangeable, in
addition tO the aromatic hydrogens. The reactivities Of the para isomers Of
hese three substituted Chlorobenzenes are consistently the highest

ompared tO those Of the meta and ortho isomers. Although the difference
n reactivity for the ortho and meta isomers Of chlorotoluene and
hloroanisole is small, the reactivity Of the meta isomer Of
hlorobenzaldehyde is very different from that Of ortho isomer. While ions
ubstituted with four deuteriums are major products for the anions Of O-
chlorobenzaldehyde, a small peak representing a maximum Of only three
deuterium substitutions is Observed for the anions Of m-
:hlorobenzaldehyde. Nevertheless, individually these isomers can still be

differentiated on the basis Of their differences in H/D exchange efficiencies

 

 

 

 

with D20
D3-substi
Product 1
observed

An
substitute
former.
indirectc
molecule
chloronitr
the [M-1
substituer
shown in
Predomin
exchange:
found to
maximum
and para
[M'll‘ of d
three H/]
chloronitr
dichlorobr
its [M1].
“Changes

the three

 

 

 

49

with D20. For the isomers Of chlorobenzoic acid, Figure 211 shows that the
D3-substituted products are Observed for both the meta and ortho isomers.

Product ions associated with a maximum Of only two H/D exchanges are

Observed for the para isomers.

An important difference between chloronitrobenzene and the other
substituted Chlorobenzenes is the absence Of non-aromatic hydrogens in the
former. A chloronitrobenzene [M-1]' ion must be formed by direct or
indirect deprotonation Of one Of the three aromatic hydrogen of a neutral
molecule. This means that while the initial reaction site Of a
chloronitrobenzene [M-1]' ion is on the aromatic ring, the equivalent site Of
the [M-1]' Of the other substituted Chlorobenzenes may be on the
substituents. The product spectra for the isomers Of chloronitrobenzene are
shown in Figure 2-12. Although ions having only one H/D exchange are the
predominant products for all the isomers, the Observed maximum H/D
exchanges are different for different isomers. Only the ortho isomer is
found to allow all three hydrogens Of its anion tO be exchanged; a
maximum Of two and one hydrogens exchanged is Observed for the meta
and para isomers, respectively. Compared tO the same reactions with the
[M—1]' Of dichlorobenzene, the relative yields for the formation Of the two and
three H/D exchange products are significantly lower in the case Of
hloronitrobenzene. Clearly, substituting one Of the two chlorines Of a

'chlorobenzene molecule with N02 substantially reduces the reactivity Of
ts [M-1]‘ ion towards the formation Of the products Of higher order H/D

xchanges. Nevertheless, the reactions are still useful for differentiating

he three isomers.

 

 

 

 

 

Conclusi

In
transmit
collision
increasec
optimum
exchange
2.8 eV a1
observed
energies

react will

forH/Dg

 

 

 

nclusions

In the TSQ-7OB triple quadrupole system, the efficiency Of
ansmitting the [M-1]' Of O-dichlorObenzene from the ion source into the
Ilision chamber is a strong function Of collision Offset voltage, but is not

ncreased beyond a collision Offset potential voltage Of around 3.0 V. The
ptimum collision Offset energy and collision pressure setting for the H/D
xchange reactions between the [M-1]' Of O-dichlorobenzene and D20 are at
.8 eV and 4.0 mtorr, respectively. NO H/D exchange reaction products are
bserved for measured collision pressures below 0.3 mtorr or collision
nergies above 5.0 eV. Among the many deuterated reagents selected to

eact with the [M-1]' of O-dichlorobenzene, D20 is the most effective reagent

or H/D exchange.

In the study Of H/D exchange reactions between D20 and the [M-1]' Of
ilorObenzene isomers substituted with NH2, OH, CH0, CH3, OCH3, COOH
1d N02 functionalities, it was found that replacing a C1 Of a
chlorobenzene molecule with one Of the functionalities substantially
:duced the reactivity Of the [M-1]'. While the para isomers Of
lorObenzaldehyde, chlorotoluene and chloroanisole allow all hydrogens Of
.e [M-1]‘ions tO be substituted with deuteriums, nO more than two
:uterium substitutions are Observed for the [M-1]' ions Of the para isomers
chloroaniline, chlorophenol, chlorobenzoic acid and chloronitrobenzene.
or the [M-1]‘ions Of chlorobenzaldehyde, chlorotoluene and chloroanisole,
euterium substitutions clearly occur on hydrogens connecting tO the

nctionalities as well as the aromatic hydrogens. The reactivity Of the [M-

 

11' ions 0
produce
that som

exchangi

 

 

51

1]' ions Of chlorobenzoic acid, chlorOphenOl and chloroaniline are tOO low tO

rOduce useful information for isomeric differentiation. It is very likely

hat some other deuterated reagents may be more suitable than D20 for

xchanging hydrogens Of substituted chlorobenzene [M-1]' ions.

References

LU

Hunt, D. F.; McEwen, C. N.; Upham, R. A., Tetrahedron Lett. 1971,
4539.

Hunt, D. F.; Sethi, S. K., J. Am. Chem. Soc. 1980, 102, 6953.

Freiser, B. S.; WOOdin, R. L.; Beauchamp, J. L., J. Am. Chem. Soc.
1975, 97, 6893.

Chakel, J. A., Ph.D dissertation, Michigan State University, 1982.

Schmit, J. P.; Dawson, P. H.; Beaulieu, N ., Org. Mass Spectrom.
1985, 20, 269

Schmit, J. P.; Beaudet, S.; Brisson, A., Org. Mass Spectrom. 1986, 21,
493.

DePuy, C, H.; Bierbaum, V. M., Acc. Chem. Res. 1981, 14, 146.

 

MECHl
IONS

Inlrodncl

Ma
reactions
studies 0
have dem
Presumal
H39 Ds“.
16,7) first
under ch.
Active hy
1”tactions
reagents.
or MeOD
exchange
allowed 1
Erichange
“Sing ion
his (9).

tenditim

CHAPTER 3

MECHANISTIC STUDY OF H/D EXCHANGE BETWEEN [M-1l'
IONS OF CHLORINATED BENZENES AND DEUTERATED
WATER OR DEUTERATED AIVIMON IA

Introduction

Mass spectrometric studies on hydrogen-deuterium (H/D) exchange
reactions in solution have been used for structure elucidation (1,2). Kinetic
studies on isotopic methane-hydrogen and methane-deuterium systems
have demonstrated the feasibility Of H/D exchanges in the gas phase (3,4,5).
Presumably, the reactant ions involved for the exchange reactions include
H3+, D3+, CH3+, CD3+, CH4+, CD4+, CH5+ and CD5+. Hunt and co-workers
(6,7) first reported H/D exchanges for organic compounds in the gas phase
under chemical ionization (CI) conditions with deuterium oxide reagent.
Active hydrogens in organic compounds were determined by H/D exchange
reactions between neutral organic molecules and ionic deuterated
reagents. Isotopic exchange for amines was Observed when either ND3 (7)
or MeOD (8) was used as the CI reagent. Differences in the number Of
exchanges Observed between ND4+ or MeOD2+ and amine molecules
allowed the differentiation Of primary, secondary and tertiary amines.
Exchange Of aromatic hydrogens was first reported by Beauchamp et a1.
using ion cyclotron resonance (ICR) spectroscOpy for protonated benzene
ions (9). This study suggested that ring protonation was a necessary

condition for exchanging aromatic hydrogens Of a cation. Further studies

52

 

 

 

by Marti:
confirmed
than on t]

Der
compound
conditions
observed
afterglow
EtOD and
afterglow
general 111
an exchar
Significant
Positional
Chotel (15

In 1
exchange
compound
benzenes l

C011firm th.

 

 

 

53

by Martinsen and Buttrill (10) on a variety Of substituted benzenes
confirmed that when protonation occurred on the substituent groups rather

than on the aromatic rings, H/D exchange products were not Observed.

Deuterium exchanges for the [M-1]' ions Of a number Of organic
compounds were first shown by Stewart et al. under flowing afterglow
conditions, using D20 as the reagent (11). Exchange reactions were also
Observed for similar types Of compounds under both CI and flowing
afterglow conditions, using other deuterated reagents such as ND3, MeOD,
EtOD and CF3CD20D (12,13). Based on the results Obtained from a flowing
afterglow instrument, Depuy and Bierbaum (14) have concluded that, as a
general rule, H/D exchange can usually be Observed between an anion and
an exchange reagent which is as much as 20 kcal less acidic. The
significance Of using H/D exchange reactions for the differentiation Of

positional isomers Of aromatic compounds was then demonstrated by

Chakel (15).

In this work, we postulate a reaction mechanism for the H/D
exchange reaction between the hydrogen on the [M-1]' ions Of aromatic
compounds and molecules Of D20 and ND3. Isomers Of chlorinated
benzenes were the model compounds chosen for this study. The results

confirm that the reaction is potentially useful for isomeric differentiation.

 

 

 

Experime

Two
experimer
the experi
mass spec
reagent w
instrumen
instrumm
quadmpOll
Finnigan i:

For
inotrumenr
insertion I
source. Fr
rates were
reservoir a
rate was
“Perimen:
Were intrr
chmmatogj
SE94 Dhas

 

 

 

Experimental

Two types Of tandem quadrupole instruments were used for all

experiments. In the cases where ND3 was used as the deuterated reagent,
the experiments were performed on a Finnigan TSQ-70B triple quadrupole
mass spectrometer. Experiments involving the use of D20 as a deuterated
reagent were later performed on a Finnigan TSQ-700 tandem quadrupole
instrument. The configuration and the components used for both the
instruments are essentially identical, with the exception that the second
quadrupole in the Finnigan TSQ-70B instrument is an octopole device in the

Finnigan model TSQ-700.

For the experiments performed on the Finnigan TSQ-70B
instrument, samples were introduced into the ion source by either a direct
insertion probe or from an external glass reservoir connected tO the ion
source. For the samples in the external glass reservoir, the sample flow
rates were controlled by two adjustable leak valves installed between the
reservoir and the ion source. As far as possible, a steady-state sample flow
rate was maintained during each individual experiment. For the
experiments performed on the Finnigan TSQ-700 instrument, samples
were introduced into the ion source through the Varian 3400 gas
chromatograph (GC). A capillary column was used (30 m, 0.25 mm id,
SE-54 phase with 0.25 um film thickness) for component separation. The
injection port Of the G0 was set at 250 °C under a splitless injection mode.
A transfer line set at 275 °C directed the column efluent into the ion source.
Helium was used as the carrier gas at a pressure Of 15 psi. The

temperature Of the GC was programmed from 80 °C to 200 °C at 10 °C/min.

 

 

 

 

 

The GC to
emerging
isomers o:

the samplr

The
source at
extracted 1
of the ta
quadrupoll
were lfltl‘Ol
ion produc
(according
thIotlghoul
source was
were majm

indicated 0

 

 

55

The GC temperature program was set such that all the selected compounds
emerging from the GC column were at least base-line separated. The
isomers Of O-dichlorobenzene were injected into the GC as mixtures. All

the samples were dissolved in hexane to a concentration Of about 10 ppm.

The [M-1]‘ ions Of the test compounds were generated by CI in the ion
source at 150 °C using NH3 as a reagent. In general, the reactant ions
extracted from the source were mass-selected by the first quadrupole Of one
Of the tandem quadrupole instruments for reaction in the second
quadrupole (Q2) or the octopole device (02) where the deuterated reagents
were introduced. The last quadrupole was set to scan for the spectra Of the
ion products Of the reactions in Q2 or 02. A collision energy at 2.8 eV
(according tO the Q2 Of 02 Offset voltage reading Of the instrument) was used
throughout all the experiments. The pressure Of the reagent gas in the ion
source was maintained at 1.5 torr. The collision pressures for ND3 and D20
were maintained at approximately 7 and 4 mtorr, respectively, except when
'ndicated otherwise. All the experiments were performed twice to test for

onsistency in the results. The replicates were compared for discrepancies
'n terms Of relative peak ratios and the absence or presence Of stray peaks.
uplicate runs consistently gave Spectra that were approximately the

ame. The data shown come from one set Of results rather than the

verage Of the replicates.

All chlorinated isomers used in the experiments were purchased
rom Ultra Scientific, North Kingston, RI. All other halogenated
ompounds were purchased from Aldrich Chemical Company, Inc.,

'lwaukee, WI. Deuterated ammonia (99.5 atom %D) was Obtained from

 

 

 

MSD Isot
purchased
chemicals

tMJcha

Low
dichlorobe
exchange
observed f
abundanc.
dichlorobe
selected n
Stlhstitutio
reagent is
leaCtant io
Peaks at 11
ions With E

by Peaks 3

 

 

56

MSD Isotopes, Montreal, Canada. Deuterated water (99+% grade) was
purchased from Cambridge Isotopes Laboratory, Woburn, MA. All the

chemicals were used without further purification.

Results and Discussion

A. Meghani stig Analysis

Low-energy collision between [M-1]‘ ions Of all the isomers Of
dichlorobenzene and deuterated water or deuterated ammonia produce H/D
exchange products. For each isomer, the same exchange products are
Observed for either D20 or ND3 reagent, though not in the same relative
abundance. The product mass spectra Of the three isomers Of
dichlorobenzene are shown in Figure 3-1 for both D20 and ND3. The
selected reactant ions at m/z 145 contain only the 35Cl isotope. Each
substitution Of a deuterium from an exchange reaction with the deuterated
reagent is indicated by the gain Of one mass unit from the mass Of the
reactant ions. Sequential H/D exchanges Of the same ions are indicated by
peaks at more than one m/z unit above the reactant ion mass, thus product
ions with all three hydrogens substituted with deuteriums are represented

y peaks at m/z 148. Although these isomers are known to produce nearly
' distinguishable EI mass spectra, the pattern Of H/D exchange products Of
he [M-1]' ion Of each isomer is quite distinctive. For the anions Of these
'somers, only O-dichlorobenzene gives a predominant peak at m/z 148 which
orresponds to ion products with all the three hydrogens exchanged with

euteriums; the predominant products for m-dichlorobenzene and p-

 

 

 

57

 

 

H1
React
100-
80'
so
40'
20'
it...
143
Figure 3-1 Product spectra Obtained from the H/D exchange reactions %
between the [M-1]' ions Of the individual isomers of 3: m
dichlorobenzene and deuterated water and deuterated g
ammonia. The reactant ions are at m/z 145. Peaks 313 higher s so
mass units are due to one or more H/D exchange reactions. *5
1-l 60-
0
> 40-
.5,
Pg 20-
o
0.1 m
143 3
no
so
co
4e
2e
0t

 

M
143 14

Relative Intensity

 

 

 

 

 

 

58
H/D Exchanges with D20 HID Exchanges with ND3
Q—dighlgrgbenzene
H/D exchange products
Reactant ion ,_L__‘
10 100
a so
6 so
4 4o
2 2o
9 l 1 | .
143 144 145 146 147 14's 149 150 143 144 145 146 147 14's 149 150
m-dichlorobgnzene
10 10
8 a
6 6
4 4
2 2
“143 144 145 14's 147 14's 149 150 "143 144 145 146 147 14's 194' 150
p-dichlorobenzene
10 100
8 so
5 so
4 4o
2 20
"143 144 145 14's 147 14's 149 1'50 "143 144 145 14's 147 14's 149 150

m/z

 

 

 

dichlorobr
respective
indicate tl
ring affec

exchange

Sinc
are essenl
believe th
essentially
Scheme I
mechanism
between th
formation 1
molecule or
hydrogen);
hydrogen 2
ctilllplerr. A
the inter-mg
removes the
one carbon
Fuother H/]
uotil no mo
With tthee l
which, acco

react‘mt ior

 

59

dichlorobenzene are ions substituted with two and one deuteriums,
respectively. The dramatic differences in the product patterns Observed
indicate that the relative positions Of the chlorine atoms on the aromatic
ring affects the ability Of certain ring hydrogens to undergo the H/D

exchange reaction.

Since the reaction products Observed with both D20 and ND3 reagents
are essentially the same (though in differing relative abundances), we
believe that the basic reaction mechanisms for the two species are
essentially the same. We postulate the reaction mechanism shown in
Scheme I for ortho dichlorobenzene [M-1]' ion and D20. In this
mechanism, a five-membered-ring reaction intermediate is formed
between the [M-1]' ion and the reagent molecule (either D20 or ND3). The
formation Of this intermediate is initiated by the attack Of the deuterated
molecule on the charge site (localized at the carbon atom with the missing

ydrogen). An electron pair on the neutral reagent molecule reacts with a
ydrogen atom adjacent tO the charge site tO form the intermediate
omplex. A successful H/D exchange occurs only if the decomplexation Of
be intermediate complex leaves a D atom at the initial charge site and
emoves the adjacent H atom. The charge site in the product ion is shifted
ne carbon atom around the ring from its position in the reactant ion.
urther H/D exchange reactions can continue by the same mechanism

til nO more adjacent aromatic hydrogens remain. Products substituted
°th three deuteriums are the result Of three successful H/D exchanges
hich, according to the proposed mechanism, requires that the initial

eactant ion contain three aromatic hydrogens adjacent or sequentially

djacent tO the initial charge site.

 

 

Cl

Cl

——O-

 

.. —r-. .

 

 

TI.
9
m
e

h
C

S

 

Thr
initial re:
positions
charge sit
The chlor
anions all
intermedi
three difil
aromatic l
of two arc
the charg.
atoms is s.
the five-m
deuterium
Predomina
fumed wit
the locati
hydrogens
interniedia
reactant 1);
those of ti
rtatthllt io:
onions, the
hydr(teens
that the Chi
other We 1]

iii] tonic c,

 

 

 

61

The number Of adjacent (or sequentially adjacent) hydrogens in the
initial reactant ion depends on the location Of the charge and the relative
positions Of the chlorine atoms on the aromatic ring. All the possible
charge sites for the three dichlorobenzene isomers are listed in Figure 3-2.
The chlorine positions and the charge locations of the ortho-substituted
anions allow sequential deuterium substitution, via the five-membered-ring
intermediate, Of all three hydrogens. For the meta-substituted anions,
three different charge sites are possible. Both structure III and IV contain
aromatic hydrogens adjacent tO the charge site. These sites allow exchange
Of two aromatic hydrogens which are adjacent or sequentially adjacent tO
the charge sites. The third hydrogen located between the two chlorine
atoms is separated from the charge sites and can not be exchanged through
the five-membered—ring intermediate. Reactant ions substituted with two
deuteriums are shown in the product spectra Of m-dichlorobenzene as the
predominant products. Structure V shows a meta-substituted anion
formed with a charge site sandwiched between two chlorine atoms. Since
the location of this charge site is remote from the three aromatic
hydrogens, deuterium substitution through the five-membered-ring
intermediate is not possible for this reactant ion. The relatively larger
reactant peak in the product spectrum Of m-dichlorobenzene compared tO
those Of the other dichlorobenzene isomers suggests that some Of the
reactant ions have the Structure V configuration. For the para-substituted
anions, there is only one possible charge site because all four aromatic
hydrogens in the original molecule are equivalent. Structure VI shows
that the charge site is adjacent tO only one hydrogen and separated from the
other two hydrogens by a chlorine atom. By the proposed mechanism, such

an ionic configuration allows only one hydrogen Of the anion to be

 

 

 

Figure 3-2 The possible charge sites for the [M-1]‘ ions Of dichlorobenzene

isomers and the consequent location and number Of

exchangeable hydrogens according tO the proposed reaction
mechanism.

0 Sim

qu—g.

 

63

OrthO-substituted anions

Cl Cl
H C1 H I Cl
H H H "
- H
I II
0 Both structure I & II allow all three hydrogens tO be

exchanged

Meta-substituted anions

Cl 9 Cl Cl
H H* H H* H* _
H Cl _ Cl H* Cl
.. H H*
III IV V

* Non-exchangeable hydrogens

0 Structure III and IV will exchange up tO 2 H's

0 Structure V will not react with D20, but may be isomerized
tO become H inside the ion source.

Para-substituted anion

 

 

Cl
*
H ‘ * N on~exchangeable
* hydrogens
H H
Cl
VI

' Structure VI allows only one hydrogen tO be exchanged

 

exchange

dichlorobe

0V4
mechanis:
than the 1
mwdd
of a D3-31
dichlorobe
dichlorobe
any D3-sr
wrong, or
which is o
t119133-sul
can breach
remaining
Of m-dichl
lilielihood
about the
substantia
dicthrObel
explanatio]
alliOIIS of
dichloroben
m'djchloroh

 

 

64

exchanged. The predominant ions in the product spectrum of p-

dichlorobenzene are substituted with only one deuterium.

Overall, the observed results correlate well with the proposed
mechanism. The predominant deuterium exchange observed is one less
than the maximum number of hydrogen atoms between the chlorine atoms

in each dichlorobenzene isomer. However, a peak representing about 12%

of a D3-substituted product is found in the product spectrum of m-
dichlorobenzene, in reactions with the D20 reagent. For the [M-1]' of m-
dichlorobenzene, the proposed mechanism would not allow the formation of
any D3-substituted product. Thus, either the proposed mechanism is
wrong, or there is an additional process, Specific to the m-dichlorobenzene,
which is occurring. Two possible additional processes that could result in
the D3-substituted product are : a different substitution mechanism that
can breach the intervening Cl atom located between the charge site and the
emaining aromatic hydrogen, and a possible isomerization of structure V
f m-dichlorobenzene to structure II of o-dichlorobenzene. Since the
ikelihood of the success of a Cl-breaching mechanism is expected to be
bout the same for both the anions of m- and p-dichlorobenzene, the
ubstantially smaller D3-substituted product peak observed for the p-
'chlorobenzene favors the latter explanation. If the isomerization
xplanation is true, it appears that the degree of isomerization for the
nions of p-dichlorobenzene is substantially lower than that of m-
'chlorobenzene. Of all the anionic structures, the structure V anions of

-dichlorobenzene may be the most susceptible to isomerization.

 

3.99m

Th4
resident i
pressure:
deuterate
sequentia
the yield c
dichlorobe
minimum
deuterater
ions may;
required 11
with D20,
seQuentin]
highest de
With lowe
SUbstitnte.
Pressures.
replace the
high collis:
substituted

F or
dependence

 

 

65

B, Mlliin - Dnn-o'o,‘ - on Prn . of I Ex

The number of collisions of an anion with a deuterated reagent while
resident in Q2 is affected by the reagent gas pressure in Q2. A higher gas
pressure allows a reactant ion to have a higher collision frequency with the
deuterated reagent molecules. For ion products derived from a step-wise
sequential process, increasing the total number of collisions can enhance
the yield of the more highly substituted ion products. For the [M-1]' ion of o-
dichlorobenzene, the formation of D3-substituted products requires a
minimum of three reactive collisions of the initial reactant ion with
deuterated molecules. At low collision pressures, D3-substituted product
ions may not be observed because the reactant ions may leave Q2 before the
required number of ion-molecule collisions can occur. For H/D exchanges
With D20, the collision pressure study shown in Figure 3-3 supports the
sequential process depicted in our proposed mechanism. Products with the

'ghest degree of deuterium substitution are not observed until products
ith lower degrees of deuterium substitution are formed. The D1-
ubstituted ions are the predominant products at very low collision
ressures. As the collision pressure increases, higher substituted ions
eplace the less substituted ions as predominant products. At a sufficiently
'gh collision pressure, most the anions of o-dichlorobenzene are fully
ubstituted before they leave Q2. These data support a mechanism that

volves the substitution of only one deuterium per reactive collision.

For the reactant ions from m-dichlorobenzene, the pressure-
ependence plot shown in Figure 3-3 indicates that following an increase in

ollision pressure, the relative intensity of the reactant ions decreases; but

 

 

 

1001
90'.
80:
70:
5°“.

402'
30:
20:
10'.

 

C)
CA

more
90:
so-
70:
so:
230-:
40:
so-
20:
10:
06

tensity

1n

‘ ° ollision
' t 1ntens1ty vs c - .
' (1 relative produc M-l] lOIlS
Figure 3-3 PIOtS for fblfrII-IVIDIIZ:change reactions beltvglelie‘?S the [
Ififeifiscllilfirobenzene isomers and D20 mo e

 

more
90:
so:
70:
50:
so:
403
so:
20:
10:
oh

omalized relatlve
n

 

normalized relative intensity

 

100
80 "
70 "
60 "*

40':

 

100
90]

 

 

Exchanges with deuterated water

‘ -ihlr nzne

‘ a --A-— reactant ion
—0— 1 H/D exchange
--I— 2 H/D exchanges
° -—-El—- 3 H/D exchanges

 

° 634° I I I I . I I j I f I 1 I"l_""'_'|
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
M m- i hl r nz n
. -—t—- reactant ion
. —O—- 1 H/D exchange

--I—' 2 W0 exchanges
———a— 3 H/D exchanges

 

. ' o a
494/ . . 3 ° ; O
a a a a a a a ——-° . , . . . , .
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
N . p-dichlorobenzene
A . O .
‘ O
O
‘ -—A— reactant ion
° —O-— 1 H/D exchange
——I— 2 HID exchanges

3 HID exchanges

  

 

 

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

collision pressure (mtorr)

 

 

the decre:
Thereafter
total ion :
result sug
reactant i
pressure or
proposed 11
which the r
contributio:
about 20% ;
of all mete
anions iden
likely occur.
aslow isom

also POssiblr

The 1
indicates th
fOfmation of
the D1-subs1
subsfituted w

reactions b‘
relatively Sm.
dichlombenz.
likely to 11nd
At the collisi

68

the decrease of the reactant ions levels off beginning at about 3.5 mtorr.
Thereafter, the contribution of the reactant ions remains at about 25% of the
total ion intensity, despite further increases in collision pressure. This
result suggests that in the case of m—dichlorobenzene, about 25% of all
reactant ions are unreactive with D20 molecules under any collision
pressure conditions. This lack of reactivity can be accounted for within our
proposed mechanism, if about 25% of the reactant ions have structure V, in
which the charge site is not adjacent to any aromatic hydrogens. Since the
contribution of the D3-substituted products in the m-dichlorobenzene plot is
about 20% at the collision pressure of 5.0 mtorr, we believe that at least 20%
of all meta-substituted reactant ions undergo isomerization to reactant
anions identical to those from o-dichlorobenzene. This isomerization most
likely occurs in the source, possibly as part of the ionization process, though
a slow isomerization of the ion, continuing thrOugh the ion lifetime in Q2 is

also possible.

The pressure—dependence plot for p-dichlorobenzene (Figure 3-3)
ndicates that an increase in collision pressure enhances primarily the
ormation of the D 1-substituted product. Even at a high collision pressure,
he D1-substituted product remains predominant. This means that ions
ubstituted with only one deuterium are generally the terminal products for
eactions between D20 and the p-dichlorobenzene [M-1]' ions. The
elatively small contribution of the D2- and D3-substituted products in the p-
ichlorobenzene plot suggests that the p-dichlorobenzene anions are less
kely to undergo isomerization than are the anions of m-dichlorobenzene.

t the collision pressure of 4.5 mtorr, over 90% of the p-dichlorobenzene

 

anions a

products.

Col
ND3 reag
involving
formation
for both I
reagents
dichlorobe
H/D excha
ND3 thanl
the H/D e:
than that t
Product for

Similar excl

Althr
between djc
does not 0‘
exchange In
3-3 & 3.4 h
reactant ior
Proposed 11,,
the Percent

97888139 as

pmdlmt, A

 

 

 

anions are converted into the expected predominant D1-substituted

products.

Collision pressure studies were also performed on reactions with

ND3 reagent. The pressure-dependence plots for H/D exchange reactions
involving ND3 reagent are shown in Figure 3-4. The trends for the
formation of different H/D exchange products for each isomer are the same
for both D20 and ND3 reagents. The major difference between the two
reagents are their relative reactivities towards the anions of the
dichlorobenzene isomers for H/D exchange reactions. To achieve a specific
H/D exchange product yield, the required collision pressure is higher for
ND3 than for D20. A comparison between Figures 3-3 and 3-4 suggests that
the H/D exchange efficiency of D20 molecules is about three times better
than that of ND3 molecules. The similarity in the trends of H/D exchange
product formation for the two deuterated reagents supports the notion that

similar exchange mechanism are followed by both reagents.

Although H/D exchange products are observed upon collision
etween dichlorobenzene anions and D20 or ND3 reagent, reaction probably
oes not occur on every single collision. Based on our proposed H/D
xchange mechanism, we have generated plots similar to those of Figures
-3 & 3-4 by computer simulation of H/D exchange reactions initiated by
eactant ions of all the structures shown in Figure 3-2 and following our
roposed mechanism. In these simulations, the number of collisions times
e percent of collisions that are reactive are used instead of the collision
rressure as the parameter to calculate the relative intensities for each ion

rroduct. A simulated collision-dependence plot for o-dichlorobenzene

 

 

 

70

Figure 3-4 Plots for normalized relative product intensity vs collision
pressure for H/D exchange reactions between the [M-1]' 1011S
of dichlorobenzene isomers and ND3 molecules.

 

normalized relative intensity

100:

8

o 8 8 8 8 8 8 8‘ 8
.0 _l_n_l_‘_Ln_J_‘__1_._L_._L_._L__L_1_.._

100j

 

o
94.

100 'l

 

01E

71
Exchanges with deuterated ammonia

100 u . ‘ g-dighlgrgbenzene
90 ‘
80 . ——A— reactant ion
‘ —-O-— 1 H/D exchange
70 ‘ —-I— 2 HID exchanges
60 ‘ . -—El— 3 H/D exchanges

 

 

O O
A ‘ .
. ,
9.0
90 .4 . ‘ —t— reactant ion
80 .: ‘ —O— 1 H/D exchange
: ‘ ‘ —-I-- 2 HID exchanges
, 70 ‘ . —a— 3 H/D exchanges
60 i. ‘

 

 

normalized relative intensity

00 10 20 30 40 5.0 60 70 80 90
100 . “ e . , p-dighlgrgbenzene

90 " ‘ ‘ o
80 " . o

" ‘ O O
70 j ‘ '
60 - ‘ , 0 —-A— reactant ion

‘ -—O— 1 H0 exchange
50 " é

. p —I— 2 HID exchanges
40 I . . 1* —-EI-— 3 HID exchanges
30 d O a ‘

. . A
20 " ‘ ‘

II . ‘
10 ' o

 

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0

collision pressure (mtorr)

 

anions h
reaction
structure
either a

assume 1
exchange
1 shown

is the sar
abundant
the numl
no new r.
is also as
for all in
Figure 3.
(top) reve
PrOduct-r
Continue
eQUiValer
0f the dg
rtraction

number c

average"

Fig
I anions
reactiVe I

 

72

anions having the structure II configuration is shown in Figure 3-5. The
reaction pathway used for the simulation is shown in Figure 3-6 for the
structure II reactant ions. In this study, a reactive collision is defined as
either a successful H/D exchange or D/D exchange. In our calculations, we
assume that there is an equal chance for an ion to undergo either a H/D
exchange or a D/D exchange where both are possible. For example, for ion
1 shown in Figure 3-6, the chance of its conversion into either ion 1' or ion 2
is the same. The collision-dependence plot of Figure 3-5 shows the relative
abundance of the reactant and various product ion masses as a function of
the number of reactive collisions. In making this plot, it was assumed that
no new reactant was added, nor products removed over the time shown. It
is also assumed that the percent of collisions that are reactive is the same
for all ion structures. A comparison of the collision dependence plot of
Figure 3-5 with the experimental pressure-dependence plot of Figure 3-3
(top) reveals that the zero input and output flux assumption made in the
product-ratio calculations is not valid under our experimental conditions.
Continued influx of reactant ions extends its high abundance into the
equivalent of higher collision numbers and reduces the relative abundance
of the deuterium-exchanged product ions. Departure of ions from the
reaction chamber before they have actually undergone the "average"
number of collisions yields finite values for their abundance seen at high

"average" collision numbers.

Figure 3-7 shows a simulated collision-dependence plot for structure

I anions of o-dichlorobenzene. These reactant ions require at least four

reactive collisions to form the D3-substituted products. The reaction

pathway used for calculating the relative product intensity of the structure

 

 

 

73

Figure 3-5 A simulated collision—dependence plot that represents a
composite of individual collision-product plots for different
percentages of reactive collision. The simulation is based on
the proposed H/D exchange reaction mechanism between the
structure II of o-dichlorobenzene and D20 or ND3 molecules.

1.0-48

 

 

74

30658 33.08% we xx; x Acommaoo mo .05 mo #3603

menworm—vw mwNF :owm m h m m v m N _. o

I I I 1 Q 4 £

 

mmwcwzoxo em m IIGII
momcwzoxo em m lllll
mmcmnoxo g H l.|Ol

c2 ”:83me fl

 

comma onmunoneofiowwé E 8305me
. o;

Kitsueiur eArqepeJ

 

 

75

Figure 3-6 The reaction pathway for the structure II anions of o-
dichlorobenzene upon reactive collisions with D20 or ND3
molecules.

 

76

Q
I m
8 m
l 5
.H a
I E
Q m r m
‘I
5 m 5 m
#0 HO

 

Figure 3-7 A simulated collision-dependence plot that represents a
composite of individual collision-product plots for different
percentages of reactive collision. The simulation is based on
the proposed H/D exchange reaction mechanism between the
structure I anions of o-dichlorobenzene and D20 or ND3

molecules.

 

-. Duh! .n—

78

Eowmwzoo 9538 as wsv x Aaowmfioo mo .05 mo “2693

mw \L. or m_. VF or NF : or

 

O O C . rOININININ'u‘M‘“ 4 4 a
momcwaoxm OE m a
momcmgoxo em N I
omcwzoxo DE H O .
com ”.5333 IT a

 

 

momma 0533033066 H 23025 3

 

Kipsuenur extremal

 

 

 

I anion
structu
difi‘erer
the coil
and H '
pressur
betwee
differer

clear pi

i
the thr
contribr
V. The
Structu
match
ions wi
50/50 r;
reactim
to ison
Sinlular
illitiate
structu
shown;
trends
in the r

fufilter

 

 

79

I anions is shown in Figure 3-8. Comparisons between plots obtained from
structure I reactant ions and those with structure II show marked
differences in the relative abundances of product ions at many points along
the collision number axis. Ideally, one could choose between structures I
and II based on the comparison of the collision-dependence plots with the
pressure-dependence plot. Unfortunately, the differences in conditions
between the experimental and simulated data are greater than the
differences between the structure I and II simulations thus precluding a

clear preference for either one.

For the anions of m-dichlorobenzene, our simulations assume that of
the three meta-substituted ionic structures shown in Figure 3—2, the
contribution of structure III is insignificant compared to structures IV and
V. The reaction pathways for the formation of various products for both
structure 111 and structure VI anions are shown in Figure 3-9. In order to
match the experimental results, we further assume that the number of
ions with structures IV and V are initially formed in the ion source in a
50/50 ratio. Before the anions enter the collision chamber for H/D exchange
reactions, 50% of the initially formed structure V anions are also assumed
to isomerize to become structure II anions. As such, the collisional
simulations of the anions of m-dichlorobenzene are based on collisions
initiated by a mixture of anions with three different structures (50%
structure IV, 25% structure V and 25% structure II). The results are
shown in Figure 3-10 as a collision-dependence plot. The product formation
trends represented in this plot compare well with the similar trends shown
in the experimental pressure-dependence plot of Figure 3-3 (middle). This

further supports the isomerization explanation and the existence of the

 

 

 

Figure 3-8 The reaction pathway for the structure. I anions of o-
dichlorobenzene upon reactive collision With D20 or ND3

molecules.

 

81

 

 

 

 

Figure 3-9 The reaction pathway for the structure III (top) and structure
IV (bottom) anions of o-dichlorobenzene upon reactive colhsmns
with D20 or ND3 molecules.

 

 

1
Cl Cl — C1 2
H H H H _ H
_> —>
H C1 — C1 D C1
D D
l D 1.
C1
H H
D C1
1'
C1 Cl 1 Cl _
H H H H H H
__> -—>
- Cl D C] V - Cl
H - D
C1 2
_ H
D Cl

 

' 3-10 A simulated collision-dependence plot that represents a
Flgure composite of individual collision-product plots for diffegent
percentages of reactive collision. The Slmulation 18 base on
the proposed H/D exchange reaction mechanism between a
mixture of different m-dichlorobenzene amons and D20 or
ND3 molecules. The mixture of anions is assumed to contain

50% structure IV, 25% structure V and 25% structure H.

Mixture of m-dichlorobenzene anions

m: t.

Amowmwzoo 95.38 mo oev x $8658 mo .95 mo 85on

or mr VF mw

I. F _

NF :owmmnomvmm F o
—. L

h . u .

 

 

. u . _L .L L . . L n . .lel. > .
llooo_oooooo _ _ _ as oo
o o o o a a .
0v... a u we
a u a O
n u a u 0

a a a n n u n a ._ o Nd

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
ad
v.0
md
0.0
momcmfixoemm tial No

mmmcmaoxogm Ill.

wmcwfixoga 0 mo
53:33.3 ldrll ad

. oé

quEm odmNQoaeoEoEAn mo 8352

Kipsuequr earners;

 

 

unreacti‘
individur

dichlorob

Th.
relatively
formatior
collision-1
reactant 1'
the form:
comparab
Plot of F
abundanc
different i
the gener
Obtained 1
data .

can

We
proPosed 1
anion to r
hl'drogen
derived f"
tetrachlm

 

86

unreactive structure for the m-dichlorobenzene anions. (Simulated
individual collision-product plots for the different ionic structures of

dichlorobenzenes are shown in the Appendix.)

The collision simulations of the anions of p-dichlorobenzene are
relatively simple because of the simplicity of the reaction pathway for the
formation of the expected predominant products. Figure 3-11 shows the
collision-dependence plot for the anions of p-dichlorobenzene, based on
reactant ions with structure VI configuration. In this case, the trends for
the formation of the predicted product shown in the simulated plot are
comparable to those represented in the experimental pressure-dependence
plot of Figure 3-3 (bottom). Overall, although the relative product
abundances represented in all the simulated collision-dependence plots are
different from those shown in the experimental pressure-dependence plots,
the general trends for the formation of various H/D exchange products
obtained from the simulated results are consistent with the experimental

data .

C. Beagtigns with Other Aromatic Isomers

We have also used isomers of other chlorinated benzenes to test the
proposed mechanism. According to the mechanism, in order for a [M-1]'
anion to undergo an H/D exchange, there must be at least one aromatic
hydrogen adjacent to the charge site on the anion. For the [M-1]' ions
derived from 1,3,5-trichlorobenzene, 1,2,4,5-tetrachlorobenzene and 1,2,3,5-

tetrachlorobenzene, none of the possible charge sites have any adjacent

 

 

Figure 3-11 A simulated collision-dependence plot that represents a
composite of individual collision-product plots for different
percentages of reactive collision. The simulation is based on
the proposed H/D exchange reaction mechanism between p-
dichlorobenzene anions (structure VI) and D20 or ND3
molecules.

Structure VI p~dichlorobenzene anion

30658 96.38 mo ©ch x 30658 mo 65 mo 35on

        
 
 

or hr wrmr I, or NF 2. owm w h o m v

     
  

mwcwfloxo GE H lOl

”EN—owes Ildl

qusuequr eApcrepel

 

 

aron
reac‘
1,2,
tetra
pred
to tl
shoe
neul
ch10:
1,2,2
the

with
mec
1,2,1

desc

$3

aromatic hydrogens and no products resulting from the H/D exchange
reactions are observed. The product spectra derived from the reaction of
1,2,3-trichlorobenzene, 1,2,4—trichlorobenzene and 1,2,3,4-
tetrachlorobenzene anions with D20, shown in Figure 3-12, exhibit
predominant exchanges of 2, 1 and 1 deuteriums, respectively. According
to the proposed mechanism, reactant ions substituted with two deuteriums
should be the predominant products for 1,2,3-trichlorobenzene because its
neutral configuration contains three sequential hydrogens between the
chlorine atoms. The original molecules of both 1,2,4-trichlorobenzene and
1,2,3,4—tetrachlorobenzene contain only two sequential hydrogens between
the chlorine atoms, the predominant products should be ions substituted
with only one deuterium. All the results are consistent with the proposed
mechanism. The small D2-substituted peak in the product spectrum of
1,2,4-trichlorobenzene could be due to isomerization of the reactant ions as
described earlier. Similar results are also observed for the same reactions
with ND3 reagent. The dramatic differences in the product patterns for
most of the isomers studied suggest that these reactions are useful for
‘ differentiating isomers of chlorinated benzenes. The results of the
i reactions are summarized in Table 3-1. Since the number of chlorines can
I be determined from the masses of different [M-1]' ions, our studies clearly
demonstrate that tandem MS along with H/D exchange reactions using
ND3 or D20 as a reagent can positively identify all the chlorinated benzenes
except two of the tetrachlorinated isomers. Product spectra similar to those
of the dichlorobenzene isomers are also obtained for the isomers of
difluorobenzene and bromochlorobenzene. This strongly suggests that the

same mechanism is operating on all halogenated benzenes.

 

 

 

 

Figure 3-12 Product spectra for the tetra and trichlorobenzene isomers that
produce H/D exchange ion products. The H/D exchange
products are due to reactions between the [M-1]' ions of the
isomers indicated and D20 molecules. No H/D exchange

products are observed for the other tetra and trichlorobenzene
isomers.

relative intensity

91

H/D exchange products
r————J————1
100- [M-1]' reactant ion
80-
Cl
Cl
60H
Cl
40— H

2m

177 178 179 180 181 182 183 184

100‘

 

 

 

80% C,
H Cl
60-
H H
Cl
40.
2e
0 mfi, .l.

 

100'
80‘
Cl
601 H Cl
H C]
40‘ Cl
20-

 

177 178 179 180 181 182 183 184

 

r1]

 

 

 

211 212 213 214 215 216 217 218

m/z

 

 

Table 3-1. Maximum number of H/D exchanges for the [ M-1]' ions of
isomers of chlorinated benzenes (in cases of no isomerization).

 

type of C1 position of Cl max. # of D
substitution on an aromatic ring substitution
mono (no isomers)
di 1,2 3
1,3 2
1,4 1
tri 1,2,3 2
1,2,4 1
1,3,5 0
tetra 1,2,3,4 1
1,2,3,5 O
1 2 4 5 0

penta (no isomers)

 

93

We have begun studies on larger chlorinated aromatic compounds.
For a molecule made up of fused benzene rings, the initial negative charge
site of an [M-1]' ion may be on a different ring than the exchanging
aromatic hydrogens. For example, sequential H/D exchanges for aromatic
hydrogens on different rings of a chlorinated naphthalene molecule require
the formation of a six—membered-ring intermediate for relocation of the
charge site from one ring to another assuming a similar mechanism is
also operating. The product spectrum of 1-chloronaphthalene shown in
Figure 3-13 indicates that reactant ions with all five hydrogens substituted
with deuteriums are the predominant products . For the [M-1]‘ anion of 1-
chloronaphthalene, substituting more than three hydrogens requires a
mechanism of relocating the charge site from one ring to the other. In this
case, the mechanism should involve the formation of a six-membered—ring
intermediate, as shown in Scheme 11. It is quite possible that the same
mechanism may also apply to chlorinated compounds with more than two

adjacent rings.

Conclusions

Using ND3 or D20 as a reagent, H/D exchange reactions are found to
occur between the reagents and the [M—1]' ions of halogenated benzenes
‘ originating from molecules with two or more sequential aromatic
hydrogens. The reactivity of D20 is about three times that of N D3 in the

exchange reactions. A mechanism involving the formation of a five-

membered-ring intermediate is consistent with the results derived from the

l
I'—_

 

 

Figure 3-13 A product spectrum for H/D exchange reactions between the
[M-1]' ions of 1-chloronaphthalene and D20 molecules.
Reactant ions having five hydrogens substituted with
deuteriums are the predominant products.

 

relative intensity

 

H/D exchange products

 

H c1
H
H @H
H
Reactant Ions
I I I L IL
160 162 164 166

m/z

 

 

 

Scheme II

 

 

 

97

isomers of all the chlorinated benzenes in this study. The results obtained
from the isomers of florinated and brominated benzenes strongly suggest
that the same mechanism is also applicable to all halogenated benzenes.
The number of deuterium substitutions for the predominant H/D exchange
products is one less than the number of sequential aromatic hydrogens in
the original molecules of chlorobenzene. It is possible that a mechanism
involving a six-membered-ring intermediate is used for exchanging

hydrogens across two or more fused benzene rings.

Computer simulations based on the proposed mechanism are also
consistent with the product formation trends derived from the experimental
results. The anions of m-dichlorobenzene are more susceptible to
isomerization than the anions of o- and p-dichlorobenzene. A comparison
among the experimental pressure—dependence plots suggests the existence
of an unreactive structure for the anions of m-dichlorobenzene Which is
predicted by our proposed mechanism. For most chlorinated benzenes, we
have demonstrated the potential of using H/D exchange reactions for

l isomeric differentiation.
l
l

References

1. Biemann, K, "Mass Spectrometry", McGraW-Hill, New York, 1962,
Chapter 5.

2. Budzikiewcz, H.; Djerassi, C.; Willians, D. H. "Structure Elucidation
of Natural Products by Mass Spectrometry", Vol 1, Holden-Day, San
Francisco, CA, 1964, Chapter 2.

3. Harrison, A. G.; Heyer, B. G., Can. J. Chem. 1973, 51, 1265.

4. Inoue, M.; Wexler, S., J. Am. Chem. Soc. 1969, 91, 5730.

 

 

 

10.

11.

12.

13.
14.
15.

98

Bowers, M. T.; Elleman, D. D., J. Am. Chem. Soc. 1970, 92, 1847.
Hunt, D. F.; McEwen, C. N.; Upham, R. A., Anal. Chem. 1972, 44,
1292.

Hunt, D. F.; McEwen, C. N.; Upham, R. A.,Tetrahedron Lett. 1971,
4539.

Blum, W.; Schlumpf, E.; Liehr J. G.; Richter, W. J., Tetrahedron
Lett. 1976, 565.

Frieser, B. S.; Woodin, R. L.; Beauchamp, J. L., J. Am. Chem. Soc.
1975, 97, 6893.

Martinson, D. P.; Buttrill, S. E. Jr., Org. Mass Spectrom. 1976, 11,
Stewart, J. H.; Shapiro, R. H.; DePuy, C. H.; Bierbaum, V. M., J.
Am. Chem. Soc. 1977, 99, 7650.

Depuy, C. H.; Bierbaum, V. M.; King, G. K.; Shapiro, R. H., J. Am.
Chem. Soc. 1978, 100, 2921.

Hunt , D. F.; Sethi, S. K. J., Am. Chem. Soc. 1980, 102, 6953.
DePuy, C. H.; Bierbaum, V. M., Acc. Chem. Res. 1981, 14, 146.

Chakel, J. A., Ph.D Dissertation, Michigan State University, East
Lansing, MI, 1982.

 

 

CHAPTER 4

EXPLORATION OF ION-MOLECULE REACTIONS IN
TANDEM MASS SPECTROMETRY FOR DISTINGUISHING
ISOMERS OF POLYCHLORODIBENZO-P-DIOXINS

Introduction

Polychlorodibenzo-p-dioxins (PCDDs) are well known for the great
concern over their occurrence in the environment. With the exception of
the octachloro-congeners, all the very toxic PCDD congeners have chlorine
substitution at the 2, 3, 7, and 8 positions (1). In particular, the extreme
toxicity of 2,3,7,8-tetrach1orodibenzo-p-dioxin (TCDD) has attracted much
1attention. For many years, GC/MS has been a standard method for
analyzing TCDDs in environmental samples. However, isomers of TCDD
often give identical or similar spectra when traditional mass spectrometric
techniques are used. Although the use of oxygen negative chemical
ionization (NCI) has been shown to provide distinct mass spectra for 1,2,3,4—
TCDD and 2,3,7,8-TCDD (2), there is no experimental evidence that the
technique is capable of differentiating among 2:2 substituted TCDD isomers
(3).

Chemical interferences are the major concerns in analyzing

 

environmental samples for 2,3,7,8-TCDD. Prior to the mass spectrometric
analysis of a sample, extensive chemical cleanup and high resolution gas

chromatography (HRGC) are necessary in order to reduce interferences to

99

 

1(X)

a reasonable level. Even so, high resolution mass spectrometry (HRMS)
with resolving power of over 10,000 is needed to resolve TCDDs from many
common chemical interferences (4,5). However, for modern double-
focusing mass spectrometers, achievement of the requisite resolving power
results in a substantial loss of instrument sensitivity. Tandem mass
spectrometry (MS/MS) was first described by Chess and Gross (6) as a
rapid-screening method for analyzing TCDD. The loss of COCL by the
metastable decomposition of the molecular ion of TCDD was monitored for
the analysis. Despite the fact that this technique offers speed and reduced
interferences from PCBs, its overall specificity and sensitivity are
significantly worse than GC/HRMS. Since then, a number of workers have
explored the use of MS/MS coupled with GO or HRGC for analyzing TCDD
(7-20). In most cases, the increased specificity of MS/MS was found to
greatly reduce chemical interferences, which were seen in HRMS when

real samples with complex matrices were analyzed (5,9,10,14,15,17,19,20).

Since the MS/MS techniques mentioned above monitor only the loss of
COCL by collisionally induced dissociation (CID) of [M]+, they are not
isomer-specific for 2,3,7,8-TCDD. From analyzing fly ash samples, Buser
and Rapper (21) found a complex isomeric mixture of at least 17 TCDDs. It
was estimated that 2,3,7,8-TCDD was present at only 1% of the total TCDD
amount in the samples. Although isomer-specific chromatographic
methods have been developed for analyzing 2,3,7,8-TCDD (22-24), lengthy
separations are still required. Kostianinen and Auriola (25,26) have

recently employed the reaction between the [M]' of TCDDs and 02 in the

collision cell of a triple quadrupole system to distinguish 1,2,3,4—,

1.2.3.6/1,2,3,7,- and 2,3,7,8-TCDD isomers.

 

 

 

 

101

In this work, we have further explored the use of reactions between
the [M-1]‘ ionsof PCDDs and different neutral molecules for distinguishing

the more toxic PCDDs from the less toxic ones.

Experimental

All the experiments were performed on the Finnigan TSQ-70B triple
quadrupole mass spectrometer (TQMS). Samples were introduced into the
Varian 3400 gas chromatograph (GC) which is directly interfaced with the
TQMS. A capillary column was used (30 m, 0.25 mm i.d., SE-54 phase with
0.25 um film thickness) to ensure the high purity of the PCDD samples
entering the TQMS. The injection port of the GC was set at 250 °C under a
splitless injection mode. The temperature was programmed from 150 °C to
300 °C at 15 °C/min. A transfer line directs the column into the ion source of
the TQMS and was set at 275 °C. Helium was used as the carrier gas at a

l pressure of 15 psi. Each injection contained 1-2 uL of sample. The ion
source temperature was held at 150 °C, the electron ionization at 70 eV, and
the electron current at 200 uA. During ammonia chemical ionization (CI),
a pressure of 1.5 torr was maintained throughout the experiments. During
the MS/MS operations, ions extracted from the source were selected by the
first quadrupole (Q1) of the TQMS to enter the second quadrupole (Q2)
where neutral molecules were introduced for ion-molecule reactions.
Product ions emerging from Q3 were then mass scanned to obtain the
product mass spectra. A somewhat better than unit resolution was tuned
for Q1 in order to ensure the purity of the selected parent ions, whereas Q3

was maintained at unit resolution. A collision energy of 3 eV (laboratory

 

 

102

energy as determined by the Q2 offset voltage) was used for all the
experiments involving ion-molecule reactions. All the experiments were
performed twice to test for consistency in the results. The replicates were
compared for discrepancies in terms of relative peak ratios and the absence
or presence of stray peaks. Duplicate runs consistently gave spectra that
were approximately the same. The data shown come from one set of results

rather than the average of the replicates.

All the PCDD congeners were purchased from AccuStandard, New
Haven, CT, except for 1,3,6,8-TCDD and congeners with six chlorines or
more, which were purchased from UltraScientific, North Kingstown, RI,
and Cambridge Isotope Laboratories (CIL), Woburn, MA, respectively. All
the congeners were obtained in toluene solution. All alcohols were obtained
from Aldrich Chemical Company Inc., Milwaukee, WI. Deuterated water.
(99.5+% grade) was also purchased from CIL. Deuterated ammonia (99.5
atom %D) was obtained from MSD Isotopes, Montreal, Canada. All the

chemicals were used without further purification.

Results and Discussion

A. Generation of Reactant Ions

A mixture of CH4/N20 was used by Oehme and Kirschmer as a
reagent to obtain the negative chemical ionization (NCI) mass spectra of
TCDD isomers (27). Most of the isomers were found to form predominantly

the [M—1]' ion due to direct or indirect deprotonation by OH‘. In our

 

 

103

experiments, NH3 was used as the NCI reagent and is found to produce a
higher [M-1]' yield than the previous study with CH4/N20. The abundance
of the [M-1]' relative to [M]' is related to the acidity of the [M]' ion which, in
turn is affected by the positions of the chlorines. The NCI spectra obtained
from the five TCDD isomers (Figure 4-1) in this study indicate that 2,3,7,8-
TCDD has the highest acidity and therefore the highest relative abundance
of [M-1]'. The acidity of 1,2,3,4-TCDD is shown to be the lowest among the
isomers by having only a very small peak at m/z 319. This is primarily due
to the lack of electron-withdrawing Cl adjacent to H in 1,2,3,4-TCDD. For
the rest of the TCDD isomers, the acidity is directly related to the number of
hydrogens at the 1, 4, 6 and 9 positions. The lower acidity of 1,3,6,8-TCDD
compared to 1,2,7 ,8-TCDD and 1,3,7,8-TCDD is the result of having only two
hydrogens at the 1, 4, 6 and 9 positions instead of three. The trends
observed in our study (summarized in Table 4-1) matches well with the
results obtained by Oehme et al. (27) using CH4/N 20 as reagents. A similar
trend is also observed for the lower chlorinated PCDD congeners.
‘ However, congeners containing five chlorines or more do not produce any
[M-1]' ions under our CI conditions because of their relatively low acidity.
1 For these higher chlorinated congeners, the predominant NCI product is
[M]' (see examples in Figure 4-2), which is mainly due to the capture of a

thermal electron inside the CI source.
B. Reagflgns m‘th DZQ

Hydrogen/deuterium (H/D) exchange reactions between the [M-1]'

ions of chlorinated benzenes and D20 were found to be distinctive for the

three dichlorinated isomers as well as the others (28,29). Upon low energy

 

 

 

104

Figure 4-1 Mass spectra of TCDD isomers derived from negative chemical
ionization (NCI) using ammonia as a reagent.

 

relative intensity

105

 

 

 

 

 

 

 

 

 

 

 

 

[M-1]‘ 321
100 2378TCDD
80 H H \\\\ 319
G 0
so 323
O a
40 H
324
20 285 l 326 339
0 ll“. I llnm lllln‘
320 340 350
322
1001
3
L2,+TCDD 320
80' H a
H Cl
H
40
301
20 285 326
I 1305 “I |
0 l l . . 1 l l I v
300 3.0 340 350
3.2
10° lfififiJCDD 32
80 H Cl 324
O H
60
H
40' 319
20_ 326
285 339
II. I
0 ' I ' 1’ —1*
300 320 340 360
321
100
LzstCDD
80- 319
H Cl
— ° ‘ u 323
so
a”.
40‘ -
J 324
20 285 I 326 339
l.l.1. i_._. l- illlim
0 I fi" ‘I’ l f
300 320 340 360
321
100
73
L3, TCDD 319
80 n a
a a 323
60
G
40 "
324
20 285 I l 326 339
1114:; i“ . 7‘ llm iJlerl r
0 300 320 340 360

m/z

 

 

106

Table 4-1 Ion intensity ratios for m/z 319 and m/z 320 derived from the
ammonia NCI mass spectra of the selected TCDD isomers.

TCDD isomers 319/ 320 # of Hs at the 1,4,6,9 positions
2378 100/ 20 4
1278 100/ 27 3
1378 100/ 25 3
1368 100/ 210 2

1234 100 / 1700 2

 

 

 

—s4__.V--

107

Figure 4-2 Mass spectra of 1,2,4,7,8- and 1,2,3,7,8rpentachlorodibenzo-p-
dioxin derived from negative chemical ionization (NCI) usmg
ammonia as a reagent.

 

108

 

 

 

 

 

 

 

 

 

 

_ [M]' 354
100 \ 358
80-
1,2,4,7,8-PentaCDD
60—
360
401 I
>3
3:4"
w 20‘ 321 337
g 31:!) 323 339 362
"*3 llllli J1 Iii! H; L.
g 0 Y'I'III'VI'I‘Ili'lll'Il'lllI'ITjIYrV‘IVIIIfI I ll'lll vllrfII‘I—rI—I—VT
0H 320 330 340 350 360 370 380
Q)
E
. 356
*3 100-
'?‘o 1,2,3,7,8—PentaCDD
s...
'I f
380 400

 

 

109

collisions with D20 the [M-1]' ions of most PCDD congeners undergo H/D
exchange reactions. The product mass spectra for the five TCDD isomers
selected for this study are shown in Figure 4-3. The exact mass for TCDD
was used to select the [M-1]' reactant ions by Q1. The [M-1]' ions selected
contain only the 35Cl and 12C isotopes that corresponds to m/z 318.9 in the
mass spectra. Product ions with deuterium substitutions are shown by
peaks with one or more m/z units above those for the reactant ions. Only
product ions indicative of one deuterium exchange are observed for 2,3,7,8-
TCDD. This product pattern is unique among the five TCDD isomers. For
both 1,2,7,8- and 1,3,7,8-TCDD, up to two hydrogens are found to be
exchanged. No H/D exchange product was observed for 1,3,6,8-TCDD.
However, product ions associated with exchange of three hydrogens in the

anions of 1,2,3,4-TCDD are found.

In this study, the reactivity of the 1,2,3,4-TCDD anions is found to be
significantly higher than the anions of the other selected isomers. In
addition, there is also a strong tendency for 1,2,3,4-TCDD to form a [M-
1+D20]' adduct which is completely absent from the spectra of the other
isomers. Although the mechanism of the exchange reactions is not known
at this point, we propose that the H/D exchange reaction is initiated by
attack of the negative charge site on the [M-1]' ions of TCDD isomers to one
of the two deuteriums of a D20 molecule. Multiple deuterium substitutions
on the reactant ion are then formed by sequential H/D exchange reactions
with different D20 molecules. Clearly, the number of exchangeable
hydrogens of a TCDD [M-1]' ion depends on the positions of the chlorines.
The formation of [M-1+D20]' and the relatively high H/D exchange

reactivity for the anions of 1,2,3,4-TCDD are probably due to the absence of

 

 

110

Figure 4-3 Product mass spectra of [M-1]' ions derived from TCDD

isomers upon hydrogen/deuterium (H/D) exchange reactions
With D20. The exchange products are indicated by peaks at
one or more m/z units above the reactant ions.

 

111

M—ll— (Reactant ion)

 

 

 

 

 

 

 

318. 9
100
2318TCDD
80
H H
so ° °
0
4o 1
20 319.9
l'—" fl I fifi IL fix 1—' I 1 Y—* I
300 310 320 330 340 350
318.9
100~
L&&&TCDD
80
H G
601 ° "
H
40'
20‘
-t3 3 o ' 3io ' 320 330 340 350
$3 313.9
100
Q) L218TCDD
l 4.: 80‘ H a
l c: _ a a
‘ .,4 so H
9* A 319.9
Q) 40 “
F> J 3239
OH 20
4g
FEE 300 310 320 330 340 350
E3 31 9
10° LazeTCDD
H G
. H
GOO ..
'—'—‘—r 1 1 Yfi—V—l—
340 350
3419
339.9
338.9
I
. . f

340 350

 

112

hindrance of Cls on the benzene ring where the charge is located. A
complete understanding of the specificity of the H/D exchange reactions
may not be known until a more comprehensive study involving all possible
isomers is done. Nevertheless, this preliminary study has demonstrated
that the H/D exchange reactions allow 2,3,7,8-TCDD to be differentiated
from 1,2,7,8-, 1,3,7,8-, 1,3,6,8- and 1,2,3,4-TCDD. An attempt to observe H/D
exchange reactions between D20 and the [M]' ions of PCDD's was
unsuccessful; no exchange products were observed. Since the [M]' ion is
an odd-electron species, we conclude that the lone-pair electrons on the [M-

1]' ions are critical for the initiation of a fruitful H/D exchange reaction.

C. Beaotions with Alcohols

When the [M-1]' ions of PCDDs react with simple alcohols, ion-
molecule adduct complexes, [M-1+alcohol]‘, are formed. Under our
experimental conditions, the reactivities of the TCDD isomers are found to
be dependent on the positions of the Cls. The alcohols selected for this
study, in ascending order of acidity, are methanol, ethanol, isopropanol
and isobutanol. The abundances of the adducts formed for various

combinations of alcohol and TCDD isomers are tabulated in Table 4-2.

The reactivity of 2,3,7,8-TCDD is shown to be consistently the lowest
among the isomers studied. According to our NCI experiment and others
(27), the acidity of 2,3,7,8-TCDD is the highest among all the TCDDs. This
means that the [M-1]' of 2,3,7,8—TCDD is the weakest base of all the anions of
TCDDs. For 2,3,7,8-TCDD, there is only one possible site for the negative

charge of the [M-1]‘ ion because the positions of all four hydrogens on the

 

 

113

o.m comm 8v cg g
EN 00H #5 cw wbmfi
mfl mu mm X... waH

 

.mpqmuowme fie): 35 mo $03 so 683 $9535 $3833-63 93 mo mmwfismonmm «iv 28mm.

 

114

molecule are equivalent. It is conceivable that adduct complexes are
formed by interactions between the lone-pair electrons on the charge site
and the H of the OH functional group of the alcohol molecules. This
explanation is supported by a significant isotopic effect observed when
alcohols with an OD functionality are used. The charge site on the [M-1]' of
2,3,7,8-TCDD is less likely to attack a proton because of its weaker basicity.
The reactions with alcohols were also extended to the [M-1]' ions of 2,3- and
2,7-dichlorodibenzo-p-dioxin (DCDD). Overall, the reactivities of 2,3,-DCDD
are also consistently lower than those of 2,7,DCDD. In the case of reactions
with methanol, the ratio of [M—1+alcohol]' : [M—1]' for 2,7-DCDD is found to
be about 10 times that of 2,3-TCDD. The differences in reactivities for the
two isomers coincide with the fact that while the structure of 2,7-DCDD
allows four hydrogens to be adjacent to the chlorines, only the two
hydrogens at the 1 and 4 positions of 2,3,DCDD are adjacent to the
chlorines. This means that most of the [M-1]' ions of 2,3-DCDD would have
a charge site at the location identical to that of a 2,3,7,8-TCDD [M-1]' ion.
The results from these two DCDD isomers appear to substantiate the above
explanation on the differences in TCDD reactivities with the alcohol

molecules.

For the alcohol reagents, an increase in reactivity is observed when
the acidity of the alcohol is increased from that of methanol to that of
isopropanol. This is exactly what is expected for an acid/base reaction. The
dramatic decrease in reactivity for isobutanol is probably due to the steric
hindrance of its structure. For the higher chlorinated PCDD congeners,

reactions between the [M]' ions and alcohol molecules do not produce any

 

115

products. The lack of the lone-pair electrons on the [M]' ions apparently do

not allow the formation of the complexes.

DBeacfiamsan‘tth

Oxygen molecules were also used as a reagent to react with the [M]'
ions of PCDDs. At a collision pressure of 0.8 mtorr and collision offset
energy of 2.8 eV, reactions between the [M]- of 1,2,3,4-TCDD and 02 produce
[M-Cl+O]' ions from oxygen/chlorine (O/Cl) exchanges. Although the ratio
of [M-Cl+O]' : [M]' for 1,2,3,4-TCDD is found to be about 3:1, the O/Cl
exchange product is not observed for the [M]' ions from 2,3,7,8-, 1,3,6,8-,
1,2,7,8- and 1,3,7,8-TCDD under our experimental conditions. It is
necessary to point out that the [M]' ions selected for the 2:2 substituted
TCDD isomers also include the 13C isotope of the [M-1]' ions, which could
not be separated by the first quadrupole mass filter. However, this should
have little or no effect on the formation of any product ions resulted from the
[M]- ions because no reaction was observed when the [M-1]' ions of PCDDs
were reacted with 02. The [M-Cl+O]' ions are obtained from the [M]' ions of
the penta-, hexa- and hepta—chlorinated PCDD congeners. Although others
(25,26) have demonstrated the potential of using 02 as a reagent to
differentiate 2,3,7,8-TCDD from 1:3 substituted TCDD isomers, we could not
produce experimental evidence to Support the potential usefulness of using
02 to differentiate 2,3,7,8-TCDD from the other 2:2 substituted TCDD
isomers. More experiments are needed in order to determine the
mechanism of the O/Cl exchange on the [M]' ions of PCDDs, and thus the
reason for the apparent lack of reactivity of the 2:2 substituted TCDDs.

 

116

Conclusions

Ion-molecule reactions occurring in the center stage of a triple
quadrupole mass spectrometer are feasible options for distinguishing the
more toxic isomers of PCDDs from the less toxic ones. For reactions
involving the formation of [M-1+a1cohol]‘ adducts, the reactivity of the [M-
1]' of 2,3,7,8-TCDD is consistently the lowest among the five TCDD isomers
studied. Of the four simple alcohols selected for these reactions, methanol
appears to allow 2,3,7,8-TCDD to be differentiated to a greater degree from
the other less toxic isomers. The H/D exchange product patterns obtained
from reactions between the [M-1]- ions of TCDD isomers and D20 permit
2,3,7,8-TCDD to be differentiated from 1,3,6,8-, 1,2,7,8-, 1,3,7,8— and 1,2,3,4-
TCDD. At this stage, it is not known whether the H/D exchange reactions
are useful to differentiate 2,3,7,8-TCDD from all other TCDD isomers. A
more comprehensive study involving all possible TCDD isomers is needed
in order to gain additional insight into the mechanism of these reactions,

and verify their applicability as analytical techniques.

The unique negative charge location on the [M-1]‘ of 2,3,7,8-TCDD
may be further utilized for a contiuous search of ion-molecule reactions
which are truly specific for 2,3,7,8—TCDD. With the selectivity of MS/MS,
the added specificity provided by ion-molecule reactions may reduce sample

cleanup and shorten GC separation time for 2,3,7,8-TCDD analysis.

 

117

References

10.

11.

12.

Clement, R. E.; Tosine, H. M., Mass Spectrom. Rev. 1988, 7, 593.

Hass, J. R.; Friesen, M. D.; Hoffman, M. K., Org. Mass Spectrom.
1979, 14, 9.

Mitchum, R. K.; Korfmacher, W. A.; Moler, G. F.; Stalling, D. L.,
Anal. Chem. 1982, 54, 719.

Tondeur, Y.; Beckert, W. F.; Billets, S.; Mitchum, R. K.,
Chemosphere 1989, 18, 119.

Tondeur, Y.; Niederhut, W. E.; Campana, J. E.; Missler, S. R.,
Biomed. Environ. Mass Spectrom. 1987, 14, 443.

Chess, E. K.; Gross, M. L., Anal. Chem. 1980, 52, 2057.
Shushan, B.; Fulford, J. E.; Thomson, B. A.; Davidson, W. R.;
Danlewych, L. M.; Ngo, A.; Nacson, S.; Taner, S. D., Int. J. Mass

Spectrom. Ion Physics 1983, 46, 225.

Shushan, B.; Tanner, S. D.; Clement, R. E.; Bobbie, B., Chemosphere
1985, 14, 843.

Clement, R. E.; Bobbie, B.; Taguchi, V., Chemosphere 1986, 15, 1147.

Charles, M. J.; Green, B.; Tondeur, Y.; Hass, J. R., Chemosphere
1989, 19, 51.

McCurvin, D. M. A.; Schellenberg, D. H.; Clement, R. E.; Taguchi,
V. Y., Chemosphere 1989, 19, 201.

McCurvin, D. M. A.; Clement, R. E.; Taguchi, V. Y.; Reiner, E. J .;
Schellenberg, D. H.; Bobbie, B. A., Chemosphere 1989, 19, 205.

 

 

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

(23)

118

Fraisse, D.; Gonnord, M. F.; Becchi, M., Rapid Commun. Mass
Spectrom. 1989, 3(3), 79.

De Jong, A. P. J. M.; Liem, A. K. D.; Den Boer, A. C.; Van Der Heeft,
E.; Marsman, J. A.; Van De Werken, G.; Wegman, R. C. C.,
Chemosphere 1989,19, 59.

Moore, C.; Moncur, J.; Jones, D.; Wright, B., Rapid Commun. Mass
Spectrom. 1990, 4(10), 418.

Reiner, E. J.; Schellenberg, D. H. Taguchi, V. Y.; Mercer, R. S.;
Townsend, J. A.; Thompson, T. S.; Clement, R. E., Chemosphere
1990, 20, 1385.

Charles, M. J .; Tondeur, Y., Environ. Sci. Technol. 1990, 24, 1856.

Charles, M. J.; Marbury, G. D., Anal. Chem. 1991, 63, 713.

Iida, J .; Takeda, T.; TakaSuga, T.; Moncur, J .; Ireland, P.; Wright,
B., J. High Res. Chromatogr. 1991, 14(2) 103.

Reiner, E. J.; Schellenberg, D. H.; Taguchi, V. Y., Environ. Sci.
Technol. 1991, 25, 110.

Buser, H. R.; Rappe, 0., Anal. Chem. 1980, 52, 2257.

O’Keefe, P. W.; Smith, R.; Meyer, C.; Hilker, D.; Aldous, K.; Jelus—
Tyror, B., J. Chromatogr. 1982, 242, 305.

Naikwadi, K. P.; Karasek, F. W., Chemosphere 1990, 20, 1379.

Tiernan, T. 0.; Garrett, J. H.; Solch, J. G.; Harden, L. A.; Lautamo,
R. M.; Freeman, R. R., Chemosphere 1990, 20, 1371.

Kostiainen, R.; Auriola, S., Rapid Commun. Mass Spectrom. 1988, 2,
135.

 

26.

27.

28.

29.

 

119

Kostiainen, R.; Auriola, S., Org. Mass Spectrom. 1990, 25, 255.
Oehme, M.; Kirschmer, P., Anal. Chem. 1984, 56, 2754.
Chakel, J. A., Ph.D Dissertation, Michigan State University 1982.

Chan, S.; Enke, C. G., paper submitted for publication.

 

CHAPTER 5

IMPROVEMENT TO THE DETECTION OF LOW-ENERGY ION
PRODUCTS IN A TRIPLE QUADRUPOLE MASS
SPECTROIVEETER - A THEORETICAL CONSIDERATION

Introduction

Ion-molecule reactions occurring at thermal or near thermal
energies are often very sensitive to structural differences. Ion products
formed by these reactions possess little or no kinetic energy. Unfortunately,
the lack of kinetic energy of the ion products creates a challenge for their
detection. In a triple quadrupole mass spectrometer (TQMS), the detection
of these ions in the second quadrupole collision chamber (Q2) depends on
the ability of the ions to reach the detector. Without a guiding field, many of
these ions would eventually be lost from the reaction chamber in directions
other than towards the detector. In order to increase the detection
efficiency of these ions, a standard technique uses a withdrawing potential
applied between Q2 and the third quadrupole (Q3) to extract the low—energy
and thermalized ions from Q2 for mass analysis by Q3. However, for
thermalized ions, ion currents are limited by the rate of diffusion along the
axis of Q2 from the point where their axial kinetic energy becomes zero
until they reach the drawout potential field that extends only a short
distance into the reaction chamber. This chapter will describe novel
designs of the second quadrupole for facilitating the movement of low-

energy ion products towards the detector.

120

 

121

Where the study of the thermochemistry of ion-molecule reactions is
not the main goal, relatively high collision pressures (z 4 mtorr) of neutral
molecules can be introduced into Q2 in order to assure an abundance of
neutral molecules for reactions. Under high pressure conditions, ions
transmitted into the reaction chamber may have multiple collisions with
the neutral molecules before reaching the exit of Q2. Such reaction
conditions increase the rate of ion-molecule reactions and thus the yield of
products. Furthermore, multiple collision conditions have a stabilization
effect on products and intermediates that would otherwise be unstable and
undetectable under single collision conditions (1). A novel ion trapping
technique (2,3) has been developed at Michigan State University by Dr.
Watson‘s and Dr. Enke's research groups for enhancing multiple collision
conditions even at low collision pressures. The detection of reaction
products is improved essentially by a net gain of product signals due to the

presence of more product ions.

The novel ion trapping technique requires no hardware modification
to a traditional TQMS instrument. The so-called trap-and-pulse mode of
this technique involves the use of an extraction lens (EL) that is placed
between Q2 and Q3 as an ion trapping and pulsing device. Usually, +100V
is applied to the extraction lens for trapping positive ions and -100V for
negative ions. In this method, parent ions of a selected mass are
continuously transmitted into Q2 for ion-molecule reactions while the
trapping potential is applied to EL. After a certain length of time, a pulsing
potential, which is opposite in polarity to the trapping potential, is applied to
extract a fraction of the trapped ions from Q2 for detection. Ion trapping

time can be varied to maximize ion signals. Normally, 90% of the

 

 

maximum intensity can be obtained within 100 ms of trapping time. This

method is necessarily limited to collecting ions of a single nominal mass-to-
charge value per pulse because the millisecond duration of the ion pulse is
too short for full range Q3 mass scanning. Nevertheless, significant
improvements in signal-to-noise ratios have been obtained over those in the
conventional data collection method. The greater residence time of ions in
Q2 made possible by extending the trap time revealed some reaction

products not observed by continuous product extraction.

Designs for the Second Quadrupole Collision Chamber of a TQMS
Instrument

The extraction lens used in both the standard and the trap-and-pulse
methods can only extract ions that are very near the exit of Q2. A high
percentage of the thermalized ions still stay in Q2 without being detected as
products. Figure 5-1 shows a potential contour diagram for the standard
ion extraction design based on calculations using SIMION (4), a computer
program for ion optics calculations. This diagram indicates that the 100 V
extraction potential applied to the lens at the end of the collision chamber
can only penetrate a very short distance into Q2. Most ion products formed
in Q2 are not directly affected by the extraction potential. An idea to assist
the thermalized ions to move towards the exit of Q2 is to create a
longitudinal electric field along the path of the ions towards that region. In
this way, ions can experience a continuous pushing force in the direction of
the Q2 exit. This is particularly important for ions that are heavily

surrounded by neutral molecules under high collision pressure conditions.

 

 

 

 

 

 

A> ooCmcfl

dmmmov mum: desomfixo 55 now Ennmwmo .3038 83:33 < Tm 939m

 

>8 o_ .8 EiEBoa
m .18 ac: .5028 m

 

124

The first idea for creating such a longitudinal electric field is to modify the
design of the Q2 rods in such a way that the rods are made of stainless steel
segments which are separated by thin layers of non-conductive Teflon
(Figure 5—2). Resistors are connected to each segment to divide the total
voltage along the rod into equal steps for each segment. If a potential
difference is applied to the ends of each rod, a potential field gradient is
created in space along the ion path of Q2. Theoretically, for a true smooth
continuous potential field gradient, an infinite number of infinitesimally
thin segments are needed in each rod. If the rods are made of too few
segments, the gradient created along the Q2 ion path would be in a step-
jump fashion. Using SIMION calculations, the potential contour diagrams
for the designs of 6-, 11-, 21—, 26- and 35-segmented Q2 rods have been
generated. The configurations used for the calculations are shown in
Figure 5-3 through Figure 5-7. In this case, all the calculations are based
on a 5—V potential difference applied to the ends of a 20-cm Q2 assembly.
Based on data extracted from the potential contour diagrams, plots of
longitudinal potential as a function of spatial distance into Q2 for the 6-, 11-,
21-, 26- and 35—segmented designs are created and shown in Figure 5-8.
The 6- and 11-segmented designs give distinct potential plateaus in the
space along the ion path of Q2. At the region of a potential plateau, ions
would experience only a very slight potential gradient. These ions may still
stay in the collision chamber if the forward momenta of the ions are too
small to allow their arrival to the next potential gradient regions. Since the
potential gradient along the ion path of Q2 is related to the change of
potential as a function of changing spatial distance inside Q2, plots that
represent the 1st derivative of the results of Figure 58 would give a direct

representation for the smoothness of the potential gradients for each

 

 

 

 

mwon 2386335 980% mg wqfiqoamwm mo 939% < Wm charm

con No concmEmwm m

 

    
  

homoswmb
-oflwmm

 

 

wvou wmquamwé mo 3&8 N0 a

633% 13ququ >-m w firs
pom Embwmmu 38:8 35:38 n no dosmpdmwcoo 2; m-m BPwE

 

 

 

 

623% 3283 >-m a £3,
wwop wmuqmfiwwméfi mo 338 NE a Sm Sapwmww anon—Goo 3.2309 m. we cosmgsmwcoo one “Tm 8de

 

 

r'

 

.3:QO Ragga >-m m fits
mwop wmucmfimwmém mo mvmfi NQ m .Sm Smbwwww 5850 EM

Egon a mo cosmgdmwcoo 2E. m-m mgswE

 

 

.vmmmgw E3283 >-m a fits
wvop vmuswammmém mo 338 Nd a you 89“.”va 38:8 $5593 a mo Goflwhrwmgo 2E. w-m 95mg

 

 

130

683% R5980: >-m m :83
m8: Egofimwmbm mo 85:: NG m pom 88mg: 88:8 33:80: a mo :ofimudmmgo BEL Wm 9:de

 

131

Figure 5-8 Potentials along the ion path for the designs. of Q2 made 0t;
different numbers of segments When a potential difference 0
5-V is applied.

 

 

potential (V)

132

6-segmented rods

 

l I l ' l I l ' l l I I I

 

 

 
 

21-segmented rods

  

 

 

 

I I Ifi' I
2 4 6 8 10 12 14 16 18 20

 
 

35-segmented rods

  

 

1 ffj’ 1 I V l 1 l ' rfi I ‘ I ' l ' I

1
0 2 4 6 8 10 12 14 16 18 20

distance into the Q2 ion path (cm)

 

 

 

133

design. The potential gradient plots shown in Figure 5-9 indicate that
increasing the number of segments for the Q2 rods allows a smoother
potential gradient. With a 5-V potential applied to the 20-cm Q2 rods, an
idea potential gradient throughout the entire Q2 ion path would be 0.25
V/cm. According to the potential gradient plots, the average potential
gradient variations for the 21-, 26- and 35—segmented designs are
approximately 40%, 20% and 5%, respectively from the idea 0.25 V/cm. For
this application, an average variation in potential gradients should not be
more than 10%. This means that at least 30 segments are needed for
creating a reasonably smooth potential gradient along a 20-cm Q2 ion path.
An extraction lens is still needed for the segmented designs because the
calculations suggest that there is a negative potential gradient near the exit
region of Q2. Nevertheless, the smoothness of the field gradient created is
clearly a function of the number of segments used to make the Q2 rods.
Mechanically, there is a limitation to the maximum number of segments
possible for rods with a specific length. If the four rods of Q2 are each made
of 50 segments, the fabrication of the device would be very labor—consuming

and mechanically tedious.

Although the idea of a segmented Q2 quadrupole is theoretically
feasible, the making of the device is very difficult. In the summer of 1988,
after discussions with Dr. E. Strangas in the electrical engineering
department and Dr. C. G. Enke, a new idea emerged for creating a smooth
field gradient within Q2 Without physically segmenting the quadrupole. In
this novel idea, a resistance wire is wounded around each Q2 rod to create
the effect of a voltage divider. High resistance wires with resistivity up to

800 ohm/emf (or 1.38x10‘6 ohm-meter) can be purchased from several

 

 

 

134

Figure 5-9 The lst derivative plots representing potential gradient vs
spatial distance into the Q2 ion path.

 

V m—T’r—w;

 

 

 

 

 

 

 

 

 

135
1.6 — 6-segmented rods
1.2 j
0.8:
0.4 -
0.0 -
-0.4 ' I ' I ' I ' I ' I ' I ' I ‘ I ' I ' I
0 2 4 6 8 10 12 14 16 18 20
0.8 11-segmented rods
0.6
0.4
0.2
0.0
-0.2
'0.4 I r I I I fir I I fi" I I I I
49 0 2 4 6 8 10 12 14 16 18 20
Cl
Q)
"-4
"d
a
,—_4
C6
-.—4
.4.)
:1
Q)
4.)
0
Q4
0 2 4 6 8 10 12 14 16 18 20

 

 

136

manufacturers. Winding the resistance wire by the conventional helical
coil method will create an inductance that will make each rod an electro—
magnet. It is certain that the magnetic field will have unwanted
interactions with ions in Q2. The problem of inductance can be eliminated
if wires are winded onto the rods by a non-circular zig-zag winding
technique (Figure 5-10). In this winding technique, wire direction around
the rod is reversed at a series of hooks installed on the back of each rod.
Looking out from the inside of the quadrupole, this wire-winding design
gives essentially the same physical features as the conventional circular
coiling method except for the absence of inductance when an RF voltage is
applied. Each winding turn of wires is equivalent to one segment of
resistance. The maximum number of turns depends on the number of
hooks on the rods. Clearly, winding 50 turns of wires on the rods of Q2 is a
lot easier and more economical than making a Q2 quadrupole with 50
different segments. For a wire-winding design that gives an equivalence of
50 segments in each rod, ions entering Q2 should experience a very gentle

longitudinal field gradient.

Because the purpose of using resistance wires for the Q2 modification
is to create the effect of a voltage divider, rods that support these wires must
not be conductors. Technically, a perfect insulator would be ideal for
making the Q2 rods. Common metallic materials such as stainless steel
and copper cannot be used to make rods in this case because of their high
electrical conductivity. In order to evaluate the feasibility of this design, it
is also necessary to obtain information about the resistance of the wires

winded on Q2 as it is related to the amount of heat generated when a

 

137

 

. 8 m:%:§ - 939m
mm 33:8 8 0H m
was, 250.2%: saw was $56050 8s a 4
23 mo 32m Mean 28 :0 wmwwwwm: 80:58.86 MERE
6o: . a 88:0: 6 .
-983 mwumfi .H 3 .

>m

 

 

 

we: NG

 

 

 

_—
-.~
-
__
.—

< a...

5

~—~..—

awe:

\\

(E

_...
—-_-_
.—
~—

 

—.-
_...—
_.-

E

—-—-~
_-—-~--
.—

 

E

\

8.: ma

 

138
potential gradient is applied. Generally, this can be calculated according to
the following equation:

resistance (R) = EAL

where p is the resistivity of the wire, L is the length of the wire, and A is the
cross-section area of the wire. For a design that uses wires with resistivity
of 800 ohm/emf and a cross-section diameter of 0.5 mm, the resistance of a
50—turn wire winded on a 0.78 cm2 x 20 cm rod is calculated to be about 11
ohm. This resistance will generate about 2.2 W of power when a 5-V
potential difference is applied. The whole quadrupole winded with the
resistance wires will generate close to 9 W of heat. For a system operated at
high vacuum conditions, without the means of conduction, dissipation of
such large amount of heat through radiation alone is too inefficient. The
only efficient means for dissipating the heat would be by conduction
through the rods. This imposes another restriction for the type of materials
that can be used to make these rods. For this application, the four rods of
Q2 must be made of a material that is both a good heat conductor and a good
electrical insulator. This restriction excludes most available materials in
the market. Upon an extensive library search, a ceramic material called
beryllium oxide (BeO) (5) is found to be both an excellent heat conductor and
electrical insulator. Figure 5-11 shows the room temperature thermal
conductivity of BeO compared to various materials. In fact, the thermal
conductivity of BeO is higher than most metallic materials. The Q2 design
made of BeO can be fabricated by manufacturers for under $1000.

 

 

 

 

139

 

.20 £82 880 ..o:H 688.80 .3 «8me88 888 9:888: 8on wmumawwv

.mEEQEE 32.2? 8 88:88 Omm mo 3:826:00 $8.88 @888388 Soom 3% 3de

 

 

 

 

 

 

 

 

 

 

 

«5:52

830 88M $3 832 :ooEm 555834 Cam 8:900
lmd
lvd
lwd
.lwd

 

 

 

70A

 

 

(go'oasumypao

 

 

 

140

For the improvement of the trap-and-pulse technique described
before, the wire-winding Q2 modification can be further refined. Instead of
an extraction lens, the last 10 turns of the resistance wire of each rod can be
used for a 10-step longitudinal field gradient either for trapping or pulsing
of ions. This could provide an environment for a less abrupt potential field
change at the Q2 exit region. Similarly, another 10-step longitudinal field
gradient can be implemented near the entrance region of Q2 for a so-called
inject-trap-and-pulse operation in which a large potential field applied at
this region fiinctions as an ion gate to allow only a packet of ions to enter the
collision chamber for trapping and pulsing. The longitudinal potential field
will be in "boat" and "chair" forms during ion trapping and pulsing,

respectively.

Conclusions

The idea of facilitating the movement of low-energy ions in Q2
towards the detector could be accomplished by creating a longitudinal
electric field gradient along the ion path inside Q2. This could be created by
making Q2 rods behave as voltage dividers. This can be accomplished by
either using segmented rods connected with resistors or rods winded with
high resistance wires. Because the smoothness of a potential field gradient
created Within Q2 depends on the number of voltage-dividing segments on
each rod, for creating rods with a high number of voltage-dividing
segments, the design of using rods winded with resistance wires is
mechanically more feasible than the other design. However, the wire-

winding design requires the use of rods made of a material that is both a

 

 

 

 

141

good electrical insulator and a good heat conductor. Beryllium oxide would

appear to be an ideal material for the application.

A smooth longitudinal potential field gradient created inside Q2

would allow the detector to sample low-energy product ions formed in every

 

region of Q2. Product spectra obtained through the proposed method
described in this chapter should represent the ion distribution in Q2 more
closely than the traditional method.

References

1. White, E. L,; Tabet J. -C.; Bursey, M. M., Org. Mass Spectrom. 1987 ,
22, 132.

2. Dolnikowski, G. G.; Kristo, M. J.; Enke, C. G.; Watson, J. T., Int. J.
Mass Spectrom. Ion Phys. 1988, 82, 1.

3. Kristo, M. J ., Ph.D Dissertation, Michigan State University, 1987.

4. Dahl, D. A; Delmore, J. E., The SIMION PC/PSZ Users's Manual,
Version 2; Informal Report EGG-CS-7233, Rev. 2, Idaho Falls
National Engineering Laboratory, 1988.

5. National Beryllia Corporation: National Beryllia Technical Bulletin
on Berlox Beo.

 

 

CHAPTER 6

CONCLUSIONS AND SUGGESTED FUTURE EXTENSIONS

The specificity of ion-molecule reactions has been well-demonstrated
in this dissertation as a mean for distinguishing isomers of halogenated
aromatic compounds. The unique features of a TQMS instrument offer
enhanced selectivity for analytical applications of ion-molecule reactions.
In particular, H/D exchange reactions are found to produce very distinct
product spectra for the [M-1]' of halogenated benzene isomers. These allow
the isomers to be differentiated when traditional mass spectrometric
techniques fail. A series of studies of H/D exchange reactions using
various deuterated reagents suggests that D20 is among the most effective
reagents for exchanging aromatic hydrogens of the [M-1]' ions of
chlorobenzene isomers. When D20 is used to react with the [M-1]' ions of
tetrachlorodibenzo-p-dioxin (TCDD) isomers, 2,3,7,8—TCDD gives an unique
H/D exchange product pattern which allows its differentiation from the
other selected TCDD isomers. In addition, alcohol reagents are also found
useful for differentiating 2,3,7,8-TCDD from the other less toxic TCDD
isomers. The results described in this dissertation strongly demonstrate
the great potential of applying ion-molecule reactions as analytical tools for
differentiating organic isomers. A tandem quadrupole mass spectrometer

is particularly well-suited for such applications.

Although studies of ion-molecule reactions have been extensive over

the past three decades, the use of ion—molecule reactions for analytical

142

143

applications is largely unexplored. This dissertation has merely explored a
very small aspect of the enormous potential of the reactions. Certainly,

there is much work still to be done. The following are some suggestions.

I. Substituting one of the two chlorines of a dichlorobenzene molecule
with other functionalities such as NH2, OH, CH0, CH3, OCH3, COOH or
N02 is found to substantially reduce the reactivity of the [M-1]' ion for H/D
exchange reactions with D20. The observed reduction in reactivity suggests
that while D20 is an excellent reagent for exchanging hydrogens of
dichlorobenzene [M-1]' ions, different deuterated reagents are needed for
effectively exchanging the hydrogens of the substituted chlorobenzene
[M-1]' ions. The study of H/D exchange reactions on the substituted
chlorobenzene [M-1]' ions should be continued by exploring the use of
different deuterated reagents such as ND3, CD4, CH3OD, CH3COOD, etc..
The results of such a study may be important in establishing the
effectiveness of using ion-molecule reactions for differentiating isomers of

the substituted Chlorobenzenes.

II. The proposed H/D exchange mechanism involving the formation of a
five-membered-ring intermediate for reactions between the [M-1]' of
chlorobenzene and D20 is consistent with the results derived from all the
isomers of Chlorobenzenes. It is quite possible that a similar mechanism is
also applicable to the [M-1]' of aromatic compounds containing two or more
fused benzene rings. Our mechanistic study may be extended to
compounds with multiple benzene rings. Isomers of chloronaphthalene
are ideal candidates for initial studies. With a better understanding of the

two-ring system, the investigation might then be continued on compounds

 

144

with three or more fused benzene rings. Information obtained from the
mechanistic studies may be useful for developing better analytical

strategies.

III. Although the ion-molecule reactions described in Chapter 4 are
found useful to differentiate 2,3,7,8-TCDD from other less toxic isomers, the
study was limited to only a few isomers. In order to investigate the
usefulness of these reactions as analytical techniques, future studies
should include all 22 TCDD isomers. Since the structures of
polychlorodibenzofurans (PCDFs) are very similar to those of
polychlorodibenzo-p-dioxins (PCDDs), the study of PCDDs may also be

extended to include isomers of PCDF.

IV. Finally, the collision cell design described in Chapter 5 is expected to
provide improvement in the detection of low-energy product ions formed

inside the collision cell; the implementation of the proposed design would

be an interesting project to explore.

 

 

 

APPENDIX

 

145

Appendix

Computer-simulated individual collision-product plots based on different
percentages of reactive collision for the H/D exchange reactions derived
from different structures of the [M—1]' ions of dichlorobenzene isomers. The
simulations are based on the proposed H/D exchange reaction mechanism

described in Chapter 3.

 

 

relative intensity

0.9 "
0.8 '
0.7"
0.6 '
0.5 "
0.4—
0.3 -
02$

0.1 ‘

 

0.0 4

 

146

Structure 1 at 5% reactive collision

fi— reactant ion
‘ . '0— lH/D exchange
A ‘ " "— 2H/Dexchanges
‘ ‘ "‘3— 3H/Dexchanges
A
A
A A . . . . . O C O C O
A . . . O
5 2
o .
. A
. A
o . ‘
. A
. A ‘
o A ‘
o . ‘
O ‘ A ‘
o
o
o
“-—-"u-u-n n u n u u n o u u u a n» o o a u u—u—u_u_u
2 4 6 8 10 1214 1618 20 22 24 26 28 30

number of collision

Figure A-1 A simulated collision-product plot for the structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 5% reactive

collision.

 

0.9 '

0.8 '

0.7 '

relative intensity

 

147

Structure I at 10% reactive collision

—'A— reactant ion
_0— 1 H/D exchange
+ 2 H/D exchanges

—fi— 3 H/D exchanges

 

4 6 810 12141618 20222426 2830

number of collision

Figure A-2 A simulated collision-product plot fonthe structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 10% reactive

collision.

 

0.9 "
0.8 -

0.7 -

relative intensity

 

 

 

148

Structure I at 20% reactive collision

‘—fi— reactant ion
—0— 1 H/D exchange
—I— 2 H/D exchanges

3 H/D exchanges

24681012141618202224262830

number of collision

Figure A-3 A simulated collision-product plot fonthe structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 20% reactive

collision.

 

 

0.7 -

relative intensity

 

 

 

149

Structure I at 25% reactive collision

—‘A— reactant ion

—0— 1 H/D exchange

+ 2 H/D exchanges
3 H/D exchanges

24681012141618202224262830

number of collision

Figure A-4 A simulated collision-product plot fonthe structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 25% react1ve

collision.

 

150

1.0 1 Structure I at 30% reactive collision

    
   
  
 

0'9 _ —fi—_ reactant ion
0.8 " —0_ 1 WD exchange
' 2 H/D exchanges

0.7 -
. 3 H/D exchanges

  

relative intensity

 

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

number of collision

Figure A-5 A simulated collision-product plot fonthe structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 30% react1ve
collision.

 

151

Structure I at 35% reactive collision

  
   
   
  
 

‘—A— reactant ion
——0— IH/D exchange

0.9 "

0.8 "
. + 2 H/D exchanges
0.7 "

- 3 H/D exchanges

  

0.6 ‘
0.5 '

0.4 '

relative intensity

 

024681012141618202224262830

number of collision

Figure A-6 A simulated collision-product plot forOthe structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 35% react1ve
collision.

 

 

152

Structure I at 40% reactive collision

1011
0.9 _ —A'_ reactant ion

‘ —O—' 1 H/D exchange
08 - —I_ 2 H/D exchanges

3 H/D exchanges

relative intensity

 

 

0 2 4 6 810121416182022 24262830

number of collision

Figure A-7 A simulated collision-product plot fonthe structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 40% react1ve

collision.

 

 

153

Structure I at 45% reactive collision

1015

' —A‘ reactant ion
0.9 '

‘ —0— l H/D exchange

0.8: —l—‘ 2 H/D exchanges

3 H/D exchanges

relative intensity

 

 

0 2 4 6 81012141618202224262830

number of collision

Figure A-8 A simulated collision-product plot fonthe structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 45% react1ve

collision.

 

   

relative intensity

 

 

154

Structure I at 50% reactive collision

reactantion
1ILGDexchange
2 H/D exchanges
3 H/D exchanges

 
     

10 12 14 16 18 2O 22 24 26 28 30

number of collision

..vv.-.'—_—w

' ‘ ' ' I of 0-
F1 re A-9 A Simulated colliswn-product plot fonthe structure .
digllillorobenzene based on an H/D exchange react1v1ty of 50% react1ve

collision.

 

155

Structure I at 55% reactive collision

1.0 'A
- —A— reactant ion

09 _ —0— 1 H/D exchange

2 H/D exchanges

3 H/D exchanges

relative intensity

 

 

0 2 4 6 8 1012 1416 1820 2224 2628 30

number of collision

Figure A-10 A simulated collision-product plot for. the structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 55% react1ve

collision.

156

1.01:

Structure I at 60% reactive collision

  

reactant ion

   
 
  
  

—0— 1 H/D exchange
—I— 2 H/D exchanges

3 H/D exchanges

relative intensity

 

   

10 12 1416 18 20 22 24 26 28 30

number of collision

Figure A-11 A simulated collision-product plot for. the structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 60% react1ve

collision.

 

157

1.0 '18

0,9 — Structure I at 65% reactive collision

   
  
  

 

0.8 "

>, 0.7 '-

3: .

a 0 6 - reactant ion

.8 ' '—0— 1 H/D exchange
S 0 5 - + 2 H/D exchanges
a.) .
E 0.4 ‘ 3 H/D exchanges
.3 .

a) 0.3 "

M I

 

0 2 4 6 81012141618202224262830

number of collision

Figure A-12 A simulated collision-product plot for. the structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 65% react1ve

collision.

 

158

1.0 1‘ -
Structure I at 70% reactive collision

0.9 '

     
 
 
 
  

reactant ion
—0— 1 H/D exchange

—I_ 2 H/D exchanges
3 H/D exchanges

relative intensity

 

1012141618 20 22 24 26 28 30

number of collision

Figure A-13 A simulated collision-product plot for the structure I of o-
dichlorobenzene based on an H/D exchange react1v1ty of 70% react1ve

collision.

159

 

Structure II at 5% reactive collision

 
    
 
 

—A— reactant ion
—0— 1 H/D exchange
—I— 2 H/D exchanges
3 H/D exchanges

  

relative intensity

 

0 2 4 6 81012141618202224262830

number of collision

 

Figure A-14 A simulated collision-product plot for the structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 5% react1ve

collision.

 

160

1-0 ‘4 Structure II at 10% reactive collision

- ‘ . —A— reactant ion
—0— 1 H/D exchange
+ 2 H/D exchanges

‘ —G— 3 H/D exchanges

0.8 ‘ ‘

relative intensity

 

 

024681012141618202224262830

number of collision

Figure A-15 A simulated collision—product plot forthe structure II of o—
dichlorobenzene based on an H/D exchange react1v1ty of 10% react1ve

collision.

 

 

161

1-0 7‘ Structure II at 20% reactive collision

‘1— reactant ion
‘—0— 1 H/D exchange
+ 2 H/D exchanges

—'13— 3 H/D exchanges

0.8 ‘

relative intensity

 

 

024681012141618202224262830

number of collision

Figure A-16 A simulated collision-product plot for'the structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 20% react1ve

collision.

 

162

Structure II at 25% reactive collision

 
    
  
 
 

“—A— reactant ion
—"0— 1 H/D exchange
—I_ 2 H/D exchanges

'—G— 3 H/D exchanges

  

relative intensity

 

0 2 4 6 81012141618202224262830

number of collision

product plot for the structure II of 0-

~ _ ‘ d collision- '
Figure A 17 A Simulate exchange reactivity of 25% react1ve

dichlorobenzene based on an H/D
collision.

 

163

Structure II at 30% reactive collision
1.0 i

    
   
  
    

'_A_ reactant ion
—0— 1 H/D exchange
—I_ 2 H/D exchanges

3 H/D exchanges

0.8 ‘

relative intensity

 

number of collision

Figure A-18 A simulated collision-product plot fonthe structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 30% react1ve
collision.

 

 

164

Structure II at 35% reactive collision

  

1.0 A ”—A— reactant ion
‘ —0—" IH/Dexchange
—'— 2H/Dexchanges
08 —E'— 3 H/D exchanges
>} II
p . D
UH u
a A can
_8 06‘ n a
$21 a a
.,—( ‘ n
@ II
> n
113 n
(U 0.4‘ u
75 o
H -4
.0
02‘ o.
. o
. .0
o. .. o. .
00‘ A“"‘““‘"“ AAAAAAAAAAA AAA A A

 

024681012141618202224262830

number of collision

Figure A-19 A simulated collision-product plot forthe structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 35% react1ve

collision.

 

 

165

1 0 1“ Structure II at 40% reactive collision

   
 

'—A'_ reactant ion
+ 1 H/D exchange

‘—"’—' 2 H/D exchanges

 
    

"—G— 3 H/D exchanges

  

relative intensity

 

0 2 4 6 810121416182022 24262830

number of collision

product plot for the structure II of 0-

Figure A-2O A simulated collision- . . .
exchange react1v1ty of 40% react1ve

dichlorobenzene based on an H/D
collision.

 

166

Structure II at 45% reactive collision
1.0 ‘A

   
 

—'A— reactant ion
—0— 1H/Dexchange

 
 

0.8 " .
+ 2 H/D exchanges

3 H/D exchanges

     

relative intensity

 

02468112141618202224262830

number of collision

Figure A-21 A simulated collision-product plot for-the structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 45% react1ve

collision.

167

Structure II at 50% reactive collision
1.0 ‘A

   
 

+ reactant ion
—0_ 1 H/D exchange
+ 2 H/D exchanges
3 H/D exchanges

   

0.8 "

 
   
    

relative intensity

 

   

18 20 22 24 26 28 30

   

10 12 14 16

number of collision

Figure A-22 A simulated collision-product plot fonthe structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 50% react1ve

collision.

168

Structure 11 at 55% reactive collision

1.0 is
—A— reactant ion

—0— 1 H/D exchange
—I— 2 H/D exchanges

3 H/D exchanges

      

  

0.8 ‘

 
    

relative intensity

 

0 2 4 6 81012141618202224262830

number of collision

' ' ' ‘ II of 0-
F1 re A—23 A Simulated colhsmn-product plot forthe structure .
dighllorobenzene based on an H/D exchange react1v1ty of 55% react1ve

collision.

 

169

1.0 ‘A

Structure II at 60% reactive collision

     
 
 
 

reactant ion
—0— 1 H/D exchange

—l—' 2 H/D exchanges
3 H/D exchanges

    

relative intensity

 

81012141618 20 22 24 26 28 30

number of collision

Figure A-24 A simulated collision—product plot forthe structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 60% react1ve

collision.

 

170

1.0 1! Structure II at 65% reactive collision

  

reactantion
—0— 1 H/D exchange

—I— 2 H/D exchanges
3 H/D exchanges

   
    
 
    

relative intensity

 

10 12 14 16 18 20 22 24 26 28 30

number of collision

Figure A-25 A simulated collision-product plot for‘the structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 65% react1ve

collision.

171

1-0 'A Structure II at 70% reactive collision

      
  
 

reactant
—O— 1 H/D exchange
——I— 2 H/D exchanges

3 H/D exchanges

     

relative intensity

 

81012141618 20 22 24 26 28 30

number of collision

Figure A-26 A simulated collision-product plot fonthe structure II of o-
dichlorobenzene based on an H/D exchange react1v1ty of 70% react1ve

collision.

 

 

172

A mixture of 55% structure VI, 25% structure V

 

1.0 i and 20% structure II at 20% reactive collision
0.8 _ —A— reactant ion
—0— 1 H/D exchange

4'3" —I_ 2 H/D exchanges
z: 0.6 _ —9— 3 H/D exchanges
Q)
4.:
$21
~I—1
Q)
.2 0.4 -
4.)
CU
l—i
Q)
f-d

0.2 ‘

° n D . I I I I I I I lfifi

 

0.0V” 5 :r-I
6

... . . .
8 10 12 1416 18 20 22 242628 30

number of collision

Figure A-27 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 20% reactive
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure II.

173

A mixture of 50% structure VI, 25% structure V
and 25% structure II at 20% reactive collision

1.0 i

0 8 .. —_A— reactant ion
—0— 1 H/D exchange
—I— 2 H/D exchanges

0 6 - —'fl_ 3 H/D exchanges

relative intensity

 

 

l w l u l v I u *I—Vfi

2 4 6 810 12 14 16 18 202224262830

0.0\’"'¥""f"l'l'l'l‘l'l'l

number of collision

Figure A-28 A simulated collision—product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 20% react1ve
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

 

174

 

1.0 i A mixture of 55% structure VI, 25% structure V
and 20% structure II at 25% reactive collision
0.8‘ —A— reactantion
—O— 1 H/D exchange
>> + 2 H/D exchanges
~13 —‘9— 3 H/D exchanges
to
g: 0.6 "
a)
4.)
Ci
0H
s
33 0.4-
cc
r—d
0)
$4
0.2 '
00<"";“"'l;i;l'l'l'l'l'l'll'l'l

 

l. '
6 810 12 1416 18 2022 24 2628 30

number of collision

Figure A-29 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m—dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 25% react1ve
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure II.

 

 

175

A mixture of 50% structure VI, 25% structure V

 

1‘0 i and 25% structure II at 25% reactive collision
08- '—A— reactant ion
——O- 1H/Dexchange
:1? A ‘—l— 2H/Dexchanges
g1) 06' —'3— 3H/Dexchanges
8 x
53
OH ‘
<1)
.> o ' ° ’ o
H ‘ - .
8 o ‘ "m...
A‘.‘6‘;“AAAAA‘AAAA
0.2" . o . .
o o .‘_ a-” a n
q nnflununauv...
0.0J;H$HT*INI'I'III'I'I'I-I'I-Ivfi
0 4 6 8 10 12 14 16 18 20 22 24 26 28 30
number of collision
Figure A-30 A simulated collision-product plot for a mixturemff
e

dichlorobenzene anions originated from m-dichlorobenzene.

calculations are based on an H/D exchange reactivity of 25% reactive
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

 

 

176

 

A mixture of 55% structure VI, 25% structure V
1.0 .1 and 20% structure II at 30% reactive collision

I —A—‘ reactant ion

0.8 " —0— 1 H/D exchange
>, —I— 2 H/D exchanges
5‘53 —‘3— 3 H/D exchanges
C1 0.6 '
Q)
4..)
Ci
--—I
Q)
E 0.4 -
£0
H
Q)
h

0.2 “

0.0 " "' $ "' T ' I ' I ' I ' I ' I ' I ' I ' I ' I ' I I ' I fl

 

 

0 2 4 6 81012141618202224262830

number of collision

Figure A-31 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 30% reactive
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure II.

0.8 "

0.6 ‘

0.4 '

relative intensity

0.2 '

 

Q‘s—m

Figure A-32 .
dichlorobenzene anions originated from m-dlchlorobenzene.

177

A mixture of 50% structure VI, 25% structure V
and 25% structure II at 30% reactive collision

—A— reactant ion
—0_ l H/D exchange
+ 2 H/D exchanges
'—G— 3 H/D exchanges

'l'l'l'l'l'l'l'l'l'l'l

T II—I—l
4 6 810 12 1416 18 20 22 2426 28 30

N-ll

number of collision

A simulated collision—product plot for a mixtureTIff
e

calculations are based on an H/D exchange reactivity of 30% reactive
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

178

A mixture of 55% structure VI, 25% structure V
and 20% structure II at 35% reactive collision

 

 

1.01
—A— reactantion
08- —0— 1H/Dexchange
+ 2H/Dexchanges
f; —G— 3H/Dexchanges
'6
Ci 0.6-
8 A
Ci
”-4
<1) 0 .0
35 0.4- . , '..
‘ o
H . ““-.AAAAAAAAAAAAAAAAAA
.
0.2- 0..
..._nuu0°un°°u
a a a a u a 9" O
a II a all ........
0.0‘HV"|1HI'Il'l'l'l'l'l'l'l‘T—fi
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
number of collision
Figure A-33 A simulated collision-product plot for a mixtureTlff
e

dichlorobenzene anions originated from m-dichlorobenzene. .
calculations are based on an H/D exchange react1v1ty of 35% react1ve
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure II.

179

A mixture of 50% structure VI, 25% structure V
and 25% structure II at 35 % reactive collision

1.0 i
—‘A— reactant ion
—0— 1 H/D exchange
08 - + 2 H/D exchanges
>3 —E'— 3 H/D exchanges
<1) 0.6 '
43
C1
~r-t
Q)
.2
+3 0.4 '
CU
H
G)
3.4
0.2 "
0.0 \ "" ‘1" "' I ' I ' I ' I ' I ' I ' I ' I ' I jfi—l

 

 

I ' I ' I '
2 4 6 8 1O 12 14 16 18 20 22 24 26 28 30

number of collision

Figure A-34 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 35% react1ve
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

 

180

A mixture of 55% structure VI, 25% structure V
1_0 i and 20% structure II at 40% reactive collision

—A— reactant ion

08" —0— 1H/Dexchange
_'— 2H/Dexchanges
3;, ‘ —9‘— 3H/Dexchanges
"-4
a 0.6-
Q)
49 A
.5
<1.) ..0
a 0.4‘ . ‘ o
13’ ‘ °
'r - ‘ ‘1
<1) A C
H ‘ ‘ A‘.“ A A A A A A A A A A A A A A A A A A A
O
0.2- ‘.
O a n a u
.Oa.nou°°uuun
n u u a 9 a '. . .
Dana 000......
I

 

 

0.04:“;Vll'l'l'l‘l'l'I'l'l'l' Iifi
2 4 6 8 101214161820 22 24 26 28 30

number of collision

Figure A-35 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 40% reactive
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure II.

181

A mixture of 50% structure VI, 25% structure V

 

 

 

1.0 1 and 25% structure II at 40% reactive collision
+ reactantion

Q8— —0— 1H/Dexchange
E9 ‘ + 2H/Dexchanges
'a —9_ 3H/Dexchanges
g 0.6-
4.3
$21 I
°H
g

o

'13 0.4-
CU
H
8 o

0.2- 0 , a . ., . .

. - n u u u u a u u I:
a II n n D a- . .
a u C
J Dunn .‘O.......
0-0 ‘: "' 'l' V ' I ' I ' I ' I ' I ' I ' I ' I ' ' I ' I ' I fl
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

number of collision

Figure A-36 A Simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 40% reactive
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

 

 

182

A mixture of 55% structure VI, 25% structure V
and 20% structure II at 45% reactive collision

 

 

 

1.0-t
—A— reactant ion

08‘ —O— 1H/D exchange

' ——I— 3H/D exchanges
.4? —G-— 2H/D exchanges
to
g 0.6— a n D D D u u n n U
4-7 n n '3 9 0
Ci
~I-(
S
oI—I
+> 0.4-
Cd
0—4
G)
h

0.2" o .

" o
O -
. a , . . .
. . O . O . . . . .
00 -;"'l'I'I'I‘I'I'l'l'l'l'I'l'l'fi
0 2 4 6 81012141618202224262830

number of collision

Figure A-37 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 45% reactive
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure II.

183

A mixture of 50% structure VI, 25% structure V
1.0 i and 25% structure II at 45% reactive collision

 
 

——A—
_o_
0.8-
——G—
3
"-1 A
«2
q, 0.6-
4.)
:1
-H
<1) ‘0
.2 ’
4&5? 0.4
r—1
0)
H

“.AAAAAAAAAAAAAAAAAAAAA

 

number of collision

reactant ion

1 H/D exchange
2 H/D exchanges
3 H/D exchanges

 

0 2 4 6 81012141618202224262830

Figure A-38 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange react1v1ty of 45% react1ve
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

 

184

A mixture of 55% structure VI, 25% structure V

1.0 fl and 20% structure II at 50% reactive collision

—A— reactant ion
—0— 1 H/D exchange
0.8 ' —I— 2 H/D exchanges
—D— 3 H/D exchanges

 
 

{>3

4.)

-H

a A

a) 0.6“

4..)

.S

g) 0 O
'5 0.4- .
CU

I—I

Q)

I-(

.

l .

‘A
‘.AAAAAAAAAAAAAAAAAAAAAA

 

 

0 2 4 6 8 10 12 1416 18 20 22 2426 28 30

number of collision

Figure A-39 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 50% reactive
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure II.

185

A mixture of 50% structure VI, 25% structure V

1“ and 25% structure II at 50% reactive collision
1.0

—_A_ reactant ion

0.8 ' —O— 1 H/D exchange
—I— 2 H/D exchanges
_'3— 3 H/D exchanges

0.6 "

 
 

0.4 '

relative intensity

‘ ‘ . A A A A A A A A A A A A A A A A A A A A A A

 

 

0 2 4 6 81012141618202224262830

number of collision

Figure A-40 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 50% react1ve
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

 

186

A mixture of 55% structure VI, 25% structure V

1 0 fl and 20% structure II at 55% reactive collision

——A—' reactant ion
_0— 1 H/D exchange
0.8 " + 2 H/D exchanges
—G_ 3 H/D exchanges

0.6‘ A

relative intensity

  

 

0 2 4 6 810 12141618 20222426 2830

number of collision

Figure A—41 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 55% react1ve
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure II.

187

A mixture of 50% structure VI, 25% structure V
and 25% structure II at 55% reactive collision

1.0 11
—A'— reactant ion
0 8 _ —'0_ 1 H/D exchange
' —I'— 2 H/D exchanges
—'3— 3 H/D exchanges
0.6 " A

relative intensity

 

 

0 2 4 6 81012141618202224262830

number of collision

Figure A-42 A simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 55% react1ve
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

 

188

A mixture of 55% structure VI, 25% structure V
1.0 fl and 20% structure II at 60% reactive collision

——A— reactant ion
—0— 1 H/D exchange

0.8 '

>‘ + 2 H/D exchanges
E —9— 3 H/D exchanges
8

p 0.6 -

g A

u—I

G)

>

'4'3

CU
I—(

(D

3.4

 

 

0 2 4 6 8 10 12 14 16 18 2O 22 24 26 28 30

number of collision

Figure A-43 A simulated collision~product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactivity of 60% reactive
collision. The mixture contains anions of 55% structure VI, 25% structure

V and 20% structure 11.

 

189

A mixture of 50% structure VI, 25% structure V
1.0 fl and 25% structure II at 60% reactive collision

—fi_ reactant ion
08 — _0_ 1 H/D exchange
"—I— 2 H/D exchanges

—E’_ 3 H/D exchanges
0.6 "

relative intensity

 

 

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

number of collision

Figure A-44 A Simulated collision-product plot for a mixture of
dichlorobenzene anions originated from m-dichlorobenzene. The
calculations are based on an H/D exchange reactiv1ty of 60% react1ve
collision. The mixture contains anions of 50% structure VI, 25% structure

V and 25% structure II.

 

 

190

 

—fi— reactant ion
—0— 1 H/D exchange

relative intensity

 

 

0.0l'l'I‘I'l'l‘I'I'I'l'I'I'I'l'Ifi—l

0 2 4 6 81012141618202224262830

 

number of collision

Figure A-45 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange react1v1ty of 5% react1ve

collision.

191

 

reactant ion

1 H/D exchange

relative intensity

 

 

 

024681012141618202224262830

number of collision

Figure A-46 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange react1v1ty of 10% react1ve

collision.

192

Structure VI at 20% reactive collision

—'A— reactant ion
—0— 1 H/D exchange

relative intensity

 

 

 

0 2 4 6 81012141618202224262830

number of collision

Figure A-47 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange react1v1ty of 20% react1ve

collision.

 

193

Structure VI at 25% reactive collision

+ reactant ion
'—0— 1H/Dexchange

relative intensity

 

 

0 2 4 6 8 10 12 14 16 18 20 2224262830

number of collision

Figure A-48 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange react1v1ty of 25% react1ve

collision.

194

Structure VI at 30% reactive collision

--£r- reactantion
'—0— 1 H/D exchange

relative intensity

 

 

024681012141618202224262830

number of collision

. . . . VI of p-
Fi re A-49 A Simulated colliSion-product plot for the structure .
dighllorobenzene based on an H/D exchange reactiv1ty of 30% reactive

collision.

195

Structure VI at 35% reactive collision

‘—A— reactant ion

—0— 1 H/D exchange

relative intensity

 

 

0 2 4 6 810 1214 1618202224 262830

number of collision

Figure A-50 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange reactiv1ty of 35% reactive

collision.

 

196

Structure VI at 40% reactive collision

—A'— reactant ion
——0— 1 H/D exchange

relative intensity

 

 

0 2 4 6 8 10 1214 1618 20 22242628 30

number of collision

Figure A-51 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange reactiv1ty of 40% reactive

collision.

 

197

Structure VI at 45% reactive collision

——fl—' reactant ion
—0— lH/Dexchange

relative intensity

 

 

0 2 4 6 8 1 12 14 16 18 20 22 24 26 28 30

number of collision

Figure A-52 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange reactiv1ty of 45% reactive

collision.

 

198

Structure VI at 50% reactive collision

——A— reactant ion
_0— 1 H/D exchange

relative intensity

 

 

0 2 4 6 8 10 12 14 16 1820 22 24 2628 30

number of collision

Figure A-53 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange reactivity of 50% reactive

collision.

199

Structure VI at 55% reactive collision

—A'— reactant ion
—0— 1 H/Dexchange

relative intensity

 

 

0 2 4 6 81012141618202224262830

number of collision

Figure A-54 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange reactivity of 55% reactive

collision.

Structure VI at 60% reactive collision

.0,

0.8 "

       

__A__.

 

reactant ion
0.6 r
—0— 1 H/D exchange

0.4 '

relative intensity

0.2 '

 

number of collision

Figure A-55 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange reactiv1ty of 60% reactive

collision.

 

Structure VI at 65% reactive collision

’—‘£F" reactantion
—O— 1 H/D exchange

relative intensity

 

 

024681012141618202224262830

number of collision

Figure A-56 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange react1v1ty of 65% reactive

collision.

 

 

Structure VI at 70% reactive collision

‘_A— reactant ion
—O—— 1 H/D exchange

relative intensity

 

 

0 2 4 6 8 10 12 1416 18 20 22 2426 28 30

number of collision

Figure A—57 A simulated collision-product plot for the structure VI of p-
dichlorobenzene based on an H/D exchange reactivity of 70% reactive

collision.

 

 

1|11111!1111111111Hi1 1 ’ HIIIIHM
312930090818

111
W
56