DYNAMICS AND KINETICS OF

ION IMPACT DISSOCIATION OF

MOLECULES

Thesis submitted to the

University of Calcutta
for the degree of
Doctor of Philosophy (Science)

by
Mridula Rani Jana
Department of Physics
University of Calcutta
Kolkata, India

2013
To My Beloved Parents
Acknowledgments

Memories of experiences I have had over the past six years flashed across my mind
while writing this thesis. It would never be possible for me to complete this long journey
of research work without the kind help, inspiration, support and wise advices from many

people around me.

First of all I would like to thank Prof. Pradip N. Ghosh and Dr. Biswajit Ray for
their constant support, academic guidance and encouragement. The wise advices and

the fruitful discussions, that I have received from them throughout my research carrier,
have enriched me a lot. I owe my sincere gratitude to both of you for accepting me as
your student.

I am indebted to Dr. C. P. Safvan who first introduced me to the world of ex-

perimental physics and who always have shown great enthusiasm, patience, and joy in
answering all my numerous questions. His great knowledge and skill of experiment and
his practical suggestion, not only for research but also for life, have helped to widen my
knowledge in various ways. I would certainly say that it has been great working with

you and I am lucky to have you as an advisor.

My warmest thanks are due to Dr. Bhas Bapat, with whom I have had many en-
lightening discussions especially during the long beam time hours. He has been very
inspiring, when it comes to scientific writing. Thank you for all the help and knowledge

you have given to me.

The work in this thesis was carried out in collaboration with two research groups
from Physical Research Laboratory (PRL), Ahmedabad and Inter University Accelera-

iii
ACKNOWLEDGMENTS iv

tor Centre (IUAC), New Delhi. I would like to give special thanks to I. A. Prajapati, S.
Banerjee, R. K. Kushawaha (now at Sorbonne University, Paris) and K. Saha of PRL

and to Dr. J. Rajput (now at Delhi University, India), P. Kumar, U. K. Rao, all the
staff members of LEIBF and Pelletron groups in IUAC. Thank you all for your great
hospitality, invaluable help and expertise in performing the experiments and analysis.

I have been immensely benefited through helpful discussions with Dr. Amit Roy. I

express my thanks to you for providing the necessary facilities at LEIBF and Pelletron
for this experimental program. I would also like to convey my thanks to all my teachers
especially, in the department of Physics of University of Calcutta, for their inspiration,
advices and encouragement.

I owe my sincere thanks to my lab–mates Ajit and Sairam for their dedicated
support and the interesting discussions I had with them. I am deeply indebted to my
colleagues and friends, S. Sunil Kumar, Diana, Vicky, Nupur, Srashti, Sunil, Narender,

Archana and Geetanjali, who have been very kind and supportive to me and have helped
me a lot in dealing with both academic and non–academic matters. They have made my
life at IUAC, an interesting one. Mabud, Priyanka, Soma, Saswati and Swarup had al-
ways kept me in high spirits with stimulating interactions in the department. I am very
much thankful to you also.

Financial support in the form of fellowship received from UGC and CSIR at vari-
ous stages of my Ph.D. career, is gratefully acknowledged.

I want to dedicate this thesis to my parents, the most important people in my life.
There are not words enough to express the gratitude I feel towards them for their un-
demanding, unselfish and never waning support throughout all my choices in life. I am
also grateful to my little brother, Mrinal, for taking care of my parents in my absence,

which has enabled me to focus on what I want to do. The words of encouragement
from him have always helped me in beating the challenges of my Ph.D. life. There is
no match for the affectionate support, well-wishes, prayers and the encouragement that
my beloved grand parents, my family members and my in laws showered on me. None
ACKNOWLEDGMENTS v

of the words are adequate to express my gratitude towards each and everyone of them.
And last, but certainly not least, I would like to thank my husband, Saikat. When I

was about to quit and give up, he convinced me to have patience and encouraged me to
continue the work. Everything that I have accomplished up to this point academically
and in life is due to his seeing potential in me. His skills in C++ and programming made
my data analysis job much easier. I want it to be documented that my success is actually
his success and that this degree has been earned by him as well. Thanks sweetheart, for

keeping me on the right track and for making everyday the best day.

———————————–

(MRIDULA RANI JANA)

Department of Physics
University of Calcutta
Kolkata
Contents

Acknowledgments iii

Synopsis x

List of Publications xiii

1 Introduction 1

1.1 Ionization and excitation of a molecule . . . . . . . . . . . . . . . . . . 2

1.1.1 Photoionization . . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.1.2 Electron impact ionization . . . . . . . . . . . . . . . . . . . . 4

1.1.3 Ion impact ionization . . . . . . . . . . . . . . . . . . . . . . . 5

1.1.4 Comparison of electron, photon and ion induced ionization of

molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

1.2 Ion induced molecular dissociation . . . . . . . . . . . . . . . . . . . . 8

1.3 Overview of ion induced molecular dissociation technique . . . . . . . 10

1.3.1 Production and acceleration of projectile ions . . . . . . . . . . 11

1.3.2 Various ion – molecule collision setup . . . . . . . . . . . . . . 15

1.3.3 Separation of fragment ions by mass spectrometers . . . . . . . 17

1.3.4 Overview of imaging detectors . . . . . . . . . . . . . . . . . . 21

1.4 Review of ion molecule collision studies . . . . . . . . . . . . . . . . . 27

1.5 Motivation and thesis layout . . . . . . . . . . . . . . . . . . . . . . . 29

vi
CONTENTS vii

2 Experimental Techniques and data analysis 41

2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

2.2 Ion beam from accelerator . . . . . . . . . . . . . . . . . . . . . . . . 42

2.2.1 Pelletron . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

2.2.2 Low Energy Ion Beam Facility (LEIBF) . . . . . . . . . . . . . 44

2.3 The molecular dissociation apparatus . . . . . . . . . . . . . . . . . . . 45

2.3.1 Ion – molecule interaction zone . . . . . . . . . . . . . . . . . 47

2.3.2 The linear time–of–flight mass spectrometer (TOFMS) . . . . . 48

2.3.3 Design of the TOFMS . . . . . . . . . . . . . . . . . . . . . . 54

2.3.4 Optimization of TOFMS extraction voltage . . . . . . . . . . . 56

2.4 Position sensitive ion imaging detector . . . . . . . . . . . . . . . . . . 58

2.5 The electron detector . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

2.6 Signal Processing and data acquisition . . . . . . . . . . . . . . . . . . 60

2.7 Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

2.7.1 Analyzing TOF spectrum . . . . . . . . . . . . . . . . . . . . . 65

2.7.2 Coincidence map . . . . . . . . . . . . . . . . . . . . . . . . . 67

2.7.3 Analyzing position spectrum . . . . . . . . . . . . . . . . . . . 70

2.7.4 Momentum calculation . . . . . . . . . . . . . . . . . . . . . . 71

2.7.5 Geometry of the parent molecular ion . . . . . . . . . . . . . . 72

2.7.6 Kinetic energy release . . . . . . . . . . . . . . . . . . . . . . 74

2.8 Efficiency of a TOFMS . . . . . . . . . . . . . . . . . . . . . . . . . . 75

3 Dissociation dynamics of carbon dioxide (CO2 ) 80

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

3.1.1 Prior work on fragmentation of CO2 . . . . . . . . . . . . . . . 81

3.2 Dissociation of CO2 by low energy ion impact: Experimental details . . 82

3.3 Analysis of dissociation dynamics of CO2 . . . . . . . . . . . . . . . . 83

3.3.1 TOF spectrum of dissociated CO2 . . . . . . . . . . . . . . . . 83

CONTENTS viii

3.3.2 Coincidence spectrum of CO2 showing various fragmentation

pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

3.4 Triple ionization and fragmentation of CO2 by swift heavy ion . . . . . 93

3.4.1 Triple fragmentation kinematics . . . . . . . . . . . . . . . . . 94

3.4.2 Ab initio quantum chemical calculation . . . . . . . . . . . . . 98

3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

4 Low energy ion induced fragmentation of carbonyl sulphide (OCS) 108

4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

4.2 Experimental details . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

4.3 TOF spectrum of dissociated OCS . . . . . . . . . . . . . . . . . . . . 110

4.4 Coincidence map showing various fragmentation pathways of OCS . . . 111

4.4.1 Dissociation of OCS2+ . . . . . . . . . . . . . . . . . . . . . . 113

4.4.2 Dissociation of OCS3+ . . . . . . . . . . . . . . . . . . . . . . 119

4.4.3 Dissociation of OCS4+ . . . . . . . . . . . . . . . . . . . . . . 126

4.5 Effect of projectile charge state on dissociation dynamics of OCS . . . . 127

4.5.1 Effect of projectile charge state on TOF distribution of corre-

lated ion-pairs . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

4.5.2 Effect of projectile charge state on KER distribution of a disso-

ciation channel . . . . . . . . . . . . . . . . . . . . . . . . . . 130

4.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

5 Fragmentation of CO2 and OCS by high energy highly charged ion impact:
A comparative study 137

5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

5.2 Experimental details . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

5.3 Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 140

5.3.1 Two–body dissociation of OCX2+ . . . . . . . . . . . . . . . . 141

5.3.2 Three–body dissociation of OCX2+ . . . . . . . . . . . . . . . 143

CONTENTS ix

5.3.3 Three–body dissociation of OCX3+ . . . . . . . . . . . . . . . 147

5.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

6 Dissociation dynamics of sulphur hexafluoride (SF6 ) 158

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158

6.2 Experimental details . . . . . . . . . . . . . . . . . . . . . . . . . . . 160

6.3 Time–of–flight spectrum of SF6 . . . . . . . . . . . . . . . . . . . . . 161
6.4 Coincidence spectra of SF6 . . . . . . . . . . . . . . . . . . . . . . . . 162
6.5 Dissociation of SF2+
6 . . . . . . . . . . . . . . . . . . . . . . . . . . . 164

6.5.1 Two–body dissociation . . . . . . . . . . . . . . . . . . . . . . 164

6.5.2 Three–body dissociation . . . . . . . . . . . . . . . . . . . . . 165
6.5.3 Many–body dissociation . . . . . . . . . . . . . . . . . . . . . 167
6.6 Dissociation of SF3+
6 . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
6.6.1 Double–ion coincidence . . . . . . . . . . . . . . . . . . . . . 171

6.6.2 Triple–ion coincidence . . . . . . . . . . . . . . . . . . . . . . 174

6.6.3 Coincidence between F+ :F+ :F+ . . . . . . . . . . . . . . . . . 177
6.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

7 Summary and Future outlook 187

Synopsis

A molecule is a quantum mechanical bound system in which electrons and nuclei are
dynamically arranged in such a way that the system remains stable. When high en-
ergy projectile ions collide with these neutral molecules several electrons are stripped
of and multiply charged transient molecular ions are formed in the excited states of

the molecule at the equilibrium internuclear distance (Re ). Nuclear repulsion energy
dominating over binding electronic energy makes the multiply charged molecular ions
unstable and compels them to dissociate instantly into copious quantities of energetic
charged atomic fragments. A multiply charged molecular ion possesses a set of config-
urations with certain symmetries. These configurations have different electronic ener-

gies, depending on which, a molecular ion follows different dissociation pathways. Un-
derstanding the dynamics and kinetics of ion induced dissociation of multiply charged
molecular ions remains an open question and continues to pose a significant theoretical
and experimental challenge. In this dissertation, fragmentation dynamics of multiply

charged molecular ions created by the impact of low as well as high energy projectile
ions on neutral molecules has been investigated to understand the underlying physics
of ion induced molecular dissociation processes. To get the properties of the electronic
states of precursor ion and to understand the energetic of the break up, it is neces-

sary to measure complete kinematic parameters of all ionic fragments. For the present
experiments we have employed multihit time–of–flight coincidence technique to gain
knowledge in dissociation of molecules having different symmetries. Toward this goal,
ion induced fragmentation of three different molecules have been studied in the present

x
synopsis xi

work: two linear triatomic molecules, one having a center of symmetry (CO2 ) and the
other with no center of symmetry (OCS), along with a highly symmetric nonlinear (oc-

tahedral) polyatomic molecule (SF6 ).

Chapter 1 introduces the subject of ion–induced molecular dissociation and gives

an extensive overview of the molecular dissociation techniques adopted by various
groups in recent years. In the course of our discussion, we mention the various ioniza-

tion and excitation techniques of a molecule and give a comparison between electron-,
photon- and ion–induced ionization processes. We have also discussed the previously
obtained experimental results on ion induced molecular dissociation starting from di-
atomics to polyatomics in this chapter.

Chapter 2 describes the ion accelerator facilities and molecular dissociation appa-
ratus used for the present experiments. The apparatus consists of a time-of–flight–mass
–spectrometer (TOFMS) equipped with a position-sensitive multi-hit detector system.

Description of each part of the setup is given in detail with special emphasis on its
multi-coincidence detection capabilities. We have presented in detail an optimization
technique of the TOFMS and discussed how the efficiency of the spectrometer has been
tested using collision-induced spectra of Ar and N2 molecules by various energy ion
beams. This chapter also contains a brief introduction to momentum imaging tech-

nique, data acquisition methodology and offline analysis technique of the collected data
to extract information on different fragmentation processes.

Chapter 3 describes the results on dissociation dynamics of CO2 by low energy

ion impact. Another result of particular importance presented in this chapter is the
mechanism behind the formation and dissociation of CO3+
2 , induced by high energy

highly charged ion beam, leading to triple–ion coincidence. The experimental results
have been complemented by ab initio quantum chemical calculations that account for

the observed kinetic energy release distribution.

In Chapter 4 we have investigated the concerted and sequential dissociation chan-

nels of OCSq+ (q= 2 – 4) formed in the collision of 150 keV Ar+ ion with neutral OCS.
synopsis xii

From the momentum correlation map we estimated the geometry of the parent molec-
ular ions prior to dissociation and inferred that bent geometrical states are involved for

most of the fragmentation channels of OCS2+ and OCS3+ ions. The experimental ki-
netic energy release values for the complete Coulomb fragmentation channels of OCS
are compared with those calculated from the Coulomb explosion model. This study
provides the first results on dissociation dynamics of OCS under the impact of low en-
ergy ions. The effect of projectile charge state on fragmentation of OCS has also been

studied here.
Chapter 5 presents a comparative study on dissociation dynamics of CO2 and OCS
by high energy highly charged ion beam. Here the role of symmetry on dissociation of
two linear triatomic molecules has been investigated. We observed that dissociation

of CO2+
2 and OCS
2+
occurs either simultaneously or in sequential ways. But triply
charged ions of the same molecules always dissociate in a concerted manner. From
triple ion fragmentation it has been observed that the kinetic energy carried away by
the O+ ion is much larger than the energy associated with the S+ ion of OCS3+ . But,

due to its symmetric structure, the two O+ ions of CO3+

2 carry equal amount of kinetic

energies.
Chapter 6 deals with the third molecule of our interest, SF6 , a highly symmetric
polyatomic molecule. The analysis of the experimental results that led to the under-
standing of some important dissociation channels of SF2+ 3+
6 and SF6 is presented in this

chapter. We have also explored the nature of explosion of three F+ ions from SF3+
6 and

estimated the geometry of the precursor molecular ion for this particular dissociation
channel.
The thesis is concluded by a short chapter (Chapter 7) summarizing all the results

we have obtained and with few suggestions for the continuation of the work.
List of Publications

1. Triple F+ ejection from SF6 bombarded by swift ions – R. K. Kushawaha, S.

Sunil Kumar, M. R. Jana, I. A. Prajapati, C. P. Safvan and B. Bapat, J. Phys.
B:At. Mol. Opt. Phys., 43, 205204 (2010).

2. Dissociation of carbonyl sulfide by 150 keV Ar+ ion impact – M. R. Jana, B.

Ray, P. N. Ghosh and C. P. Safvan, J. Phys. B:At. Mol. Opt. Phys., 43, 215207
(2010).

3. Ion-induced triple fragmentation of CO3+ + + +

2 into C + O + O – M. R. Jana, P.

N. Ghosh, B. Bapat, R. K. Kushawaha, K. Saha, I. A. Prajapati and C. P. Safvan,

Phys. Rev. A., 84, 062715 (2011).

4. Fragmentation of OCS by 5 MeV u−1 Si12+ ions – M. R. Jana, P. N. Ghosh, B.

Ray, B. Bapat, R. K. Kushawaha, K. Saha, I. A. Prajapati and C. P. Safvan (in
preparation).

xiii
Chapter 1

Introduction

Atomic and molecular processes occur all around us. Elementary collisions involving

atoms, ions and molecules take place in many gaseous and plasma environments. A
large number of polyatomic molecules or molecular species have been discovered in
the interstellar medium and atmosphere of many planets [1]. Several exotic molecules
or radicals have been found in space which are not normally available on the earth.
The origin of many of these molecules has yet to be explained. Various kinds of satel-

lites, space stations and manned airships are operated in the earth space, surrounded
by the solar wind and controlled by the earth’s magnetic field. Therefore collisions be-
tween the particles of solar wind and other heavy particles are possible in this region.
In ion molecule interaction, molecular intermediates may form which act as precursors

for formation of other stable molecules. Hence ion impact on molecules, which are
abundant in the astrophysical environment, needs to be studied for understanding the
evolution of our galaxy and even the origin of life on earth [2]. The interaction of ions
with atoms and/or molecules is not only of fundamental importance but also essential

for detail understanding of a broad range of large-scale physical systems such as, plan-
etary atmospheres, astrophysical plasmas, semiconductor processing plasmas, fusion
plasmas and gas discharge lasers. Collision of ions with matter and with biological
molecules is important to ion beam therapy as well [3]. The radio–biological effects of

1
Introduction 2

highly charged ion beams are of interest for tumor therapy and for radiation protection
in space. Dissociation of small molecules, like, building blocks of DNA, plays key roles

in understanding the radiation induced damage of living tissues [4]. Fragmentation of

H2 O is also important in biological tissues because water radiolysis leads to cell death
or mutation [5, 6].

Ion atom collision studies are often simpler compared to ion molecule collisions.
The number of electrons captured by the projectile ion from an atom can be easily ob-
tained from the charge state of the recoil ion using coincidence between emitted elec-

trons, charge exchanged projectile and recoil. But this fundamental information may be
lost in case of ion molecule collision. When a polyatomic molecule is excited by some
perturbation and dissociates due to extreme instability, many electrons and atomic or
molecular fragments may form. The complete kinematic information about each of the
fragments ejected from the dissociating molecule is difficult to measure by traditional

techniques. Using multi-hit coincidence techniques, the lost information can be recov-
ered, provided no fragments are lost from the system. In this way if proper analysis is
made of the fragmentation patterns, molecular targets can give additional information
on the energy transfer to the target accompanying multiple capture processes.

1.1 Ionization and excitation of a molecule

A molecule may be thought of as a collection of nuclei moving in the mean field of

electrons, with overall charge neutrality. It may possess many sorts of energies like ro-
tational, vibrational and electronic by virtue of its rotation about the centre of mass, pe-
riodic displacement of constituent atoms from their equilibrium position and unceasing

motion of associated electrons respectively [7]. When sufficient energy is transferred to

a neutral molecule, one or few electrons may be ejected from it, resulting into the for-
mation of a singly or multiply charged molecular ion. The energy may be transferred
to the molecule by photon or charged particle impact on it. The excited or ionized
Introduction 3

molecule then rearranges itself to attain a minimum energy configuration. In the ab-
sence of a stable configuration, or when the configuration is not bound, the molecular

species disintegrates into ionic or neutral fragments. A crucial component in molecu-

lar fragmentation study is the method and the source of ionization. A broad variety of
sources are available, suitable for different techniques and for probing different proper-
ties of the electronic structure of the molecule. The target molecule can be bombarded
by photons, lasers, electrons or ions.

1.1.1 Photoionization

Photons are electromagnetic radiation with zero mass, zero charge, and a velocity which

is equal to the speed of light. Being electrically neutral, it does not lose energy via Cou-
lomb interaction with molecular electrons as do the charged particles. Instead, they un-
dergo a “catastrophic” interaction that radically alters the properties of the incident pho-
ton [8]. In all cases of practical interest, the interaction results in partial or total transfer
of incident photon energy to the electrons of the constituent atoms. There are three ma-

jor mechanisms by which photons interact with matter: Photoelectric effect, Compton
scattering and Pair production. Absorption of photon, below ionization threshold en-
ergy, involves transition of a single electron from one electronic state to other obeying
the angular momentum selection rule. The photoelectric interaction occurs when the

low energy photon (in general a few eV to a few keV) imparts all its energy to an orbital
electron and vanishes. If the energy of the photon is greater than the binding energy
of the electron, then the electron gets ejected from its host atom. Compton process is
important in the photon energy range 100 keV to 10 MeV. In Compton scattering, the

incoming photon is deflected from its original path by interaction with an electron. The
photon transfers a portion of its energy to the electron which is then ejected as a recoil.
When the incoming photon energy exceeds 1.02 MeV, it may interact with the matter
by a process called Pair production. In this mechanism of energy transfer, an electron–
Introduction 4

positron pair is created with the annihilation of a photon, passing near the nucleus of
an atom. Photons, required for photoionization of a molecule, are generally available

from discharge lamp, laser or storage rings emitting synchrotron radiation. Absorption
of photon by neutral molecule results in excitation of the molecule which undergoes re-
laxation either by emission of light or emission of electrons. At low intensity, removal
of electron is achieved when the energy of a photon exceeds the ionization potential.
At higher intensities, possibility of multiphoton absorption increases. If the intensity

is further increased, the potential energy surfaces of the molecule are suppressed to the
point where an electron tunnels through the potential barrier. This type of ionization is
known as ‘tunneling ionization’. Advent of high intensity, ultrashort laser pulses (more
recently, attosecond pulses) and improving laser technology have opened up new vistas

for experimental research into multiple ionization of gas phase molecules. Using lasers
to image and control complex chemical reactions remains an intense pursuit of con-
temporary molecular physics [9]. Molecular fragmentation has been investigated using
synchrotron radiation [10, 11] and femtosecond lasers [12]. Photoelectron spectroscopy

is used to measure the energy of an ejected photoelectron and to define the exact state
of the molecular ion, created by the photoionization event. In photoelectron-photoion
coincidence (PEPICO) measurement, the energy analyzed electron is recorded together
with the photoion. This provides knowledge on how the molecular fragmentation pro-
cesses depend on the electronic transitions [13]. Bandrauk et. al. had given a recent

update on photon induced processes in their book [14].

1.1.2 Electron impact ionization

When a neutral molecule is hit by an energetic electron carrying several tens of electron-
volts (eV) of kinetic energy, some of the energy of the electron is transferred to the
neutral molecule. If the electron, in terms of energy transfer, collides less effectively
with a molecule, it will bring the molecule into an electronically excited state without
Introduction 5

ionizing it but in case of very effective interaction the energy transferred may exceed
the ionization energy resulting into the formation of a molecular ion. Therefore electron

ionization technique can be used to investigate the formation and fragmentation of mul-
tiply charged molecules. The electron could also be captured by the neutral molecule to
form a negative radical ion. However, it is very unlikely to occur with electrons of high
energy. It proceeds effectively with thermal electrons [15]. Wang et. al. [16] analyzed
various dissociation channels of singly and doubly charged alkanes, as well as triply

charged propane and n-butane induced by 200 eV electron. Electron impact studies
had recently been extended to perform recoil ion momentum spectroscopy (RIMS) for
ionization and dissociation of molecules using a position sensitive detector coupled to a
time of flight setup [17, 18]. With electron bombardment on a sample one can study the

Auger electron spectra or electron energy loss spectra of the chosen samples. In Auger
electron spectroscopy an electron emitted from the target during the decay of a core-
hole state is analyzed. Whereas in electron energy loss spectroscopy the inelastically
scattered original electron and its loss of energy are measured [19].

1.1.3 Ion impact ionization

Ion impact is another efficient way to produce molecular ions in different excited states
[20, 21, 22, 23]. In collision of highly charged ions (HCI) with neutral molecules several
electrons may be stripped of and multiply charged transient molecular ions may be
formed in the excited states of the molecule at the equilibrium inter-nuclear distance

(Re ). Charge transfer processes between projectile ions (Iq+ ) and neutral molecules
(AB) may take place in multiple ways:

I q+ + AB → I q+ + AB + + e (Direct ionization) (1.1)

I q+ + AB → I (q−1)+ + AB + (Capture) (1.2)

I q+ + AB → I (q−1)+ + AB 2+ + e (T ransf er ionization) (1.3)

Introduction 6

Figure 1.1: Archetypal potential energy curves of a diatomic molecule (XY) along with
the schematic description of the kinetic energy release between the fragments X and Y.

I q+ + AB → I (q−2)+ + AB 2+ (1.4)

→ I (q−1)+ + AB 2+ + e (Capture autoionization) (1.5)

The transfer ionization process is defined as a process where one electron is transferred
from the target to the projectile and at the same time another electron is removed from

the target molecule. Ionization weakens the bond within the ion as compared to the
precursor neutral and results in longer bond lengths [15]. The nuclear repulsion energy
dominating over binding electronic energy makes the multiply charged molecular ions
unstable and compels them to dissociate instantly into copious quantities of energetic

charged atomic fragments. In dissociation of molecular ion, most of the excess energy
is converted into the kinetic energy of the fragments. This situation can be best viewed
by looking at the potential energy curve of a diatomic molecule (which will be replaced
by potential energy surfaces in case of polyatomic molecules) where energy is plotted
Introduction 7

as a function of internuclear separation. The stable, ground state potential energy curve
of a neutral diatomic molecule (curve A in Figure 1.1) and excitation of it to different

dissociative states (curves B and C in Figure 1.1) of multiply ionized precursor have
been shown in Figure 1.1. In principle, many repulsive states are possible in a single
precursor but which state will be accessed depends on the strength of the perturbation.
The further fate of the excited ion depends on the shape of its potential energy curve.
The molecular ions excited to curve B in Figure 1.1 possess no minimum on their po-

tential energy curve, therefore dissociate spontaneously as soon as they are formed. If
excited to the metastable state (curve C in Figure 1.1), there is a chance to observe
an undissociated molecular ion. It may exist there for a short period of time before
tunneling through the potential barrier releasing energy and dissociate.

1.1.4 Comparison of electron, photon and ion induced ionization of

molecules

An electron can be characterized by its mass, charge and magnetic moment, all of which

are fixed in magnitude and also by its energy and momentum, which are variable. On the
other hand, photons do not have mass and charge, but they have momentum. Relative to
the electron, photon has mostly energy, but not very much momentum. A considerable
advantage of photo ionization method, in comparison with electron ionization, is that

the amount of energy transferred to the target molecule is precisely known for most
of the cases. Inelastic collisions between an electron and a neutral molecule usually
result in the formation of a positively charged ion. Electron ionization serves the same
purpose as photoionization, but the situations are different in both the cases. There is

no restriction on the proportion of electron energy that may be transferred to the tar-
get molecule during collisions between electrons and molecules. The photons can be
viewed as packets of energy (quanta) that interact with atoms or molecules to produce
ionization even though they themselves possess no net electrical charge. Photoion-
Introduction 8

ization is also different from ion induced ionization of a molecule. In case of single
photon ionization energy deposited on the molecule is known and equals to the photon

energy. Therefore the interaction process is energy and momentum selective. By choos-
ing photon energy, we may select the molecular electron to be ionized. But for multi
photon ionization (for e.g. laser induced ionization) and charged particle interaction,
the situation is different. The amount of energy deposited on the molecule depends on
the number of absorbed photons in case of laser interaction and on impact parameter

for charged particle collisions. Recent studies [24, 25, 26] have started exploring the
similarities and differences between laser-induced and ion-induced intense fields for un-
derstanding the dynamics governing the behavior of matter in strong fields. Generally
when we perturb a molecule by a charged particle or photon, molecular ion is produced

in different excited states. The effect of ionization induces a change in the mean field
of electrons, seen by the nuclei which lead to rearrangement of electrons and nuclei in
a molecule causing dissociation of it into charged and neutral fragments. The chemical
forces originating from the remaining electrons govern this dissociation. The current

work focuses on the study of dynamics of formation and subsequent dissociation of

multiply charged molecular ions produced in the interaction of neutral molecules with
strong electric fields generated by various projectile ions. Apart from the fact that ion–
molecule collision is common in nature; it is also of more general interest to study them.
Investigations of these interactions form the basis of obtaining a better understanding

of the underlying physics involved.

1.2 Ion induced molecular dissociation

The interaction time, ‘tuned’ by the energy of the ion beam, plays an important role in
ion impact dissociation of molecules. In collision of fast ions with molecules, the role
played by the ion can be viewed as a point charge passing by the molecule in attosec-
ond time scale which is much smaller than the typical vibrational and rotational time
Introduction 9

periods of a molecule. The molecular excitation, for such a small interaction time, can
be considered as a Franck Condon process where electrons move much faster than the

nuclei because of their mass difference. Therefore the electrons can nearly immediately
adjust their positions to the new nuclear configuration when the nuclei move. Accord-
ing to the Born-Oppenheimer approximation, this implies that nuclear and electronic
motions are independent of each other and permits decoupling of rotational, vibrational
and electronic degrees of freedom of a molecule. Since ionization process is purely

electronic excitation, it takes place much faster compared to the vibrational motion of
the nuclei which means that the positions of the atoms and the internuclear separation
remain unaltered during ionization. In the potential energy diagram such transitions are
represented as ‘vertical transitions’ and are denoted by vertical lines (Fig. 1.1).

For slow highly charged ions the interaction times are in the sub femtosecond
regime. In collisions with fast projectiles, direct ionization is the dominant process

of electron removal in contrast to the case of low energy projectiles where electron
capture and transfer ionization are the dominant processes. Thus slow ion impact on
neutral molecule is an effective source for studying dissociative ionization. Ion induced
molecular fragmentation at different energy regimes has shown that electron removal
due to capture process transfers less excitation to the target than direct ionization [27].

At low impact velocities, pre and post collision interaction effects involving the projec-
tile play an important role in the overall collision dynamics. It may also induce some
polarizability in the target molecule.

The kinetic energy release in a dissociation of molecular ion is a measure of the

repulsive state potential energy curve. For a Frank-Condon excitation the difference
between the potential energy of the unstable molecular ion at equilibrium inter-nuclear

distance and the formation energies of its dissociated fragments, appears as kinetic en-
ergy of the dissociated fragments. If the fragments are formed in the ground states, their
formation energy will be least making the kinetic energy release maximum. The dis-
tribution of kinetic energy can be broadened, depending on the slope of repulsive state
Introduction 10

of precursor ion to which it has been excited. The high density of electronic states in
many multiply charged molecular ions complicates the theoretical estimation of kinetic

energy release values and demands use of ab initio quantum chemical calculations well
beyond simple Hartree Fock level. Attempts have been made to develop methods to
quantum mechanically calculate multidimensional potential energy surfaces of differ-
ent electronic states of a molecular ion to gain deeper insights into the metastability,
branching ratios between various fragmentation pathways and distributions of kinetic

energy released upon their fragmentation.

1.3 Overview of ion induced molecular dissociation tech-

nique

In studies of ion induced molecular fragmentation, it is interesting to perform exper-

iments in which all the fragment ions produced after collision are detected in coinci-
dence. Such studies are expected to provide not only the valuable information about

the charge states and the potential energy surfaces of the intermediate multiply charged
molecular ions, but also shed light on the mechanism and dynamics of their excitation,
ionization and fragmentation processes. Production of multiply charged molecular ions
and study of fragmentation processes that occur on timescales as small as few tens

of femtosecond demand instrumentation techniques capable of detecting coincidence

spectra of two or more energetic fragment ions with adequate mass-energy resolution
and collection efficiency. Mass spectrometry is one of the most important tools, which
can be used to get information about the structures and dynamics of the fragmenting

molecular ions. This technique is difficult to place in the methods of spectroscopy,

because it separates ions based on their mass to charge ratio in the static electric or
magnetic field and does not need the interaction with an electromagnetic field (except
in the case of Fourier Mass Spectrometry). Mass spectrometry can be considered as a
Introduction 11

three step procedure: the first step is to create ions from the neutral molecules by bom-
barding it with charged particles or photons. In the second step the ions are separated

according to their mass to charge ratio using the electric and magnetic fields and finally
the intensity of the separated ions are detected electronically.

1.3.1 Production and acceleration of projectile ions

Structure of a molecule or multiply charged molecular ion cannot be explored using

ordinary light wave because of its long wavelength. To probe into the structure of an
atom or molecule one has to use particles with very short wavelengths, comparable to
the size of the atom or molecule under observation. The wave-particle duality of matter

suggests that most of the particles around us have very large wave lengths. According to
De Broglie’s hypothesis, the wavelength (λ) of a particle is inversely proportional to its
momentum (p). Therefore an increase in the momentum of a particle will decrease the
wavelength associated with it. This principle is used extensively in accelerator physics
to produce ion beams of very high speed and energy. A particle accelerator is a sci-

entific apparatus used to accelerate different particles like electrons, protons or ions so
that they achieve high energy. Particle accelerators are like microscopes which help
physicists probe into the structure of matter. It uses electromagnetic fields to propel
electrically charged particles to high speeds. A cathode ray tube is the simplest form of

a particle accelerator which accelerates electrons along the tube using an electric field.
The problem of achieving high energy highly charged ion beam involves two aspects:
production of positively charged ions and acceleration through necessary potential dif-
ference. The most essential and integral part of a particle accelerator is the ion source.

An ion source is a device that produces stream of ions, especially for the use in particle
accelerators. A large variety of ion sources are available all around the world for pro-
ducing positive as well as negative ions. The basic principle of ion production is similar
in different types of ion sources that have been developed to meet the demands of the
Introduction 12

variety of ions and accelerator types in use. Here we will discuss two ion sources, used
extensively for the production of positively charged ions in ion – molecule collision

experiments.

Source of Negative Ions by Cesium Sputtering (SNICS)

Figure 1.2: Schematic diagram of SNICS.

It is a sputter ion source capable of delivering anions from a solid sample by bom-
barding it with Cs+ ions (Fig. 1.2). In the discharge chamber of the ion source a
reservoir is kept at high temperature in which Cs is heated to produce the vapor. A

hot conical ionizer is maintained at a positive potential inside the chamber. The cath-
ode used in SNICS is usually a Cs coated cylindrical section of Cu with a small cavity
drilled onto the cylinder axis. Powder containing the material from which the beam is
desired is loaded into the cavity of the cathode. The entire arrangement is maintained at

a high vacuum. The Cs vapor from the reservoir comes into the enclosed area between
the cold cathode and heated ionizer. Some of the Cs adheres to the cool surface of the
cathode while some hit the hot surface of the ionizer and boils away immediately being
ionized. The ionized Cs then gets accelerated towards the cathode by the applied elec-
Introduction 13

tric field and sputters atoms out of the powdered sample. Some materials preferentially
sputter negative ions, while some sputter neutral or positive particles which pick up

electrons while passing through the cold, condensed layer of Cs and produce negative
ions. Since the entire ion source is operated below the ground potential and the extractor
is kept at high positive potential with respect to the whole ion source, the negative ions
are accelerated and extracted out of the source. Instead of a single cathode, now SNICS
ion source with multi-cathode (MC-SNICS) is being used where 40 cathodes can be put

together on a wheel and one can quickly switch to various cathodes containing different
materials.

A tandem van de Graaff accelerator is then used to accelerate the ions generated in

the ion source. In a van de Graaff accelerator, a moving belt transports charge to a high
voltage terminal, which forms one end of an accelerating column. In tandem van de
Graaff, the negatively charged ions produced by the negative ion source are accelerated
from ground to the positive potential at the centre electrode of the accelerator. In the
centre, the negative ions are stripped of electrons by passage through a foil or gas vol-

ume to become positive ions. After stripping, the positive ions are accelerated as they
move from the positive potential on the centre electrode back to ground. This principle
doubles the energy of the emerging beam for any given terminal voltage. The pressure
vessel is often filled with an insulating gas. Because of its high reliability and suffi-

cient beam output SNICS sources are used extensively in tandem van de Graaff type
accelerator, for example, the Pelletron. The drawback of SNICS lies in the limited time
of operation before maintenance. After some days of use the cathode sample might be
sputtered away and needs to be renewed. Also it cannot produce ion beams from the

noble gases.
Introduction 14

Electron Cyclotron Resonance Ion Source (ECRIS)

The SNICS provides multiply charged ion beams of very high energy but ion beams
of elements with small or zero electron affinity, as e.g. the noble gases, cannot be
generated. The highly charged ion beams of various elements, including the noble
gases, can also be produced from the Electron Cyclotron Resonance ion source.

Solenoid coils

Plasma

Sextupole

Figure 1.3: Working principle of ECRIS.

It can be described as a plasma reservoir for electrons, neutral atoms and ions.
Production of highly charged ions occurs in a sequence of ionization steps caused by
electron impact (Fig. 1.3). An electron, in a static and uniform magnetic field, moves
in a spiral orbit due to the Lorentz force. The linear motion of the electron along the

magnetic field lines, superimposed with the circular motion in the plane perpendicular
to the same field, results in above mentioned cyclotron motion, with gyration frequency
ω =2πf=(e.B)/me where e is the electronic charge, me is the electron mass and B is the
given magnetic field strength. When microwave of same frequency is injected from

an external ultra high frequency traveling wave tube (TWT) amplifier, through a rect-
angular waveguide, into a volume containing low pressure gas, then the electrons are
resonantly accelerated or decelerated, depending on the phase of their transverse veloc-
ity component with respect to the electric field [28, 29]. The ECR ion source makes
Introduction 15

use of this Electron Cyclotron Resonance condition to heat the plasma. The injected
microwaves heat free electrons in the gas which in turn collide with the atoms or mole-

cules of the gas in the volume and cause ionization. Highly charged ions are produced
in successive step-by-step collisions. The ECR ion source confines the ions for long
enough time so that multiple collisions, leading to multiple ionization, take place. The
plasma confinement in ECR ion source is achieved by the superposition of an axial
magnetic field, produced by a pair of solenoid coils and radial magnetic field produced

by sextupole magnet. The low gas pressure in the source is maintained to avoid any
recombination. Since the magnetic confinement is not perfect, few electrons leave the
plasma in the axial direction. To maintain plasma neutrality, few ions also escape from
the plasma. By using suitable extraction electrode one can extract those ions from the

ECR ion source [30].

1.3.2 Various ion – molecule collision setup

The study of the molecular fragmentation involves kinematically complete detection of

correlated fragments in the dissociation of a molecule which yields direct information

on the structure of the parent molecular ion. In most of the ion – molecule collision
setup, the highly charged projectile ions, generated from various ion sources, interact
with an effusive gas target in orthogonal geometry. The heart of the dissociation appa-

ratus is a time-of-flight (TOF) spectrometer combined to a position encoding detector

for measurement of recoil ion momentum vector. Recoil ion momentum spectroscopy
(RIMS) is a very powerful tool in collision studies where a uniform electric field is
applied to guide the ions and electrons formed in the ion molecule interaction to a po-

sition sensitive detector, placed perpendicular to the applied electric field. From the
initial position and flight duration, motion of the fragment ion can be completely de-
termined. RIMS have been successfully implemented for studying fragmentation of
various molecules [31, 32]. This method enables one to do a kinematically complete
Introduction 16

measurement of all fragments, giving unprecedented information on multi–fold differ-

ential cross–sections [18].

Ehrich et. al. have [21] dispersed the projectile ions by a pair of deflection plates
and detected them with a one-dimensional position sensitive detector to perform the
post–collision analysis. From the TOF recorded in coincidence with the projectile– or
electron–signal trigger, additional information about the number of electrons captured

by the projectile ion can be obtained for each event.

For studying fixed–in–space molecule one may also use COLTRIMS setup where
COLTRIMS stand for COLd Target Recoil Ion Momentum Spectroscopy and is a tech-

nique developed in University of Frankfurt by the group of Prof. H. Schmidt-Böking

[33]. Here we can measure the time of flights of both, recoil ions and ejected electrons,
using a precooled and localized supersonic gas jet and two large diameter position sensi-
tive detectors. COLTRIMS prepares intrinsically cold molecules in their rotational and

vibrational ground states. So their interaction with photons, electrons and projectile
ions show great utility in studying molecular dissociation dynamics. A solenoidal mag-
netic field around the analyzer is applied in the COLTRIMS apparatus which causes a
spiraling motion of the electrons around their paths towards the detector. This increases
the accuracy of energy measurements and confines the electrons inside the analyzer.

“Reaction Microscopes” are innovative many-particle momentum spectrometers,

rapidly developing to measure the vector momenta of several ions and electrons result-
ing from a molecular fragmentation. Applying combined parallel electric and magnetic

fields, all emitted electrons and ions are simultaneously projected on opposite posi-
tion and time sensitive detectors reaching a detection efficiency of close to 100 % [34].
Evolving from RIMS and COLTRIMS, reaction microscopes mark the decisive step
forward to investigate many–particle quantum dynamics occurring when atomic and

molecular systems or even surfaces and solids are exposed to time-dependent external
electromagnetic fields.

Instead of using a fixed target one may use a fast (MeV) molecular beam from
Introduction 17

electrostatic accelerator and study the dissociation of the molecular beam in a thin foil
or gas. This technique is called the Coulomb–explosion technique and is first intro-

duced by Vager et al [35]. Here the collision fragments are detected down-stream with
specially developed multi-particle position and time sensitive detectors [36]. Through
detection of position and TOF of the Coulomb exploding fragments, the stereo chemical
structure of the molecular ions constituting the incident beam can be deduced.

1.3.3 Separation of fragment ions by mass spectrometers

The static properties of a stable molecule, like bond length, rotational and vibrational
frequencies in the ground state, etc. can be understood in great detail using traditional

spectroscopic techniques, for example Raman, IR, UV-Vis, fluorescence, NMR, ESR,
etc.. But studying the dynamic properties of a many body system is still a major
challenge for theory as well as experiment. Study of excitation and dissociation of a
molecule under various perturbations, release of kinetic energies in the dissociation pro-
cesses, structure of precursor molecular ion prior to dissociation, bond angle, bond re-

arrangement and orientation of the target molecule in presence of the ion beam demand
a powerful spectroscopic technique. The present day setups, with multihit coincidence
device, have the capability of detecting two or more correlated fragment ions, gener-
ated in the breakup of a molecular ion. In the dissociation apparatus, highly charged

projectile ion beam interacts with the target molecular gas jet in orthogonal geome-
try. In the interaction of projectile ions with target molecules, electrons and recoil ions
are produced. The fragment ions from the dissociated molecules are separated from
the electrons and extracted by a homogeneous weak electric field applied to the mass

spectrometer placed inside the chamber perpendicular to the plane containing the pro-
jectile beam and the target molecular gas jet. The fragment ions of various mass (m)
and charge state (q) are separated on the basis of their flight times in the spectrometer.
q
Since the TOF of an ion is proportional to the m/q, from the acquired TOF spectrum
Introduction 18

we can identify the dissociated fragment ions. Depending upon the requirement of mass
resolution various types of mass spectrometers can be used like Single Focusing Mass

Spectrometer, Double Focusing Mass Spectrometer, Quadrupole Mass Spectrometer

and Time-of-Flight Mass Spectrometer [37].

In Single Focusing Mass Spectrometer, the separation of ions is effected by the

application of a magnetic field perpendicular to the motion of the ions. It actually

works as a momentum analyzer rather than a mass analyzer. Since the magnetic sector
separates the ions on basis of their momentum, ions with little difference in translational
energy are not focused at the same point in Single Focusing Mass Spectrometer. This
limitation is overcome in Double Focusing Mass Spectrometer, where the ions are lead
through a radial electrostatic field prior to magnetic separation. Therefore, ions with

same kinetic energy are only fed to the magnetic sector and thus velocity focusing is
achieved. But as a cost of increased resolution more ions are lost in Double Focusing
Mass Spectrometer, thereby decreasing its sensitivity. Still it is the classical instrument
against which other mass spectrometers are compared. But these two spectrometers are

not well suited for pulsed ionization methods.

Among the two other mass spectrometers, there are at least two advantages of us-
ing TOFMS over Quadrupole Mass Spectrometer (QMS). In TOFMS, the entire mass
spectrum can be recorded in a single shot, when a pulsed acceleration field is used

whereas for QMS we have to scan through the entire mass range to record a spectrum.
This gives faster acquisition time of typically few µs, compared to QMS where single
scan time is typically of the order of a few seconds. Also in QMS, the ion trajectory
along the spectrometer axis is perturbed by the application of RF and dc frequencies

to opposing pairs of quadrupole rods so that only the ions with specific m/q reach the
detector for a given level of applied RF and dc voltages.
Introduction 19

Time Of Flight Mass Spectrometer (TOFMS)

Time of Flight Mass Spectrometers (TOFMS) are best suitable for the detection and
identification of charged fragments. These are based on simple mass separation prin-
ciple. Suppose an ion of mass m and charge q, formed at a particular position at a
particular time, is accelerated by a homogeneous electric field with potential V. The
kinetic energy of the ion, leaving the point of its formation, will be

1
KE = qV = mv 2 (1.6)
2

where v is the velocity of the ion. If L be the path length traveled by the ion in time T,
then
L
v= (1.7)
T

Combining the above equations we get

s
m 1
T =L . (1.8)
q 2V

which suggests that the flight time of the ion is actually proportional to the square root
of its mass to charge ratio [38]. This principle is used in TOFMS to separate ions of
various types.

Figure 1.4: Schematic of a linear time–of–flight–mass–spectrometer (TOFMS).

Introduction 20

The linear TOFMS is a simple device consists of an ion source and an ion collector
fixed at the two ends of an evacuated tube (Fig. 1.4). Ions formed in collision of highly

charged ions with molecules are extracted out from the interaction zone and are made
to move towards the collector under a constant or a pulsed electric field. All the accel-
erated ions then enter a field free drift region. A detector at the end of the field-free
region records a signal as it is struck by an ion. The total time of flight of the ion is
q
proportional to m/q where q is the charge state and m is the mass of ion. The above

description of a time-of-flight mass spectrometer is very simplified. Wiley and McLaren

[39, 40] observed that ions of a particular mass to charge ratio would reach the detector
with a spread in arrival times. This is due to the uncertainty in the formation time of
the ions, location in the extraction region and initial kinetic energy of the ions resulting

in the reduced resolution. They devised an instrument incorporating a pulsed two-grid

ion source, to compensate for temporal, spatial and initial kinetic energy distributions.
Due to the effects of uncertainty in the time and location of formation and also in the
initial kinetic energies, ions of different m/q ratio would take different time to reach

the detector. This would result into a spread in ion arrival time and would reduce the
resolution of the TOFMS. To compensate the temporal spread due to the initial kinetic
energies of the ions, one may use “reflectron” in a TOFMS, in which ions are passed
through a mirror or reflector to reverse their flight directions [41]. It consists of a de-
celerating and reflecting field. A linear field reflectron allows ions with higher kinetic

energies to penetrate deeper into the decelerating field compared to the ions with lower
kinetic energies. Therefore the deeper penetrating ions will spend more time within the
reflecting field and finally catch up with the ions having smaller kinetic energies further
down the flight path. Thus a reflectron decreases the spread in ion TOF for a packet

of ions and improves the resolution of the TOFMS. Michael Guilhaus has given a nice
overview on principles and instrumentation in TOFMS in a tutorial published in [38].
The recent experimental approach to study molecular fragmentation processes utilizes
modern multi–hit detection capability. A position sensitive multi–hit detection system
Introduction 21

gives information on complete momentum balance in the molecular break up. From
this information one can determine the initial momentum of the fragment ion generated

in the ion molecule interaction and the mutual angle between the correlated fragment
ions. For example, the geometry of core-excited BF3 and excited CO2+
2 molecular ion

have been investigated using the momentum imaging technique [42, 43].

1.3.4 Overview of imaging detectors

A multi–hit detector system is required to measure the complete momentum vectors

of all ionic fragments after the dissociation of complex molecules induced by photon,
electron or ion impact. In the detector system developed by Ali et. al., the fragments are

collected in an electrostatic field and detected with a position-sensitive micro-channel

plate detector using a fast timing delay-line readout [44]. The electrons and the frag-
ment ions produced in the molecular breakup can be detected using different detectors.
Since the electron’s charge is too small to create a usable pulse, the detectors for elec-
tron must multiply the electrons. Here we would like to mention a few detectors which

can be used for breakup studies.

Channel Electron Multipliers (CEM)

Channel Electron Multipliers (or Channeltrons) are detectors, which respond to the

charged particles, hard and soft X-rays, and ultraviolet radiation [45]. It has a high sur-
face resistance which forms a continuous dynode when potential is applied between the
input and output end. The dynode emits secondary electrons and through this process
makes CEM capable of detecting a particle or photon. The primary particles, entered

in its funnel-shaped input aperture, generate secondary electrons that are accelerated
down the channel by a positive bias. Upon striking the interior surface of the chan-
nel walls, these electrons keep generating secondary electrons. The resulting avalanche
process produces an output pulse of charge containing up to 108 electrons with dura-
Introduction 22

Figure 1.5: Working principle of a channel electron multiplier (CEM).

tion (FWHM) of about 8 nanoseconds. The output pulses can be detected easily by
using a preamplifier and a constant fraction discriminator to generate a NIM standard
timing pulse. The standard CEMs are capable of detecting over 5 million particles per
second. Figure 1.5 illustrates the basic operation of a CEM. The electron avalanche

that is generated by a primary particle follows the channel to the positively charged end
and is collected by an anode. The curvature of the channel is necessary to prevent ion
feedback caused by the high electron density at the end of the channel. In a straight
channel ions would pick up too much kinetic energy and generate additional secondary

electrons leading to an unstable operation of the CEM.

Micro Channel Plate detector (MCP)

Micro Channel plates (MCP) are widely used for detecting single particles such as elec-
trons, ions and photons. It is a high resolution 2D-imaging and timing device used for

the detection of charged particle or photon at high rates with multi-hit capability. MCPs
are used in a large variety of applications including, imaging spectroscopy, electron
spectroscopy, mass spectrometry, astronomy, atomic and molecular collision studies,
cluster physics etc. It has unique properties like high gain, high spatial and temporal
Introduction 23

resolutions, etc. which make it a better choice for Time-of-Flight Mass Spectrometry
[46].

Electroding on each face

V
+

Channels
Glass structure
Primary
Single channel radiation
V

Secondary
electrons

Figure 1.6: Schematic and working principle of a micro channel plate detector (MCP).

Figure (1.6) shows the basic structure of an MCP and its working principle. MCP

is a specially fabricated plate that amplifies electron signal similar to secondary electron
multiplier (SEM). It is basically an array of a large number of single channel electron
multipliers. The channels are made of special glass with typical channel diameter in the
range 25 µm. The inside of each channel is coated with high-resistance semiconduc-

tor which serves as secondary electron multiplier. The array of glass microchannels is
electrically connected in parallel by metal electrodes on the opposite faces of the plate
for high voltage operation in ultra high vacuum conditions. A single incident particle
(ion, electron, photon etc.) enters a channel and emits 2–3 electrons from the channel
Introduction 24

wall. These electrons are accelerated down the channel by a positive bias voltage and
travel along the parabolic trajectories until they strike the channel surface and produce

more secondary electrons. This cascade process, repeated many times along the chan-
nel, finally produces a pulse of as much as 106 electrons, which emerge from the rear
of the plate. When two or more MCPs are operated in series, from a single input event,
a pulse of 108 or more electrons can be generated at the output. Because of the con-
finement in the individual channels, the spatial patterns of electron pulses at the rear of

the MCP preserve the pattern (image) of the particles, incident on the front surface of
MCP. MCPs have large surface area compared to the channeltrons and the advantage
of using MCP is that after electron multiplication, the charge is emitted from a small
spot at the back surface of the MCP. Therefore, MCPs can be used in so-called position

sensitive detectors. The gain of MCP is limited by ion feedback, current saturation and
(c) charge saturation. The basic problem in electron multiplication is the ion-feedback
instability at high operating voltages and high background pressure. The positive ions,
produced in the collision between electrons and residual gas atoms inside the channel,

are accelerated towards the front surface of MCP and may produce secondary electrons
if sufficiently energetic. The ions striking the wall may also damage the channel. The
positive ion feedback also produces noise which depends on the gain, pressure, nature
of the residual gas and surface properties of the channel wall. The straight geometry of
the microchannels is responsible for this. Therefore the MCPs, used in position sensi-

tive detectors, have a bias angle of about 7◦ ± 2◦ . It is advantageous to put two such
MCPs in Chevron configuration, turned 180◦ along their axes with respect to each other
or three MCPs (Z-Stack configuration) to get better performance, since the feedback
positive ions get trapped at the interface of the two plates. The other effects can be

minimized if the system is operated at relatively low potentials and low pressure (below
106 Torr). These combinations also provide higher gain.
Introduction 25

Position sensitive detectors

From the position information of charged fragments one can know the spatial spread in
the trajectory of ions which can be converted into the momentum components of frag-
ments. The output signals from MCP are collected by several means, including metal or
multi-metal anodes, one- or two- dimensional resistive anode, wedge and strip anode,

delay-line Readout or a CCD camera. The signal from the primary electron shower of
the MCP is used to determine the time-of-flight of the fragment ions. Most of the posi-
tion sensitive detectors used in electron and ion spectrometers are based on MCPs, but
the method of determining the coordinates of the detected particle varies. A glass anode
plate coated with fluorescent material can be put behind the MCP, on which the electron

avalanches create spots of visible light. A CCD frame grabber coupled to a computer,
images the luminescence produced on the screen. Though it can detect the position of
a number of electrons simultaneously, but due to the relatively slow electronic readout
of the optical image, it is not very useful for obtaining timing information. One can

also use resistive anode detector in which the anode plate behind the MCP is uniformly
coated with a resistive material. The electron pulse is collected from the four corners
of the anode and by comparing the pulse heights at the four corners, the position on the
surface, from where the pulse originated, can be determined.

Special segmented anodes with crossed-wire structure combined with advanced

electronic readouts are also used for position information. A crossed-wire detector
consists of independent sets of anodes and amplifiers; thus in principle it can detect two
particles that hit at exactly the same time, but its position resolution is limited by the
anode pixel size. In the above systems, the charge deposited on the different electrodes

is collected and measured in order to obtain usually good position resolution and fair
timing resolution [47]. The Wedge and Strip anode offers good spatial resolution and
fast timing signal. The segmented anode consists of electrically separated area with
a wedge and strip structure at typical periodicity of 1.5 mm. Since the area of the
Introduction 26

wedges and stripes depends linearly on the x and y position, the pulse height of the
signals picked up at both electrodes are proportional to the position of the charge cloud

centroid. The position information is extracted by normalizing the pulse height of the
wedge and stripe segments to the collected total charge which varies from event to event
[48].

Figure 1.7: Complete configurational side and top view of delay line anode detector
(DLD) along with its operation as a position sensitive detector.

The delay line anode is a unique technique which can be used to extract the posi-
tion of charge cloud centroid with good position resolution. Delay line is a bare copper

wire wound across opposite edges of an insulating plate. Two crossed pairs of delay
lines, isolated from one another, give a 2-D grid (Fig. 1.7). The corresponding ends of
the delay-line for each dimension are located on the opposite corners of the array (rear
side). In order to avoid serous pickup problems associated with a single wire, a pair of

wires is used for each coordinate of DLD. Both cables pick up an identical noise via
capacitive coupling with the surrounding. Their signals differ only when a real charge
is deposited on one but not on the other wire. One of the wires is biased with a more
positive potential to ensure collection of electrons from the cloud by it only while the

other wire is basically used for monitoring the noise. The end of these cables is fed into
a fast differential amplifier which subtracts the noise from the signal. The differential
signal allows us to determine the time difference between both ends of the cable. Ali et.
al. developed a multi-hit detector system comprising a micro channel plate (MCP) de-
Introduction 27

tector equipped with a delay-line-anode for complete momentum analysis in molecular

fragmentation study and got a position resolution better than 0.1 mm.

1.4 Review of ion molecule collision studies

Research into ion molecule collision processes stretches back to the beginning of the
previous century. Interaction of electrons and positive ions with variety of molecules

has been investigated. This thesis concerns the study of fragmentation dynamics of
highly charged molecular ions formed in the collision of energetic ions with neutral
molecules. A refined analysis of the ion induced molecular fragmentation plays an im-
portant role in assessing the validity of theoretically computed potential energy (PE)

surfaces of molecular ions [49]. In the first momentum spectroscopy experiment, Ali
et. al. studied the single ionization of He by Ar8+ ion beam [45]. In this experiment, the
azimuthal and polar angles of the scattered projectile were determined in coincidence
with the three dimensional momentum vector of the recoil ion. Momentum and energy
conservation relations were used to extract the energy loss of the projectile and momen-

tum of the ejected electrons. Multihit coincidence techniques using a position sensitive
detector have been used to study the dissociation dynamics of diatomic [27, 50, 51, 52];
triatomic [6, 23, 53, 54] and polyatomic [55, 56, 57] molecules in collision with highly
charged ions. Since electron rearrangement is much faster than the dissociation pro-

cess, the transient molecular ion charge is preferentially equally shared between the
two fragment ions of diatomic molecule [51, 58, 59, 60]. Rajput et. al. have com-
pared the kinetic energy release distributions for the isoelectronic species N2 [51] and
CO [61]. The multiply charged molecules of these two diatomic molecules were gen-

erated in collisions of neutral parents with slow highly charged ions. Ab initio quantum
chemical calculations were performed to investigate the low lying electronic states of
these multiply charged molecules to establish correlation between the experimentally
determined kinetic energy release distributions to the states of these molecular ions.
Introduction 28

The introduction of position sensitive detectors in TOFMS have opened up the subject
of molecular orientation studies with respect to the beam direction due to the influence

of collision processes. For a diatomic molecule oriented along the beam, probability
of large energy deposition is greater compared to the molecule oriented perpendicular
to the beam. The projected electron density sampled by the projectile is larger for tra-
jectories close to the molecular axis. First systematic experimental study of orientation
effect in multiple ionization was done by Kabachnik et. al. for He ion collisions with N2

molecules at the energies 100 – 300 keV [62]. Recently alignment effect has been stud-
ied for highly charged projectile ions colliding with N2 and O2 molecules at very high
energy [63]. Mizuno et. al. found that double ionization of CO2+ induces isotropic
molecular fragmentation for both electron capture and electron loss collisions, while

triple ionization of CO3+ induces highly anisotropic fragmentation during electron loss
collisions [64]. Becker et. al. [47] measured the double ionization and fragmentation of
H2 by 200 keV H+ and He+ ions and found that the kinetic energy spectra of correlated
protons emitted from the collisions was identical, regardless of whether H+ or He+ was

used as the projectile ion.

Leach [65] has considered the formation and destruction of doubly charged poly-
atomic molecules in interstellar space and has concluded that such species can be

formed by interaction of molecules with high-energy photons, cosmic-ray particles and

by energetic shock-heated particles. De et. al. have studied dissociation dynamics of
methanol [66] and C2 H2 [67] by low energy ion impact. The most striking feature of
methanol experiment was the formation of H+ +
3 and H2 by proton coagulation in dou-

bly charged methanol. Fragmentation of C2 H2 indicates a competitive bond breaking

scenario for the completely dissociated molecule. It also implies that the breakage of
bonds of highly charged molecular complex does not always occur simultaneously but
there is a tough competition from the sequential breakdown as well.

The measurement of momentum vector of each fragment is important to character-

ize the fragmentation dynamics in case of a triatomic molecule. When a bent triatomic
Introduction 29

molecule Coulomb explodes, the molecular fragments move in a plane, not along the
axis of the detector. The geometry modification of CO2 molecule in ion induced break

up has been explored by many groups [68, 69]. Adoui et. al. have investigated the
concerted and sequential nature of dissociation process for two fragmentation channels
of multiply charged CO2 [68]. The bend angle distribution of the (C2+ :O2+ :O2+ ) frag-
mentation channel of CO6+
2 , as observed by Sanderson et. al., is extremely close to the

zero-point motion of a neutral molecule when perturbed by 120 keV Ar8+ ions [69].

They also found that the kinetic energy of carbon ions increases with the bending of the
molecular skeleton. The interaction of He2+ with H2 O and CO2 is of interest in stud-
ies of astrophysics, comets, the aurora and corona discharges. Abu-Haija et. al. have
studied the double differential cross sections, in energy and angle, for collisions of the

solar wind He2+ ions with cometary neutrals, H2 O and CO2 by means of translational
energy-gain spectroscopy [70]. Their results indicate that dissociative transfer ioniza-
tion channels of H2 O are populated at the lowest collision energy, whereas contributions
from non-dissociative single electron capture increase with the increasing collision en-

ergy. For the CO2 , dissociative transfer ionization is dominant over the entire collision
energy region. Fragmentation of bent NO2 was studied by Shiromaru et. al. [71] and
that of water molecules have been studied by Werner et. al. [6, 72]. Rajput et. al.
[73, 74] have shown that dissociation of isolated water molecules by 450 keV Ar9+ ion
impact leads to the production of highly excited, fast, neutral hydrogen atoms having

transient Rydberg molecular ions as parents. Nishide [75] and Rajgara et al. [23] have
investigated fragmentation dynamics of CS2 using 120 keV Ar ions.

1.5 Motivation and thesis layout

Dynamics and kinetics of an unstable molecular ion in presence of various perturba-

tions is still a concern of present day research. Understanding molecular fragmenta-
tion dynamics by obtaining detailed potential energy surfaces of the transient species
Introduction 30

is always important. Ion-induced molecular fragmentation studies in different energy

regimes have shown that electron removal due to capture results in weaker excitation

of the target than direct ionization and these energy regimes give information about the
different excited electronic states of a given multiply charged molecular ion. Molecular
parameters such as bond length, bond angle etc. may get modified during the evolu-
tion from ground state of the neutral to the dissociative state of the molecular ion. The
minimum energy state of the system or stability of the molecular ion is achieved by the

geometrical deformation and reorganization of bond within the molecular ion.

The current work focuses on experimental investigation of the interaction of var-

ious energy ion beams with three neutral gas molecules: CO2 , OCS and SF6 . It is a

well known fact that the symmetry of a molecular ion is often different from that of the
ground state of neutral molecule. Fragmentation of three molecules, having different
symmetries, have been reviewed in the present section. Two linear triatomic molecules,
one with a center of symmetry (CO2 ) and the other with no center of symmetry (OCS)

have been chosen to investigate the effect of ion impact on the geometry and dissoci-
ation scheme of the molecules, along with a highly symmetric nonlinear (octahedral)
polyatomic species (SF6 ).

CO2 is a simple, linear, triatomic molecule, having a centre of symmetry. Carbon

dioxide (CO2 ) is naturally present in the Earth’s atmosphere and is the most discussed
greenhouse gas due to its connection to our everyday life. Global warming and increase
of CO2 has been predicted to have consequences on not just rainfall or temperature, but

on plant life as well.

Carbonyl Sulphide (OCS) is the most abundant sulphur compound naturally present
in the atmosphere. According to Yuichiro Ueno et al [76], it was the most likely can-

didate for an Archean atmospheric blanket and was the perfect greenhouse gas. The
shielding by OCS is an archaic analogue to the current ozone layer that protects us
from ultraviolet radiation. But unlike ozone, OCS would have kept the planet warm.
Unlike CO2 , OCS is highly asymmetric and is more prone to various geometry mod-
Introduction 31

ifications under strong fields. Like all triatomic molecules with 16 valence electrons,
OCS is linear in its ground state and bent in its first excited state [77].

Sulphur hexafluoride (SF6 ) is a highly symmetric molecule. It is therefore, a mat-

ter of primary concern to obtain a thorough understanding of the nature and the prop-
erties of the fragmentation product of SF6 under various perturbations. SF6 is a non-

flammable man–made gas with excellent dielectric and arc-quenching properties. It oc-
curs naturally in common granitic rocks and fluorite minerals and is released extremely
slowly through weathering processes. It is an infrared (IR) absorber and not removed
from the earth’s atmosphere rapidly. These properties make SF6 a potent greenhouse
gas. Due to its chemical inertness it is benign with regard to stratospheric ozone de-

pletion. Owing to its special properties, namely, high dielectric strength, chemical in-
ertness and high saturation vapor pressure at a room temperature, it has received wide
applications in industry as a gaseous insulator in high-voltage electrostatic generators,
transformers, condensers, and cables. In a large molecule like SF6 , the dissociation of a

multiply charged state proceeds via large number of pathways, and discerning these is
a challenge. There can be several fragments, and they may be emitted in a stepwise or
a direct manner.

The goal of my thesis is to study the fragmentation dynamics of the above three gas
molecules of different geometry and symmetry (CO2 , OCS and SF6 ) by the impact of
various energy ion beams. The experiments have been performed in an attempt to illus-
trate the possible dynamics of these gases in the atmosphere and to understand the basic

physics of the ion molecule collision processes. A time of flight mass spectrometer
(TOFMS) has been used to study the dynamics and kinetics of ion impact dissociation
of molecules, such as, identification of channels, kinetic energy release (KER), angu-
lar distribution of the fragment ions, bond distortions and geometry modifications. Ab
initio quantum chemical calculations have also been performed in few cases in order to

deduce the excited states of the parent molecular ion as well as fragment ions.

In Chapter 2 we have introduced briefly the particle accelerators used for the
Introduction 32

production of highly charged energetic ion beam followed by the experimental set up
used for performing the ion induced molecular dissociation experiments reported in

this thesis. The principle and operation of all the detectors, data analysis methodology,
measurement of kinematic parameters and estimation of the geometry of molecular ions
are discussed in this chapter.

The following chapters introduce the molecules of interest in the present work.

The experimental results for ion induced molecular dissociation of CO2 , OCS and SF6
are described in detail after a brief discussion on the specific motivation for studying
these species.

Chapter 3 introduces the first molecule of interest, CO2 . We have studied multi-
ple ionization and fragmentation of CO2 induced by low as well as high energy highly
charged ion impact using time of flight mass spectroscopy (TOFMS) in multihit co-
incidence mode. In addition to the study of various dissociation channels of multiply

charged CO2 , some results on kinetic energy release invoked our interest in carrying out
complementary quantum chemical calculations. We have performed some preliminary
calculations using the GAMESS software package to generate CO3+
2 potential energy

surfaces and some conclusions have been derived based on the results.

In Chapter 4, first comprehensive study on ion induced dissociation of OCS has

been presented. We have analyzed the concerted and sequential dissociation channels
of OCSq+ (where q = 2 – 4) by low energy singly charged ion beam and estimated the
geometry of the precursor molecular ion from momentum correlation map. Another

result of particular importance presented in this chapter is the effect of projectile charge
state on dissociation dynamics of OCS.

A multiply charged molecular ion possesses a set of configurations with certain

symmetries. All these configurations have different electronic energies depending on

which a molecular ion follows different dissociation pathways. In comparison with
the symmetric triatomic molecules CO2 , OCS is more prone to various modifications
of shape under strong fields and in chemical processes. We have investigated this by
Introduction 33

comparing similar dissociation channels of both OCS and CO2 in Chapter 5 using high
energy highly charged ion beam.

In Chapter 6 we have explored the fragmentation dynamics of multiply ionized

SF6 molecule induced by high energy ion beam from Pelletron. The analysis of the ex-
perimental results that led to the understanding of some important dissociation channels
of SF2+ 3+
6 and SF6 is presented in this chapter. The main focus of the present study on

SF6 is to determine the nature of explosion of the three F+ ions from SF3+
6 and geometry

optimization of the precursor molecular ion from the angular correlation diagram.
Chapter 7 gives a brief conclusion and discussion of the thesis, few suggestions
for the continuation of the reported work and new possibilities that have opened up for
further studies in this field.
Bibliography

[1] E. Herbst, Annu. Rev. Phys. Chem. 46, 27 (1995)

[2] P. Ehrenfreund, W. Irvine, L. Becker, J. Blank, J. R. Brucato, L. Colangeli,

S. Derenne, D. Despois, A. Dutrey, H. Fraaije, A. Lazcano, T. Owen and F.

Robert, an International Space Science Institute ISSI-Team, Rep. Prog. Phys.

65 1427 (2002)

[3] E. R. Pike, P. C. Sabatier (Eds.), Scattering:scattering and inverse scattering

in pure and applied science, Volume 2, Academic Press, London (2002)

[4] R. Téoule, Int. J. Radiat. Biol. 51, 573 (1987)

[5] J. E. Biaglow, Radiation Chemistry: Principles and Applications, Farhataziz

and M. A. J. Rodgers (Eds.), pp. 527, VCH Publishers, New York (1987)

[6] U. Werner, K. Beckord, J. Becker and H. O. Lutz, Phys. Rev. Lett. 74, 1962
(1995)

[7] B. H. Bransden and C. J. Joachain, Physics of atoms and molecules, 2nd edi-

tion, Pearson Education, Harlow (2003)

[8] E. A. Varella (Ed.), Conservation Science for the Cultural Heritage: Applica-
tions of Instrumental Analysis, Springer, Heidelberg (2012)

[9] K. Dota, M. Garg, A. K. Tiwari, J. A. Dharmadhikari, A. K. Dharmadhikari

and D. Mathur, Phys. Rev. Lett. 108, 073602 (2012)

34
Introduction 35

[10] S. Sunil Kumar, P. C. Deshmukh, R. K. Kushawaha, V. Sharma, I. A. Prajapati,

K. P. Subramanian and B. Bapat, Phys. Rev. A 78, 062706 (2008)

[11] P. Baltzer, M. Lundqvist, B. Wannberg, L. Karlsson, M. Larsson, M. A. Hayes,

J. B. West, M. R. F. Siggels, A. C. Parr and J. L. Dehmer, J. Phys. B: At. Mol.
Opt. Phys. 27, 4915 (1994)

[12] F. A. Rajgara, M. Krishnamurthy, and D. Mathur, Phys. Rev. A 68, 023407

(2003)

[13] R. Feng, R. G. Cavell and A. P. Hitchcock, Journal of Electron Spectroscopy

and Related Phenomena 144, 231 (2005)

[14] A. D. Bandrauk, Y. Fujimura, R. J. Gordon (Eds.), Laser Control and Manipu-

lation of Molecules, ACS Symposium Series 821, American Chemical Society,

Washington, DC (2002)

[15] J. Gross, Mass Spectrometry, 2nd edition, Springer-Verlag Berlin Heidelberg

(2011)

[16] P. Wang and C.R. Vidal, Chem. Phys. 280, 309 (2002)

[17] R. Singh, P. Bhatt, N. Yadav and R. Shanker, Meas. Sci. Technol. 22, 055901
(2011)

[18] V. Sharma and B. Bapat, Eur. Phys. J. D 37, 223 (2006)

[19] J. Hillier and R. F. Baker, J. Appl. Phys 15, 663 (1944)

[20] B. Seigmann, U. Werner, and H. Lutz, Aust. J. Phys. 52, 545 (1999)

[21] M. Ehrich, B. Seigmann, U.Werner, and H. Lebius, Radiat. Phys. Chem. 68,
127 (2003)
Introduction 36

[22] H. O. Folkerts, R. Hoekstra, and R. Morgenstern, Phys. Rev. Lett. 77, 3339
(1996)

[23] F. A. Rajgara, M. Krishnamurthy, D. Mathur, T. Nishide, T. Kitamura, H. Shi-

romaru, Y. Achiba, and N. Kobayashi, Phys. Rev. A 64, 032712 (2001)

[24] D. Mathur, Phys. Rev. A 63, 032502 (2001)

[25] V. R. Bhardwaj, F. A. Rajgara, K. Vijayalakshmi, V. Kumarappan, D. Mathur

and A. K. Sinha, Phys. Rev. A 58, 3849 (1998)

[26] T. Schlathölter, R. Hoekstra, S. Zamith, Y. Ni, H. G. Muller and M. J. J.

Vrakking, Phys. Rev. Lett. 94, 233001 (2005)

[27] M. Tarisien, L. Adoui, F. Frémont, D. Leliévre, L. Guillaume, J. -Y. Chesnel,

H. Zhang, A. Dubois, D. Mathur, Sanjay Kumar, M. Krishnamurthy and A.
Cassimi, J. Phys. B: At. Mol. Opt. Phys. 33, L11 (2000)

[28] D. Kanjilal, T. Madhu, G. Rodrigues, U. K. Rao, C. P. Safvan and A. Roy, Ind.

J. of Pure and Appl. Phys. 39, 25 (2001)

[29] D. Kanjilal, G. Rodrigues, U. K. Rao, C. P. Safvan, P. Kumar, A. Roy and G.

K. Mehta, Proceedings of APAC, Beijing, China (2001)

[30] R. Geller, Electron Cyclotron Resonance Ion Sources and ECR Plasmas, IOP,
Bristol (1996)

[31] K. Ueda and J. H. D. Eland, J. Phys. B: At. Mol. Opt.Phys. 38, S839 (2005)

[32] Imaging in Molecular Dynamics: Technology and Applications, B. Whitaker

(Ed.), Cambridge University Press, Cambridge (2003)

[33] R. Dorner, V. Mergel, O. Jagutzki, L. Spielberger, J. Ullrich, R. Moshanmer,

H. Schmidt-Böcking, Phys. Rep. 330, 95 (2000)
Introduction 37

[34] J. Ullrich, R. Moshammer, A. Dorn, R. Döorner, L. Ph. H. Schmidt and H.

Schmidt-Böcking, Rep. Prog. Phys. 66, 1463 (2003)

[35] Z. Vager, D. S. Gemmell, B. J. Zabransky, Phys. Rev. A 14, 638 (1976)

[36] D.S. Gemmell, Chem. Rev. 80, 301 (1980)

[37] P. M. van Galen, Mass Spectrometry @ the Organic Chemistry Department

(2005)

[38] M. Guilhaus, J. Mass Spectrom. 30, 1519 (1995)

[39] W. C. Wiley, I.H. McLaren, Rev. Sci. Instrum. 26, 1150 (1955)

[40] A. R. Bottrill, Ph.D. Thesis, University of Warwick (2000)

[41] B. A. Mamyrin, V. I. Karataev, D. V. Shmikk,V. A. Zagulin, Sov. Phys. JETP

37, 45 (1973)

[42] K. Ueda, A. De Fanis, N. Saito, M. Machida, K. Kubozuka, H. Chiba, Y. Mu-

ramatu, Y. Sato, A. Czasch, O. Jaguzki, R. Dörner, A. Cassimi, M. Kitajima,

T. Furuta, H. Tanaka, S. L. Sorensen, K. Okada, S. Tanimoto, K. Ikejiri, Y.
Tamenori, H. Ohashi and I. Koyano, Chem. Phys. 289, 135 (2003)

[43] B. Bapat and V. Sharma, J. Phys. B: At. Mol. Opt. Phys. 40, 13 (2007)

[44] I. Ali, R. Dörner, O. Jagutzki, S. Nüttgens, V. Mergel, L. Spielberger, Kh.

Khayyat, T. Vogt, H. Bräuning, K. Ullmann, R. Moshammer, J. Ullrich, S.
Hagmann, K. Ō. Groeneveld, C. L. Cocke, H. Schmidt-Böcking, Nucl. In-
strum. Methods Phys. Res., Sect. B 149, 490 (1999)

[45] R. Ali, V. Frohne, C. L. Cocke, M. Stöckli, S. Cheng and M. L. A. Raphaelian,

Phys. Rev. Lett. 69, 2491 (1992)

[46] http://www.roentdek.com
Introduction 38

[47] J. Becker, K. Beckord, U. Werner and H. O. Lutz, Nucl. Instrum. Methods

Phys. Res., Sect. A 337, 409 (1994)

[48] J. S. Lapington and H. E. Schwarz, IEEE Transactions on Nuclear Science 33,

288 (1986)

[49] D. Mathur, Phys. Rep. 225, 193 (1993)

[50] B. Siegmann, U. Werner, R. Mann, N. M. Kabachnik and H. O. Lutz, Phys.

Rev. A 62, 022718 (2000)

[51] J. Rajput, S. De, A. Roy and C. P. Safvan, Phys. Rev. A 74, 032701 (2006)

[52] G. Laurent, J. Fernández, S. Legendre, M. Tarisien, L. Adoui, A. Cassimi, X.

Fléchard, F. Frémont, B. Gervais, E. Giglio, J. P. Grandin, and F. Martı́n, Phys.

Rev. Lett. 96, 173201 (2006)

[53] L. Adoui, T. Muranaka, M. Tarisien, S. Legendre, G. Laurent, A. Cassimi, J.

-Y. Chesnel, X. Fléchard, F. Frémont, B. Gervais, E. Giglio and D. Hennecart,

Nucl. Instrum. Methods Phys. Res., Sect. B 245, 94 (2006) and references
therein

[54] S. Legendre, E. Giglio, M. Tarisien, A. Cassimi, B. Gervais and L. Adoui, J.

Phys. B: At. Mol. Opt. Phys. 38, L233 (2005)

[55] F. A. Rajgara, M. Krishnamurthy, D. Mathur, T. Nishide, H. Shiromaru and N.

Kobayashi, J. Phys. B: At. Mol. Opt. Phys. 37, 1699 (2004)

[56] G. Veshapidze, M. Nomura, T. Nishide, F.A. Rajgara, H. Shiromaru, Y. Achiba

and N. Kobayashi, J. Phys. B: At. Mol. Opt. Phys. 37, 2969 (2004)

[57] B. Siegmann, U. Werner and R. Mann, Nucl. Instrum. Methods Phys. Res.,
Sect. B 233, 182 (2005)
Introduction 39

[58] K. Wohrer, G. Sampoll, R. L. Watson, M. Chabot, O. Heber and V. Horvat,

Phys. Rev. A 46, 3929 (1992)

[59] I. Ben-Itzhak, S. G. Ginther, V. Krishnamurthi and K. D. Carnes, Phys. Rev. A

51, 391 (1995)

[60] L. Adoui, C. Caraby, A. Cassimi, D. Leliévre, J. P. Grandin and A. Dubois, J.

Phys. B: At. Mol. Opt. Phys. 32, 631 (1999)

[61] J. Rajput and C. P. Safvan, Phys. Rev. A 75, 062709 (2007)

[62] N. M. Kabachnik, V. N. Kondratyev, Z. Roller-Lutz, and H.O. Lutz, Phys. Rev.

A 57, 990 (1998)

[63] B. Siegmann, U.Werner, R. Mann, Z. Kaliman, N. M. Kabachnik, and H.O.

Lutz, Phys. Rev. A 65, 010704 (2002)

[64] T. Mizuno,T. Majima, H. Tsuchida, Y. Nakai and A. Itoh, J. Phys.: Conf. Ser.

58, 173, (2007)

[65] S. Leach, Molecular Astrophysics: State of the Art and Future Directions, G.
H. F. Diercksen, W. F. Huebner and P. W. Langhoff (Eds.) D. Reidel Publishing

company, NATO ASI Series 157, 353 (1985)

[66] S. De, J. Rajput, A. Roy, P. N. Ghosh and C. P. Safvan, Phys. Rev. Lett. 97,
213201 (2006)

[67] S. De, J. Rajput, A. Roy, P. N. Ghosh and C. P. Safvan, J. Chem. Phys. 127,
051101 (2007)

[68] L. Adoui, M. Tarisien, J. Rangama, P. Sobocinsky, A. Cassimi, J.-Y. Ches-

nel, F. Fremont, B. Gervais, A. Dubois, M. Krishnamurthy, S. Kumar and D.
Mathur, Phys. Scr. T. 92, 89 (2001)
Introduction 40

[69] J. H. Sanderson, T. Nishide, H. Shiromaru, Y. Achiba, and N. Kobayashi, Phys.

Rev. A 59, 4817 (1999)

[70] O. Abu-Haija, E.Y. Kamber and S.M. Ferguson, Nucl. Instrum. Methods Phys.
Res., Sect. B 205, 634 (2003)

[71] H. Shiromaru, T. Nishide, T. Kitamura, J. H. Sanderson, Y. Achiba, and N.

Kobayashi, Phys. Scr. T. 80, 110 (1999)

[72] U. Werner, K. Beckord, J. Becker, H.O. Folkerts, and H.O. Lutz, Nucl. In-
strum. Methods Phys. Res., Sect. B 98, 385 (1995)

[73] J. Rajput and C. P. Safvan, J. Chem. Phys. 134, 201101 (2011)

[74] J. Rajput and C. P. Safvan, Phys Rev. A 84, 052704 (2011)

[75] T. Nishide, F. A. Rajgara, T. Kitamura, H. Shiromaru, Y. Achiba, and N.

Kobayashi, Phys. Scr. T. 92, 415 (2001)

[76] Y. Ueno, M. S. Johnson, S. O. Danielache, C. Eskebjerg, A. Pandey and N.

Yoshida, PNAS 106, 14784 (2009)

[77] A. Sugita, M. Mashino, M. Kawasaki, Y. Matsumi, R. Bersohn, G. Trott-

Kriegeskorte and K. Gericke, J. Chem. Phys. 112, 7095 (2000)
Chapter 2

Experimental Techniques and data

analysis

2.1 Introduction

The experiments presented in this thesis were carried out at Inter University Accelera-
tor Centre (IUAC), New Delhi, India. In the following an account of the experimental
techniques that are employed in this work have been presented. The chapter begins
with the accelerator facilities of IUAC from which various energy ion beams have been

generated for conducting molecular dissociation experiments. In the subsequent sec-

tions, details of the ion–molecule collision setup is presented along with the multi-hit,
position-sensitive, time of flight measurement system. The last part of the chapter de-
scribes the data acquisition procedure and explains how the detailed information on

fragmentation dynamics of multiply charged molecular ions are extracted from the col-
lected data.

41
Experimental Techniques and data analysis 42

2.2 Ion beam from accelerator

The projectile ion beam, required for the ion–molecule collision experiment, is pro-

duced in an ion source and accelerated to required energy before entering into the col-
lision chamber in which actual collisions take place. The energy of the ion beam plays
key role in molecular dissociation processes. The dynamics of a fragmenting molecular
ion changes with the energy of the projectile ion beam. The experiments, presented in

this thesis, have been carried out using ion beams of very high or low energies. The
high energy highly charged ions are generated from a tandem Van de Graaff type ac-
celerator, called Pelletron, whereas, comparatively low energy ions are produced from
Electron Cyclotron Resonance (ECR) Ion Source based Low Energy Ion Beam Facility

(LEIBF). In the following subsections we will discuss these two accelerator facilities of
IUAC which we have used in the present work.

2.2.1 Pelletron

The Pelletron of IUAC produces DC as well as pulsed ion beams of several of ele-

ments having energy up to 15 MV per charge state [1]. The principle of acceleration of
ions in Pelletron is shown in Figure 2.1. In the Pelletron, negative ions of the projec-
tile material are produced from the Multi–Cathode Source of Negative Ions by Cesium
Sputtering (MC-SNICS), already described in Chapter 1. The beam of negative ions,

produced by MC-SNICS, is first analyzed by a 90◦ injector magnet so that the de-
sired ions of 1− charge state are separated from other contaminated isotopes and charge
states. The analyzed beam is then injected into strong electrical field inside an accel-
erator tank filled with insulating SF6 gas. High voltage of up to 15 MV is generated

in the middle of the tank which is known as terminal. Inside the tank the negative ions
are accelerated while traveling from the column top of the tank to the positive terminal,
through the accelerating columns. The terminal is charged by a set of double Pelletron
chain. The corona-based voltage grading system provides strong electrical potential to
Experimental Techniques and data analysis 43

Figure 2.1: Schematic diagram of Pelletron showing ion acceleration principle. Cour-
tesy: IUAC website

the negative ions and accelerates them. Inside the terminal the negative ions are passed

through either a carbon foil or gas strippers where some of the electrons are stripped
from the negative ions due to collisions with carbon or gas atoms. The positive ions
thus formed are repelled away from the positive terminal, and accelerated further to the
ground potential at the bottom of the tank. After the ions leave the Pelletron accelerator,

a particular beam of ion is selected using a 90◦ analyzing magnet which also bends the
beam into the horizontal plane. Finally with the help of a switching magnet, the high
energy highly charged ion beam is directed into one of the six beam lines in different
experimental areas of the beam hall. The ion – molecule collision set up is connected

at the exit port of the General Purpose Scattering Chamber (GPSC), installed at the 45◦
beam line in Beam Hall I of IUAC.
Experimental Techniques and data analysis 44

2.2.2 Low Energy Ion Beam Facility (LEIBF)

Pumping Faraday
station cup
Slit PSTV

Quadrupole doublet

Steerer BPM BPM Faraday cup

ECR source Accelerating

Analyzing
Quadrupole triplet

magnet
column
Slit
Steerer
Pumping station Pumping station

BPM (Beam profile monitor) Liquid droplet

Four−jaw slit chamber

150 Beam line

Faraday cup
900 Beam line

Quadrupole triplet

Pumping station

Scanner

Material Science chamber

Atomic and Molecular

Physics chamber
(ATMOL)
Electrostatic analyzer

Figure 2.2: Schematic diagram of LEIBF. Various experimental stations connected to

different beam lines have been shown. The molecular dissociation setup (ATMOL) is
situated at the end of 90o beam line.

The Low Energy Ion Beam Facility (LEIBF) at IUAC also provides multiply charged
ion beams at a wide range of energies (a few keV to about an MeV) for ion molecule col-

lision experiments. The positively charged ions are produced from a 10 GHz Nanogan
type Electron Cyclotron Resonance (ECR) ion source (from Panteknik, France), based
on a fully permanent magnet (NdFeB), designed for radial and axial confinement of
the plasma. The principle of ion production in an ECR ion source has already been

discussed in chapter 1. The ECR ion source, along with the UHF transmitter, vac-
uum components and all peripheral electronics, placed on a high voltage platform [2, 3]
provides multiply charged positive ions. These are controlled through optical fiber com-
munication in multiplexed mode. The source, together with its extraction assembly, is
Experimental Techniques and data analysis 45

coupled to a pumping section, which also houses a three-element einzel lens. The einzel
lens and the following accelerating tubes focus the ion beam at the object plane of an

analyzer magnet. The ion source, einzel lens, gate valve, accelerating column, pump-
ing station, electrostatic steerer, double slit, beam profile monitor (BPM), Faraday cup
and a fast acting pneumatic straight through valve (PSTV) coupled to the dipole mag-
net through a flexible bellow, constitutes the pre-analyzer section of the LEIBF. The
ions extracted from the ECR ion source are charge and mass selected by the analyzing

(switching) dipole magnet and transported to the experimental chamber through one of
the two beam lines (one at 90◦ and the other at 15◦ ) coupled to the magnet. The beam
transport system comprises of electrostatic quadrupole for beam focusing, electrostatic
or magnetic steerer for beam steering, beam profile monitor (BPM), four-jaw slit and

Faraday cup (FC) for beam diagnostics along with pumping stations, isolation valves
and flexible bellows for keeping the entire beam line under high vacuum. The LEIBF
has been upgraded with enhanced platform voltage of 400 kV. Therefore the maximum
available beam energy is now 400 keV/q where q is the charge state of the projectile

ion. Figure 2.2 shows the two experimental beam lines of LEIBF. The ion induced
molecular dissociation set up is situated at the end of the 90◦ beam line.

2.3 The molecular dissociation apparatus

The mass and charge selected projectile ions produced either by ECR ion source in
LEIBF or MC–SNICS in Pelletron, are focused and accelerated to the required collision
energy before it enters into the experimental chamber. At the centre of the chamber pro-
jectile ions interact with neutral molecules, effusing from a grounded needle located at

right angles to the beam direction, causing dissociation. The single collision condition
is maintained during the experiments for which the experimental chamber is operated
at a base pressure of the order of 10−8 Torr without the target gas and with gas load it
may reach up to 10−7 Torr. A dedicated turbo molecular pump connected to one port of
Experimental Techniques and data analysis 46

Figure 2.3: Schematic of the experimental setup showing coincidences between frag-
ments/recoils, electrons and projectile ions. The ion beam crosses the target molecular
beam at the interaction zone, where ions and electrons are formed. The electrons and
ions are extracted towards their respective detectors by means of a uniform electric field
applied to the time–of–flight mass spectrometer (TOFMS). The start pulse is provided
by electron signal from CEM. The time–of–flight information is extracted by the stop
pulses from the ion MCP signals and the position information is available from the
delay-line (DLD) signals.

the experimental chamber takes care of the vacuum inside the experimental chamber.
In collision with high energy projectile ions, the neutral molecules are first ionized and
then they dissociate. The scattered projectile ions can be charged–state analyzed by an
electrostatic deflector and detected by a channel electron multiplier (CEM or channel-

tron). Details of the parallel plate electrostatic analyzer for post collision projectile ions
have been discussed in the work of S. De [4]. The fragment ions generated in the above
collisions are separated in a time of flight mass spectrometer (TOFMS) by application
of a homogeneous electric field, which guides and accelerates the ions before they enter

into a field free drift region. The recoil ions are finally detected by a position sensitive
detector kept at the exit of the time of flight assembly. Another detector is mounted op-
posite to the fragment ion TOF direction which detects the electrons emitted by either
the projectile or the target particle. The fast output pulses obtained from the electron
Experimental Techniques and data analysis 47

detector serves as the starting trigger for the data acquisition whereas the fragment ions
provide the stop signals. The TOF spectrum is acquired in multi-hit coincidence mode

by a time to digital converter (TDC) interfaced to a computer. Using recoil ion momen-
tum spectroscopy, it is possible to carry out the complete measurements of all fragment
ions generated in a molecular dissociation process. Momentum of each correlated frag-
ment ion is calculated from the TOF and position information recorded simultaneously
during data acquisition. Figure 2.3 shows the schematic of the experimental setup, dif-

ferent parts of which are described in detail in the following subsections.

2.3.1 Ion – molecule interaction zone

Figure 2.4: The ion molecule interaction zone.

The ion molecule interaction zone is defined by the region where projectile ion
beam interacts with target molecular gas jet. In order to have a good momentum reso-
lution and high coincidence rates, the overlap volume of the two beams should be very
Experimental Techniques and data analysis 48

good and lie on the spectrometer axis, exactly half way between the two extraction elec-
trodes. In our set up, the projectiles are made to cross, at right angles, a molecular gas

jet effusing from a grounded hypodermic needle having an inner diameter of 0.9 mm
and an outer diameter of 1.1 mm (Fig. 2.4). The gas flow is controlled by a fine-control
all-metal needle valve and a target density of about 1013 cm−3 is maintained. Ionization
of the effusive gas is brought about by highly charged energetic ions. Two collimators
of diameters of the order of 1 mm were used at the beam entrance and exit ports of

the chamber to minimize the ionization volume for various target-projectile combina-
tions. If the interaction takes place in a uniform electric field, applied to the TOFMS,
the ion beam trajectory may get deflected. In order to get a good overlap between the
ion beam and gas jet at the collision centre, we have positioned the hypodermic needle

a little below the ion beam trajectory and steered the ion beam in horizontal and vertical
planes. After having a coarse optimization, the fine optimization of the interaction zone
is achieved by minimizing the full width at half maximum (FWHM) of TOF profile
of a parent molecular ion and obtaining highest intensity at the centre of the position

spectrum of ions produced in the ion – molecule interactions.

2.3.2 The linear time–of–flight mass spectrometer (TOFMS)

The fragment ions and ejected electrons produced in the ion molecule collision pro-

cesses are separated by a homogeneous electric field, applied perpendicular to the de-
tector plane. A linear TOFMS has been used in all the experiments, performed in this
thesis, which can be operated either in single field or double field mode.

Single field TOFMS

The single field linear TOFMS consists of two regions: extraction region (s) and field
free drift region (D) (Fig. 2.5). The extraction region is enclosed between two elec-
trodes: one is pusher, maintained at positive potential and the other is puller, maintained
Experimental Techniques and data analysis 49

Figure 2.5: Schematic of single field linear time–of–flight mass spectrometer (TOFMS).

at negative or ground potential for ion extraction. The puller plate is attached to the drift
tube and is nearer to the detector. The fragment ions are produced in the interaction re-
gion in presence of an electric field (Es =V/s where V = V1 - V2 ). After extraction the
ions are accelerated towards the field free drift tube where they are time separated on
basis of their mass–to–charge (m/q) ratio. The TOF measurement is triggered by the

ejected electron signal and stopped by the ion signal from the detector placed at the exit
of the drift tube.

Mass Resolution of TOFMS

The ability of a TOFMS to distinguish between two ions of nearly equal mass is ex-
pressed in terms of mass resolution. Two adjacent mass peaks corresponding to two
ions of masses m and m+1 are just resolved when their TOF separation (Tm+1 – Tm )
is greater than or equal to the full width at half maximum (FWHM) of mass peak. The

mass resolution is defined by m/∆m where ∆m is the peak width necessary for separa-
tion at mass m. Ions of same mass and initial kinetic energy, formed at same time but
different locations in the interaction region will cause spatial distribution in TOF spec-
trum. Two ions of same m/q, formed at the same time and location, will also broaden
Experimental Techniques and data analysis 50

the TOF peak if they are emitted with different energies or same energy but in different
directions. Both, spatial and kinetic energy distributions of TOF will degrade the mass

resolution of TOFMS which is generally expressed by

1
m/∆m = .t/∆t (2.1)
2

The time interval ∆t in the above equation will be constant for two ions having same
mass and initial kinetic energy but formed at different instants of time. Thus the ini-
tial space and energy distributions of the ions also influence the mass resolution of a
TOFMS. By the use of two field linear TOFMS one can improve the mass resolution of

single field TOFMS where Wiley-McLaren space focusing condition compensates for
the finite spatial extent [5].

Double field TOFMS

Figure 2.6: Schematic of double field linear time–of–flight mass spectrometer

(TOFMS).

The double field TOFMS introduces one new region (acceleration region (d)) in
between extraction and drift regions. This increases the flexibility of the double field
Experimental Techniques and data analysis 51

TOFMS and gives better resolution. Shown in Figure (2.6) is the schematic diagram
of a double field linear TOFMS. The TOF of the ion and its location of impact on the

detector can be completely determined from the applied electric field, initial position of
ion formation and the initial kinetic energy associated with it. To derive the flight time
of ions, let us consider an ion, created in the interaction region, having charge q, mass
m and initial kinetic energy zero. The electric fields in the extraction, acceleration and
drift regions are given by

Es = 2V1 /s (2.2)

Ed = (V2 − V1 )/d (2.3)

ED = 0 (2.4)

respectively. The accelerations gained by the ion in the above regions will be

as = (qEs )/m (2.5)

ad = (qEd )/m (2.6)

aD = 0 (2.7)

If the time of flight of the ion in the extraction region (s), acceleration region (d) and

drift region (D) of the linear TOFMS be ts , td and tD respectively then

s s
m s
ts = (2.8)
q Es
s
m 1 q q
td = [− sEs + sEs + 2dEd ] (2.9)
q Ed
s
m D
tD = √ (2.10)
q sEs + 2dEd

Therefore the total flight time T0 of the zero energy ion will be given by the sum of Ts ,
Experimental Techniques and data analysis 52

Td and TD

T0 = Ts + Td + TD (2.11)
s s √ √
m s sEs sEs + 2dEd D
T0 = { − + +√ } (2.12)
q Es Ed Ed sEs + 2dEd

If the fragment ion is produced with an initial kinetic energy, in a random direction

with respect to the TOFMS axis, then in the above analytical calculation, we have to
incorporate the component of velocity of the ion along the TOF axis (vz ). Then the time
of flight (T) of the ion will be given by

q
vz q 1 1 vz2 + as s + 2ad d D
T =− + vz2 + as s{ − } + +q (2.13)
as as ad ad 2
vz + as s + 2ad d

In case of a practical TOFMS, the above equation will be converted to

q
T = A + B m/q (2.14)

where the constant A arises due to the instrumental effect and electronic delays in the
cables. Equation 2.14 is known as calibration equation and is used to identify different
fragment ions from the TOF spectrum under the same extraction as well as acceleration

fields. The deviation of the fragment ion from the spectrometer axis depends on its
component of velocity, perpendicular to the applied extraction field.

In a real experimental situation, ions are not formed in a point region, but in a re-
gion having certain dimensions. This affects the flight time of the ion and reduces the
time resolution of TOFMS by broadening the peak in the TOF spectrum. It would be

desirable to minimize this dependency so that identical ions, created at different points
within the source volume, arrive at the detector simultaneously.
Experimental Techniques and data analysis 53

Space focusing condition

The resolution of TOFMS depends on the initial spatial and kinetic energy distribution
of the created ions. Suppose two ions of same mass and charge are formed at the same
time with same initial kinetic energy but at two different points s1 ± ∆ in the interaction

region of the TOFMS (Fig. 2.6). Now these two ions will experience different potential
gradient and gain different energies due to their origin of formation. The ion starting
from s1 + ∆ will be accelerated to higher kinetic energy compared to the ion starting
from s1 − ∆. The ion formed at the back of the source (s1 + ∆) will enter the drift
region later but eventually pass the ion started from s1 − ∆ due to larger velocity. The

maximum resolution of the TOFMS would be obtained at the point where the overtaking
has taken place. In practice by adjusting the extraction (Es ) and acceleration (Ed ) fields
and the path lengths (s, d) in different regions of TOFMS, we can achieve a space focus
plane where ions of any given mass arrive at the same time. The condition (dT/ds)0,s0

= 0 gives the position where two ions formed at s1 ± ∆ pass each other. The above
condition results into
!
3/2 1 d
D= 2s0 k0 1− 1/2 s
(2.15)
k0 + k0 0

where the initial kinetic energy of the ion is zero, s = s0 and

k0 = (s0 Es + dEd )/s0 Es (2.16)

Since s0 , d and D are fixed, the space focusing is achieved by adjusting the Ed /Es ratio
in double field TOFMS. For single field TOFMS, d = 0. Therefore, the space focusing
condition becomes purely geometric in this case with D = 2s0 . Our TOFMS is designed

to satisfy this condition.

Experimental Techniques and data analysis 54

2.3.3 Design of the TOFMS

-V2 -V2 -V1 +V1

MCP
CEM
& DLD
z-axis

Drift tube (D)

Resistance
Acceleration Extraction
region (d) region (s)

Figure 2.7: Schematic representation of a time–of–flight mass spectrometer.

The design of TOFMS is based on the principle of simultaneous measurement

of ion TOF and spatial spread transverse to the TOF direction,under the action of a
uniform electric field. The spectrometer is assembled inside the ion – molecule collision

chamber which is made of stainless steel (S.S.). The electric field is generated by a set
of parallel rings (electrodes) in a potential divider arrangement. All the rings of TOFMS
are made of S.S. rings of 50 mm inner diameter and 90 mm outer diameter. The two
electrodes (pusher and puller) in the TOFMS, used in the LEIBF facility, has separation

of 44 mm. After extraction, the ions are accelerated through a distance of 80 mm

before they enter into the field free drift tube of length 200 mm. Four equally spaced
co-axial rings (similar to the extraction electrodes), with opening of 50 mm diameter
and covered with fine nickel mesh of 95% transparency, form the acceleration region

of the TOFMS (Fig. 2.7) [4]. Well shielded high voltage (HV) cables are connected to
the electrodes by spot welding for the application of voltages. To maintain the uniform
electric field in the extraction as well as acceleration region we have used a step-down
resistor network outside the chamber. It is used to vary and set voltages on each of the
Experimental Techniques and data analysis 55

Figure 2.8: Actual photograph of a time of flight mass spectrometer.

electrode through a multi-pin electrical feedthrough. To avoid field penetration in the

drift zone, especially from the high voltage of the MCP front, the final electrode before
the detector is also covered with mesh.

In the other TOFMS, used in Pelletron beam line, uniform electric field is applied
over a length of 99 mm. Ions travel through this field, followed by a field free drift
region of 198 mm, before hitting the detector. The photograph of a TOFMS is shown
in Figure 2.8.
Experimental Techniques and data analysis 56

2.3.4 Optimization of TOFMS extraction voltage

15000
178 V/cm

128 V/cm

85 V/cm
12000
TOF (Channel number)

9000

6000

3000

2 3 4 5 6 7
(m/q)

Figure 2.9: Variation of time–of–flight (TOF) as a function of fragment mass to charge

ratio and applied extraction field.

Optimization of TOFMS is crucial in collision induced molecular dissociation

studies. The TOFMS allows three-dimensional (3D) momentum imaging of a frag-
ment ion from the measured TOF and transverse deflections. The width of the TOF
peak is proportional to the energy spread of the ions. However in real experiments the

finite size of the source and the inherent thermal distribution of velocities of molecules
lead to the broadening of the ion peaks. Optimization of TOFMS demands tuning of
the ion beam for proper overlap with the target gas jet and locating the ion – molecule
overlap volume correctly with respect to the spectrometer axis for good position and

TOF spectra. Since the ion beam enters the spectrometer normal to the extraction field,
it suffers a deviation. By using electrostatic deflectors, we sweep the ion beam in the
horizontal and transverse directions to position it on the target even when the extraction
field is on. But this may produce a large number of secondary electrons due to the beam
Experimental Techniques and data analysis 57

grazing the extraction ring. The deviation problem of the ion beam is solved by intro-
ducing a new compensating plate which partially compensates the effect of extraction

field on the ion beam trajectory. Smaller the overlap volume of ion beam and target gas
better will be the momentum resolution for the spectrometer. Therefore the capillary
of the needle should have a suitable diameter and aspect ratio. It is important to ensure
that the overlap of the ion beam and molecular beam occurs within about 0.5 mm of the
axis of the spectrometer in order to obtain high coincidence efficiency. The position of

the overlap along the extraction axis is equally critical, in order to meet the space focus
condition. The electric field applied to the extraction electrodes is chosen according to
the goal of the experiment. For angle and energy resolved measurements small extrac-
tion voltages are required to ensure better momentum resolution. But then it suffers

from low collection efficiency, since only few ions emitted in a cone around the axis of
the TOFMS reach the detector. On the other hand a high extraction field, though good
for identification of energetic fragment ions with high collection efficiency, but deterio-
rates the momentum resolution and accuracy of the kinetic energy release distributions

(KERD). Thus an optimum value of extraction voltage is required for the TOFMS. Final
optimization is done on the basis of the TOF and position spectra. An atomic gas (e.g.
argon) is used for this purpose to ensure negligible momentum width of the produced
ions. The goal is to reduce the FWHM of the parent ion peak and minimize the position
of the same in a very small area by tuning the ion beam trajectory and extraction field.

A good position spectrum with highest intensity at the centre is also desirable to en-
sure good position resolution of the TOFMS. We have found that the extraction field of
100-200 V/cm is reasonable for our spectrometer to achieve maximum momentum res-
olution and ion collection efficiency for a range of target molecules. Figure 2.9 shows

how the TOF of various ion species vary with the applied extraction field.
Experimental Techniques and data analysis 58

2.4 Position sensitive ion imaging detector

The recoil ions produced in the fragmentation of multiply charged molecular ions in-

duced by energetic ion beams are detected by a position-sensitive detector placed at

the exit of fragment ion TOFMS. The ion detector used in our experiment consists of
twin micro channel plates (MCP) in chevron configuration with a delay line anode be-
hind it [6, 7]. The position sensitive delay line detector used in the present experiments

is developed by Schmidt-Böcking and co-workers and is commercially available from

Roentdek GmbH [8]. The micro channel plates used in the LEIBF facility have diam-
eters of 50 mm whereas the MCPs used in Pelletron experiments are much larger (80
mm in diameter). The advantage of using chevron MCP over the straight channel MCP

is significantly the more gain at a given voltage. Typical position resolution of the mi-
cro channel plate detector is less than 0.1 mm; multi-hit dead time is about 10–20 ns
and the counting rate is about 1 MHz. The electron gain obtained was of the order 107
at the typical operating voltage of 2500 V. The MCPs detect the incoming recoil ions
and produce the signals well separated from noise; which are later used as timing sig-

nals. The delay line anode provides position information of the recoil ions. An electron
shower, from the back of the MCP, causes image-charge pulses to travel to the ends of
the wires, giving four pulses for every particle hit. The position at which the electron
cloud, corresponding to a fragment ion, impinged on the DLD is encoded by the signal

arrival time difference at both ends of each parallel pair delay line, for each dimension
independently. If the four arrival times are x1 , x2 , y1 and y2 along the X-axis and Y-axis
respectively, then the x and y coordinates of a single event will be (x1 – x2 , y1 – y2 ) in
time units. The single pitch propagation time on the delay line is about 0.73ns (per 1

mm) for our DLD and the correspondence between position and time in the two dimen-
sional image is twice this value i.e. about 1.46ns. The sum of the arrival times (within
the time resolution of about one ns) for each event is constant along one wire giving, x1
+ x2 = y1 + y2 . Only those events that obey this condition are taken into account, and
Experimental Techniques and data analysis 59

all others are ignored. If more than one ion pulse strikes the MCP assembly, then each
delay line produces a series of pulses and the above criterion allows one to correctly as-

sign individual pulses to individual ions, thereby distinguishing between true and false
events. This is known as time-sum condition. The total propagation time for DLD40
along one dimension is 38 ns with a base width of about 5 ns. Figure (2.10) shows the
photograph of the DLD based MCP used in the present experiments.

Figure 2.10: Photograph of the MCP equipped with position sensitive delay line detec-
tor.

2.5 The electron detector

For detection of ejected electrons we have used a Channel electron multiplier (CEM)
in the LEIBF set up and a micro channel plate of 40 mm diameter in the Pelletron set
up, both mounted opposite to the fragment TOFMS for our experiments. Electrons

are extracted from the interaction zone by the application of a positive potential to the
extraction electrode (pusher). The extracted electrons in the LEIBF setup enter into a
metal cylinder through a hole of 5 mm diameter and are finally detected by a Channel-
tron (SJUTS, KBL 15 RS), mounted inside the cylindrical box. The cylinder shields
Experimental Techniques and data analysis 60

Figure 2.11: Photograph of the channeltron, used in one of the present experiments
reported in this thesis, along with its cylindrical housing.

the electron detector from external interferences and unwanted scattered particles thus
enabling only the real event electrons getting detected for the trigger of data acquisition.
Figure 2.11 shows the CEM along with the metallic box.

2.6 Signal Processing and data acquisition

The data acquisition system in the LEIBF setup is shown in Figure (2.12). It is based on
the time of flight (t) and the position (x, y) measurement of each ion produced in the ion–

molecule collision. In order to measure (x, y, t), all the charged particles are detected in
coincidence with the ejected electrons produced in an ion-molecule collision event. In
the present experiment, six signals are produced: an ion signal from the MCP, an elec-
tron signal and four signals from the delay line anode. The preamplifier amplifies the

weak signal from a detector and drives it through a cable that connects the pre-amplifier
with the rest of the equipment. Since the input signal at the preamplifier is generally
weak, therefore pre-amplifiers are normally mounted as close as possible to the detector
so as to minimize the cable length. Signals from the MCP and DLD40 are decoupled
Experimental Techniques and data analysis 61

Figure 2.12: Data acquisition methodology: electronic setup of the time– and position–
sensitive multihit detector system. Different abbreviations mean the following. DA:
differential amplifier, CFD: constant fraction discriminator, TA: timing amplifier, PA:
pre-amplifier, D: delay, NEC: NIM-ECL converter, TDC: time-to-digital converter.

R-C circuit
Electron CEM

+ HV

Electron Signal for triggering DAQ

Figure 2.13: The decoupler circuit used to decouple electron signal from the applied dc
voltage.

from dc voltages on the wire using RC decoupling and special transformer (FT12–TP)
circuits and finally fed to a six-channel differential amplifier (DLATR6 from Roent-
Dek). The amplified signals are then fed to the constant fraction discriminators (CFD)
Experimental Techniques and data analysis 62

inside the DLATR6 box to discriminate the true signals from noise by proper setting of
threshold. The CFD gives out fast ECL signals which are directly taken to a 32-channel

multihit time-to-digital Converter (TDC; LeCroy 3377). We can also get the analogue
outputs of the amplified signals from DLATR6, before the CFD-stage, for monitoring
the noise level and signal quality from the delay line. The electron signal from the
Channeltron is first ac coupled by an R–C circuit (Fig. 2.13) and then taken to a pream-
plifier of gain 200 (ORTEC VT120A). The amplified electron signal is discriminated

from the noise by using an external CFD (ORTEC 935). The ORTEC 935 CFD deliv-
ers a fast negative NIM digital signal with timing properties almost independent from
the pulse height for signals above a selectable threshold level. The output signal from
the CFD is converted to differential ECL signal by a 16-channel NIM-ECL converter

(LeCroy 4616) and finally fed to the input of the TDC. This was necessary because our
LeCroy 3377 TDC can take only ECL inputs. The electron signal serves as the master
START signal (common trigger) for every event, triggering five channels of the TDC:
four channels are stopped by delay-line signals, giving the ion position whereas the fifth

one is stopped by the ion MCP signal, giving the time-of-flight of the ion. The 16-bit
TDC (Lecroy 3377) has a minimum timing resolution of 500 ps, maximum range of 32
microseconds and multihit pulse pair resolution of 10 ns. It can take up to 16 hits for a
single trigger in the selectable time range and work in both common start and common
stop mode. Digitized outputs of the five channels are read event by event and stored as

a list in a file called list-mode file. The five analogue signals from DLATR6 (MCP and
four delay line signals) are also discriminated from noise using external CFDs and fed
into five different channels of TDC after NIM-ECL conversion. These signals are used
to cross check the TOF and position data independent of DLATR6. Data from the TDC

is collected in list mode through a home built Ethernet based CAMAC crate controller
[10]. The data acquisition program is home-written in C language on Linux operating
system.

The five ion signals from the MCP based delay line detector are fed to ATR19
Experimental Techniques and data analysis 63

(an amplifier–discriminator module from RoentDek) in the Pelletron setup which hosts
three DLATR differential timing amplifier and CFD boards, each with two independent

channels. The outputs of ATR19 allow verification of the analog signals after the first
amplification stage on the DLATR board and of the NIM or differential ECL timing
output signals. A LeCroy 133MTD TDC (500 ps resolution and 32 µs range) has been
used for data acquisition in the Pelletron setup which has a PC interface written in MS
Visual C++ language. The digitized data obtained from TDC are stored in a list mode

file using CoboldPC software.

The event list in the list mode file consists of the data for position information (x1;
y1; x2; y2) and for time information t. These entries are used to determine (x; y; t)
triplet. In practice, the false coincidence counts will be recorded if the event rate is

high. Usually we keep the event rate as low as possible. Most of the data presented in
this thesis is acquired with event rate ranging from 100 – 300 Hz. In this case the false
coincidence counts are expected to be low and with time sum condition we can get only
true coincidence counts from the list mode file.

2.7 Data analysis

Initial momentum components of all the detected ions produced in the ion-molecule
interaction can be obtained from the list mode file if the position of arrival or deflection

from the spectrometer axis is known besides the TOF of that ion.

The data analysis is done in combination with ROOT [11] from CERN and the
interface programs necessary to convert the binary data into ROOT specific format de-
veloped in-house. ROOT is an object oriented framework for large scale data analysis,

a C++ replacement of the PAW program developed at CERN. It is a free open source
product which provides programming interface to use in own applications as well as
graphical user interface for interactive data analysis. The ROOT data analysis frame-
work is written in C++ and heavily relies on it. The binary raw data recorded in list file
Experimental Techniques and data analysis 64

Define data format of each event in ROOT file

No. of TDC channels active
No. of hits allowed per channel

Start of individual file process

Input : raw datafile
Output : datafile.root

No
Header found ?

Loop till Yes; LeCroy TDC

‘end of file’
Initialize all TDC channels and
no. of hits per channel to ’zero’

Check for data word

Read all LeCroy TDC channels
Store multiple hit data per channel

Write timing data in ROOT file

Figure 2.14: Flowchart showing conversion of raw binary data into ROOT format.

is first converted into ROOT specific format using a C program. Figure (2.14) shows

the flowchart for this conversion. The ROOT files thus obtained enable us to refer each
event independent of other and make it possible to do an event by event analysis of
the acquired multihit data. To extract valuable information like position coordinate and
TOF of each fragment produced in ion induced molecular dissociation, we operate dif-

ferent C++ programs, written in ROOT environment, on the above ROOT data files.
The data read from the LeCroy TDC, used in LEIBF setup, is in LIFO (Last In First
Out) order, the hit that is read first is actually the last particle to arrive in the multi event
buffer. Therefore, while analyzing the data we reversed the order in our program to
Experimental Techniques and data analysis 65

make hits for each event chronological in real time. Only those events are taken into
account for which both position and TOF values are recorded and have satisfied the

time sum condition.

2.7.1 Analyzing TOF spectrum

800 Ar 2+
700 Ar 4+
counts (arbitrary unit)

600 Ar 3+
500

400

300

200 Ar 5+ N2+
2

100 N2+
N+ N+2 Ar +
0
4000 5000 6000 7000 8000 9000 10000
Time of flight (0.5 ns / channel)

Figure 2.15: TOF spectrum of Ar ions produced in the interaction of neutral Ar mole-
cules with 60 kV/q Ar6+ ions. Presence of N2 ions are actually coming from the back-
ground.

The raw TOF spectrum can be seen as histogram of events in the channel corre-
sponding to the MCP front. In order to analyze the ion molecule collision data, the first
step is to identify the ion species in the TOF spectrum and find out the point (T0 ) corre-

sponding to zero initial kinetic energy. For this purpose TOF spectrum of a monoatomic
gas (for example, Ar, as shown in Figure 2.15) is recorded in crossed beams mode, un-
der the same experimental condition. The mean positions of Arn+ (n=1–5) TOF peaks
q
are taken and linearly fitted as a function of m/q (Fig. 2.16) where m and q corre-
Experimental Techniques and data analysis 66

(m/q) = 0.039 + 0.00127 TOF

6

(m/q)

2
2000 3000 4000 5000
Time of Flight (ns)

Figure 2.16: Calibration curve drawn from the TOF spectrum of Arn+ (n=1–5) ions
shown in Figure 2.15.

spond to the mass and charge state of the Ar ions. This provides the calibration equation

s
m
= constant + slope × T OF (2.17)
q

from which we can identify various fragments, generated with zero or negligible kinetic
energy, during the ion molecule interaction. The T0 information of an ion is used to
determine the z component of momentum of a fragment ion.

Figure (2.17) shows the typical TOF spectrum of dissociative nitrogen molecules
caused by the impact of 60 kV/q Ar6+ ion beam on neutral nitrogen in an extraction
field of 190 V/cm. Apart from the peaks due to undissociated molecular ions N+
2,

N2+ + 2+ 3+
2 , double peaks of N , N , N , etc. are also observed in the TOF spectrum.

For all fragment ions, the peak appearing at lower channel number corresponds to the
fragment emitted along the extraction field while the other peak appearing at higher
channel number corresponds to the same ion but ejected opposite to the field direction.
The fragment ions, emitted perpendicular to the TOF axis with high energy, are actually
Experimental Techniques and data analysis 67

1600

1400
counts (arbitrary uniit) N2+
2
1200

1000

800

600
N2+ +
400 N
3+
N
200 N4+ N+2

0
3000 4000 5000 6000 7000 8000
TOF of first fragment (0.5 ns / channel)

Figure 2.17: TOF spectrum of ions arising from 60 kV/q Ar6+ impact on neutral N2
molecules at an extraction field of 190 V cm−1 .

lost from the system because of the finite size of the detector, thereby giving a dip in
between the two peaks. The time difference between the forward and backward directed

fragment ions signifies the kinetic energy carried away by the fragment. The FWHM
of the N+
2 TOF peak is 9 ns while the modal TOF is 4133 ns. This gives the mass

resolution of m/∆m = 230 for our TOFMS.

2.7.2 Coincidence map

In most of the molecular fragmentation, two fragments ejected may have different mass

and velocity. Ions ejected from one dissociation event carry valuable information about
the precursor state. The same ion can also be ejected from many other events but their
energies will be different in all the cases. Therefore detection of ions from TOF spec-
trum alone cannot give information about a particular dissociation event. If multiple

ions are ejected from a single event then the TOFs of the two fragment ions may have a
correlation in the ion-ion TOF plot. To distinguish between different possible fragmen-
tation pathways and to know at which stage charge separation takes place, one has to
measure all the fragment ions in coincidence along with their momenum correlations.
Experimental Techniques and data analysis 68

These correlations can be best visualized by plotting a two dimensional histogram with
the TOF of the first particle hit being the X axis and that of the second particle hit being

the Y axis. The histogram representing interparticle correlation is referred to as coin-

cidence map which is an useful tool for identification and separation of multi-particle
events. The underlying principle behind coincidence technique is the time ordered de-
tection of each fragment ions ejected from a particular event.

+
TOF of second fragment (0.5 ns / channel)

N3+ N2+ O2+ N +

O
140
+
O
6000 120
N+
100

5000 80

O2+ 60

4000 N2+ 40

20
N3+
3000
3000 3500 4000 4500 5000 5500 6000 6500
TOF of first fragment (0.5 ns / channel)

Figure 2.18: Ion–ion coincidence map of two–particle events observed in the collision
of 60 kV/q Ar6+ ions with neutral N2 molecules at an extraction field of 190 V cm−1 .
Due to the presence of oxygen gas as impurity in the N2 cylinder, coincidence islands
corresponding to dissociation of O2 ions are also seen in the above coincidence map.

Figure (2.18) shows the typical coincidence map for fragmentation of N2 in colli-
sion with 60 kV/q Ar6+ ions. Each point in the coincidence map is due to the coincident
detection of two fragment ions. The dark patches, known as coincidence islands, signify
correlated ions generated in a molecular dissociation and represent fragmentation path-

ways of various precursors. For example, we have identified five dissociation channels
of multiply charged N2 molecular ions from Figure 2.18. They are:

N22+ → N + + N + (2.18)

N23+ → N 2+ + N + (2.19)
Experimental Techniques and data analysis 69

N24+ → N 2+ + N 2+ (2.20)

N24+ → N 3+ + N + (2.21)

N25+ → N 3+ + N 2+ (2.22)

The dynamic and kinetic information of a particular dissociation channel can be ob-

tained by putting a selective cut on the corresponding coincidence island. In the frag-
mentation of polyatomic molecules, usually many distinct dissociative events occur, the
pattern of which is reflected in various coincidence islands. The identification of these
islands is accomplished on the basis of calibration equation.

The geometrical properties of a coincidence island, i.e. the shape, size, orien-
tation, etc., contain useful information about the dynamics of molecular dissociation,
such as kinetic energy release (KER) distribution, ejection direction of the fragment
ions, dissociation sequence, etc [12, 13]. Theoretical analysis of coincidence mapping

in mass spectrum has shown that the structure of the coincidence island is a momen-
tum contour, the length reflects the kinetic energy of the ion pair and the slope gives
information about the charge and momentum of the involved particles. We can define
the slope of the coincidence island between the TOF distributions of correlated ionic

species in the coincidence map as:

Py Qx
tan φ = (2.23)
Px Qy

where φ is the angle made by the major axis of the nearly elliptical coincidence island
with the horizontal TOF axis i.e. x-axis and P, Q are the momentum and charge state
of the ions displayed along the x- and y- TOF axis. Conservation of linear momentum
in a two–body dissociation of molecular dication into two singly charged ions yields a

slope of -1 of the coincidence island.

In case of three-body fragmentation, the measured slope is influenced by the mo-

menta of the dissociation products. A multiply charged ion may dissociate into several
Experimental Techniques and data analysis 70

fragment ions plus one or two neutrals. Such neutrals remain undetected by the charged
particle detectors. Therefore the amount of momentum carried away by them cannot

be measured. According to Eland [12], three-body fragmentation leading to two ions

and one neutral, can be of concerted (C) or sequential type. Sequential fragmentation
can be categorized as initial charge separation S(i) and deferred charge separation S(d).
In concerted fragmentation all three fragments are created simultaneously. In the S(i)
reaction, the total charge of the parent molecular ion is first shared by two fragment

ions and later one of the fragment ions emits a neutral particle. Whereas in S(d), neutral
particle is emitted in the very first step, followed by a charge separation in the second
stage of the reaction. No single prescription can be applied to determine the slope of the
dumb-bell like or elliptical coincidence island arising from a concerted fragmentation.

In S(i) fragmentation, the slope of the best fit-line is given by -(q1 /q2 )[m2 /(m2 +m3 )] or
-(q1 /q2 )[(m1 +m3 )/m1 ], where the ions are labeled 1, 2 in order of their detection and
3 refers to the neutral fragment. Which of the two values is to be used depends on
the intermediate diatomic cation. In S(d) fragmentation, the slope of the best-fit line is

-(q1 /q2 ). For sequential dissociation, the shape of the coincidence islands are parallelo-
grams.

2.7.3 Analyzing position spectrum

Apart from the axial motion in a TOFMS, ions also suffer deflections in a plane perpen-
dicular to the TOF axis due to their transverse velocity components and fall at different
positions on the detector. The position spectrum of an ion is basically a two dimen-
sional histogram of its x and y deflections under the time sum condition and is obtained

by setting a window on the TOF spectrum of the ion of interest. To determine the po-
sition in one dimension, we find out the delay in arrival times between two pulses from
the two ends of DLD along that dimension. Figure 2.19 shows the position image of
dissociating N2 under the impact of 60 kV/q Ar6+ projectile ions. The intense spot in
Experimental Techniques and data analysis 71

80
400

60
350

Y Position (channel no.)

40
300

20
250

0
200

-20 150

-40 100

-60 50

-80
-80 -60 -40 -20 0 20 40 60 80
X Position (channel no.)

Figure 2.19: Position image from fragmentation of N2 in collision with 60 kV/q Ar6+
projectiles. This is a 2D histogram of the x– and y– deflections of the recoil ions.

the position image is actually the centroid of interaction volume projected on the detec-
tor plane. The transverse components of velocity of an ion can be determined from the
position of impact of the fragment ions on MCP using the relations

vx = (x − x0 )/t (2.24)

vy = (y − y0 )/t (2.25)

where (x0 , y0 ) is the reference point obtained from the observed image (corresponding
to the most intense spot in Figure 2.19).

2.7.4 Momentum calculation

The kinetic energy release distribution as well as geometry of the parent molecular ion
can be estimated from the three momentum components of correlated fragment ions.
Experimental Techniques and data analysis 72

Assuming that the kinetic energy gained in acceleration is much larger than the kinetic
energy released upon dissociation, the time difference between the zero energy and non-

zero initial velocity ions (T0 -T) can be written in a simplified form. Thus from equations
2.12 and 2.13 we get the component of momentum of the fragment ion parallel to the
TOFMS axis (say, pz ):
pz = mvz = qEs (T0 − T ) (2.26)

where q is the charge state of the ion, Es is the extraction electric field, T is the flight
time of the ion proportional to the component of initial momentum along TOF axis and
T0 is the standard flight time of the same ion initially at rest and determined from the
calibration equation. It is noted that this z–component of momentum is only the first

order term which is dominant and good enough. The other terms are neglected. The
two other components of momentum, parallel to the detector plane and perpendicular
to the TOFMS axis are determined from the position and TOF information of the ion
having mass m.

px = mvx (2.27)

py = mvy (2.28)

The total momentum of each fragment ion thus can be obtained from the relation

q
| p |= px 2 + py 2 + pz 2 (2.29)

All the above calculations are performed using simple programming which extracts the
time and position coordinates of each fragment ions from the list mode data file.

2.7.5 Geometry of the parent molecular ion

When a molecular ion breaks up, the fragments fly apart in a direction governed by
the geometry of the molecular ion at the instant of break-up. Thus the geometry of the
Experimental Techniques and data analysis 73

200 11

150 10
st +
1 N
9
100

Momentum (a. u.)

2ndN+ 8
50
7
0
6
-50
5
-100
4

-150 3

-200 2
-200 -150 -100 -50 0 50 100 150 200
Momentum (a. u.)

Figure 2.20: Correlated momentum map of N+ –N+ coincidence island showing ejec-
tion of two N+ ions with equal and opposite momentum. The arrow indicates the ejec-
tion direction and blue histogram represents momentum distribution of the first N+ ion.
The momentum of 1st N+ is taken as the reference and the components of momentum
of 2nd N+ , parallel and perpendicular to it are plotted.

parent molecular ion at the instant of decay can be estimated, if we examine the linear

momenta of all the fragments detected in coincidence. Also from the observed complete
momentum of each fragment, we can visualize the ejection pattern of the fragments in
two–body or three-body molecular break up. The angular information between any two
fragments can be used to obtain the geometry of the parent molecular ion prior to its

fragmentation. If the momentum vectors of the ions flying apart be p~A and p~B , then the
free angle between them will given by

p~A .~pB
αA:B = cos−1 (2.30)
| pA || pB |

In all analysis of momentum and angular distributions, we take the direction of the
ion arriving first as the reference and plot the longitudinal and transverse momentum
Experimental Techniques and data analysis 74

components of the other fragments with respect to the reference in a two dimensional
histogram. Figure 2.20 shows the momentum correlation between the two N+ ions

emitted from the dissociation of N2+

2 ions where momentum distribution of the second

N+ ion has been plotted with respect to the momentum of first N+ ion (the continuous
blue line). Since it is a pure two–body dissociation, the two N+ ions have been ejected
with equal and opposite momentum. Although the neutral fragment in the three-body
break-up is not detected, its momentum can be determined by the momentum balance

in the centre-of-mass frame. Consequently, the angle between the neutral and charged
fragment can also be obtained. Thus from the momentum map, one can identify the
sequential as well as concerted break ups.

2.7.6 Kinetic energy release

The multiply charged molecular ions formed in the ion molecule interactions are in-
variably unstable and dissociates instantly into copious quantities of energetic charged
atomic fragments. The sum of the kinetic energies of the fragment ions formed in the
dissociation process is expressed as Kinetic Energy Release (KER). The KER reflects

the properties of the ground state wave function of the molecule as well as the shape of
the potential energy surfaces in which the molecular ion has been formed. In principle
molecular ions may have many repulsive states. The strength of the perturbation decides
which repulsive potential energy will be accessed by the precursor. It is not possible to
have all the information about various excited states of fragmenting ion but the distribu-

tion of KER may put some light on some of the excited states. Thus, the investigation
of KER is useful for understanding the dissociation dynamics of molecules.

The width of the TOF peak is proportional to the energy spread of the correspond-
ing ion. Finite size of the source and inherent thermal distribution of velocities are
responsible for the broadening of TOF peak of an ion. So often it is not possible to
use the peak width as a means of kinetic energy calculation. The time difference (△Ts )
Experimental Techniques and data analysis 75

between the forward and backward emitted fragment ions is the measure of fragment
kinetic energy in the centre of mass frame through the relation

2(2mUker )1/2
△Ts = (2.31)
qE

where q is the ionic charge, E is the extraction field used in our TOFMS and m is the
mass of the ion. In case of two body Coulomb break up the KER is given by

q1 q2
KER = 14.4 (2.32)
R

where R is the inter atomic distance in Å unit and q1 , q2 are the charges of the fragment
ions. Experimentally, the kinetic energy of a fragment ion can be obtained from its
measured momentum using the relation

p2
KE = (2.33)
2m

where p is the momentum of a fragment of mass m.

2.8 Efficiency of a TOFMS

The efficiency of a TOFMS depends on the applied extraction field, initial velocity

of the fragment ions as well as efficiency of the detector. An ion having a large initial
velocity perpendicular to the TOF axis may get lost from the system because of the finite
size of the detector, which restricts the detection of ion from all possible directions. We
have checked the efficiency of a TOFMS analytically considering its actual geometry (s

= 19 mm; d = 114 mm; D = 250 mm) and the voltages applied for extraction of fragment
ions. At 170 V/cm extraction field, we have found that the collection efficiency is
100% for N2+ ion having kinetic energy less than 6.6 eV irrespective of the direction
of emission of the ions. But when we increase the extraction field the high energy ions
Experimental Techniques and data analysis 76

start reaching the detector. For instance, at 200 V/cm, N2+ ion of energy up to 7.9 eV
reached the detector. It is not only true for N2+ ions but for all doubly charged ions

generated in the ion molecule collision because the collection efficiency depends only
on the charge and not on the mass of the fragment ion.

TOF of second fragment (0.5 ns / channel)

TOF of second fragment (0.5 ns / channel)

6200 1800
N2+:N+

6000 1700

5800 1600

5600 1500

5400 1400
3800 4000 4200 4400 1000 1050 1100 1150 1200 1250
TOF of first fragment (0.5 ns / channel) TOF of first fragment (0.5 ns / channel)

(a) (b)

Figure 2.21: Coincidence island of N2+ :N+ obtained using (a) 190 V/cm (long) and (b)
540 V/cm (short) extraction field in the TOFMS.

Enhancement of extraction field increases the acceptance solid angle of the TOFMS.

By increasing the potential difference in the extraction region or decreasing the flight
path of the ion, we can increase the extraction field. Since the MCP is operated at -2150
V, the maximum voltage that could be applied to the drift tube of the TOFMS would be
less than the above value. Therefore by increasing the potential difference we cannot

increase the field to the desired value. This is possible only by shortening the length of
the TOFMS where the length of the extraction and acceleration regions are cut down
to 19 mm with a drift region of only 57 mm. The resistance chain, used as a potential
divider network is also modified accordingly. The changed geometry of the TOFMS

increased the extraction field without any application of higher potential. We found that
in the short TOFMS, MCP can detect N2+ ions up to 59 eV energy emitted perpendicu-
lar to the TOF axis when the extraction field is 340 V/cm. Even if we use the same 170
V/cm field, N2+ ions of energy 29 eV will get detected in the short TOFMS.
Experimental Techniques and data analysis 77

We have also verified the results experimentally from ion impact dissociation of
N2 molecules. Figures 2.21(a) and 2.21(b) show coincidence island of N2+ and N+ for

a long and short TOFMS. In the long TOFMS, where we have applied a field of 190
V/cm, the high energy fragment ions emitted at 90◦ are lost whereas short TOFMS (540
V/cm) has been able to guide most of the fragment ions towards the detector. Therefore
the coincidence island is dumb-bell shaped in the former case and it is a straight line in
the latter.
Bibliography

[1] D. Kanjilal, S. Chopra, M. M. Narayanan, I. S. Iyer, V. Jha, R. Joshi and S.K.

Datta, Nucl. Instrum. Methods Phys. Res., Sect. A 328, 97 (1993)

[2] G. Rodrigues, P. Kumar, U. K. Rao, C. P. Safvan, D. Kanjilal and A. Roy, Pro-

ceedings of 15th International workshop on ECR ion sources ECRIS, Finland

(2002)

[3] P. Kumar, G. Rodrigues, U. K. Rao, C. P. Safvan, D. Kanjilal and A. Roy,

Pramana 59, 805 (2002)

[4] S. De, P. N. Ghosh, A. Roy and C. P. Safvan, Nucl. Instrum. Methods Phys.

Res., Sect. B 243, 435 (2006)

[5] W. C. Wiley and I. H. McLaren, Rev. Sci. Instrum. 26, 1150 (1955)

[6] I. Ali, R. Dörner, O. Jagutzki, S. Nüttgens, V. Mergel, L. Spielberger, Kh.

Khayyat, T. Vogt, H. Bräuning, K. Ullmann, R. Moshammer, J. Ullrich, S.

Hagmann, K. Ō. Groeneveld, C. L. Cocke, H. Schmidt-Böcking, Nucl. In-
strum. Methods Phys. Res., Sect. B 149, 490 (1999)

[7] O. Jagutzki, V. Mergel, K. Ullmann-Pfleger, L. Spielberger, U. Spillmann,

R. Dörner, H. Schmidt-Böcking, Nucl. Instrum. Methods Phys. Res., Sect. A

477, 244 (2002)

[8] http://www.roentdek.com

78
Experimental Techniques and data analysis 79

[9] CHANNELTRON Electron Multiplier Handbook For Mass Spectrometry Ap-

plications, GALILEO

[10] E. T. Subramaniam, K. Rani, B. P. Ajith Kumar, R. K. Bhowmik, Rev. Sci.

Instrum 77, 096102 (2006)

[11] http://root.cern.ch

[12] J. H. D. Eland, Mol. Phys. 61, 725 (1987)

[13] M. R. Bruce, L. Mi, C. R. Sporleder and R. A. Bonham, J. Phys. B: At. Mol.

Opt. Phys. 27, 5773 (1994)
Chapter 3

Dissociation dynamics of carbon

dioxide (CO2)

3.1 Introduction

Highly charged ion molecule interaction is an efficient way to produce molecular ions
in excited states which due to their extreme instability dissociate into various energetic

fragments. The energetic of fragmentation of diatomic molecules is often simple and

can be explained by the coulomb explosion model (CEM). CEM assumes that, in case
of a diatomic molecule, the final kinetic energy of the fragments is equal to the initial
Coulomb repulsion energy. But it does not take into account the electron correlations

which are important in determining the KER. Thus CEM is found to be inadequate to
explain experimental values of the kinetic energy released in the molecular dissociation
process [1]. For triatomics, the fragmentation pattern is far more complex. Besides
two–body fragmentation, three–body fragmentation and two sequential two–body frag-

mentations are also possible. If one of the fragments is neutral, then the kinematics
of three–body dissociation can only be deduced indirectly, provided the momenta of
the two ionic fragments are completely determined. Here we present the dissociation
mechanism of multiply charged CO2 molecular ions by using time of flight (TOF) co-

80
Dissociation dynamics of carbon di oxide (CO2 ) 81

incidence technique.

3.1.1 Prior work on fragmentation of CO2

Fragmentation of CO2 is a much studied topic, with synchrotron radiation [2, 3], elec-
tron impact [4], fast heavy ion beams [5], slow highly charged ions [6], as well as with
intense lasers [7, 8, 9] having been employed as perturbations. In an earlier experiment

on fragmentation of CO2 induced by 5.9 MeV u−1 Xe18+ and Xe43+ ions, Siegmann
et al. [5] employed a Coulomb explosion model for the description of the fragmen-
tation process behind C+ +O+ +O+ channel. Though the measured angular spectra of
the ions were reproduced reasonably well by this model, it failed to explain the ob-

served energy distribution. In another experiment with very low energy (120 keV) Ar8+
projectile, Sanderson et al. [10] observed modification of molecular geometry in the
form of widening of the zero point bend–angle distribution. Recoil ions, produced in
100-MeV collisions of Si8+ with CO2 , have been studied by Mathur et. al. using
energy-discriminated mass spectrometry. Kinetic energy distributions of atomic recoil

ions formed in the dissociation of highly charged CO2 molecular ions show that the
peripheral O ions possess larger kinetic energies compared to the central C ion, as ex-
pected from multiply charged, linear CO2 [11] molecule. The relatively large energies
imparted to C ions are explained in terms of a sequential Coulomb explosion model. Ba-

pat and Sharma [12] have reported the bent geometry of dissociating CO2+
2 formed due

to electron impact, by a combination of multiple-ion coincidence and momentum imag-

ing techniques. In a recent experiment, Singh et. al. [3] have studied the formation and
fragmentation of CO3+
2 by 200 eV photons from the INDUS-1 synchrotron, using coin-

cidence mapping. They observed triple-ion coincidence C+ :O+ :O+ resulting from the
complete dissociation of CO3+
2 but could not identify whether it follows via concerted

or sequential process. Photoelectron spectroscopy of CO2 [13] has revealed many pre-
viously unknown vibrational levels of the X Π2g state of molecular ions. Tian and Vidal
Dissociation dynamics of carbon di oxide (CO2 ) 82

[4] reported the dissociation mechanism involved in the single to quadruple ionization
of CO2 due to electron impact. Correct description of dissociation mechanism requires

knowledge of potential energy surfaces involved in the fragmentation process. Many

body fragmentation proceeds via regions of densely populated potential energy sur-
faces which support many transitions [6]. Ion-induced molecular fragmentation studies
in different energy regimes have shown that electron removal due to capture results in
weaker excitation of the target than direct ionization [1] and these energy regimes give

information about various excited electronic states of a given multiply charged molecu-
lar ion. Many theoretical studies for the electronic states of neutral [14, 15], singly [16]
and doubly charged [17, 18] CO2 molecular ions have been reported in the literature,
but the potential energy surfaces of CO3+
2 are largely unexplored.

In the present chapter we have discussed dissociation dynamics of CO2 using time–

of–flight coincidence technique. Various dissociation channels leading to double–ion

coincidences have been identified by analyzing the coincidence spectrum. For analyz-
ing the triple–ion coincidence of CO2 we have employed momentum imaging technique
and performed ab initio quantum chemical calculations in order to understand the dy-

namics of complete dissociation of CO3+

2 [19].

3.2 Dissociation of CO2 by low energy ion impact: Ex-

perimental details

We have studied multiple ionization and fragmentation of CO2 induced by low energy
Ar8+ ion impact using the technique of TOF mass spectrometry. The experiments were

carried out at the Low Energy Ion Beam Facility (LEIBF) at the Inter University Ac-
celerator Centre (IUAC), New Delhi, India. 1.2 MeV Ar8+ ions produced from the
ECR ion source interact with neutral CO2 molecules at the centre of the experimental
chamber. The fragment ions were extracted in the linear TOFMS by application of a
Dissociation dynamics of carbon di oxide (CO2 ) 83

uniform electric field perpendicular to the ion beam and molecular gas jet. The inter-
action region was 44 mm long followed by an acceleration region of 80 mm and field

free drift region of 200 mm. At the exit, ions were detected by the MCP detector and
the electrons were detected by a channeltron in the opposite direction of TOFMS. The
electron signal provided the trigger for multihit coincidence data acquisition. The TOF
measurement was effected by an ejected electron-ion coincidence and recorded by a
multi-stop time-to-digital converter which allowed us to investigate different dissocia-

tion channels of multiply charged CO2 . From the coincidence spectra between various
fragment ions, we separated different dissociation channels of CO2 molecular ions and
using geometrical properties of the coincidence islands, we gained information about
the momentum balance in the fragmentation processes. Details of the set up and data

analysis methodology have already been described in Chapter 2 of this thesis.

3.3 Analysis of dissociation dynamics of CO2

3.3.1 TOF spectrum of dissociated CO2

C2+
25000
counts (arbitrary unit)

20000
+
C
15000
+
CO2

10000
3+
C /O4+ CO2+
+ 2
O
5000 3+
O O2+ +
CO
0
3000 4000 5000 6000 7000 8000 9000
TOF of first fragment (0.5 ns / channel)

Figure 3.1: Time-of-flight spectrum of dissociative CO2 .

Dissociation dynamics of carbon di oxide (CO2 ) 84

The dissociation products observed from the TOF spectrum (Fig. 3.1) of CO2
molecules shows very sharp peaks of undissociated CO+ 2+
2 and CO2 molecular ions.

Existence of molecular dication, CO2+

2 , shows the metastability of such ionic species.

Coulomb explosion of highly charged parent molecular ions generally produces frag-
ment ions with large kinetic energy releases. The broad peak of CO+ present in Figure
3.1 indicates a large kinetic energy release associated with its formation. Complete rup-
ture of O-C-O skeleton producing charged atomic species such as Cq+ and Oq+ ions

(where q varies from 1 to 4) was also evident. The spectra showed a very broad struc-
ture of O+ ion with four peaks and a shoulder at lower mass number. These are due
to the formation of oxygen ions with different kinetic energies through a number of
dissociation channels of excited CO2 molecular ions.

3.3.2 Coincidence spectrum of CO2 showing various fragmentation

pathways

From the offline analysis of the acquired multihit data, we have generated the coinci-

dence map (Fig. 3.2): a two-dimensional histogram between the TOF of the first frag-
ment ion versus that of the second. This gives evidence for formation and dissociation
of multiply charged CO2 molecular ions having charge states 2 to 6 where contributions
from CO5+
2 (coincidence of C
3+
and O2+ ) and CO6+
2 (coincidence of C
3+
and O3+ ) are

very negligible. In the present section we will discuss the possible fragmentation mech-
anism of some of the dissociation channels of COq+
2 from the geometrical properties of

various coincidence islands [20].

Formation of CO+ :O+

From the above coincidence map we had separated a two–body fragmentation channel
of CO2+
2 dissociating into

CO22+ → CO + + O + (3.1)
Dissociation dynamics of carbon di oxide (CO2 ) 85

C3+/O4+O3+C2+ O2+ C+ O+

TOF of second fragment (0.5 ns / channel)

8000 60

CO+
7000 50

6000 40
+
O
30
5000
C+

20
4000 O2+
C2+
10
3000 3+
O
2000 3000 4000 5000 6000 7000

TOF of first fragment (0.5 ns / channel)

Figure 3.2: Coincidence map of the dissociative products of CO2 . Islands correspond-
ing to different breakup channels are shown.

The CO+ :O+ coincidence island is narrow and aligned along θ = – 45◦ or – 1 slope.
Being a pure two–body dissociation, CO+ and O+ ions are ejected in opposite directions

with equal momenta |Px | = |Py |. A long tail is observed originating from the CO+ :O+
coincidence island, extending up to the limit t1 =t2 =3247 ns, along the forward diagonal
of the coincidence map. In the vicinity of the forward diagonal, a V–shaped scatter is
also observed at the end of the tail. The main island of CO+ :O+ is actually due to the
prompt dissociation of CO2+ + + 2+
2 into CO +O whereas the metastable CO2 ions, which

survived at least for 3.2 µs in the TOFMS, are responsible for the long tail [21].

Formation of C+ :Oq+ (where q = 1 to 3)

Other than CO+ :O+ coincidence island, we observed correlation between various dis-
sociation channels corresponding to Cq+ (q = 1 to 3) and Oq+ (q = 1 to 4) ions. It is seen
from Figure 3.2 that the C+ :O+ coincidence has a star-like structure. We separated the
Dissociation dynamics of carbon di oxide (CO2 ) 86

C+ O+

TOF of second fragment (0.5 ns / channel)

6000 6200 10

TOF of third fragment (0.5 ns / channel)

45
9
40
5800 6000 8
35
7

5600 30 5800 6
O+
25 5

5400 20 5600 4
O+
15 3

5200 5400 2
10
1
5
5000 5200 0
4400 4600 4800 5000 5000 5200 5400 5600 5800 6000
TOF of first fragment (0.5 ns / channel) TOF of second fragment (0.5 ns / channel)
(a) (b)

Figure 3.3: (a) Star–like coincidence island between C+ :O+ correlated ion–pair and (b)
the corresponding triple ion coincidence map arising from 1.2 MeV Ar8+ impact on
neutral CO2 molecules.

specific area of the C+ :O+ coincidence island from Figure 3.2 by drawing a polygon

around it (shown in Figure 3.3(a)) and then noted down all available information corre-
sponding to C+ :O+ correlated ion pair by operating only on the area surrounded by the
polygon. We also extracted the third fragment of the C+ + O+ dissociation channel and
plotted another histogram between the TOF of the second fragment versus that of the

third (Fig. 3.3(b)). From the above triple ion coincidence map we got the evidence for
dissociation of triply charged CO2 molecular ions in the following manner:

CO23+ → C + + O + + O + (3.2)

We observed intensity difference in Figures 3.3(a) and 3.3(b) which indicates that every
C+ :O+ coincident event does not contain an O+ ion as third hit. Few O fragments

always remain undetected. This can be the consequence of charge neutrality of the
O fragment or finite efficiency of our detector system which is unable to detect high
energy fragment ions ejected perpendicular to the applied extraction field. Therefore
the C+ :O+ coincidence island may also originate from the dissociation channel of CO2+
2
Dissociation dynamics of carbon di oxide (CO2 ) 87

mentioned below:
CO22+ → C + + O + + O (3.3)

Three–body fragmentation of a triatomic molecule into three atomic or ionic fragments

may take place either by simultaneous breaking of the bonds (the two C–O bonds in
case of CO2 ), called a concerted break up (Equations 3.2 and 3.3), or by a sequential

break up of the two bonds. In a sequential break up the time interval between the two
successive bond breaks is of the order of rotational time period. In case of a sequential
fragmentation, the measured slope is influenced by the momenta of the intermediate
dissociation products. From Figure 3.3(a) we measured the slope of the C+ :O+ coinci-
dence island and found that the main island, aligned along C+ line, has slope -3.7 and

the two less intense blobs are having slope of zero and -1. Though the parallelogram
shape of the island suggests an initial charge separation or S(i) type of dissociation of
CO2+
2 the slope does not match with the predicted value given by Eland [20]. There-

fore we may assume that it is due to the concerted break up of CO2+ 3+

2 and CO2 into

C+ :O+ :O and C+ :O+ :O+ respectively where the two O fragments move in opposite
directions with almost equal momenta leaving C nearly at rest. A dissociation like

CO22+ → CO 2+ + O (3.4)
ց
C + + O+ + O (3.5)

results in the C+ :O+ coincidence island of slope -1 (intensity is very low). In the above
reaction slow ejection of neutral O from CO2+
2 is followed by dissociation of remaining

CO2+ ion into C+ + O+ with almost equal and opposite momenta. This is known as
deferred charge separation or S(d) type of reaction. The other less intense island of
C+ :O+ having zero slope suggests ejection of O+ ion with almost zero kinetic energy
where most of the kinetic energy release is shared between the C+ and the other O

fragment.
Dissociation dynamics of carbon di oxide (CO2 ) 88

O2+ O3+

TOF of second fragment (0.5 ns / channel)

TOF of second fragment (0.5 ns / channel)

5200 5100

5000
5000
4900

4800 C+ 4800
C+
4700

4600
4600

4500
4400

4400
3400 3600 3800 4000 4200 2800 2900 3000 3100 3200 3300 3400 3500
TOF of first fragment (0.5 ns / channel) TOF of first fragment (0.5 ns / channel)
(a) (b)

Figure 3.4: Coincidence islands corresponding to (a) O2+ :C+ and (b) O3+ :C+ .

Figures 3.4(a) and 3.4(b) show coincidence islands of O2+ and O3+ with C+ . Both
the islands are dumb–bell shaped having slopes -0.2 and -0.3 respectively. From the
triple ion coincidence map, we found that most of the third fragments are neutral O for

C+ :Oq+ (q = 1, 2) correlated ion pairs, thus contribution from C+ +Oq+ +O+ channel is
very small. If these reactions were to proceed via initial charge separation, where Oq+
is separated from the parent molecular ion in the first step followed by dissociation of
CO+ into C+ +O, one predicts slopes of -0.8 and -1.3 for C+ :O2+ and C+ :O3+ coin-
cidence islands respectively. The more negative peak slopes measured in the present

work suggests that the correlated momentum carried away by the central carbon ion is
much lower than is predicted for S(i) fragmentation. Therefore the possible dissociation
mechanism behind these two correlated ion pairs is concerted break up through

CO23+ → C + + O 2+ + O (3.6)

CO24+ → C + + O 3+ + O (3.7)

For breakup of a molecular bond having different atoms, the energy release will
be asymmetrically distributed between the two, the lighter mass taking away the larger
Dissociation dynamics of carbon di oxide (CO2 ) 89

share. Also a multiply charged molecular ion may dissociate in fragments with sym-
metric or asymmetric charge distribution where the energy release of one process can

be quite different from that of the other. By projecting a specific coincidence island of
the 2D coincidence map on either the first-hit or the second-hit axis, we get the time
difference plot between the forward and backward scattered fragments. The projection
made on the first-hit or the second-hit TOF axis helps in getting the information related
to fragments with different masses or charge states. We can even calculate the kinetic

energies for these fragments individually for a selected dissociation channel.

+ +
C :O
2+ +
6000 +
C :O
2+ C :O
500 2+ 2+
+ 3+
C :O C :O

5000
400
counts (arbitrary unit)

counts (arbitrary unit)

4000

300
3000

200
2000

1000 100

0 0
0 50 100 150 200 0 50 100 150 200
KE (eV) KE (eV)

(a) (b)

Figure 3.5: Kinetic energy distributions of Oq+ ions (where q = 1 to 3) detected in

coincidence with (a) C+ and (b) C2+ .

The kinetic energy distributions of the Oq+ ions (q = 1 to 3) detected in coincidence

with C+ have been shown in Figure 3.5(a). For dissociation pathways producing asym-
metric charge distribution, we get relatively higher kinetic energy values for O ions with
higher charge state (O3+ ). The most probable kinetic energies of C+ ions in the above
dissociation channels, corresponding to C+ :Oq+ (q = 1 to 3) ion pairs, are close to zero

as also evident from the slope of the coincidence islands. The maximum kinetic energy
carried away by the C+ ions is 10 eV for the above dissociation channels.
Dissociation dynamics of carbon di oxide (CO2 ) 90

Formation of C2+ :Oq+ (where q = 1, 2)

In the coincidence spectrum (Fig. 3.2) we have also observed two C2+ :Oq+ coincidence

islands where q = 1, 2. Both the islands are lying along the C2+ line which indicates
that the two O fragments are ejected with almost equal and opposite momenta leaving
the C2+ ion at rest. The kinetic energy distributions of O+ and O2+ ions detected in
coincidence with C2+ have been shown in Figure 3.5(b). Like C+ :Oq+ dissociation
channels here also the kinetic energy of oxygen ion is lower in case of C2+ + O+ channel

compared to C2+ + O2+ . These coincidence islands cannot be formed due to sequential
breakup of CO3+ 4+
2 and CO2 because in those cases the predicted slopes would be -4.7

and -2.3 respectively. Therefore the possible dissociation mechanism is concerted and
the fragmentation channels are

CO23+ → C 2+ + O + + O (3.8)

CO24+ → C 2+ + O 2+ + O (3.9)

assuming the third fragment to be neutral.

Formation of Om+ :On+ (where m, n = 1, 2)

We have observed and separated three correlated ion pairs corresponding to Om+ :On+

where m, n = 1, 2 in the two dimensional coincidence map (Fig. 3.2). All the coinci-
dence islands are very narrow and dumb-bell shaped which suggests that no momentum
has been taken away by the third fragment i. e. C. We have checked that for all three of
them no carbon ion has been detected as third hit. The slopes of the two coincidence is-

lands corresponding to symmetric charge distribution are -1 where as for O2+ :O+ chan-
nel it is -2. These measurements are consistent with concerted dissociation of COq+
2

through

CO22+ → C + O + + O + (3.10)
Dissociation dynamics of carbon di oxide (CO2 ) 91

+ +
O :O
2+ +
O :O
2+ 2+
O :O
1500

counts (arbitrary unit)

1000

500

0
0 50 100 150 200
KE (eV)

Figure 3.6: Kinetic energy distribution of the first oxygen ion detected in coincidence
with the second for Om+ +C+On+ (where m, n = 1, 2) dissociation channels.

CO23+ → C + O 2+ + O + (3.11)

CO24+ → C + O 2+ + O 2+ (3.12)

In the above dissociation channels the total charge of the parent molecular ion is shared
between the two peripheral oxygen ions either symmetrically or asymmetrically and
they get separated instantaneously with equal and opposite momentum leaving the cen-
tral C atom at rest. The kinetic energy distribution of the first oxygen ion detected in

the above dissociation channels are shown in Figure 3.6. The kinetic energy distribution
of the other oxygen fragment is also very similar. It is clear from the above figure that
the kinetic energy of oxygen ion increases with increasing charge state of the parent
molecular ion indicating higher repulsion among the constituents.

The measured kinetic energies of the Oq+ ions are compared with that calculated

from pure Coulomb explosion model (CEM). The value of KER is predicted by CEM
according to the relation :
q1 × q2
E(eV ) = 14.4 (3.13)
R(Å)
Dissociation dynamics of carbon di oxide (CO2 ) 92

Table 3.1: Experimentally measured KE values of Oq+ ions for various fragmentation
channels of CO2 induced by 1.2 MeV Ar8+ ion impact. These values are compared with
that from pure Coulomb explosion model.

Dissociation Ion Fragment KE (eV) ECoul

products Eprob Emax (eV)

C+ + O+ O+ 7.0 50.0 12.4

C+ + O2+ O2+ 25.4 100.0 24.8

C+ + O3+ O3+ 50.0 150.0 37.2

C2+ + O+ O+ 18.4 75.0 24.8

C2+ + O2+ O2+ 51.5 125.0 49.6

O+ + O+ O+ 8.3 40.0 6.2

O+ + O2+ O2+ 29.2 100.0 12.4

O2+ + O2+ O3+ 47.0 100.0 24.8

where q1 and q2 are the charges on the two fragment ions and R (Å) is the internuclear
distance of the neutral parent molecule. We have calculated the theoretical values of
KER taking the equilibrium geometry of neutral CO2 : RC−O = 1.162 Å, RO−O = 2.324

Å. Table 3.1 lists the theoretical KER values calculated from CEM along with the most
probable (Eprob) and maximum (Emax ) kinetic energies (in eV) of Oq+ ions for various
dissociation channels of CO2 , producing C+ :Oq+ , C2+ :Oq+ and Om+ :On+ coincidence
islands due to 1.2 MeV Ar8+ ion impact. The oxygen ionic charge state varies from

1 to 3. For all the above channels, we observed a gradual increase in both the Eprob
and Emax values for the oxygen ions having higher charge states. On comparing the
most probable kinetic energy of O ion with the theoretical KER value, we observed that
except for C+ :O+ and C2+ :O+ channels the measured Eprob s are higher than that pre-
Dissociation dynamics of carbon di oxide (CO2 ) 93

dicted by pure CEM. This discrepancy may arise due to the oversimplified assumption
of localized point charges on the atoms. The effect of overlapping electron clouds must

be considered to account for the KER of such complex precursor molecular ionic states.

3.4 Triple ionization and fragmentation of CO2 by swift

heavy ion

The geometry of CO3+

2 and its dissociation mechanism leading to triple ion coincidence

have been investigated in the following section using complementary experimental and

theoretical techniques [19]. The momenta of all the charged fragments are measured
and the angle between any two fragment vectors are obtained. In the experiment a
beam of neutral CO2 molecules crossed a beam of 5 MeV u−1 28 Si12+ ions from the Pel-
letron accelerator of Inter University Accelerator Centre (IUAC). The projectile speed
translated to an interaction time of 17 attoseconds, which was much shorter not only

than the typical rotational and vibrational time periods of CO2 , but also than its typical
fragmentation time (around 10 fs). We thus had access to the molecular ion in its vi-
brationally frozen state devoid of any interference from the projectile electric field. The
molecular core can be assumed to be frozen during the interaction, and the kinetic en-

ergy distribution of the fragments can be expected to reflect the properties of the states
of the multiply ionized molecule. For additional information on the excited states of
CO3+
2 and the corresponding dissociation limits, ab initio quantum chemical calcula-

tions have been performed. We compared the results of dissociation dynamics of CO3+
2

into C+ + O+ + O+ (using 5 MeV u−1 Si12+ ) with those of Siegmann et al. (5.9 MeV
u−1 Xe18+ and Xe43+ ) [5], Neumann et al. (3.2 keV u−1 Ar8+ ) [6] and Adoui et. al. (8
MeV u−1 Ni24+ ) [22], which involve collisions with widely differing interaction times
and perturbation strengths (see Table 3.2).
Dissociation dynamics of carbon di oxide (CO2 ) 94

Table 3.2: The interaction time and perturbation strength for various energy projectile
ions used in different experiments: 3.2 keV u−1 Ar8+ [6], 5 MeV u−1 Si12+ [19], 5.9
MeV u−1 Xe18+ and Xe43+ [5] and 8 MeV u−1 Ni24+ [22].

Projectile Energy Interaction Perturbation

ion (MeV u−1 ) time (as) strength

Ar8+ 3.2 × 10−3 680 22.30

Si12+ 5.0 17 0.85

Xe18+ 5.9 16 1.17

Xe43+ 5.9 16 2.80

Ni24+ 8.0 13 1.34

3.4.1 Triple fragmentation kinematics

CO3+
2 may fragment via various channels, with several fragment combinations being

possible. From the analyzed data we discern the following four fragment triplets (the
neutral being undetected, but deduced by analysis) (i) C2+ :O+ :O, (ii) C:O2+ :O+ , (iii)
C+ :O2+ :O and (iv) C+ :O+ :O+ .

From the triple ion coincidence map (Fig. 3.7) we found O+ to be the third ion,
implying a C+ :O+ :O+ break-up. While the C+ :O+ coincidence may arise from the
break up of molecular ions other than CO3+
2 , we can separate out the contribution of

CO3+ + +
2 in the C :O coincidence by applying suitable conditions on the event list. This

channel is the dominant triple fragmentation channel of CO3+

2 in the present case, and

is the focus of our attention. The detail about the other dissociative channels will be
discussed in Chapter 5 where we will compare dissociation dynamics of CO3+
2 with

OCS3+ .

Figure 3.8(a) shows momentum correlation between the three fragments of CO3+
2 ,
Dissociation dynamics of carbon di oxide (CO2 ) 95

C+ 3200
O+

3000

TOF of second fragment (ns)

TOF of third fragment (ns)

O+ 3100

2900

3000

2800
O+

2900

2700

2800
2400 2500 2600 2700 2700 2800 2900 3000
TOF of first fragment (ns) TOF of second fragment (ns)
(a) (b)

Figure 3.7: Coincidence map of (a) first and second fragments and (b) second and
third fragments corresponding to CO3+ + + +
2 → C :O :O channel. The solid lines drawn
correspond to the peak of the respective ions in the TOF spectrum.

where the arrow represents direction of the momentum of the first fragment (C+ ) and
is taken as the reference direction. The components of momentum of the second and
third fragments (O+ in both the cases) parallel and perpendicular to it are obtained,

and the events are histogrammed as 2-D momentum plot. The distributions of both
O+ ions are single lobed and nearly identical. The similarity is further seen in the
individual kinetic energy spectra of the two O+ ions (Fig. 3.8(b)). The O+ kinetic
energy distributions are broad and peak around 15 eV. In contrast, the corresponding

C+ kinetic energy distribution is narrow and peaks at slightly over 1 eV. Our result is in
very good agreement with that obtained by Siegmann et. al. where 5.9 MeV u−1 Xe43+
were used as projectiles [5].

If the two oxygen ions separate out symmetrically, the central carbon ion will

have a zero momentum in case the separation takes place from a linear geometry. The
observed non-zero kinetic energy of C+ is a consequence of the O–C–O bond angle in
the fragmentation complex being < 180◦. The most probable O–C–O bond angle in
an ensemble of CO2 molecules, taking into account zero-point vibrational excitations
Dissociation dynamics of carbon di oxide (CO2 ) 96

300 7 40

st +
1 O 6 35
200
C+

counts (arbitrary unit)

30
Momentum (a. u.)
5
100
25
4
C+
0 20
3 1stO+
15
-100
2
10
-200
2ndO+
nd + 1
2 O 5

-300 0 0
-300 -200 -100 0 100 200 300 0 5 10 15 20 25 30 35 40 45 50
Momentum (a. u.) Fragment KE (eV)
(a) (b)

Figure 3.8: (a) Correlated momentum map of C+ and two O+ ions originated from
dissociation of CO3+
2 . (b) Kinetic energy distributions of individual ions for the same
channel.

is 172.5◦ [5]. The asymptotic angle between the momentum vectors of the two O+ ions
originating from such a state will, in case of concerted dissociation, be smaller than the
initial O–C–O bond angle, due to the Coulomb repulsion between the three fragments,

resulting in a distribution with the modal value being < 172.5◦. This is clearly seen in
the triple-ionization results of Kushawaha et. al. [2], Siegmann et. al. [5] and Adoui
et. al. [22]. There may be other reasons for broadening, such as dissociation from a
bent electronic state. In the present case the distribution of the angle between the two
O+ momentum vectors peaks at about 165◦ (Fig. 3.9(a)) with a long tail extending to

angles as small as 60◦ , and a minor peak around 135◦ . The magnitude of the momenta
carried away by the two O+ ions are nearly identical. The combined implications of the
observations regarding the relative emission angles and the kinetic energy distributions
is that we have here a concerted fragmentation, from linear as well as bent states of

CO3+
2 .

Results from Siegmann et. al. (5.9 MeV u−1 Xe18+ and Xe43+ impact) show that
in case of O+ +C+ +O+ fragmentation, the distribution of the O+ :O+ angle peaks at
Dissociation dynamics of carbon di oxide (CO2 ) 97

120◦ [5] for both charge states, and those distributions are broader than in the present
case. On the other hand, the results of Adoui et. al. show a peak around 165◦ , tailing

off at about 120◦ [22], and the distribution is narrower than the present case. This
comparison shows, that both, projectile charge state and projectile speed, influence the
distribution of the O+ :O+ angle. The distribution of this angle in all cases deviates
from that expected in a pure Coulomb explosion model, which means that excited state
potential energy surfaces play a significant role in determining the angular distributions.

The observed influence of the projectile charge and speed on the angular distributions
is consistent with the fact, that different projectile charge states and speeds will lead to
different distributions of excited states.

50 50

40 40
counts (arbitrary unit)

counts (arbitrary unit)

30 30

20 20

10 10

0 0
60 80 100 120 140 160 180 0 10 20 30 40 50 60 70 80 90 100
+ +
O : O Angle (degrees) KER (eV)
(a) (b)

Figure 3.9: (a) Distribution of the angle between the momentum vectors of two O+ ions
and (b) KER of CO3+ + +
2 dissociating into C +O +O .
+

The KER distribution (Fig. 3.9(b)) of CO3+ + + +

2 → C + O + O channel shows

signature of three peaks at 20.7 eV, 26.8 eV and 30.8 eV with some minor higher energy

peaks in the region 15–55 eV. These peaks can have their origin in the shallow electronic
states of the parent molecular ion. We have analyzed the KER distribution as a function
of O+ :O+ angle in the two ranges 60◦ –140◦ and 140◦ –180◦ and found that the KER
distributions in the two cases are not significantly different. The KER distributions in
Dissociation dynamics of carbon di oxide (CO2 ) 98

the other swift ion collision cases (Siegmann et. al. [5], Adoui et. al. [22]) and the
present case are quite similar. Within the range of projectile charge states for swift

ions compared here, there is no significant dependence of the KER distributions on the
projectile charge state. The case of slow collisions (Neumann et. al. [6]) is different,
with the KER spectrum showing a broad peak around 20.6 eV with a high energy tail
extended up to 40 eV. The data of Neumann et. al., shows a mixture of processes. In
that case the interaction time was much longer than present, allowing for vibrational

and rotational relaxation during the collision. In particular, when analyzed as a function
of the KER, their data show the following. For KER = 17 ± 0.5eV, the break up is
predominantly sequential type where ejection of one O+ from CO3+
2 leaves behind a

rotating CO2+ ion which fragments after a time delay. The momentum distribution in

the Newton diagram [6] is a proof of this sequential mechanism. On the other hand,
concerted break-up is the dominant mechanism for KER = 31 ± 0.5eV and KER =
38±0.5eV. Within the range of perturbations compared here, the KER distribution is the
same for all swift collisions (14a.u. < v < 18a.u.), irrespective of the projectile charge

(12 < q < 43). In slow collisions, excited states that lead to sequential dissociation are
accessed, leading to a different KER distribution as compared to swift collisions.

3.4.2 Ab initio quantum chemical calculation

A simple Coulomb explosion model of the triple ion break-up, assuming an initial sep-
aration of 120 pm between the C and O nuclei in the CO3+
2 molecular ion in a linear

state yields a KER of 30 eV [3]. Under the same approximation, but from a bent state,
the KER would be higher. The range of KER values arises from various possibilities:
bent initial states, crossing of CO2 and CO3+
2 potential energy surfaces etc. These are

investigated by ab initio computations of the potential energy curves for the low-lying

states of the CO3+

2 molecular ion.

The term ab-initio means calculations from first principles, in this case it means
Dissociation dynamics of carbon di oxide (CO2 ) 99

the solution of the time-independent Schrödinger equation

Htotalψtotal = Etotalψtotal (3.14)

where ψtotal is a total wave function for all particles in the molecule, and Etotal is

the total energy of the system. For a molecule with N nuclei and n electrons the
Hamiltonian is

N N n
X 1 X ZkZk′ X 1 2
Ĥtotal = − ∇2k + − ∇i
k=1 2Mk
~ ~
k>k′ |Rk − Rk′ | i=1 2
N X
X n Zk n
X 1
− + (3.15)
k=1 i=1 |Rk
~ − r~i| i>j |~
ri − r~j |

~k are the mass, charge and the position vector,

in atomic units, where Mk, Zk and R
respectively, of the kth nucleus; r~i is the position of ith electron.

As the nuclei are thousands of times more massive than the electrons, they move
much slower compared to the electrons. Therefore we may assume that the electrons
adjust themselves to new nuclear positions much faster than the nuclei can adjust to
new electronic positions. According to Born-Oppenheimer approximation the elec-

tronic motion can be separated from the nuclear motion and the electronic structure
calculations can be performed for fixed nuclear configurations by solving the electronic
Schrödinger equation:
Ĥelψel = Eelψel (3.16)

where Ĥel is the electronic Hamiltonian operator given by

n
X 1 XN Zk n
X 1
Ĥel = − ∇2i − + (3.17)
i=1 2 ~ ~i| i>j |~
k=1 |Rk − r ri − r~j |

Eel is the electronic energy and ψel is the electronic wave function in the field of fixed
nuclei. The potential energy of the system for a given, fixed, nuclear configuration is
Dissociation dynamics of carbon di oxide (CO2 ) 100

then equal to the electronic energy plus the internuclear repulsion energy:

N
ZkZk′
~1 , R
~2 , . . . , R~N ) = Eel +
X
Epot(R (3.18)
k>k′
~k − R~k′ |
|R

Since the second term is a constant, the problem reduces to solving for Eel and ψel,

thereafter denoted by ψ and E respectively.

For an n-electron system we associate the n electrons with an orthonormal set of

one-electron functions ψ̃1 , ψ̃2 , . . . , ψ̃n, called spin orbitals. Each ψ̃i is assumed to
have the form
ψ̃i(~
r, σ) = ψi(~
r )χi(σ) (3.19)

in which the orbital ψi(~

r) describes the motion of electrons in space and the spin
function χi(σ) describes its spin state. Being fermions, the total n electron wave
function is the antisymmetrized product of one-electron wave functions, given in the
form of a determinant (the Slater determinant) :
 
 

 ψ̃1 (1) ψ̃2 (1) . . . ψ̃n(1) 

 
 

−1/2 
ψ̃1 (2) ψ̃2 (2) . . . ψ̃n(2) 

Φ= (n!) .. .. ..
 (3.20)
 .. 

 . . . . 

 
 

 ψ̃1 (n) ψ̃2 (n) . . . ψ̃n(n) 


If we substitute the above wave function in the expression for energy we get

n n n
Φ∗ĤΦdτ 1X 1X
R
X
E= R ∗ = hi + Ji,j − Ki,j (3.21)
Φ Φdτ i=1 2 i,j 2 i,j

where the orbital energies hi, the Coulomb integrals Ji,j and the exchange integrals
Ki,j are defined by
Z
hi = ψi∗(~
r)ĥψi(~
r ) d~
r (3.22)
Dissociation dynamics of carbon di oxide (CO2 ) 101

Z ψi∗(~
ri)ψi(~
ri)ψj∗(r~j )ψj (r~j )
Ji,j = d~
ridr~j (3.23)
|~
ri − r~j |

Z ψi∗(~
ri)ψj (~
ri)ψj∗(r~j )ψi(r~j )
Ki,j = δ(msi, msj ) d~
ridr~j (3.24)
|~
ri − r~j |

The most popular method in solving this is to use the variational principle and vary
the one-electron wave functions to obtain an approximate solution to the Schrödinger
equation. For a molecular system of any size, direct solution of the above equations
is impractical, and the molecular orbitals are approximated as a linear combination of

atomic orbitals (LCAO approach). In this approach, each molecular orbital is of the
form
M
X
ψi(~
r) = Cµiφµ(~
r) (3.25)
µ=1

where φµ are atomic orbitals.

In the variational principle first an initial guess wave function is built up using
the LCAO techniques. In the Restricted Hartree Fock (RHF) method, it is assumed
that the orbital energies and the coefficients Cµi are independent of the spin state. In
the unrestricted method (UHF) the spin states are explicitly taken into account and

the coefficients then depend upon the spin state of the electron and are denoted by
α
Cµi where α denotes the spin state. The electrons are assigned to the orbitals ψi of
lowest energies. The RHF method takes the electrons in pairs for each spatial orbital,
whereas the UHF wave function has an implicit spin dependence. Finally the energy
α
of the system is calculated iteratively and the coefficients Cµi or Cµi are varied till the

minimum energy is obtained.

Regarding the dissociation energy and the nature of the dissociation products, the
RHF model can yield an incorrect description of the system. RHF dissociation energies
are generally too small and, in some cases, negative values are obtained, indicating
Dissociation dynamics of carbon di oxide (CO2 ) 102

that the RHF molecule is unstable towards dissociation [23]. Unrestricted Hartree-Fock
(UHF) theory is the most common molecular orbital method for open shell molecules

where the number of electrons of each spin are not equal.

The Hartree-Fock technique effectively treats each electron motion as being gov-
erned by the average potential created by the N nuclei and the remaining (n − 1)
electrons. As such, the HF technique neglects the instantaneous repulsions between
pair of electrons. The contribution of such instantaneous repulsions to the total energy

is called correlation energy. The method dealing with correlation energy is known as
Configuration Interaction (CI).

These calculations give us the potential energy at a fixed nuclear configuration (in
the case of diatomic, at a fixed internuclear distance R). The calculations, repeated for

different nuclear configurations for triatomic molecule like CO2 yield a potential energy
surface which characterizes the molecule.

Computed KER values

Ab inito quantum chemical calculations have been performed using the GAMESS suite

of programs [24] to determine the potential energy of triply charged CO2 . We used a
6-311G basis set including diffused sp(L) shell. The 6-311G type basis set consists
of 6 Gaussian type orbitals (GTO) in the K shell of each atom, L shell had 3, the M
and N shells had 1 GTO each. The single point potential energies of CO3+
2 in doublet

and quadrate states were calculated using self-consistent field unrestricted Hartree-Fock
(UHF) method along with the formation energies of C+ and O+ ions. Considering
Frank-Condon (vertical) transition we then calculated the most probable KER for the
dissociative decay of each of these molecular states into different states of C+ and two

O+ ions.

Table 3.3 lists possible molecular states of CO3+

2 dissociating into various states of

C+ , O+ and O+ along with the KER values obtained by UHF calculation. In these cal-
culations CO3+
2 is taken to have a linear geometry. The calculated KER values span the
Dissociation dynamics of carbon di oxide (CO2 ) 103

Table 3.3: The possible molecular states of linear CO3+ + +

2 dissociating into C + O + O
+

along with the theoretically calculated values of KER.

Molecular Release energy Dissociation

states (eV) limit

2
Π 20.6 C+ (4 P ) + O+ (2 D) + O+ (2 D)

24.4 C+ (2 P ) + O+ (2 D) + O+ (2 D)

25.8 C+ (4 P ) + O+ (4 S) + O+ (2 D)

29.3 C+ (2 P ) + O+ (4 S) + O+ (2 D)

34.5 C+ (2 P ) + O+ (4 S) + O+ (4 S)

4
Σ 15.9 C+ (4 P ) + O+ (2 D) + O+ (2 D)

19.7 C+ (2 P ) + O+ (2 D) + O+ (2 D)

24.6 C+ (2 P ) + O+ (2 D) + O+ (4 S)

26.3 C+ (4 P ) + O+ (4 S) + O+ (4 S)

range of energies observed in the experiment, and the dominant fragmentation channels
may be tagged to the observed KER from Table 3.3. With our preliminary calcula-
tions we found the existence of shallow (possibly metastable) electronic states of CO3+
2

molecular ion whose dissociation can lead to sharp peaks in the KER spectra. The
appearance of peaks in the KER distribution, whose origin lies in these shallow elec-
tronic states, depends on the population of other excited states of the parent molecular
ion. The high energy components (36.7 eV, 42.7 eV, 50.0 eV, etc.) cannot be explained

from the low lying 4 Σ and 2 Π states of CO3+

2 . These may arise from the higher ex-

cited states of the precursor molecular ion. A close inspection of the KER distribution
for the C+ :O+ :O+ channel seen in 3.2 keV u−1 Ar8+ impact on CO2 shows a broad
peak around 20.6 eV and other less prominent peaks at 19 eV, 22 eV and 24 eV. These
Dissociation dynamics of carbon di oxide (CO2 ) 104

peaks lie well within the range of our computed values considering linear structure of
the CO3+
2 ion.

3.5 Summary

In this chapter we have analyzed various coincidence islands of multiply charged CO2
molecular ions, formed in the collision of neutral CO2 with 1.2 MeV Ar8+ ion beam.
From the coincidence map we observed complete two–body dissociation of CO2+
2 into

CO+ +O+ . Analyzing the slope and shape of the coincidence island we deduced whether
the fragmentation process is concerted or sequential in nature. For the present energy
range we conclude that the three–body fragmentation of COq+
2 (where q = 2–4) usually

proceeds in a concerted manner though sequential dissociation of CO2+

2 is also observed

in case of C+ :O+ coincidence island. For all channels of COq+

2 (where q = 2, 3), except

C+ +O+ +O and C2+ +O+ +O, the experimental KER values are found to be higher than
the same predicted by the CE model. We observed that the total kinetic energy, released
in the dissociation of CO2 molecular ion, is mostly carried away by the terminal atoms.
Moreover, the kinetic energy, with which the O ion is emitted, increases with the in-

creasing charge state of the parent. The triple ion fragmentation channel (C+ +O+ +O+ )
of CO3+
2 , formed by the high energy highly charged ion (5 MeV u
−1
Si12+ ) impact on
CO2 , has been investigated in detail in the present scope. From the kinetic energy re-
lease distribution we found that the geometry of CO3+
2 is linear in this case which is

well supported by the ab initio quantum chemical calculations. However participation

of bent electronic state is also visible in the angular distribution between the momen-
tum vectors of the two O+ ions. Comparing the break–up pattern of CO3+
2 induced by

various perturbations we found that swift perturbations almost exclusively lead to con-
certed triple-ion break-up, whereas significant probability for stepwise break-up is seen

with slow perturbations.

Bibliography

[1] M. Tarisien, L. Adoui, F. Frémont, D. Leliévre, L. Guillaume, J-Y. Ches-

nel, H. Zhang, A. Dubois, D. Mathur, S. Kumar, M. Krishnamurthy and A.

Cassimi, J. Phys. B: AT. Mol. Opt. Phys. 33, L11 (2000)

[2] R. K. Kushawaha, S. Sunil Kumar, I. A. Prajapati, K. P. Subramanian and B.

Bapat, J. Phys. B: At. Mol. Opt. Phys. 42, 105201 (2009)

[3] R. K. Singh, G. S. Lodha, V. Sharma, I. A. Prajapati, K. P. Subramanian and

B. Bapat, Phys. Rev. A 74, 022708 (2006)

[4] C. Tian and C. R. Vidal, Phys. Rev. A 58, 3783 (1998)

[5] B. Siegmann, U. Werner, H. O. Lutz and R. Mann, J. Phys. B: At. Mol. Opt.

Phys. 35, 3755 (2002)

[6] N. Neumann, D. Hant, L. Ph. H. Schmidt, J. Titze, T. Jahnke, A. Czasch,

M. S. Schoffler, K. Kreidi, O. Jagutzki, H. Schmidt-Bocking and R. Dorner
Phys. Rev. Lett. 104, 103201 (2010)

[7] C. Cornaggia, M. Schmidt and D. Normand, J. Phys. B: At. Mol. Opt. Phys.

27, L123 (1994)

[8] A. Hishikawa, A. Iwamae and K. Yamanouchi, Phys. Rev. Lett. 83, 1127
(1999)

105
Dissociation dynamics of carbon di oxide (CO2 ) 106

[9] G. Ravindra Kumar, P. Gross, C. P. Safvan, F. A. Rajgara and D. Mathur,

Phys. Rev. A 53, 3098 (1996)

[10] J. H. Sanderson, T. Nishide, H. Shiromaru, Y. Achiba and N. Kobayashi,

Phys. Rev. A 59, 4817 (1999)

[11] D. Mathur, E. Krishnakumar, F. A. Rajgara, U. T. Raheja and V. Krishna-

murthi, J. Phys. B: At. Mol. Opt. Phys. 25, 2997 (1992)

[12] B. Bapat and V. Sharma, J. Phys. B: At. Mol. Opt. Phys. 40, 13 (2007)

[13] V. Kumar and E. Krishnakumar, J. Chem. Phys. 78, 46 (1983)

[14] S. S. Xantheas, S. T. Elbert and K. Ruedenberg, Chem. Phys. Lett. 166, 39

(1990)

[15] N. W. Winter, C. F. Bender and W. A. Goddard III, Chem. Phys. Lett. 20, 489
(1973)

[16] K. E. McCulloh, J. Chem. Phys. 59, 4250 (1973)

[17] M. Hochlaf, F. R. Bennett, G. Chambaud and P Rosmus, J. Phys. B: At. Mol.

Opt. Phys. 31, 2163 (1998)

[18] H. Hogreve, J. Phys. B: At. Mol. Opt. Phys. 28, L263 (1995)

[19] M. R. Jana, P. N. Ghosh, B. Bapat, R. K. Kushawaha, K. Saha, I. A. Prajapati

and C. P. Safvan, Phys. Rev. A 84, 062715 (2011)

[20] J. H. D. Eland, Mol. Phys. 61, 725 (1987)

[21] V. Sharma, B. Bapat, J. Mondal, M. Hochlaf, K. Giri and N. Sathyamurthy,

J. Chem. Phys. A 111, 10205 (2007)
Dissociation dynamics of carbon di oxide (CO2 ) 107

[22] L. Adoui, T. Muranaka, M. Tarisien, S. Legendre, G. Laurent, A. Cassimi, J-

Y. Chesnel, X. Flechard, F. Frémont, B. Gervais, E. Giglio and D. Hennecart,

Nucl. Instrum. Methods Phys. Res., Sect. B 245, 94 (2006)

[23] V. R. Marathe and D. Mathur, Physics of Ion Impact Phenomena, D Mathur

(Ed.), Berlin : Springer-Verlag, Chapter 8 (1991)

[24] M. W. Schmidt et al., J. Comput. Chem. 14, 1347 (1993)

Chapter 4

Low energy ion induced fragmentation

of carbonyl sulphide (OCS)

4.1 Introduction

Carbonyl sulphide (OCS) is the most abundant sulphur species in the atmosphere with
immense biological, chemical and astrophysical significance. It is present in the Earth’s

atmosphere, volcanic gases, ice cores in Antarctica and the interstellar medium. It cat-
alyzes the formation of biological molecules and played an important role in the origin
of life on our planet [1]. Electron impact [2, 3, 4, 5], photon impact [6, 7, 8, 9] and
laser induced [10, 11] dissociation of OCS have been widely investigated by various

groups, but hardly any report has been found on the dissociation dynamics of OCS by
ion impact. In the present chapter, ion impact dissociation of OCS is investigated be-
cause of its dissimilarity to the previously discussed triatomic molecule (CO2 ). Though
both, CO2 and OCS, are linear triatomic molecules having polar bonds, the former does

not possess any dipole moment but the later has permanent dipole moment and is more
prone to various geometry modifications under strong fields. In the following sections
dissociation dynamics of multiply charged OCS molecular ions have been discussed
using the time of flight mass spectroscopy and momentum imaging techniques [12].

108
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 109

4.2 Experimental details

The experiment has been carried out at the Low Energy Ion Beam Facility (LEIBF) of

Inter-University Accelerator Centre (IUAC), New Delhi, India. In the experiment, 150
keV, mass and charge selected Ar+ projectiles, produced from an ECR ion source, were
transported to the collision chamber where they interact with the neutral OCS molecules
effusing from a needle at right angle to the ion beam. Typical operating pressures were

in the range of 7 × 10−7 Torr whereas the experimental chamber was kept at a base pres-
sure of 9 × 10−8 Torr. The dissociated fragments were extracted from the interaction
region into the linear TOFMS by applying a uniform electric field (16.36 V/mm) per-
pendicular to both the ion beam and molecular gas jet. The ion and electron extraction

electrodes were separated by 44 mm. After extraction the ions entered the acceleration
region of 80 mm followed by a field free drift region of 200 mm. At the exit of TOFMS,
the dissociation products were detected by a position sensitive detector that consists of
twin micro-channel plates (MCP) of 50 mm diameter in chevron configuration with
a delay-line anode (DLD40) behind the MCP. Ejected electrons were extracted in the

opposite direction of the TOFMS and were detected by a channel electron multiplier de-
tector which acted as the trigger for starting the coincidence data acquisition. The stop
signals were recorded from the MCP. The TOF spectrum was acquired in the multi-
hit mode by a time-to-digital converter (TDC) interfaced to a computer where multiple

fragment ions were recorded in coincidence to obtain information on correlated disso-

ciation products. The recorded data were analyzed off-line. The observed flight time t
and the position (x, y) information for each ion in an event were transformed to yield the
three momentum components. Further information about the fragmentation processes

such as the angle of ejection, kinetic energy of fragments, etc. were obtained from the
fragment ion momentum data.
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 110

4.3 TOF spectrum of dissociated OCS

C+
3500

3000
counts (arbitrary unit)

2500

2000 S+

1500 O+/S2+
CO+ OCS+
1000
OCS2+
2+
C
500 O2+/S4+ CS+ OC34S+
0
1500 2000 2500 3000 3500 4000 4500 5000 5500
TOF of first fragment (ns)

Figure 4.1: Time-of-flight spectrum of dissociative OCS under the impact of 150 keV
Ar+ ion.

The dissociation products observed in the TOF spectrum (Fig. 4.1) of OCS shows

undissociated molecular ions like OCS+ and OCS2+ along with CO+ , CS+ , singly and
doubly charged S, O and C ions. Presence of OC34 S+ with intensity corresponding to
the natural abundance (4.21%) of 34 S is also visible in the TOF spectrum of dissociating
OCS.

The fragment ions generated in the Coulomb explosion of highly charged parent
molecular ions generally possess large kinetic energies. In the TOF spectrum, the broad

peaks of CO+ and CS+ exhibit large kinetic energies associated with them. Since the
mass to charge ratio of O+ and S2+ ions are same, they are inseparable in the TOF spec-
trum. The peak splitting of O+ (or S2+ ) and O2+ (or S4+ ) demonstrates that these ions
are expelled with relatively large kinetic energy in the forward and backward directions
to the detector.

The sharp peak on the right side of the O+ /S2+ peak is attributable to water molec-
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 111

+
C2+ O2+ C +
O /S2+
+
CO 45
+
CS
4500 40

TOF of second fragment (ns)

34 + 35
S
4000
32 + 30
S
+ 25
3500 CO
20

3000 15
+
O /S2+
10
2500 +
C 5

1500 2000 2500 3000 3500

TOF of first fragment (ns)

Figure 4.2: Coincidence map of the dissociative products of OCS under the impact of
150 keV Ar+ ion. Islands corresponding to different breakup channels are shown.

ular ion (H2 O+ ), coming from the background.

4.4 Coincidence map showing various fragmentation path-

ways of OCS

To closely investigate the branching ratios of different fragmentation channels and to

identify the two–body and three–body dissociation pathways of multiply charged OCS

molecular ions, we generated the two–dimensional coincidence map from the off–line
analysis of the acquired multihit data.

Figure 4.2 shows the coincidence map of all the correlated fragment ions dissociat-
ing from multiply charged OCS molecular ions formed in the collision of neutral OCS

with 150 keV Ar+ ions. From the coincidence map, we separated correlations between
various dissociation channels corresponding to the formation of C+ , C2+ , O+ , O2+ and
S+ , S2+ ions. These correlations, extracted from the coincidence map, give evidence for
formation and dissociation of OCSq+ , where q=1-4. A great deal of information about
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 112

Figure 4.3: Relative intensities of various dissociation channels arising from different
precursor molecular ions of OCS under the impact of 150 keV Ar+ projectiles. The
counts have been normalized to the count of OCS+ .

various dissociation pathways can be extracted from the analysis of the above correla-
tion maps. Following the prescription of Eland [13] (the details have been explained in
section 2.7.2), we exploited the slope and shape of the coincidence islands to resolve the
mechanism of fragmentation of OCS molecular ions formed in the present experiment.

The relative intensities of various two-body and three-body dissociation channels

of OCS are shown in Figure 4.3. We measured the intensity of the coincidence pairs

by putting selective cuts on different dissociation channels of the coincidence map. A

drawback in our intensity measurement comes from the lack of consideration of higher
electron detection probability when multiple ionization releases a large number of elec-
trons and also due to a discrimination against fast fragments emitted perpendicular to

the TOF axis. We observed a large difference in relative intensity between break-up of
doubly and triply charged OCS molecular ions. It is evident from Figure 4.3 that in the
present energy range, collision of Ar+ with OCS yields more singly and doubly charged
OCS ions compared to triply and quadruply charged ones.
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 113

4.4.1 Dissociation of OCS2+

Two–body dissociation of OCS2+

400
1400 220
300
200
1200
counts (arbitrary unit)

200 180
+
S

Momentum (a. u.)

1000 160
100
+ 140
800 CO
0 120

600 100
-100
80
400
-200 60

40
200 -300
20

0 -400
0 5 10 15 20 25 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.4: (a) Kinetic energy release distribution and (b) momentum map of the frag-
ments contributing to the coincidence island CO+ :S+ . The momentum of CO+ is taken
as the reference and the components of momentum of S+ ion parallel and perpendicular
to it are plotted.

400
18
300
80
16
counts (arbitrary unit)

200
Momentum (a. u.)

14
60
CS+
100
12
O+
0
10
40
-100
8

-200 6
20
-300 4

0 -400 2
0 10 20 30 40 50 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.5: (a) Kinetic energy release distribution and (b) momentum map of the frag-
ments contributing to the coincidence island O+ :CS+ . The momentum of O+ is taken as
the reference and the components of momentum of CS+ ion parallel and perpendicular
to it are plotted.

From the coincidence map (Fig. 4.2) we separated two complete two-body frag-
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 114

mentation pathways of doubly charged OCS molecular ions:

OCS 2+ → CO + + S + (4.1)

OCS 2+ → O + + CS + (4.2)

In the 200 eV electron-impact dissociation of OCS metastable OCS2+ had been formed

dissociating into CO+ :S+ [2]. We observed undissociated OCS2+ peak in the TOF spec-
trum, but signature of its metastable states, displayed as a long tail extending from the
CO+ :S+ coincidence island to the diagonal of the coincidence spectrum, is not visible
in the present experiment. The auger–electron-ion coincidence measurement done by

Kaneyasu et. al. [14] revealed that OCS2+ states, lying around a binding energy of 32
eV are mostly stable against dissociation, whereas the states lying around 34 and 42
eV, dissociate predominantly into CO+ +S+ and O+C+ +S+ , respectively. Both CO+ :S+
and O+ :CS+ coincidence islands are narrow, having slope of -1 indicating pure two–

body Coulomb explosion. The kinetic energy release (KER) distributions for these two
fragmentation channels have been shown in Figures 4.4(a) and 4.5(a). For CO+ +S+ dis-
sociation channel, we found the KER has a mean value of 7.8 eV and for O+ +CS+ , 9.8
eV. In laser field induced ionization and fragmentation of OCS, Bryan and co-workers
[10] obtained the KER of CO+ +S+ channel as 4.1 eV but they did not observe the

O+ + CS+ channel of OCS2+ . It is due to the fact that OCS2+ →CO+ +S+ is more fa-
vored than OCS2+ →O+ +CS+ as evident from Figure 4.3. The momentum correlations
(Figures 4.4(b) and 4.5(b)) between the two fragments also suggest separation of them
with equal and opposite momenta |Px | = |Py |. No evidence for formation of the OS

ion has been found in our studies, though it was formed in the 200 eV electron impact
dissociation of OCS [2].

In analyzing various dissociation channels, care has been taken to separate chan-
nels with neutral fragments, by setting appropriate conditions on the list mode file.
Although neutral fragments produced in the three–body break up are not detected by
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 115

our detector, its momentum has been determined from the momentum balance in the
centre–of–mass frame. Three three-body dissociation channels of OCS2+ have been

found in our studies which we will discuss in the following sections.

Concerted fragmentation of OCS2+

Due to the asymmetric nature of OCS, it is expected that even for a perfectly linear
OCS molecule, momentum conservation would result in a non-zero momentum of the
central C ion [11]. It is evident from the following concerted type dissociation channel

of OCS2+ :

OCS 2+ → O + + C + S + (4.3)

400
500
300 50
S+
400
counts (arbitrary unit)

200
Momentum (a. u.)

40
100
300 +
O
0 30

200 -100
20
-200 C
100
-300 10

0 -400
0 20 40 60 80 100 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.6: (a) Kinetic energy release distribution and (b) momentum map of the frag-
ments contributing to the coincidence island O+ :C:S+ . The momentum of O+ is taken
as the reference and the components of momentum of S+ and C, parallel and perpen-
dicular to it are plotted.

The O+ :S+ coincidence island, as observed from Figure 4.2 is narrow and dumb–
bell shaped having slope nearly equal to -1. In the above dissociation channel the O+
and S+ ions are ejected from the two ends of the molecular ion. If we assume that
the two ions have been ejected with equal and opposite momenta from a linear precur-
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 116

sor state of OCS2+ then C should remain at rest. But momentum correlation diagram
(Fig. 4.6(b)) shows that the momentum carried away by C is not negligible; also the

momentum distribution of S+ with respect to O+ gives a distinct lobe around 1580 and
the C atom is preferentially emitted at 970 to the O+ ion. All these observations sug-
gest bending of OCS2+ in the field generated by 150 keV Ar+ ion. Upon dissociation
the total charge is shared between the two peripheral atoms and they are ejected with
nearly equal momentum. Because of the bending the C fragment is ejected with non-

zero momentum giving the O+ +C+S+ dissociation channel. It is a concerted type of

dissociation where the mean value of the kinetic energy release is 13 eV.

In the coincidence map, just above the O+ :S+ coincidence, another island with a
smaller intensity is also observed corresponding to the coincidence image of O+ with
34 +
S , an isotope of S (not visible in Figure 4.2 due to the strong background).

Sequential dissociation of OCS2+

OCS2+ may dissociate in a sequential manner:

OCS 2+ → CO + + S + (4.4)
ց
O + C+ + S+ (4.5)

Figure 4.7(b) shows the momentum distributions of O and S+ fragments relative

to the momentum vector of C+ ion corresponding to above dissociation channel. Both

distributions are very broad and no particular angle can be assigned to them. It is clear
from Figure 4.2, that the C+ :S+ island is a parallelogram and the measured slope is -2.3
± 0.1, which exactly matches with the predicted slope for S(i) fragmentation of OCS2+ .

Thus we can conclude that OCS2+ first dissociates into CO+ +S+ where CO+ , in a
subsequent stage, releases the neutral O and becomes C+ . Due to this dual step nature
of dissociation, the momentum of these fragments has a large distribution resulting in a
broad momentum image. The KER corresponding to this channel is 14 eV (Fig. 4.7(a)).
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 117

400
70
300 +
400
S 60

counts (arbitrary unit)

200

Momentum (a. u.)

50
100
300
+
C 40
0

200
-100 30

-200 O 20
100
-300
10

0 -400
0 20 40 60 80 100 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.7: (a) Kinetic energy release distribution and (b) momentum map of the frag-
ments contributing to the coincidence island O:C+ :S+ . The momentum of C+ is taken
as the reference and the components of momentum of S+ and O, parallel and perpen-
dicular to it are plotted.

Another three–body dissociation channel of OCS2+ as observed from Figure 4.2 is

given by

OCS 2+ → O + + CS + (4.6)
ց
O+ + C + + S (4.7)

where the slope of the O+ :C+ coincidence island is -3.7±0.1 and can be explained by
the S(i) mechanism. But this particular coincidence island may have contamination of
OCS3+ dissociating into O+C+ +S2+ , though the slope does not suggest that possibility.
Detail discussion on this channel is given in the next section.

Figure 4.3 shows that fragmentation of OCS2+ into the O+C+ +S+ channel is more

favorable than O+ +C+ +S channel as also observed by Millie et al [15]. According to

Safvan et al this is due to the fact that maximum Coulomb repulsion is expected to
occur when the internuclear electronic charge shielding is minimized [16].

The kinetic energy released in the above dissociation channels of OCS2+ has been
compared with that obtained from pure Coulomb explosion model (CEM). We have
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 118

calculated the theoretical values of KER taking the equilibrium geometry of neutral
OCS, RC−O = 1.16 Å, RC−S = 1.56 Å, RO−S = 2.72 Å. Table 4.1 lists the experimentally

observed most probable (Eexpt expt

prob ) and maximum (Emax ) kinetic energy release (KER) for

different dissociation channels of OCS2+ along with the theoretically obtained KER
values (Etheo
Coul ). Under this model, two KER values (5.3 and 9.2 eV) are possible for

fragmentation of OCS2+ into the CO+ +S+ channel as the point charge on the CO+
complex can be assumed to stay either near the C atom or near the O atom. Similarly

for OCS2+ dissociating into O+ +CS+ , the theoretical KER values will be 5.3 and 12.4
eV. On comparing the most probable and maximum energy values with the theoretical
KER values calculated from pure CEM, we observe that the experimental KER values
are much higher than the CEM predictions for the three–body dissociation channels of

OCS2+ .

Table 4.1: KER values and dissociation schemes for various fragmentation channels of
OCS2+ induced by ion impact. These values are compared with that from pure Coulomb
explosion model (see text).

Dissociation Fragmentation KER (eV)

channels scheme Eexpt

prob Eexpt
max Etheo
Coul

CO+ + S+ two–body 7.8 15 5.3 and 9.2

O+ + CS+ two–body 9.8 20 5.3 and 12.4

O+ + C + S+ concerted 13 70 5.3

O + C+ + S+ S(i) 14 80 9.2

O+ + C+ + S S(i) - - 12.4
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 119

4.4.2 Dissociation of OCS3+

Two–body dissociation of OCS3+

In collision of neutral OCS with slow (velocity of the order of 0.4 a.u.) Ar+ projectiles,
triply and quadruply charged OCS molecular ions are also formed though the intensities

are very low. The field, generated by the projectile, influences the break-up dynamics
of these precursors. Here we will discuss dissociation mechanisms of OCS3+ in detail.
We have observed eight coincidence islands in the two dimensional coincidence map of
OCS corresponding to the dissociation of OCS3+ (Fig. 4.2).

400 12
60
11
300
50 10
counts (arbitrary unit)

200
Momentum (a. u.)
+ 9
40
CO
100
8
S2+
30 0 7

6
-100
20
5
-200
4
10
-300
3

0 -400 2
0 10 20 30 40 50 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.8: (a) Kinetic energy release distribution and (b) momentum map of the frag-
ments contributing to the coincidence island CO+ :S2+ . The momentum of S2+ is taken
as the reference and the components of momentum of CO+ ion parallel and perpendic-
ular to it are plotted.

The only two-body fragmentation channel of OCS3+ observed in the present ex-

periment is:

OCS 3+ → CO + + S 2+ (4.8)

The corresponding coincidence island is narrow and aligned along the slope -2. The
most probable KER of this channel is 13 eV (Fig. 4.8(a)) whereas CEM suggested
KER values are 10.6 and 18.5 eV. Being a two–body fragmentation of OCS3+ , here
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 120

also CO+ and S2+ are ejected with equal and opposite momenta and it is clearly visible
in the momentum image 4.8(b).

Concerted fragmentation of OCS3+

400 8
35
300
30 7
O+
counts (arbitrary unit)

200

Momentum (a. u.)

25 6
100
20 C+
0 5

15
-100
4
10
-200
+
S
3
5 -300

0 -400 2
0 20 40 60 80 100 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.9: (a) Kinetic energy release distribution and (b) momentum map of the frag-
ments contributing to the coincidence island O+ :C+ :S+ . The momentum of C+ is taken
as the reference and the components of momentum of the O+ and S+ ions parallel and
perpendicular to it are plotted.

q
The ion time-of-flight (TOF) being proportional to m/q (where q is the charge
state and m is the mass of the ion), it is very difficult to distinguish two species having
same mass to charge ratio (for example O+ and S2+ ). Thus in the coincidence map, the
C+ :O+ /S2+ island may have contribution from four different dissociation channels of

which only two may originate from OCS3+ .

OCS 2+ → O + + C + + S (4.9)

OCS 3+ → O + + C + + S + (4.10)

OCS 3+ → O + C + + S 2+ (4.11)

OCS 4+ → O + + C + + S 2+ (4.12)

Using triple ion coincidence technique we can separate O+ +C+ +S+ from the rest of the
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 121

channels by setting appropriate conditions on the events. Momentum analysis corre-

sponding to this channel shows that the magnitude of momentum of O+ and S+ ions

are almost equal but the angle between them is 158◦ . Also the C+ ion is ejected with
non-zero momentum subtending an angle of approximately 80◦ to the O+ ion (see Fig.
4.9(a)). The above momentum correlation indicates that it is a concerted breakup where
the KER distribution peaks at around 23 eV (Fig. 4.9(b)) and is greater than the value
obtained by Laksman et al. [17].

Similarly the coincidence island corresponding to C2+ :O+ /S2+ may contain four
dissociation channels of different parent molecular ions.

OCS 3+ → O + + C 2+ + S (4.13)

OCS 4+ → O + + C 2+ + S + (4.14)

OCS 4+ → O + C 2+ + S 2+ (4.15)

OCS 5+ → O + + C 2+ + S 2+ (4.16)

The only contribution from OCS3+ in this case comes from O+ :C2+ :S channel which is
inseparable from OCS4+ → O + C2+ + S2+ channel due to same m/q value of O+ and
S2+ .

The correlation between O+ and S2+ shows two different coincidence islands of
different slopes but due to same m/q value they appear as the same peak in the TOF

spectrum (Fig. 4.1). In the fragmentation of OCS3+ into O+ +C+S2+ , either O+ or S2+
arrive to the detector first. Depending upon their sequence of arrival, we expect two
coincidence islands of two different slopes. Leaving C at rest, when O+ comes before
S2+ , the slope of the coincidence island becomes -0.5 but when S2+ comes earlier, the

slope takes the value of -2.0. In the coincidence map we found that both the islands
are broad in structure and their slopes slight deviate from the above expected values,
which suggests not inconsiderable amount of momentum carried away by the central C
atom (Fig. 4.10(b)). Both the coincidence islands are elliptical in shape irrespective of
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 122

400

90 12
300
80
O+ 10

counts (arbitrary unit)

200

Momentum (a. u.)

70

60 100 8

50
S2+
0
6
40
-100
30 4
-200
20 C
2
-300
10

0 -400 0
0 20 40 60 80 100 120 140 160 180 200 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.10: (a) Kinetic energy release distribution and (b) momentum map of the frag-
ments contributing to the coincidence island O+ :C:S2+ . The momentum of S2+ is taken
as the reference and the components of momentum of O+ and C fragments parallel and
perpendicular to it are plotted.

whether O+ or S2+ comes first, and originate from the following dissociation channel

of OCS3+

OCS 3+ → O + + C + S 2+ (4.17)

Here we have assumed that the undetected fragment is neutral C. The angle between the
momentum vectors of O+ and S2+ is ∼ 158◦ and indicates concerted breakup of OCS3+
where the C fragment ejects at an angle of 85o with respect to the S2+ ion (Fig. 4.10(b))

and this is irrespective of the detection sequence of O+ and S2+ . This strongly suggests
that a bent dissociative state of OCS3+ responsible for the three-body break-up into
O+ :C:S2+ . The arrival sequence of O+ and S2+ only gives two different coincidence
islands of different slopes. We have also plotted the KER spectra for this channel and
found that the most probable KER occurs at 23 eV (Fig. 4.10(b)).

Another three-body dissociation channel of OCS3+ is:

OCS 3+ → O + C + + S 2+ (4.18)
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 123

The C+ :S2+ coincidence island is broad and has a slope of -3.7±0.1, which cannot be
explained by S(i) or S(d) fragmentation of OCS3+ molecular ions into O+C+ +S2+ . Here

the third fragment is neutral and remains undetected by the TOFMS. Unlike the previ-
ous dissociation channel, O+ :C:S2+ , here we cannot separate O+ :C+ :S and O:C+ :S2+
channels from the coincidence island, though they originate from two different precur-
sor ions. The slope of -3.7±0.1 can only arise from OCS2+ dissociating into O+ :C+ :S
by the S(i) mechanism, but the possibility of concerted fragmentation of OCS3+ into

O:C+ :S2+ cannot be eliminated.

400 8
25
300
7
counts (arbitrary unit)

20 200 +
Momentum (a. u.) S
6
100
15
0
O2+ 5

10 -100
4
-200
C
5
3
-300

0 -400 2
0 50 100 150 200 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.11: (a) Kinetic energy release distribution and (b) momentum map of the frag-
ments contributing to the coincidence island O2+ :C:S+ . The momentum of O2+ is taken
as the reference and the components of momentum of S+ and C fragments parallel and
perpendicular to it are plotted.

OCS3+ can also dissociate in the following manner:

OCS 3+ → O 2+ + C + S + (4.19)

The O2+ :S+ coincidence island is dumb-bell shaped. Momentum correlation between
the three fragments suggests that the O2+ :S+ angle is 158◦ and the KER distribution
shows a sharp peak at 23 eV (Fig. 4.11).

In resonant core electron excitation [17], Laksman et al. have shown that the angle
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 124

between the O and S momentum vectors is 160◦ (see Fig. 4(b) in [17]) in the three-
body dissociation of OCS2+ . The same angle for the OCS3+ → O+ +C+ +S+ channel is

nearly 158◦ . No other dissociation channels of OCS3+ has been discussed in their study.
The authors reported that the C 1s → π ∗ resonant excitation induces a bending mode
in OCS2+ and OCS3+ and concerted dissociation is the primary fragmentation process
for three-body break-ups. The kinetic energy released in the reaction is independent
of the excitation energy and reflects Coulomb repulsion between the fragments after

ionization. In our study, we observed that the angle between the momentum vectors
of the O and S ions is 158◦ and the fragment charge distribution is either symmetric
(O+ +C+ +S+ ) or it is shared between the two peripheral atoms of OCS3+ (O2+ +C+S+
and O+ +C+S2+ ). This behavior is very much similar to the behavior observed in the

triple fragmentation of OCS2+ studied by Laksman et al. The three fragmentation chan-
nels of OCS3+ , discussed above, are concerted with the most probable KER = 23 eV
for all three of them (see Table 4.2).

400 9

50
300 8
counts (arbitrary unit)

40 200
Momentum (a. u.)

100
C+
6
30 O2+
0
5
20 -100

S 4
-200
10
-300 3

0 -400 2
0 20 40 60 80 100 120 140 160 180 200 -400 -300 -200 -100 0 100 200 300 400
KER (eV) Momentum (a. u.)

(a) (b)

Figure 4.12: (a) Kinetic energy release distributions and (b) momentum map of the
fragments contributing to the coincidence island O2+ :C+ :S. The momentum of O2+ is
taken as the reference and the components of momentum of C+ and S fragments parallel
and perpendicular to it are plotted.
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 125

Sequential dissociation of OCS3+

We have also got the evidence for sequential dissociation channels of OCS3+ . The
weak coincidence islands of O2+ :C+ and C2+ :S+ are parallelograms and have slopes
-0.5 and -4.6 respectively which agree well with the theoretical interpretation given for
S(i) type of fragmentation [18]. The measured slopes of these two coincidence islands
are influenced by the momenta of the intermediate dissociation products.

OCS 3+ → OC 2+ + S + (4.20)
ց
O + C 2+ + S + (4.21)

OCS 3+ → O 2+ + CS + (4.22)
ց
O 2+ + C + + S (4.23)

Figure 4.12(b) shows the momentum distributions of C+ and S fragments relative to

the momentum vector of the O2+ ion. Despite the poor statistics, both distributions are

very broad and no particular angle can be assigned to them. The momentum contri-
bution from the intermediate CS+ ion, generated in the above two step fragmentation
process, results in such a broad momentum image. The KER spectra for this channel
has been shown in Figure 4.12(a). The other dissociation channel also exhibits the same
tendency to a S(i) fragmentation (not presented here). In both the fragmentation chan-

nels we observe that the total charge is shared between the central C atom and one of the
peripheral atoms. But no evidence for S(d) fragmentation is found in our experiment.

Most probable and maximum KER values along with theoretical values calculated
from Coulomb explosion model are listed in table 4.2 for various dissociation channels

of OCS3+ . We observed that both Eexpt expt

prob and Emax values for the above dissociation

channels are much higher than that predicted by pure CEM.

Low energy ion induced fragmentation of carbonyl sulphide (OCS) 126

Table 4.2: KER values and dissociation schemes for various fragmentation channels of
OCS3+ induced by ion impact. These values are compared with that from pure Coulomb
explosion model.

Dissociation Fragmentation KER (eV)

channels scheme Eexpt

prob Eexpt
max Etheo
Coul

CO+ + S2+ two–body 13 40 10.6 and 18.5

C+ + O+ + S+ three–body 23 60 -

O+ + C + S2+ concerted 23 120 10.6

O2+ + C + S+ concerted 23 130 10.6

O2+ + C+ + S S(i) 37.5 140 24.8

O + C2+ + S+ S(i) - - 18.5

4.4.3 Dissociation of OCS4+

No two–body fragmentation channel of OCS4+ has been found in the coincidence map.
However, we have been able to identify the reaction

OCS 4+ → O 2+ + C + S 2+ (4.24)

in which O2+ and S2+ move in the opposite directions with equal momenta but neutral
C stays at rest, giving the coincidence island a slope of -1. We have also observed
coincidence of C2+ with O+ /S2+ . No third hit signal has been detected in the triple ion

coincidence map corresponding to this coincidence island. The C2+ :O+ /S2+ correlated
fragments may originate from two dissociation channels–

OCS 4+ → C 2+ + S 2+ + O (4.25)

OCS 3+ → C 2+ + O + + S (4.26)
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 127

by concerted fragmentation because the experimental slope of the coincidence island is

-3.0 which cannot be explained by the S(i) or S(d) type of fragmentation. OCS4+ may

also dissociate in the following manner:

OCS 4+ → O 2+ + C + + S + (4.27)

but the statistics is too limited for all these dissociation channels to be discussed.

4.5 Effect of projectile charge state on dissociation dy-

namics of OCS

The effect of projectile charge state on the dissociation dynamics of OCS has been in-
vestigated in the present section. We have ionized the OCS molecules by the impact
of Arq+ (where q = 6–8) projectiles having velocity of 1 a.u. and studied the disso-
ciation dynamics of multiply charged OCS molecular ions into few selected channels.
By keeping the projectile velocity constant, we have actually frozen the ion–molecule

interaction time for all charge states of Ar ion beam.

The TOF spectra of dissociating OCS (Fig. 4.13) show stable existence of undisso-
ciated parent molecular ions (OCS+ and OCS2+ ) along with various atomic fragments
having charge states 1–3. On careful observation we found that the peak intensities

of all the fragment ions are relatively large in the case of Ar8+ projectile. It is already
known that in collision with fast projectiles, direct ionization dominates when compared
to the case of low energy projectiles, where electron capture and transfer ionization are
the dominant modes. But for the present experiment, since velocity of all the projectile

ions are same, the dependence of fragment ion’s TOF distribution on projectile charge
states suggests that the degree of ionization of the parent molecule increases with the
increasing charge state of the projectile ion, resulting into highly charged atomic frag-
ments, even when the energy of the beam is kept constant.
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 128

6+
Ar
20000
7+

2+
Ar
C
8+

16000 Ar

counts (arbitrary unit)

12000 C
+

8000 3+
2+ 4+

O O /S + 2+ 2+

O /S OCS

+
+
OCS
4000
3+
C CO

+
S

0
1000 2000 3000 4000 5000
TOF of first fragment (ns)

Figure 4.13: TOF spectrum of dissociative OCS with different projectile charge states
having same velocity (1 a. u.). All spectra are normalized to OCS+ peak for all projec-
tile charge states.

3+ 3+ + +
5000
C O C2+ O2+ C +
O /S2+ CO
50
+
CS
TOF of Second Fragment (ns)

4500 45
34 + 40
4000
S
32 +
S 35
+
3500 CO 30

25
3000 +
O /S2+ 20
+
2500 C 15
2+
2000 O 10

5
1500
1000 1500 2000 2500 3000 3500 4000
TOF of First Fragment (ns)

Figure 4.14: Two-dimensional coincidence map showing inter particle correlation in

dissociation of OCS by Ar8+ ion impact. Velocity of Ar8+ is equal to 1 a. u.

Figure 4.14 shows the correlation diagram between various fragment ions of OCS
by 1 a. u. Ar8+ projectiles. We have observed several coincidence islands in the
coincidence map corresponding to the dissociation of OCSq+ where q can be 2 to at
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 129

least 5 (C3+ :O2+ ). Some neutral fragments were also produced along with the ionic
ones in the dissociation of OCS which we could not detect by the MCP detector. Also

due to insufficient detector efficiency, some fragment ions were lost from the system.
These are the reasons for which we have chosen only those dissociation channels of
OCS in the present study where all the fragment ions have been detected (Fig. 4.14)
and we have studied the effect of projectile charge state on these dissociation channels.
These are–

OCS 2+ → CO + + S + (4.28)

OCS 2+ → O + + CS + (4.29)

OCS 3+ → CO + + S 2+ (4.30)

4.5.1 Effect of projectile charge state on TOF distribution of corre-

lated ion-pairs

6+
Ar 6+
Ar
7+
Ar 7+
Ar
8+
Ar 8+
Ar

300 300
counts

counts

200 200

100 100

0 0
3600 3800 3800 3900 4000 4100
+ + + + + +
TOF of CO from CO -S channel TOF of S from CO -S channel

(a) (b)

Figure 4.15: Forward–backward TOF peaks of (a) CO+ and (b) S+ , corresponding to
OCS2+ → CO+ + S+ , projected on their respective time axes. All spectra have been
normalized to the maximum count.

Figure 4.15 shows projection of CO+ :S+ coincidence island on the first- and second–
hit axes in the dissociated OCS coincidence map (Fig. 4.14). The forward and backward
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 130

peaks in the projected TOF distribution of each fragment ion are associated with equal
energy fragment ions emitted along and opposite to the direction of applied extraction

field (or the TOF axis) in the TOFMS. The fragments, emitted with zero kinetic energy
or with small initial velocity perpendicular to the TOF axis, are expected to appear in
the middle of the forward–backward peaks. The high energy fragments, ejected per-
pendicular to the TOF axis, may not reach the detector and will be lost from the system.
On careful observation in the region between the two peaks, we found significant en-

hancement in the intensity of TOF distribution with the increasing charge state of the
projectiles. This can be the consequence of formation of low energy fragment ions in
the dissociation of OCS2+ when the projectile charge state is high.

Similar distributions have been observed in the projected TOF of the two other

two-body dissociation channels: OCS2+ → O+ + CS+ (Figures 4.16(a) and 4.16(b)) and
OCS3+ → S2+ + CO+ (Figures 4.16(c) and 4.16(d)). In both the cases, dissociation of
OCS yields low energy fragment ions in the interaction with highly charged projectiles.

4.5.2 Effect of projectile charge state on KER distribution of a dis-

sociation channel

The width of the peak in a TOF distribution is proportional to the energy spread of
the corresponding ion. But finite size of the source and inherent thermal distribution
of velocities often broaden the TOF distribution of a fragment ion. Therefore it is not
possible to use the peak width as a means of calculating the ion energy. The time
difference between the forward and backward emitted fragment ions is the measure of

fragment kinetic energy (Uker ) in the centre of mass frame.

Figure 4.17 shows the KER distributions for various dissociation channels of OCS2+
and OCS3+ obtained from the projected TOF spectra for that particular channel. We de-
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 131

6+
6+
Ar Ar
40 Ar
7+
7+
Ar
8+
Ar 8+
35 40 Ar

30

30
25

counts

counts
20
20
15

10
10

0 0
2500 2600 2700 2800 2900 3000 4400 4500 4600 4700 4800 4900
+ + + + + +
TOF of O from O -CS channel TOF of CS from O -CS channel

(a) (b)

6+ 6+
Ar Ar
7+ 7+
Ar Ar
8+ 8+
Ar Ar

300

160
counts

counts
200

80
100

0 0
2800 3500 3600 3700 3800
2+ + 2+ + + 2+
TOF of S from CO -S TOF of CO from CO -S

(c) (d)

Figure 4.16: Forward–backward TOF peaks of (a) O+ and (b) CS+ corresponding to
O+ :CS+ → OCS2+ and (c) S2+ and (d) CO+ corresponding to OCS3+ → CO+ + S2+ ,
projected on their respective time axes. All spectra have been normalized to the maxi-
mum count.

duced the KE of each fragment ion using the equation

2(2mUker )1/2
△Ts = (4.31)
qE

where Uker is the kinetic energy of the fragment ion produced in a particular disso-

ciation process, △Ts is the time separation between forward and backward emitted
fragment peaks, q is the ionic charge, E the extraction field used in the TOFMS and
m is the mass of the ion. In all the above KER spectra, we observed that the inten-
sity of high energy component decreases gradually with the increasing charge state of
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 132

2000 6+
Ar
6+
Ar
7+ 100 7+
Ar Ar
8+ 8+
Ar Ar
1500 80

counts (arbitrary unit)

counts (arbitrary unit)

60
1000

40

500
20

0 0
0 3 6 9 12 15 0 5 10 15 20
KER (eV) KER (eV)

(a) (b)

6+
1000 Ar
7+
Ar
800 8+
Ar
counts (arbitrary unit)

600

400

200

0
0 5 10 15 20 25
KER (eV)

(c)

Figure 4.17: The KER spectra for (a) CO+ +S+ , (b) O+ +CS+ and (c) S2+ +CO+ disso-
ciation channels of OCS2+ and OCS3+ . All spectra have been normalized to the counts
corresponding to zero energy.

the projectile ion. In a molecular dissociation, when fragments are produced in higher
excited states, the resultant KER becomes lower compared to the case where the same
fragments are produced in their respective ground states. On the other hand, when the
same parent molecular ion dissociates from an excited state, producing fragment ions

in their respective ground states, one can expect higher KER as compared to the case
where dissociation of parent takes place from the ground state. Thus the significant
lowering of the high energy component in the KER distribution with increasing charge
state of the projectile, as observed in all the above KER spectra, hints towards the con-
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 133

tribution of either less excited molecular states or formation of fragment ions in high
excited states.
3(qE)2 σt2
hK σ i = (4.32)
2m

According to Busch [19], the mean kinetic energy of a fragment ion is related to
σt2 , the variance of TOF distribution, through equation 4.32. For all the two-body dis-

sociation channels of OCS2+ and OCS3+ , we have derived the mean kinetic energy of
each of the fragment ion for a particular channel. These are presented in Table 4.3. This
clearly implies that the formation and dissociation of multiply charged OCS molecular
ions are very much dependent on the projectile charge states. With the increasing pro-
jectile charge state, the KER decreases, as a consequence fragment ions are ejected with

less KE, for a particular dissociation channel. This can be due to the formation of mul-
tiply charged molecular ions in lower excited states or dissociation of the same ion into
highly excited fragment ions. To get a deeper insight into the actual processes one has
to look into the molecular potential energy surfaces and the corresponding dissociation

limits.

Table 4.3: Estimation of fragment KE and mean KER (in eV) for various dissociation
channels of OCS2+ and OCS3+ , derived from the variance of TOF distribution

Projectile CO+ +S+ O+ +CS+ S2+ +CO+

charge KE of KE of KE of

state CO+ S+ O+ CS+ S2+ CO+

Ar6+ 5.76 5.12 9.03 4.24 10.18 10.95

Ar7+ 4.78 4.17 7.83 3.86 9.90 10.05

Ar8+ 4.19 3.65 6.51 3.02 8.54 8.71

Low energy ion induced fragmentation of carbonyl sulphide (OCS) 134

4.6 Summary

In this chapter, we have studied dissociation dynamics of multiply charged OCS molec-

ular ions formed in the collisions of neutral OCS molecules with 150 keV Ar+ ions. We
have used a linear TOF mass spectrometer equipped with a position-sensitive, multihit
detector to measure the momentum vectors of the fragment ions. From the coincidence
map we observed three complete two–body dissociation channels of OCS: OCS2+ →

CO+ +S+ , OCS2+ → O+ +CS+ and OCS3+ → CO+ +S2+ . It is found that the channel
CO+ +S+ is more favored than the channel O+ +CS+ . By measuring the slopes of var-
ious coincidence islands and studying momentum correlations between fragment ions,
the dissociation pattern is predicted for various channels of OCSq+ . The O+C+ +S+ ,

O+ +C+ +S, O+C2+ +S+ and O2+ +C+ +S channels are found to arise from sequential
break–up of the parent molecular ions, whereas O+ +C+S+ , O+ +C+S2+ , O2+ +C+S+
and C+ O+ +S+ result from pure concerted processes. We have estimated the geome-
tries of the precursor ions for each of the above cases and found that the dissociation
does not always take place from a linear parent. We measured a bent angle of ∼ 158◦

for both OCS2+ and OCS3+ in our studies. The KER distributions for the above disso-
ciation channels are also calculated and it is found that the CE model underestimates
our experimental KER values for the complete Coulomb fragmentation of OCS2+ and
OCS3+ due to its oversimplified point–charge assumption. Significant projectile charge

state dependence of the fragment ion kinetic energy distributions have been observed
in the dissociation of OCS molecular ions. We kept the projectile velocity constant at 1
a.u. to ensure same ion-molecule interaction time during the experiment and observed
that the mean kinetic energy of a fragment ion decreases gradually with the increasing

charge state of the projectile ion beam. This can be a consequence of either formation
of multiply charged molecular ion in lower excited state or dissociation of the same ion
into highly excited fragment ions.
Bibliography

[1] G. Bilalbegovic, Eur. Phys. J. D 49, 43 (2008)

[2] P. Wang and C. R. Vidal, J. Chem. Phys. 118, 5383 (2003)

[3] J. E. Hudson, C. Vallance and P. W. Harland, J. Phys. B: At. Mol. Opt. Phys.

37, 445 (2004)

[4] W. Kedzierski, J. Borbely and J. W. McConkey, J. Phys. B: At. Mol. Opt.

Phys. 34, 4027 (2001)

[5] S. M. Barnett, N. J. Mason and W. R. Newell, J. Phys. B: At. Mol. Opt. Phys.
25, 1307 (1992)

[6] T. Masuoka, I. Koyano and N. Saito, J. Chem. Phys. 97, 2392 (1992)

[7] M. L. Lipciuc and M. H. M. Janssen, J. Chem. Phys. 126, 194318 (2007)

[8] A. Sugita et al, J. Chem. Phys. 112, 7095 (2000)

[9] T. Masuoka, I. Koyano and N. Saito, Phys. Rev. A 44, 4309 (1991)

[10] W. A. Bryan, W. R. Newell, J. H. Sanderson and A. J. Langley, Phys. Rev. A

74, 053409 (2006)

[11] J. H. Sanderson, T. R. J. Goodworth, A. EL-Zein, W. A. Bryan, W. R. Newell,

A. J. Langley and P. F. Taday, Phys. Rev. A 65, 043403 (2002)

135
Low energy ion induced fragmentation of carbonyl sulphide (OCS) 136

[12] M. R. Jana, B. Ray, P. N. Ghosh and C. P. Safvan, J. Phys. B: At. Mol. Opt.
Phys. 43, 215207 (2010)

[13] J. H. D. Eland, Mol. Phys. 61, 725 (1987)

[14] T. Kaneyasu, M. Ito, Y. Hikosaka and E. Shigemasa, J. Korean Phys. Soc. 54,
371 (2009)

[15] P. Millie, I. Nenner, P. Archirel, P. Lablanquie, P. Fournier and J. H. D. Eland,

J. Chem. Phys. 84, 1259 (1986)

[16] C. P. Safvan, M. Krishnamurthy and D. Mathur, J. Phys. B: At. Mol. Opt.

Phys. 26, L837 (1993)

[17] J. Laksman, D. Céolin, M. Gisselbrecht and S. L. Sorensen, J. Chem. Phys.

133, 144314 (2010)

[18] R. K. Singh, G. S. Lodha, V. Sharma, I. A. Prajapati, K. P. Subramanian and

B. Bapat, Phys. Rev. A 74, 022708 (2006)

[19] F. von Busch, J. Phys. B: At. Mol. Opt. Phys. 34, 431 (2001)
Chapter 5

Fragmentation of CO2 and OCS by

high energy highly charged ion impact:
A comparative study

5.1 Introduction

The OCS molecule in its ground state is a triatomic linear molecule, chemically held to-
gether by two double bonds between the carbon atom and the sulphur and oxygen atoms

respectively. OCS is thus closely related to CO2 , which in the same way is bonded by
two double bonds. The difference between CO2 and OCS lies in their symmetry prop-
erties. CO2 is symmetric through the central carbon atom, thus belonging to the D∞v
symmetry group whereas OCS belongs to the C∞v symmetry group. Both, CO2 and

OCS, share common valence structure with outer orbital configuration σ 2g σ 2u π 4u π 4g in

their respective ground states [1]. Since OCS is asymmetric, conservation of momen-
tum, even for a perfectly straight molecule, results in nonzero momentum distribution
for the central C fragment.

Mathur et al [2] have measured energy discriminated mass spectra for recoil ions
produced in 100 MeV collisions of Si8+ with linear triatomics CS2 and CO2 using a

137
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 138

crossed-beams apparatus incorporating a quadrupole mass spectrometer and a retard-

ing potential energy analyzer. From the kinetic energy measurements of atomic recoil

ions they observed that the peripheral ions (S in case of CS2 and O in the case of
CO2 target) possess larger energies than the ions of the central C atom; the relatively
large energies imparted to C ions are explained in terms of a sequential dissociation
model involving formation of an intermediate molecular ion with inner shell vacancies.
Photoion–photoion coincidence spectrum (PIPICO) of CO2+
2 , formed by 30.4 nm pho-

toionization, contains a single peak corresponding to the formation of CO+ and O+ with
an energy release centered about 6.0 eV, which is only slightly more than the energy re-
lease found in the dissociation of metastable CO2+
2 . Dissociation of OCS
2+
produced
by the same 30.4 nm photoionization is dominated by the channel leading to CO+ + S+

in comparison with CS+ + O+ . The very strong preference for CO+ + S+ as products
rather than CS+ + O+ can be due to the specific features of OCS2+ surface [3]. Millie
et al have calculated the term schemes of CO2+ 2+
2 , CS2 and OCS
2+
and compared the
calculated energies with various experimental data obtained from Auger spectra, double

charge transfer, and photoionization including the PIPICO technique. They have con-
firmed that the triplet ground states of these ions are well populated by photoionization,
while the excited singlet states are revealed most clearly by the double charge transfer
technique [1].

In the previous chapters we have already discussed dissociation dynamics of CO2

and OCS by low energy ion impact using time of flight coincidence technique. Here we

will compare the results of ion induced fragmentation of OCS with the same for CO2 in
order to know the effect of replacing one oxygen atom by one sulphur atom.

5.2 Experimental details

In the experiment we probed the dynamics of dissociation of CO2 and OCS molecules
separately by the impact of 5 MeV u−1 28
Si12+ ions from Pelletron [4] under the same
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 139

experimental condition. The projectile ions, transported to the collision chamber, inter-
acted with the beam of neutral molecules effusing from a capillary, in a crossed-beams

geometry. An ion current of a few pA was maintained during the experiments. The
low current was necessary to keep the accidental coincidences at an acceptable rate.
The vacuum in the main chamber was 7 × 10−8 mbar without any target molecular
beam and around 5 × 10−7 mbar with the target loaded. The target gas number density
was approximately 1012 – 1013 cm−3 . The stagnation pressure behind the capillary was

approximately 1.4 mbar, and the delivery lines were maintained at room temperature.

In the experiment coincidence momentum imaging technique was employed which

is based on the principle of simultaneous measurement of ion time-of-flight (TOF) and

spatial spread transverse to the flight direction under the action of a uniform electric
field. The field of strength 140 V cm−1 was applied over a length of 99 mm, in a di-
rection perpendicular to the plane formed by the crossed ion beam and target molecular

gas jet. Fragment ions travel through this field, followed by a field-free drift region of
length 198 mm before hitting the detector. At the exit of the drift tube, the ionic frag-
ments were detected by a micro channel plate detector (MCP) equipped with a position
sensitive delay line detector. Ejected electrons are extracted in the direction opposite
to the ions and detected using another MCP. The event rate during data acquisition was

approximately 500–600 Hz.

The recorded data were analyzed off-line. The observed flight time t and the po-

sition (x, y) information for each ion, recorded in an event, were transformed to yield
three momentum components of each ion. The multihit capability of the acquisition
system permitted detection and determination of momentum of all three fragment ions
produced from the same event. From the multihit data, further information about the

fragmentation process such as angle of ejection, kinetic energy release, momentum cor-
relations between fragments, etc. were derived.
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 140

5.3 Results and discussion

C2+ O2+ C+ O+ 4500

C2+ O2+ C+ O+/S2+
120
4000 +
CO 80
TOF of second fragment (ns) S+

TOF of second fragment (ns)

100 70
4000
CO+
3500 60
80

50
3500
60
40
3000 O+
40 30
3000 O+/S2+
+
C 20
2500 20
+
C 10
2500
2000 2500 3000 3500 2000 2500 3000
TOF of first fragment (ns) TOF of first fragment (ns)

(a) (b)

Figure 5.1: (a) Two dimensional coincidence map of dissociative CO2 . (b) Coincidence
map of OCS showing various dissociation channels corresponding to Cq+ , Oq+ and Sq+
ions (where q = 1, 2). The projectile was 5 MeV u−1 Si12+ for both CO2 and OCS.

We observed a wide range of dissociation products from the TOF spectrum of

CO2 and OCS starting from the undissociated molecular ions, CO+ , CS+ (in case of
OCS only), to complete rupture of C–O and C–S bonds of the respective molecules.

Figure 5.1 shows the two–dimensional coincidence map of dissociative CO2 and OCS
by 5 MeV u−1 Si12+ ion impact. It is a 2D histogram between the TOF of the first
fragment ion versus that of the second and is useful for identification and separation
of multi-particle events representing inter–particle correlations. From the coincidence

map of the fragment ions, we separated various dissociation channels corresponding

to carbon, oxygen and sulphur ionic fragments of doubly and triply ionized CO2 and
OCS molecular ions. Due to the number of possible two–body and three–body frag-
mentation channels, and the fact that there is a charge to mass ratio degeneracy between

Oq+ and S2q+ ions of dissociative OCS, the data sets were parsed for each channel in
terms of their ion states. According to the charge state distribution among the fragment
ions, we have categorized seven types of dissociation channels of CO2 and OCS, here
after denoted by OCX where X = O for CO2 and S for OCS. These are (i) CO+ +X+
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 141

and O+ +CX+ , (ii) C+ +X+ +O, (iii) C+O+ +X+ , (iv) C+O2+ +X+ and C+O+ +X2+ , (v)
C2+ +X+ +O, (vi) C+ +O2+ +X and (vii) C+ +O+ +X+ . In the following sections we will

study the angular distribution between the two terminal atoms, kinetic energy release
during the fragmentation process and momentum correlations between all the fragments
for each of the above dissociation channels of CO2 and OCS molecular ions.

5.3.1 Two–body dissociation of OCX2+

OCX2+ → CO+ +X+ and O+ +CX+

The only two–body dissociation channel of CO2+

2 found in the present experiment is

CO22+ → CO + + O + (5.1)

Unlike CO2 , OCS consists of three different atoms. Therefore two–body dissociation
of OCS2+ proceeds either via O–CS or via OC–S bond break:

OCS 2+ → CO + + S + (5.2)

OCS 2+ → O + + CS + (5.3)

The CO+ :O+ coincidence island of CO2 (Fig. 5.1(a)) is narrow, dumb bell shaped
and has a long tail corresponding to metastable CO2+
2 (already discussed in Chapter 3).

But no such metastable feature has been observed in the CO+ :S+ and O+ :CS+ coinci-
dence islands of OCS (not shown in Figure 5.1(b)). Being pure two–body dissociation,
the momentum correlations between the two fragment ions of CO2+
2 (Fig. 5.2(a)) and

OCS2+ (Figures 5.2(b) and 5.2(c)), indicate separation of them with equal and oppo-

site momenta. From NIST data base [5] we found that the bond dissociation energy of
the diatomic molecule CO is 11.09 eV while that of CS is 7.36 eV in their respective
ground states. Therefore, in case of OCS, it would be easier to break the OC–S bond
rather than the O–CS bond. The maximum kinetic energy liberated in the C–O bond
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 142

300 300

200 200
S+
CO+

Momentum (a. u.)

Momentum (a. u.)

100 100

+
O CO+
0 0

-100 -100

-200 -200

-300 -300
-300 -200 -100 0 100 200 300 -300 -200 -100 0 100 200 300
Momentum (a. u.) Momentum (a. u.)

(a) (b)

300

+ +
O + CO

200
CS+ 7000
+ +
CO + S
+ +
O + CS
Momentum (a. u.)

6000
100

counts (arbitrary unit)

5000
O+
0 4000

3000
-100
2000

-200 1000

0
0 10 20 30 40 50
-300
-300 -200 -100 0 100 200 300 KER (eV)

Momentum (a. u.)

(c) (d)

Figure 5.2: Momentum map of the (a) CO+ +O+ (b) CO+ +S+ and (c) CS+ +O+ dis-
sociation channels. The momentum of the first hit is taken as the reference and the
components of momentum of the other ion parallel and perpendicular to it are plotted.
(d) Kinetic energy release distributions for two–body dissociation channels of OCX2+
where X = O for CO2 and S for OCS.

break of CO2+
2 is found to be 20 eV while the most probable kinetic energy release

(KER) is 6.3 eV for the dissociation channel given by equation 5.1 (Fig. 5.2(d)). The
most probable KER for CO+ +S+ channel, observed in the present experiment, is 4.5
eV and it is 5.5 eV for O+ +CS+ channel (Fig. 5.2(d)). Thus it is confirmed from Figure

5.2(d) that OCS2+ → CO+ +S+ is more favored than OCS2+ → O+ +CS+ . Comparing
the experimentally obtained KER values with the CEM predictions we found very good
agreement in case of CO+ +O+ (Table 3.1) and O+ +CS+ (Table 4.1) channels. But the
most probable KER for CO+ +S+ channel is much lower compared to its theoretical
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 143

estimation. The KER distributions are similar for CO+ +O+ and CO+ +S+ channels but
the same distribution for O+ +CS+ channel is different. The maximum KER is much

higher in this case which indicates involvement of highly excited molecular states.

5.3.2 Three–body dissociation of OCX2+

As the positive charge can be distributed onto the fragments in many different ways
there exist numerous fragmentation channels of triatomic molecular ions. Dissociation
of doubly charged triatomic molecule into three fragments, always produces a neutral

atom which is never detected by the usual charge particle detector.

OCX2+ → C+ +X+ +O

In the present experiment we observed two dissociation channels of CO2+

2 resulting

into C+ :O+ and O+ :O+ correlated ion pairs. The C+ :O+ coincidence island may have
originated from the two dissociation channels mentioned below:

CO22+ → C + + O + + O (5.4)

CO23+ → C + + O + + O + (5.5)

We separated the above two channels of doubly and triply charged CO2 molecular ions
by analyzing the list mode data and sorting out those events which led to ion pairs
only. The events in which ion pairs are formed arise from doubly ionized precursors.
The shape of the island is a parallelogram which indicates sequential dissociation of

the parent molecular ion but the slope of -3.7 (Fig. 5.1(a)) does not agree with it [6].
Dissociation channels of OCS2+ , similar to the charge distribution of equation 5.4, are

OCS 2+ → C + + O + + S (5.6)

OCS 2+ → C + + S + + O (5.7)
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 144

Due to the same m/q value of O+ and S2+ ions, equation 5.6 will be contaminated with

OCS 3+ → C + + S 2+ + O (5.8)

The C+ :O+ /S2+ coincidence island in Figure 5.1(b) is also a parallelogram having slope

-3.7. This can be explained by the sequential dissociation of OCS2+ in the following
manner:

OCS 2+ → O + + CS + (5.9)
ց
O+ + C + + S (5.10)

However, possibility of concerted break up of OCS3+ into equation 5.8 cannot be ruled
out.

The C+ :S+ coincidence island corresponding to equation 5.7 is another parallelo-

gram having slope -2.3 (Fig. 5.1(b)) and arises due to sequential or S(i) dissociation of

OCS2+ in the following manner:

OCS 2+ → CO + + S + (5.11)
ց
C+ + O + S+ (5.12)

The momentum correlations between various fragments of CO2+

2 dissociating into

C+ +O+ +O and OCS2+ dissociating into C+ +S+ +O have been shown in Figures 5.3(a)
and 5.3(b) respectively. The momentum distributions of O+ (in case of CO2 ), S+ (in
case of OCS) and neutral O are broad and single lobed relative to the momentum of C+

ion which also indicate sequential dissociation of the parent molecular ions. In CO2+
2 ,

the two oxygen fragments are emitted with almost equal momenta but in OCS2+ , mag-
nitude of momenta are not same for S+ and O fragments. We observed that for OCS2+
the momenta of C+ and O are small compared to the momentum of S+ ion (Fig. 5.3(b)).
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 145

300 300
S+
+
200
O
200

Momentum (a. u.)

Momentum (a. u.)

100 100
+
C C+
0 0

-100 -100

-200
O -200 O

-300 -300
-300 -200 -100 0 100 200 300 -300 -200 -100 0 100 200 300
Momentum (a. u.) Momentum (a. u.)

(a) (b)

C
+
+ O
+
+ O C
+
+ O
+
+ O
C
+ +
+ S + O 4000 C
+ +
+ S + O
3500
2000
3000

counts (arbitrary unit)

counts (arbitrary unit)

2500
1500

2000

1000 1500

1000
500
500

0
0
0 20 40 60 80 100 120 140 160 180
0 20 40 60 80 100
KER (eV) X
+
: O Angle (degrees)

(c) (d)

Figure 5.3: Momentum map of the (a) C+ +O+ +O and (b) C+ +S+ +O dissociation chan-
nels. The momentum of the C+ is taken as the reference and the components of mo-
mentum of the other fragments, parallel and perpendicular to C+ , are plotted. (c) Com-
parison of the two KER distributions obtained by double coincidence event OCX2+ →
C+ +X+ +O (where X = O for CO2 and S for OCS). (d) Distribution of the angle between
the two momentum vectors of the two peripheral fragments of OCX2+ .

We have plotted the angle between the two momentum vectors of the two terminal atoms
and found that the most probable angle between O+ and O of CO2+ ◦
2 is 170 and between

S+ and O of OCS2+ is 166◦ (Fig. 5.3(d)). The tails of the distributions are extended up
to 60◦ for both the dissociation channels. Result of 1300 eV electron impact dissocia-

tion of CO2+ + + +
2 into C +O +O shows that the emission of O and O is not symmetric

with respect to the emission of C+ ion [7]. According to the authors, the above disso-
ciation is due to an asymmetric stretch where the neutral O causes the trajectory of the
C+ ion to deviate. The kinetic energy release distributions for the above two dissoci-
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 146

ation channels (C+ +O+ +O and C+ +S+ +O) are also shown in Figure 5.3(c). Though
the maximum energy release is same for both of them, significant difference has been

observed in the most probable KER values. The KER distribution of C+ +S+ +O shows
two clear peaks at 11.2 eV and 25.3 eV with a long tail extending up to 80 eV. The same
distribution of C+ +O+ +O has a single broad peak around 22.6 eV.

OCX2+ → C+O+ +X+

300 300
S+
nd +
200 2 O
200
Momentum (a. u.)

Momentum (a. u.)

100 100

1stO+ O+
0 0

-100 -100
C
-200 -200
C

-300 -300
-300 -200 -100 0 100 200 300 -300 -200 -100 0 100 200 300
Momentum (a. u.) Momentum (a. u.)

(a) (b)

C +
+ O + O
+
C +
+ O + O
+

C +
+ O
+
+ S C +
+ O
+
+ S

1400
2000

1200
counts (arbitrary unit)
counts (arbitrary unit)

1000 1500

800
1000
600

400
500

200

0 0
0 20 40 60 80 100 0 20 40 60 80 100 120 140 160 180
+ +
KER (eV) O : X Angle (degrees)

(c) (d)

Figure 5.4: Momentum map of the (a) C+O+ +O+ and (b) C+O+ +S+ dissociation chan-
nels. The momentum of the first O+ is taken as the reference and the components of
momentum of the other fragments, parallel and perpendicular to O+ , are plotted. (c)
Comparison of the two KER distributions obtained by double coincidence event OCX2+
→ C+O+ +X+ (where X = O for CO2 and S for OCS). (d) Distribution of the angle be-
tween the two momentum vectors of the two terminal ions of OCX2+ .

In the present experiment, from coincidence map, we have identified and separated
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 147

two dissociation channels of CO2+

2 and OCS
2+
where the total charge of the parent
molecular ion is shared between the two terminal atoms:

CO22+ → C + O + + O + (5.13)

OCS 2+ → C + O + + S + (5.14)

Both O+ :O+ and O+ :S+ coincidence islands are narrow having slope of -1. The mo-
mentum distributions of the O+ (in case of CO2 ), S+ (in case of OCS) and C fragments
are single lobed with respect to the momentum of the first O+ ion (shown in Figures
5.4(a) and 5.4(b)) indicating strong correlations among them during the fragmentation

processes. From the momentum correlation maps, we found that the two peripheral
ions are ejected in opposite directions with almost equal magnitude of momenta and
the central, neutral C atom stays at rest. The asymptotic angle between the momentum
vectors of the two fragment ions is 161◦ for both CO2+
2 and OCS
2+
and can be as small

as 60◦ . Thus we may conclude that equations 5.13 and 5.14 proceed via concerted dis-
sociation of CO2+ 2+ + +
2 and OCS . The KER distribution for C+O +S has two peaks:

one at 15 eV and the other at 22.8 eV. The 22.8 eV peak can also be found in the distri-
bution of C+O+ +O+ along with other high energy peaks (29 eV, 35 eV, etc.). However
the maximum KER is same for both of them (Fig. 5.4(c)). Our observation matches

partially with the electron impact dissociation of CO2+ + +

2 into C+O +O [7]. The au-

thors observed that in this case, the two bonds of CO2+

2 broke simultaneously due to the

symmetric stretch of the molecular ion. They reported two possible angles between the
momentum vectors of two O+ ions: 80◦ and 160◦ . The mean kinetic energy release for

the 160◦ case was nearly 20.3 ± 4.8 eV.

5.3.3 Three–body dissociation of OCX3+

In the present experiment we have observed three correlated ion pairs corresponding to
O2+ :O+ , C2+ :O+ and C+ :O2+ in the coincidence map of CO2 (Fig. 5.1(a)) where the
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 148

charge distribution is asymmetric.

OCX3+ → C+O2+ +X+ and C+O+ +X2+

300 300 400

+
S2+
+ S 300
200 O 200
200
Momentum (a. u.)

Momentum (a. u.)

Momentum (a. u.)

100 100
100

O2+ O2+ O+
0 0 0

-100
-100 C -100
-200
C
-200 -200
-300 C

-300 -300 -400

-300 -200 -100 0 100 200 300 -300 -200 -100 0 100 200 300 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.) Momentum (a. u.)

(a) (b) (c)

C + O
2+
+ O
+

C + O
2+
+ O
+

80 C 2+
+ S + O
+
200 C + O
2+ +
+ S

C + O
2+
+ S
+

C + S
2+ +
+ O

70

60
counts (arbitrary unit)

150
counts (arbitrary unit)

50

40 100

30

20 50

10

0 0
0 40 80 120 160 200 0 20 40 60 80 100 120 140 160 180
2+ + + 2+
KER (eV) O : X or O : X Angle (degrees)

(d) (e)

Figure 5.5: Momentum map of the (a) C+O2+ +O+ , (b) C+O2+ +S+ and (c) C+S2+ +O+
dissociation channels. The momentum of the first ion is taken as the reference and the
components of momentum of the other fragments, parallel and perpendicular to it, are
plotted. (d) Comparison of the three KER distributions obtained by double coincidence
event OCX3+ → C+O2+ +X+ and C+O+ +X2+ (where X = O for CO2 and S for OCS).
(e) Distribution of the angle between the two momentum vectors of the two terminal
ions of OCX3+ .

The O2+ :O+ coincidence island of dissociative CO2 is very narrow (Fig. 5.1(a))

and has a slope of -2. This suggests separation of the two oxygen ions with equal and
opposite momenta while the C fragment remains at rest. From the triple ion coincidence
map, we confirmed that C is neutral in this case and the possible dissociation channel is

CO23+ → C + O 2+ + O + (5.15)
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 149

Even from the momentum correlation map, we found that the momentum of C is negli-
gible and the O+ ion is emitted with equal and opposite momentum relative to the O2+

ion (Fig. 5.5(a)). The angle between O2+ and O+ is around 160◦ and has a long tail
extending up to 0◦ (Fig. 5.5(e)). Figure 5.5(a) indicates concerted break up of CO3+
2

into C+O2+ +O+ . Photofragmentation of CO3+

2 in the above channel (Eq. 5.15) results

from concerted as well as sequential dissociation of CO3+

2 as suggested by Singh et al

[8].

From coincidence map of OCS, shown in Figure 5.1(b), two islands similar to

O2+ :O+ of CO2 have been identified. These are O2+ :S+ and S2+ :O+ corresponding to
the channels

OCS 3+ → C + O 2+ + S + (5.16)

OCS 3+ → C + S 2+ + O + (5.17)

where the central C atom has been ejected as neutral and the total charge of the parent
molecular ion is asymmetrically shared between the two peripheral atoms. The O2+ :S+

island is very narrow, like a straight line of slope -2 but the S2+ :O+ island is very broad
(Fig. 5.1(b)). A close inspection of S2+ :O+ coincidence island reveals two blobs of two
different slopes: -2 and -0.5. For both dissociation channels (Equations 5.16 and 5.17),
C atom possesses negligible amount of momentum and the terminal ions escape with

equal momenta in the opposite directions (Figures 5.5(b) and 5.5(c)). The asymptotic
angles between O2+ :S+ and S2+ :O+ momentum vectors are 163◦ (Fig. 5.5(e)) suggest-
ing concerted break–up of OCS3+ from a bent dissociative state. The kinetic energy
release distributions for equations 5.15, 5.16 and 5.17 are shown in Figure 5.5(d). The

most probable KER for CO3+ 2+ +

2 dissociating into C+O +O is very broad and peaks at

around 37.5 eV. The intensities of KER distributions are highest at 29.7 eV and 42.8 eV
for OCS3+ dissociating into C+O2+ +S+ and C+O+ +S2+ respectively.
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 150

OCX3+ → C2+ +X+ +O

300 300
S+
O+
200 200
Momentum (a. u.)

Momentum (a. u.)

100 100
C2+
C2+
0 0

-100 -100

-200 O -200 O

-300 -300
-300 -200 -100 0 100 200 300 -300 -200 -100 0 100 200 300
Momentum (a. u.) Momentum (a. u.)

(a) (b)

C
2+
+ O
+
+ O C
2+
+ O
+
+ O
120 C
2+
+ S
+
+ O 160 C
2+
+ S
+
+ O
140
100
counts (arbitrary unit) 120
counts (arbitrary unit)

80 100

80
60

60
40
40

20 20

0
0
0 20 40 60 80 100 120 140 160 180
0 25 50 75 100 125 150 175 200
KER (eV) X
+
: O Angle (degrees)

(c) (d)

Figure 5.6: Momentum map of the (a) C2+ +O+ +O and (b) C2+ +S+ +O dissociation
channels. The momentum of the C2+ is taken as the reference and the components
of momentum of the other fragments, parallel and perpendicular to C2+ , are plotted.
(c) Comparison of the two KER distributions obtained by double coincidence event
OCX3+ → C2+ +X+ +O (where X = O for CO2 and S for OCS). (d) Distribution of the
angle between the two momentum vectors of the two terminal ions of OCX3+ .

The C2+ :O+ island in the coincidence map of CO2 (Fig. 5.1(a)) arises from disso-
ciation of CO2 in the following manners:

CO23+ → C 2+ + O + + O (5.18)

CO24+ → C 2+ + O + + O + (5.19)

The contribution of CO4+

2 in the above coincidence island is very small. We have sep-
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 151

arated the C2+ +O+ +O channel of CO3+

2 from the above coincidence island by putting

appropriate condition on the data file. Similar coincidence island, corresponding to

C2+ :S+ , has also been extracted from the coincidence map of OCS. Due to the con-
tamination from C2+ +S2+ +O, we could not separate OCS3+ → C2+ +O+ +S channel
from the coincidence map of OCS. Therefore the comparison has been made between
equation 5.18 and
OCS 3+ → C 2+ + S + + O (5.20)

Both C2+ :O+ and C2+ :S+ coincidence islands in Figures 5.1(a) and 5.1(b) have infinite
slope. While C2+ :O+ island is dumb–bell shaped, C2+ :S+ island is a parallelogram.
Alignment of the two islands along C2+ line suggests that most of the kinetic energies

have been carried away by the peripheral atoms. The momentum distributions of O+
(for CO3+ + 3+
2 ) and S (for OCS ) ions relative to the momentum of C
2+
ion have been
shown in Figures 5.6(a) and 5.6(b) along with the momentum distribution of the neutral
O fragment. It is clear from these two figures that the two terminal atoms are ejected

with almost equal momenta making an angle of either 162◦ (for OCS3+ ) or 166◦ (for
CO3+ 3+ 3+
2 ) between them. Due to this small bending in CO2 and OCS , C
2+
is ejected
with non–zero momentum in both the cases. Singh et al have also concluded that the
C2+ :O+ coincidence island, observed in the photofragmentation of CO3+
2 , arises due to

the concerted break up [8]. The kinetic energy release distributions for C2+ +X+ +O type
dissociation channel of CO3+
2 and OCS
3+
are similar with most probable KER equal to
38 eV.

OCX3+ → C+ +O2+ +X

We have also observed dissociation of CO3+

2 and OCS
3+
in the following manners:

CO23+ → C + + O 2+ + O (5.21)

OCS 3+ → C + + O 2+ + S (5.22)
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 152

400 400

C+ C+

Momentum (a. u.)

Momentum (a. u.)

200 200

O2+ O2+
0 0

-200 O -200 S

-400 -400

-400 -200 0 200 400 -400 -200 0 200 400

Momentum (a. u.) Momentum (a. u.)

(a) (b)
+ 2+ + 2+
C +O +O C +O +O
+ 2+ + 2+
C +O +S C +O +S

60
250

50

counts (arbitrary uniit)

counts (arbitrary unit)

200

40
150
30

100
20

50
10

0 0
0 50 100 150 200 0 20 40 60 80 100 120 140 160 180
2+
KER (eV) O : X Angle (degrees)

(c) (d)

Figure 5.7: Momentum map of the (a) C+ +O2+ +O and (b) C+ +O2+ +S dissociation
channels. The momentum of the O2+ is taken as the reference and the components
of momentum of the other fragments, parallel and perpendicular to O2+ , are plotted.
(c) Comparison of the two KER distributions obtained by double coincidence event
OCX3+ → C+ +O2+ +X (where X = O for CO2 and S for OCS). (d) Distribution of the
angle between the two momentum vectors of the two terminal ions of OCX3+ .

OCS 3+ → C + + S 2+ + O (5.23)

Though m/q of O2+ and S4+ are same, we have ruled out the possibility of C+ :S4+
coincidence in OCS because no S3+ or S4+ ion has been detected by the TOFMS in

the present experiment. Dissociation of OCS3+ into C+ +S2+ +O is inseparable from

OCS2+ → C+ +O+ +S, hence beyond the scope of comparison. Here we have compared
equations 5.21 and 5.22 and the results are shown in Figure 5.7. We have selected only
those channels from the entire data set for which the third fragment is neutral. The
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 153

O2+ :C+ coincidence islands in Figures 5.1(a) and 5.1(b) resemble parallelograms with
slope very near to zero. This slope suggests that the C+ ion stays nearly at rest, while

O2+ emerges with comparatively larger momentum. The momentum distribution of

C+ and the undetected fragment are shown in Figures 5.7(a) and 5.7(b) relative to the
momentum vector of O2+ ion. These distributions are single lobed indicating strong
correlations between the dissociative fragments. From the momentum correlation dia-
grams we observed that O2+ and the other undetected fragment (O in case of CO2 and

S in case of OCS) dissociate in opposite directions with equal momenta leaving C+ at

rest. Thus we may conclude that it is the instantaneous break up of CO3+
2 and OCS
3+

which gives rise to C+ +O2+ +X type dissociation channels. Our prediction does not
match with the result of Singh et al where they have suggested that sequential fragmen-

tation of CO3+ +
2 alone leads to the C :O
2+
coincidence island [8]. The kinetic energy
release (Fig. 5.7(c)) and the angular (Fig. 5.7(d)) distributions are similar for the above
two dissociation channels of CO3+ 3+
2 and OCS . The most probable KER is 52.5 eV and

the asymptotic angle between the momentum vectors of O2+ and X is nearly 166◦ for

OCS3+ and 170◦ for CO3+ ◦

2 with small tails extending up to 100 .

OCX3+ → C+ +O+ +X+

Triple ion fragmentation of CO3+

2 and OCS
3+
have been observed in the present exper-
iment:

CO23+ → C + + O + + O + (5.24)

OCS 3+ → C + + O + + S + (5.25)

Using triple ion coincidence technique, we have separated the above two fragmentation
channels from the C+ :O+ coincidence islands (Figures 5.1(a) and 5.1(b)). Momentum
correlations between various fragment ions are shown in Figures 5.8(a) and 5.8(b). The
momentum maps show ejection of the two peripheral ions with almost equal and oppo-
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 154

300 300

1 O st + O+
200 200

Momentum (a. u.)

Momentum (a. u.)

100 100

+
C C+
0 0

-100 -100

-200 -200
2ndO+ S+

-300 -300
-300 -200 -100 0 100 200 300 -300 -200 -100 0 100 200 300
Momentum (a. u.) Momentum (a. u.)

(a) (b)
+ + +
+ + +
C + O + O C + O + O

70 C
+ +
+ O + S
+
60 C
+ +
+ O + S
+

60 50

counts (arbitrary unit)

counts (arbitrary unit)

50
40

40
30
30
20
20

10
10

0 0
0 10 20 30 40 50 60 70 80 90 100 0 20 40 60 80 100 120 140 160 180
+ +
KER (eV) O : X Angle (degrees)

(c) (d)

40 250

35
C+ C+
200
counts (arbitrary unit)

counts (arbitrary unit)

30

25
150 S+
20
1stO+ O+
100
15

10
2ndO+ 50
5

0 0
0 5 10 15 20 25 30 35 40 45 50 0 5 10 15 20 25 30 35 40 45 50
Fragment KE (eV) Fragment KE (eV)

(e) (f)

Figure 5.8: Momentum map of the (a) C+ +O+ +O+ and (b) C+ +O+ +S+ dissociation
channels. The momentum of C+ is taken as the reference and the components of mo-
mentum of the other fragments, parallel and perpendicular to C+ , are plotted. (c) Com-
parison of the two KER distributions obtained by triple coincidence event OCX3+ →
C+ +O+ +X+ (where X = O for CO2 and S for OCS). (d) Distribution of the angle be-
tween the two momentum vectors of the two terminal ions of OCX3+ . (e, f) Fragment
kinetic energy distributions for C+ , O+ and X+ of CO3+
2 and OCS .
3+
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 155

site momenta where the momentum taken away by C+ is not negligible. The O+ :X+
angular distributions peak at 158◦ for OCS3+ and 165◦ for CO3+
2 . Both the distributions

are extended up to 60◦ . The hump at around 135◦ may arise from some bent dissociative
states of CO3+
2 and OCS
3+
(Fig. 5.8(d)). Because of the symmetry in the constituent
atoms, the two O+ ions from CO3+
2 carry equal amount of kinetic energies (Fig. 5.8(e)),

but the kinetic energy shared by O+ ion of OCS3+ is greater than that of S+ ion (Fig.
5.8(f)). In both cases, kinetic energy of C+ ion is not zero due to the small bending in

the parent molecular ion. The kinetic energy release distribution is almost similar for
triple ion fragmentation channels of OCS3+ and CO3+ 3+
2 . The KER distribution of CO2

→ C+ +O+ +O+ shows signature of four peaks at 20.7 eV, 26.8 eV, 30.8 eV and 42.8 eV
(Fig. 5.8(c)). The 26.8 and 42.8 eV peaks are also present in the KER distribution of

OCS3+ → C+ +O+ +S+ along with two other peaks at 22.8 eV and 33 eV.

5.4 Summary

In this chapter, we have done a comparative study of dissociation mechanisms of doubly

and triply charged CO2 and OCS molecular ions, formed in the collision of respective
neutral molecules with 5 MeV u−1 Si12+ ions. Being heteronuclear, the number of dis-
sociation channels of OCS is greater than that of CO2 . Since m/q value of O+ and S2+
ions are same, some of the dissociation channels of OCS are inseparable too. The only

two–body dissociation channel observed in the fragmentation of CO2 is CO+ +O+ . But
we have identified three two–body dissociation channels of OCS leading to CO+ +S+ ,
O+ +CS+ and CO+ +S2+ of which the CO+ +S+ channel is more favorable than O+ +CS+
channel. The measured KER values for both CO+ +O+ and O+ +CS+ channels were

well accorded with the theoretical results predicted by CEM. We have analyzed all the
three–body dissociation channels of doubly and triply charged CO2 and OCS molecu-
lar ions and found that the three–body dissociation of OCX2+ proceeds via concerted
as well as sequential manner but OCX3+ always dissociates in a concerted way. The
Fragmentation of CO2 and OCS by high energy highly charged ion impact: A comparative study 156

asymptotic angle between O+ :O+ and O+ :S+ momentum vectors are same (∼ 160◦ ) for
concerted fragmentation of CO2+
2 and OCS
2+
into C+O+ +O+ and C+O+ +S+ respec-

tively. But in case of sequential dissociation, the momenta of the terminal fragments
include an angle around 170◦ for CO2+ + + ◦
2 → C +O +O and around 166 for OCS
2+

→ C+ +S+ +O channels. We observed that the angle between the momentum vectors
of two terminal atoms is always smaller for dissociation channels of OCS3+ except
C+O2+ +X+ and C+O+ +X2+ . From triple ion fragmentation of CO3+
2 we observed that

the two O+ ions carry equal amount of kinetic energies. But due to its hetero–nuclear
structure, O+ , ejected from OCS3+ , possesses much larger kinetic energy compared to
the S+ ion. However the total kinetic energy release distribution is similar for both of
them. Distribution of angle between the momentum vectors of the two terminal ions

show participation of bent excited states of both CO3+

2 and OCS
3+
in the triple ion
break–up. Analyzing all three–body dissociation channels, we conclude that the max-
imum kinetic energy release is same for similar dissociation dissociation channels of
CO2 and OCS. But the most probable KER is same only for C2+ +X+ +O (∼ 38 eV) and

C+ +O2+ +X (∼ 52.5 eV) channels. For others, it is generally lower in case of OCS.
Bibliography

[1] P. Millie, I. Nenner, P. Archirel, P. Lablanquie, P. Fournier and J. H. D. Eland,

J. Chem. Phys. 84, 1259 (1986)

[2] D. Mathur, E. Krishnakumar, F. A. Rajgara, U. T. Raheja and V. Krishna-

murthi, J. Phys. B: At. Mol. Opt. Phys. 25, 2997 (1992)

[3] D. M. Curtis and J. H. D. Eland, Int. J. Mass Spectrom. Ion Proc. 63, 241
(1985)

[4] D. Kanjilal, S. Chopra, M. M. Narayanan, I. S. Iyer, V. Jha, R. Joshi and S.

K. Datta, Nucl. Instum. Methods Phys. Res., Sect. A 328, 97 (1993)

[5] http://www.nist.gov/data/nsrds/

[6] J. H. D. Eland, Mol. Phys. 61, 725 (1987)

[7] B. Bapat and V. Sharma, J. Phys. B: At. Mol. Opt. Phys. 40, 13 (2007)

[8] R. K. Singh, G. S. Lodha, V. Sharma, I. A. Prajapati, K. P. Subramanian and

B. Bapat, Phys. Rev. A 74, 022708 (2006)

157
Chapter 6

Dissociation dynamics of sulphur

hexafluoride (SF6)

6.1 Introduction

SF6 is the smallest molecule of Oh symmetry. The ground state valence electronic con-

figuration of SF6 is given by (4a1g )2 (3t1u )6 (2eg )4 (5a1g )2 (4t1u )6 (1t2g )6 (3eg )4 (1t2u )6
(5t1u )6 (1t1g )6 [1]. It can be considered the prototype of a number of species in which
a central atom is surrounded by a cage of electronegative atoms. SF6 molecule has six
fluorine atoms placed symmetrically around a sulphur atom. The symmetrical arrange-

ment leads to extreme stability of the molecule which is very desirable characteristic for
an insulating gas. In recent years, SF6 has been widely discussed in the environmen-
tal arena. It is a very potent greenhouse gas having an atmospheric lifetime of around
3,200 years, and a global warming potential (100-year horizon) of 23,900 times that of

CO2 [2].

Study of fragmentation of SF6 is very important in connection with the early stages
of high voltage break down in insulators using SF6 gas and in connection with etching
procedures [3]. Fragmentation of multiply ionized SF6 induced by a charged particle
has been studied in the past, with the main focus being on establishment of the dis-

158
Dissociation dynamics of sulphur hexafluoride (SF6 ) 159

sociative ionization channels [1, 4, 5]. Shanker et al [4] studied partial dissociative
ionization of SF6 by electron impact using an ejected electron-ion coincidence tech-

nique. In a recent work photofragmentation dynamics of triply ionized SF6 [6] have
been investigated. Masuoka et al studied dissociative and double photoionization of
SF6 in the 75 – 125 eV region and observed that with increasing photon energy inten-
sity of SF+ +
5 peak decreases while that of F increases [8]. Also in the TOF spectrum

of SF6 alternation of intensity of SF+ 2+

n and SFn peaks with even and odd n values have

been observed [6, 7, 8]. Al–Nasir et al [9] reported the presence of SF2+
5 in the electron

induced fragmentation of SF6 . But it was not found by Lange et al [1] and Kumar et
al [6] in their studies due to its short life time. Eland et al. have observed metastable
SF2+
4 having life times of tens of ns to several µs in their studies [10]. Intramolecular

bond rearrangement in SF6 has also been observed by many groups which shows the
formation of F+ 2+ 3+
2 ion [1, 6]. Break up of SF6 and SF6 into various channels have

been investigated using momentum correlation between the fragment ions [1, 6, 11]. It
has been observed that dissociation of SF2+ + +
6 into F +SF4 +F proceeds via initial charge

separation, in the case of electron impact ionization of SF6 [11] but the same channel
when induced by highly charged ions, results from deferred charge separation [1]. It is
due to the fact that high energy ions do not contribute much to inner-shell ionization of
molecules, whereas, electrons of sufficient energy can penetrate deep into the molecule
causing inner-shell ionization. In another study a large number of super excited states

of SF6 molecule have been observed which decompose to fragment ions with excess ki-
netic energies. This is because of the overlapping of super excited states with electronic
states of SF+
6 ion [12].

In the present chapter, we will study dissociation dynamics of SF6 molecules in

order to understand the influence of high symmetry on the formation and dissociation
of multiply charged SF6 molecular ions. The experiment was carried out using high
energy highly charged ion beam from the Pelletron accelerator of Inter university Ac-
celerator Centre, in collaboration with Physical Research Laboratory, Ahmadabad and
Dissociation dynamics of sulphur hexafluoride (SF6 ) 160

Indian Institute of Technology Madras, Chennai. By analyzing the time–of–flight and

coincidence spectra of dissociated SF6 molecular ions we found various dissociation

channels of SF2+ 3+
6 and SF6 leading to double– and triple– ion coincidences. We have

also discussed the coincidence of three F+ ions ejected from dissociating SF3+
6 and

based on the analysis of the correlated momentum vectors of the three F+ ions, we have
suggested a method of determining the sequence of fragmentation and geometry of the
precursor molecular ion.

6.2 Experimental details

The experiment was performed using ion beams from the Pelletron accelerator at the
Inter University Accelerator Centre. C5+ ion beam of energy 5 MeV u−1 was used as
projectile and an ion current of a few pA was maintained during the experiments to
keep the false coincidences at an acceptable rate. The beam spot size at the target was

2 mm2. The vacuum in the main chamber was 3 x 10−8 mbar without the target molec-
ular gas and around 6 x 10−6 mbar with the target loaded. The molecular beam was
generated by effusion of SF6 through a fine stainless steel capillary with 0.15 mm inner
diameter. The stagnation pressure behind the capillary was approximately 0.3 mbar,

and the delivery lines were maintained at room temperature. The target number density
was approximately 1012 cm−3 , ensuring single collision conditions and the mean free
path of the molecules was about 2 mm, which is over ten times the capillary diameter.
Under the above circumstances clustering of molecules was unlikely.

For guiding the ions and electrons generated during collision between target mole-
cules (SF6 ) and projectile (C5+ ), an electric field is applied to a set of parallel rings in a

potential divider arrangement. To ensure field uniformity all rings at the ends have wire
meshes on them. Ions travel through this field applied over a length of 99 mm followed
by a field free drift region of length 198 mm, before hitting the detector. Electrons are
guided in the opposite direction to another detector by the same field. Time of flight
Dissociation dynamics of sulphur hexafluoride (SF6 ) 161

of ions is determined by an ejected electron–ion coincidence technique and recorded

by a multi–stop time–to–digital converter. The timing and position data were written

event–by–event into a list mode file. The event rate during data acquisition was 200 Hz
which in the absence of target gas reduced to about 0.1 Hz. The electron(s) ejected in
the fragmentation process were used for triggering the flight time measurement. But
no momentum information about the electron has been extracted in the present experi-
ment since the focus of this study is on the fragmentation dynamics of the intermediate

multiply charged molecular ion.

6.3 Time–of–flight spectrum of SF6

TOF of second fragment (0.5 ns / channel)

×103
100

90

80 +
SF5
70

60
F+
50
SF2+
4 +
40 + + SF3
S SF
30 SF2+
2
2+
20 SF SF2+
S2+ N+2
3
+ SF4
+
F2+ 3+ SF2
10
S
0
6000 8000 10000 12000 14000 16000 18000 20000
TOF of first fragment (0.5 ns / channel)

Figure 6.1: Time-of-flight spectrum of dissociative SF6 under the impact of 5 MeV u−1
C5+ ion.

The time-of-flight (TOF) spectrum of dissociated SF6 induced by 5 MeV u−1 C5+

ion is depicted in Figure 6.1. It is similar to the observed spectra obtained in 2 MeV
He2+ [1], photon [6] and electron [7] induced dissociation of SF6 . No parent molecular
ion has been observed due to their extreme instability and also because bound state of
SF2+ q+
6 does not exist [11]. In the TOF spectrum we observed SFn (where n = 1 to 5 and q
Dissociation dynamics of sulphur hexafluoride (SF6 ) 162

= 1, 2) molecular ions along with singly and doubly charged fluorine and sulphur atoms.
There is a clear indication of periodic alternation of relative intensities of SF+ 2+
n and SFn

ions with even and odd values of n. The SF+

n peaks corresponding to odd n values are

more intense than the ions having even number of F atoms attached to S+ and vice versa
for SF2+
n . The same behavior has been observed in the photon [6] and electron [7, 11]

induced fragmentation of SF6 . The possible explanation, based on valence bond model,
has been proposed by Hitchcock and van der Wiel [7]. According to them the SF+
n

and SF2+
n fragment ions are thought to be constructed from singly and doubly charged

sulphur ions having valence electron configuration 3s2 3p3 and 3s2 3p2 respectively.
In the absence of hybridization, three fluorine atoms can be associated with S+ and
two to S2+ , thus explaining the formation and stability of SF+ 2+
3 and SF2 ions. The S
+

ion in its hybridized configuration, sp3 d, accommodates five fluorine atoms whereas
S2+ favors addition of four F atoms in the sp3 hybridized orbitals, thus resulting in
moderately stable SF+ 2+ 2+
5 and SF4 ions respectively. We have not seen any SF5 peak in

our experiment, though Al–Nasir et al. reported the presence of it following electron

impact on SF6 [9]. Lange et al. challenged the above observation of Al–Nasir et al. [9]
and claimed that it cannot be true because SF2+
5 can never be stable for microseconds

[1].

6.4 Coincidence spectra of SF6

The interparticle correlations between various fragment ions can be estimated from
the coincidence spectra of the target molecule. Figure 6.2 shows two 2D coincidence
maps showing correlations between TOF of the first fragment vs. second (Fig. 6.2(a))

and between the second fragment vs. third (Fig. 6.2(b)), from the ionization of SF6
molecules by 5 MeV u−1 C5+ ions. The main islands in Figure 6.2(a) are due to the
coincident detection of F+ :SF+ + 2+
n (n = 0–5) and F :SFn (n = 0–3) ions. But no coin-

cidence of F+ :SF2+ + 2+
5 and F :SF4 has been found in the present experiment. We have
Dissociation dynamics of sulphur hexafluoride (SF6 ) 163

F2+ S2+ F+ F+2

250
10000 SF+5 9000
50
TOF of second fragment (ns) SF+4 F+:F+:SF+3

TOF of third fragment (ns)

200
8000
SF+3 F+:F+:SF+2 40
8000
SF+2 150 7000

F+:F+:SF+ 30
SF+ 6000
100
6000 SF2+
3 20
SF2+
2 F+:F+:S+
5000
S+ 50
SF2+ F+:F+:F+ 10
4000
4000
F+
2000 3000 4000 5000 6000 2000 2500 3000 3500 4000 4500 5000 5500 6000
TOF of first fragment (ns) TOF of second fragment (ns)
(a) (b)

Figure 6.2: (a)Double and (b) triple (when the first hit is F+ ) ion coincidence map of
dissociative SF6 showing interparticle correlation.

also observed correlation between F+ +

2 and SF3 in Figure 6.2(a) which is the signature

of intramolecular bond rearrangement in SF2+

6 . Coincidence of F
2+
with S+ and F+ are

also prominent in the coincidence spectrum of SF6 .

Besides double ion coincidences, we have identified and separated five triple ion
coincidence channels, shown in Figure 6.2(b), for which two F+ ions have been detected

as first as well as second hits. The F+ :F+ :SF+

4 ion triplet has not been observed in any

experiment ruling out the possibility of complete triple ion fragmentation of SF3+
6 .

While in the case of fragmentation of triatomic molecular ions a prescription for

determining the fragmentation sequence can usually be worked out, no straightforward

prescription exists in the case of polyatomic molecules [10]. With currently available
techniques, it is not possible to determine the kinematic parameters of all fragments
from many-body dissociation of a molecule. It is primarily because the neutral frag-

ments remain undetected by the charged particle detector and the efficiency of record-
ing a multi-fold coincidence falls with the order of the coincidence. A reduction in effi-
ciency occurs due to the finite dead time of the detector and associated electronics. The
coincidence efficiency generally gets worse for very energetic events in which one or
Dissociation dynamics of sulphur hexafluoride (SF6 ) 164

more fragments may be lost from the system because the spectrometer extraction volt-
ages are inadequate for directing them to the detector, especially when the direction of

their initial velocities is perpendicular to extraction field direction. Thus, our inferences
on fragmentation sequence and on the role of the neutral fragments in the dissociation
processes are primarily based on the momentum correlation between various fragment
ions.

In the following sections, we will present the analysis of double and triple ion
coincidences resulting from the dissociation of SF2+ 3+
6 and SF6 . From the prescription of

Eland [13] and from momentum correlation diagrams we will try to resolve the possible
dissociation mechanism behind each of the coincidence islands observed in Figure 6.2.

6.5 Dissociation of SF2+

6

6.5.1 Two–body dissociation

400
350
300 35
300
counts (arbitrary unit)

200 30
SF+5
Momentum (a. u.)

250
100 25
+
F 200
0 20
150
-100
15
100
-200
10

-300 50
5
-400 0
-400 -300 -200 -100 0 100 200 300 400 0 5 10 15 20 25
Momentum (a. u.) KER (eV)

(a) (b)

Figure 6.3: (a) Correlated momentum map of F+ :SF+ 5 channel arising from dissociation
of SF2+
6 . In this distribution the direction of F+
is taken as the reference and the compo-
+
nents of momentum of SF5 ion parallel and perpendicular to F+ momentum are plotted.
(b) Kinetic energy release distribution for SF2+ + +
6 → F + SF5 dissociation channel.

The only two–body dissociation channel of SF2+

6 , found in the present experiment
Dissociation dynamics of sulphur hexafluoride (SF6 ) 165

is:
SF62+ → F + + SF5+ (6.1)

where the two fragment ions are ejected with equal and opposite momenta giving rise
to a bar–shaped coincidence island of slope -1 (Fig. 6.2(a)). This is also reflected in
the momentum correlation diagram where longitudinal and transverse components of

momentum of SF+ +
5 has been plotted with respect to the momentum of F ion (shown

in Figure 6.3(a)). A maximum energy of 20 eV is released during the break up of SF2+

6

into F+ +SF+
5 channel as observed from the kinetic energy release (KER) distribution

(Fig. 6.3(b)). The most probable KER is around 5.5 eV and is nearly the same as
obtained by Lange et al. [1].

6.5.2 Three–body dissociation

From the two–dimensional coincidence map, shown in Figure 6.2(a), we have separated

two three–body dissociation channels of SF2+

6 .

400 400
18
14
300 300
16
12
200
SF+4 200
SF+3
Momentum (a. u.)

Momentum (a. u.)

14
100 10 100
+ 12
F F+2
0 0
8 10

-100 F -100 8
6 F
-200 -200 6
4
-300 -300 4

-400 2 -400 2
-400 -300 -200 -100 0 100 200 300 400 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.)

(a) (b)

Figure 6.4: Correlated momentum map of (a) F+ +SF+ + +

4 +F and (b) F2 +SF3 +F channels
2+
arising from dissociation of SF6 . In these distributions the direction of F+ and F+ 2
are taken as the references and the components of momentum of SF+ n ions parallel
and perpendicular to F+ and F+ 2 momentum are plotted. Momentum distribution of the
undetected neutral F has also been plotted in the lower half of the above two histograms.
Dissociation dynamics of sulphur hexafluoride (SF6 ) 166

A bar–shaped coincidence island, observed just below the F+ :SF+

5 coincidence,

corresponds to F+ :SF+
4 ion pair (Fig. 6.2(a)). The slope of the island is -1 and the

intensity is extremely low. The narrow width suggests a simple two–body break up of
SF2+
5 with no momentum carried away by the third fragment, a neutral F in this case.

The momentum distribution of SF+

4 is found to be anticorrelated to the momentum of

F+ ion (Fig. 6.4(a)). We have also determined the momentum of the neutral F fragment,
as negative of sum of momentum of F+ and SF+
4 ions and plotted it in the lower half

of Figure 6.4(a). It is clear from this figure that momentum of neutral F is almost zero
with respect to F+ and SF+
4 . Thus the possible fragmentation scheme is deferred charge

separation or S(d). The F+ :SF+

4 coincidence island, formed in the 2 MeV He
2+
[1]
induced dissociation of SF6 , also results from a sequential S(d)–type decay but the same

produced in 1300 eV electron [11] impact experiment proceeds via S(i) mechanism.

We have observed another three–body dissociation channel in the coincidence map

of SF6 (Fig. 6.2(a)) in which, due to intramolecular bond rearrangement, an F+
2 ion has

been formed. The slope of the F+ +

2 :SF3 coincidence island is -1 and it is very narrow in

shape. The third fragment F is neutral in this case also and possess no momentum as
observed from Figure 6.4(b). The F+ +
2 and SF3 ions share equal and opposite momenta

suggesting a two–body break up of SF2+

5 . We may assume that the F–F association

takes place within SF2+

5 in the intermediate step of the reaction. According to Lange

et al., the F+ + 2+
2 :SF3 island may form by a process via SF5 or an almost instantaneous

three–body decay, including perhaps a very short–living F+ +

3 or SF4 intermediate [1].

The above discussions on F+ +SF+ + +

4 +F and F2 +SF3 +F channels lead us to the con-

clusion that due to extreme instability, a zero energy neutral F fragment is emitted from
SF2+ 2+
6 as soon as it is formed. The resulting multiply charged complex, SF5 , later dis-

sociates in a two-body manner giving rise to two coincident ion pairs: F+ :SF+
4 and

F+ +
2 :SF3 . Thus the possible fragmentation scheme behind these two coincidence islands
Dissociation dynamics of sulphur hexafluoride (SF6 ) 167

are as follows:

SF62+ → SF52+ + F (6.2)

ց
F + + SF4+ (6.3)
ց
F2+ + SF3+ (6.4)

6.5.3 Many–body dissociation

The coincidence islands corresponding to F+ :SF+

n (where n = 0–3) ion pairs, result from

many-body dissociation of SF6 in which at least two F fragments remain undetected

(Fig. 6.2(a)). The islands are elliptical in shapes and the slopes are nearly -1 except
for F+ :S+ , in which, the slope is almost zero. The spread of counts in the F+ :SF+
n

coincidence islands and subsequently the island width increases with the loss of more

number of F atoms from the parent molecular ion.

Eland et al. have observed metastable states of SF2+

4 having lifetime of tens of ns to

several µs [10]. A weak tail has also been found by Lange et al. at the right bottom end
of F+ :SF+ 2+
3 coincidence peak directed towards the position of SF4 self coincidences

[1]. The signature of metastable states of SF2+

4 is also observed in our experiment but

not clearly visible in Figure 6.2(a) due to strong background removal.

It has been suggested in all previous experiments [1, 10, 11] that F+ :SF+
3 coinci-

dence island is formed due to deferred charge separation from SF2+

6 through

SF62+ → SF42+ + 2F (6.5)

ց
F + + SF3+ (6.6)

Removal of one neutral F from SF+ + +

3 results in F :SF2 coincidence island as observed

in ion [1] and electron [11] impact dissociation of SF6 . Bapat et al. observed that
emission of an F2 fragment (and not 2F) from SF+ +
3 leaves behind an SF ion detected
Dissociation dynamics of sulphur hexafluoride (SF6 ) 168

in coincidence with F+ in the electron induced fragmentation of SF6 [11]. They also
observed that the momentum distributions of the two F+ ions corresponding to F+ :SF+
2

and F+ :SF+ +
3 coincidences are similar. But the distribution of momentum of SF2 is

narrower than the distribution of SF+ [11] ion, both detected in coincidence with F+
ion. Thus the possible fragmentation sequence for F+ :SF+
n (where n = 1–3) coincidence

islands, as prescribed by Bapat et al., is given by:

SF42+ → F + + SF3+ (6.7)

ց
SF2+ + F (6.8)
ց
SF + + F2 (6.9)

If we assume such a process, momentum distribution of all F+ ions detected in co-

+
incidence with SF+
n (where n = 1–3) would be same and SF3 would have a narrow

momentum distribution compared to SF+ +

2 and SF . However Lange et al. suggested a

different fragmentation pathway for F+ :SF+ coincidence, given by:

SF62+ → SF32+ + 3F (6.10)

ց
F + + SF2+ (6.11)
ց
SF + + F (6.12)

with both intermediates being sufficiently long living to prevent any correlation of mo-
menta [1].

Momentum distributions of F+ and SF+ + +

n ions corresponding to F :SFn (n = 0–

3) coincidence islands have been shown in Figures 6.5(a) and 6.5(b) respectively for

the present experiment with 5 MeV u−1 C5+ ion as projectile. We observed that the
magnitude of momentum of F+ ion increases gradually with the increasing number of
undetected F fragments along with the width of the distributions. Considering the di-
rection of momentum of F+ ion as reference we plotted the longitudinal and transverse
Dissociation dynamics of sulphur hexafluoride (SF6 ) 169

+ +
F :S
+ +
+ + F :S
F :SF
+ +
+ + F :SF
F :SF
800
2 + +

Normalized counts (arbitrary unit)

F :SF
+ + 1000 2

Normalized counts (arbitrary unit)

F :SF
3 + +
F :SF
3

800
600

600
400

400

200
200

0
0
0 100 200 300 400
0 100 200 300 400
+
Momentum of F (a. u.) Momentum of second fragment (a. u.)

(a) (b)

Figure 6.5: Momentum distributions of (a) F+ and (b) SF+ + +

n corresponding to F :SFn
coincidence islands (where n = 0–3) observed in Figure 6.2(a).

components of momentum of SF+

n (n=0–3) ions in Figure 6.6. We also estimated the

total momentum carried away by all the undetected F fragments (negative sum of mo-
mentum of F+ and SF+
n ions in the centre of mass frame) and plotted it in the lower

half of the above momentum maps (Fig. 6.6). The distribution of momentum of both,
SF+ +
n and undetected Fs, are single lobed with respect to F indicating strong correla-

tions among them. The resultant momenta of the undetected fragments are negligible
in case of F+ :SF+ + +
3 and F :SF2 correlated ion pairs (Figures 6.6(a) and 6.6(b)). This

observation leads to the conclusion that dissociation of SF2+ + +

6 into F :SFn (where n =

2, 3) proceeds in two steps. In the first step, two or three neutral F atoms are emitted

from SF2+ 2+ 2+
6 with almost zero momentum leaving behind SF4 and SF3 molecular ions

respectively, which later dissociate in two–body manners giving rise to F+ :SF+

3 and

F+ :SF+
2 coincidence islands.

The above fragmentation sequence is not applicable for the other two dissociation

channels F+ :SF+ and F+ :S+ . In the momentum correlation map, we found that the
distribution of momentum of SF+ is quite broad and the total momentum taken away
by the four fluorine atoms is not zero (Fig. 6.6(c)). Momenta of SF+ +
2 and SF have

exactly same distributions and are broader compared to SF+

3 ion (Fig. 6.5(b)). However
Dissociation dynamics of sulphur hexafluoride (SF6 ) 170

400 350 400 100

90
300
300
300
SF+2
SF+3 80
200 200

Momentum (a. u.)

Momentum (a. u.)

250 70
100 100
60
200
0 F+ 0
F+
50
150
-100 -100 40

100 30
-200 -200
2F 20
-300 50 -300 3F
10

-400 -400
-400 -300 -200 -100 0 100 200 300 400 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.)

(a) (b)

400 400
240
220 900
300 300
S+
+ 200 800
SF
200 200
180
Momentum (a. u.)

Momentum (a. u.)

700
100 160 100 600
140
0 0 500
120
F+ F+
100 400
-100 -100
80 300
-200 60 -200
200
-300
4F 40
-300
5F
100
20
-400 -400 0
-400 -300 -200 -100 0 100 200 300 400 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.)

(c) (d)

Figure 6.6: Correlated momentum map of SF+ + + +

n relative to F ion for F :SFn (where n =
0–3) coincidence islands arising from dissociation of SF2+6 . The resultant momentum of
all the undetected neutral Fs have also been plotted in the lower half of each momentum
correlation map.

momenta of their correlated F+ ions are not same (Fig. 6.5(a)) which rules out the
possibility of formation of F+ :SF+ coincidence island via equations 6.9 and 6.12. The
non–zero momentum of all the undetected fragments suggests energetic emission of

four fluorine atoms from SF2+

6 . But we are unable to determine the actual dissociation

scheme for F+ +SF+ coincidence.

The slope of the F+ :S+ coincidence island is almost zero which indicates that
most of the momentum (kinetic energy as well) is carried away by the fluorine atoms
keeping S+ at rest. In the present experiment we found that the total momentum of the
Dissociation dynamics of sulphur hexafluoride (SF6 ) 171

undetected F fragments are very large compared to the momentum of the S+ ion (Fig.
6.6(d)). Also the distribution of momentum of S+ is single lobed with respect to F+

and indicates a strong correlation between them. Our observations lead to the similar
conclusion suggested by Lange et al. [1]. They assumed an instantaneous dissociation
of SF2+
6 into its atomic constituents where removal of two electrons from SF6 breaks all

its bonds at once. The main remaining internuclear force in that case would be Coulomb
repulsion between S+ and F+ ions. While F+ leaves the molecule immediately, S+

collides with the F atom, positioned on opposite side. The momentum ratio between
S+ and F+ , observed by the authors, suggests that either the parent molecule was bent
at the instant of ionization due to thermal vibration or the neutral F already had some
momentum when hit by the S+ ion.

6.6 Dissociation of SF3+

6

In this section we investigated the double– and triple–ion coincidence islands resulting
from the fragmentation of SF3+
6 .

6.6.1 Double–ion coincidence

Four coincidence islands corresponding to F+ :SF2+

n (where n = 0–3) have been ob-

served in the two-dimensional coincidence map of SF6 (Fig. 6.2(a)). No F2+ :SF+
5,

F+ :SF2+ + 2+
5 or F :SF4 coincidence has been found in our experiment. Since the flight

time difference between F+ and S2+ ions is very small, the S2+ :F+ and F+ :F+ coinci-
dence islands appear very close to each other. In addition to the above coincidences one
can also see coincidences of F2+ with S+ and F+ in Figure 6.2(a). The F2+ :F+ island is

very noisy, preventing one to draw any useful conclusion for it. The F2+ :S+ and S2+ :F+
coincidence islands appear almost along the TOF–line of corresponding S ion. These
two correlated ion pairs also originate from instantaneous explosion of SF3+
6 where S
2+

and S+ loose their momentum through collision with neutral fluorine atom.
Dissociation dynamics of sulphur hexafluoride (SF6 ) 172

400 18 400

SF2+
2
300 16 300 100

200 SF2+ 14 200

Momentum (a. u.)

Momentum (a. u.)

3
80
100 12 100
F+ F+
0 10 0 60

-100 8 -100
40

-200 2F 6 -200

20
-300 4 -300 3F
-400 2 -400
-400 -300 -200 -100 0 100 200 300 400 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.)

(a) (b)

400
+ 2+
F :SF
30
300 300
+
F :SF
2+

2+ 2

SF +
F :SF
2+

Normalized counts (arbitrary unit)

3

200 25
Momentum (a. u.)

100 200

F+ 20
0

15
-100
100

-200 4F 10

-300
0
5 0 100 200 300
-400 +
-400 -300 -200 -100 0 100 200 300 400 Momentum of F (a.u.)

Momentum (a. u.)

(c) (d)

Figure 6.7: (a)–(c) Correlated momentum map of SF2+ + +

n relative to F ion for F :SFn
2+
3+
(where n = 1–3) coincidence islands arising from dissociation of SF6 . The resultant
momentum of all the undetected neutral Fs have also been plotted in the lower half
of each momentum correlation map. (d) Momentum distribution of the first F+ ion
detected in coincidence with SF2+
n (where n = 1–3) ions.

The F+ :SF2+
n (where n = 1–3) coincidence islands are bar–shaped and the widths

of the islands decrease with the increasing size of the observed dication. The F+ :SF2+
3

island is the narrowest among them having slope of -0.5. The slope slightly decreases
for F+ :SF2+ +
2 and F :SF
2+
coincidence islands. We have not observed any third ionic
fragment for the above three coincidence channels. Thus the precursor of F+ :SF2+
n

(where n = 1–3) coincidences must be solely SF3+

6 , which breaks either sequentially
Dissociation dynamics of sulphur hexafluoride (SF6 ) 173

or in a concerted way. Momentum distributions of the SF2+

n ions, plotted relative to

the momentum of F+ ion, show single distinct lobes in Figure 6.7. In the lower half

of Figure 6.7 we have also shown the longitudinal and transverse components of resul-
tant momentum of the undetected F fragments. The slope and size of the coincidence
island and the zero momentum, carried away by the two neutral F atoms in F+ :SF2+
3

coincidence (Fig. 6.7(a)), indicate the following dissociation pathway

SF63+ → SF43+ + 2F (6.13)

ց
F + + SF32+ (6.14)

where F+ and SF2+

3 are found to be ejected with almost equal and opposite momenta.

We observed similar momentum distributions for F+ :SF2+

2 coincidence island (Fig.
6.7(b)), thus resulting from deferred charge separation of SF3+
6 via:

SF63+ → SF33+ + 3F (6.15)

ց
F + + SF22+ (6.16)

The observed slope and width of the F+ :SF2+ coincidence island suggest that in this

fragmentation channel at least one step is fast, leading to a certain amount of correlated
momenta of the neutral fragments. The momentum correlation map (Fig. 6.7(c)) also
shows not inconsiderable amount of momentum carried away by the four fluorine atoms
during the fragmentation process. Thus the proposed dissociation sequence behind this

coincidence island is:

SF63+ → SF33+ + 3F (6.17)

ց
F + + SF22+ (6.18)
ց
SF 2+ + F (6.19)
Dissociation dynamics of sulphur hexafluoride (SF6 ) 174

The neutral F taking away some momentum while separating from SF2+
2 causes the

distribution of momentum of SF2+ to be broad and not at 180◦ with respect to the

momentum of F+ ion. We observed that the momentum distributions of the F+ ions

for the above three dissociation channels (F+ :SF2+
n where n=1–3) are almost similar

(Fig. 6.7(d)) which is consistent with the preceding discussion. Intensity of F+ :SF2+
2

momentum distribution is the maximum compared to the other two ion-pairs which
indicates maximum stability of this ion-pair.

6.6.2 Triple–ion coincidence

Now we will discuss the triple–ion coincidence channels of SF3+

6 . We observed four

triple ion coincidence channels corresponding to F+ :F+ :SF+

n (n = 0–3) where the first

two fragments detected are two F+ ions (Fig. 6.2(b)). No F+ :F+ :SF+
4 coincidence has

been observed, ruling out the possibility of a concerted breakup of SF3+

6 into three ionic

fragments. The intensities of the F+ :F+ :SF+

n (where n = 0–3) coincidences increases

with the decreasing size of SF+

n ion, indicating higher stability of ions with fewer F

atoms. We have determined the kinematics of the above reactions from the correlated
ion momentum map. The momentum distribution of the second F+ ion has been plotted
relative to the momentum of the first F+ in the left panel of Figure 6.8 whereas mo-
mentum distributions of SF+
n ions (where n = 0–3) have been shown in the right panel

of Figure 6.8. We have also calculated the sum of the momentum vectors of the three
correlated fragment ions detected in the above coincidences. This gives the estimation
of residual momentum carried away by the undetected neutral fragments. The longitu-
dinal and transverse momentum distributions of the neutral Fs are shown in the lower

half of the right panel of Figure 6.8. From the vector correlation between the momenta
of the two F+ ions we observed that the direction of emission of the two F+ ions is pre-
served across all the dissociation channels. The two F+ ions are ejected either (nearly)
opposite or perpendicular to each other. The above correlation between the emission
Dissociation dynamics of sulphur hexafluoride (SF6 ) 175

400 7 400 8

300 6 300 SF+3 7

200 + 200 6
F

Momentum (a. u.)

Momentum (a. u.)

5

100 100 5
4
F+
0 +
0 4
F 3
-100 -100 3

2
-200 -200 F 2

-300 1 -300 1

-400 0 -400 0
-400 -300 -200 -100 0 100 200 300 400 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.)

(a) (b)

400 14 400 9

300 300 8
12 SF+2
200 200 7
F+
Momentum (a. u.)

Momentum (a. u.)

10

100 100 6
8
F+ F+
0 0 5
6
-100 -100 4

4
-200 -200 3
2F
-300 2 -300 2

-400 0 -400 1
-400 -300 -200 -100 0 100 200 300 400 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.)

(c) (d)

400 400 22

20
300 30 300
SF+ 18
200 200
F+ 25
Momentum (a. u.)

Momentum (a. u.)

16
100 100
14
+ 20 +
F F
0 0 12

15 10
-100 -100
8
-200 10 -200 3F
6
-300 -300
5 4

-400 -400 2
-400 -300 -200 -100 0 100 200 300 400 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.)

(e) (f)

400 400
100
300 300 100
S+
200 F+ 80 200
Momentum (a. u.)

Momentum (a. u.)

80

100 100

F+ 60 F+ 60
0 0

-100 40 -100 40

-200 -200
20
4F 20
-300 -300

-400 -400 0
-400 -300 -200 -100 0 100 200 300 400 -400 -300 -200 -100 0 100 200 300 400
Momentum (a. u.) Momentum (a. u.)

(g) (h)

Figure 6.8: Momentum distributions of (left panel) the second F+ , (right panel) SF+ n
and undetected Fs relative to first F+ ion for F+ :F+ :SF+
n (where n = 0–3) coincidence
islands.
Dissociation dynamics of sulphur hexafluoride (SF6 ) 176

of F+ ions will be maintained across all the channels only if they are ejected simulta-
neously. Furthermore, the magnitude of momenta of the two F+ ions are also nearly

the same in all dissociation channels (Fig. 6.9). From Figure 6.8(b) we found that the
neutral F is ejected with zero momentum for the F+ :F+ :SF+
3 coincidence. These ob-

servations lead us to the conclusion that dissociation of SF3+

6 proceeds through various

stages. In the first step one neutral F is emitted from SF3+

6 with zero kinetic energy.

The intermediate SF3+

5 thus formed is very unstable and dissociates immediately into

F+ :F+ :SF+ 3+
3 . Kumar et al. have also studied the complete dissociation sequence of SF6

formed by photoionization of SF6 [6]. Based on the scalar triple product of the ion mo-
mentum vectors, they proposed the following fragmentation scheme behind F+ :F+ :SF+
n

coincidences:

SF63+ → SF53+ + F (6.20)

ց
F + + F + + SF3+ (6.21)
ց
SF2+ + F (6.22)
ց
SF + + F (6.23)
ց
S+ + F (6.24)

In the present experiment the spread of momentum distribution of SF+

n decreases

with its decreasing size. On the other hand resultant momentum, associated with the
neutral F fragments, increases with the increasing number of undetected Fs which is
consistent with the scheme proposed by Kumar et al. [6]. Thus we may infer that
ejection of two F+ ions from SF3+ +
5 occurs violently but the loss of neutral Fs from SF3

occurs very slowly in various directions and brings SF+ +

3 back to stable S ion. The

momentum distribution of the first F+ ion corresponding to the F+ :F+ :SF+

n dissociation

channels are shown in Figure 6.9. We found that they are almost similar and supports
the above decay mechanism.
Dissociation dynamics of sulphur hexafluoride (SF6 ) 177

+ + +
F :F :S
+ + +
F :F :SF
+ + +

120 F :F :SF
2

Normalized counts (arbitrary unit)

+ + +
F :F :SF
3

100

80

60

40

20

0
0 100 200 300 400
+
Momentum of F (a. u.)

Figure 6.9: Momentum distribution of the first F+ ion corresponding to the F+ :F+ :SF+
n
coincidence islands (where n = 0–3) observed in Figure 6.2(b).

6.6.3 Coincidence between F+ :F+ :F+

400
+[3] 50
F +[1]
400 F
300 +[2]
F
40
counts (arbitrary unit)

+[2] 200
F
Momentum (a. u.)

300
100
30

+[1] 0
200 F
-100 20

-200
100
10
+[3]
-300 F
0 -400 0
0 10 20 30 40 50 -400 -300 -200 -100 0 100 200 300 400
Fragment KE (eV) Momentum (a. u.)

(a) (b)

Figure 6.10: (a) Kinetic energy distribution of three F+ ions detected in a three–fold
coincidence. Ions are labeled according to the descending order of their kinetic ener-
gies, event by event. Histogramming is done after the labeling. (b) Momentum map
of the three F+ ions in coincidence. The most energetic ion in a given event is taken
as the reference, and the momentum components of the others (less energetic) parallel
and perpendicular to it are plotted. The red histogram in the first quadrant is the scalar
momentum distribution of the first ion.

In this section we have reported the results of analysis of the triple F+ coincidence
Dissociation dynamics of sulphur hexafluoride (SF6 ) 178

resulting from SF3+

6 [14]. In Figure 6.2(b) we observed the coincidence between three

F+ ions, which is indicative of a very energetic fragmentation of the precursor. In ana-

lyzing the triple ion coincidence F+ :F+ :F+ , care has been taken to exclude the channels
in which a fourth ion hit (such as F+ or SF+
m , where m = 0–3) might be present. The

three ions in this coincidence are of same species but their kinetic energies are differ-
ent. The arrival sequence of the three F+ ions depends on the relative magnitudes of the
component of their momenta along the spectrometer axis. For the present analysis we

have labeled the three F+ ions as F+[i] (where i = 1–3) in the descending order of their
kinetic energies (Fig. 6.10(a)).

Precursor to triple F+ ejection

Figure 6.10(b) shows correlated momentum distributions of the three F+ ions in co-
incidence where the red histogram represents the momentum distribution of the most
energetic F+[1] ion. The strongest feature in this correlated momentum map is the pres-

ence of distinct lobes, which indicates a strong correlation between the momenta of the
three F+ ions. Thus ejection of three F+ ions is simultaneous, and is accompanied by
no more than one neutral (as participation of several neutrals would lead to a wide range
of momentum sharing). With this constraint, there are two possible sequences leading

to the observed triple F+ coincidence.

SF63+ → 3F + + SF3 (6.25)

SF63+ → SF33+ + 3F (6.26)

ց
3F + + S (6.27)

Since the separation of three F+ is a very energetic process, some energy will
be transferred to internal degrees of freedom of the non-ionic portion of exploding
complex. It is worthwhile here to note, that SF3 has three unpaired electrons, and is
inherently unstable [7]. Hence, the first pathway which suggests formation of an inter-
Dissociation dynamics of sulphur hexafluoride (SF6 ) 179

Figure 6.11: Two likely configurations (planar and tetrahedral) of SF3+

3 arising from the
3+
loss of three F atoms from SF6 , which is proposed to be the first step leading to the
triple F+ coincidence. The bonds shown by the dashed lines in the structure on the right
indicate that the corresponding F atoms are behind the plane of the paper.

mediate SF3 , is untenable. Instability of SF3 implies that the momentum sharing will
be between several neutral partners in addition to the ions, inhibiting strong momentum

correlation between the ionic fragments. On the other hand, in the second pathway the
neutral residue from the energetic explosion of SF3+
3 is an atomic fragment, irrespec-

tive of the manner in which neutrals are lost in the stage(s) prior to charge separation.
The neutral S will carry some momentum, but since the primary separation mechanism

is the (monopole–monopole) Coulomb repulsion between the F+ ions, which is much

stronger than the induced dipole–monopole S–F+ forces, the influence of S on the mo-
mentum correlation between the three F+ ions would be small. Thus, the first of the
two pathways appears to be unlikely and we propose that the precursor to the observed

F+ :F+ :F+ coincidence is SF3+ 3+

3 . The 3Fs are ejected in the first step and SF3 ion is

formed, which explodes, following the repulsive potential energy surface of the SF3+
3

ion. This “lifetime” could be of the order of picoseconds, which is much longer than
the collision timescales, which is tens of attoseconds.

Three F atoms can be selected in 20 ways from the SF6 molecule, i.e. 20 geometric
possibilities of triple F ejection exist. Of these, 12 possibilities involve triple F+ ejection
in a plane and 8 possibilities involve ejection in three mutually orthogonal directions.
The coplanar ejection of neutrals leaves behind SF3+
3 in a coplanar geometry, while
Dissociation dynamics of sulphur hexafluoride (SF6 ) 180

orthogonal ejection of neutrals leaves behind a tetrahedral SF3+

3 . Ejection of the three

F+ ions can thus take place from two conformations of SF3+

3 , which are shown in

Figure 6.11.

Mutual ejection angles

counts (arbitrary unit)

-1 -0.5 0 0.5 1
co-planarity parameter (α)

Figure 6.12: Distribution of the coplanarity parameter α. Peak at zero corresponds

to the coplanar ejection and the peaks close to -1.0 and 1.0 correspond to orthogonal
ejection of the three F+ ions.

In the laboratory frame of reference, the ejection of the three F+ ions may be in
any direction. However, their mutual angles will, in general, not be random, and will
be governed by the kinematics and the geometry of the precursor. The initial structure

of the precursor can be estimated by measuring the relative angles of the three ion
momentum vectors. In particular, the coplanarity, or otherwise, of these fragments can
be determined using their scalar triple product (α),

α = pˆ3 . pˆ2 × pˆ1 (6.28)

Dissociation dynamics of sulphur hexafluoride (SF6 ) 181

in which, p̂i is the unit vector along the momentum of the ith ion. For coplanar ejec-
tion, the distribution of α must be narrow with a peak at zero. Figure 6.12 shows the

distribution of the scalar triple product of the three momentum vectors of three F+ ions.
There is a clear peak at zero, implying coplanar ejection of the three F+ . There are
smaller peaks close to 1 and -1, which indicate mutually perpendicular emission. This
distribution is consistent with the picture of the precursor emerging out of the arguments
in the previous subsection. By setting appropriate conditions on the list mode data for

the triple coincidence, we can sort out events involving coplanar and orthogonal ejec-
tion, and study the dynamics of each category of ejection. We accept events as being
coplanar if |α| ≤ 0.25. Likewise the condition for orthogonal ejection is 0.75 ≤ |α|
≤ 1.0. Increasing the acceptance range for either category has only a slight effect on

the structures in the angular distributions which will be discussed next. The effect is a
gradual increase in the average counts across the angular range, without significantly
altering the number of counts in the peaks of the distribution.

Orthogonal triple F+ ejection

180

10
160 100

140
counts (arbitrary unit)

8 80
120
(degree)

100 6 60

80
13

4
χ

60 40

40
2 20
20

0 0 0
0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180
χ12 (degree) χ (degree)
23

(a) (b)

Figure 6.13: (a) Correlation map of the distribution of free angles between the three F+
ions for orthogonal ejection. (b) Distribution of the free angle between F+[2] and F+[3] .
Dissociation dynamics of sulphur hexafluoride (SF6 ) 182

The correlated angular map for orthogonal ejection of three F+ ions is shown in
Figure 6.13(a). In addition to this 2-D map, the free angle between F+[2] and F+[3] ions

(χ23 ), is needed to determine the conformation of the precursor (Fig. 6.13(b)). The
distributions of the three mutual ejection angles (χ12 , χ23 , χ13 ) peak nearly at the same
value (95◦ ). This supports the proposition, that one of the precursor conformations for
the triple F+ decay channel is a tetrahedral SF3+
3 , as shown in Figure 6.11. These angles

are the asymptotic angles between the ion momentum vectors, which are different from

the angles between the various S–F bonds in the SF3+

3 precursor. In general backward

mapping of the ion trajectories will lead to precise information about the initial bond
angles. The trajectory calculation rests on the assumption that the bond lengths in the
dissociation precursor are the same as the neutral bond lengths. In the present case,

such an assumption may be too drastic, as three F atoms are already lost. In the absence
of a reasonably well defined initial condition, a simple trajectory back calculation is
not possible here. However, the asymptotic angles will, in general, be greater than the
initial angles, due to the Coulomb repulsion, i.e., in this case the ejection is from a near

orthogonal geometry. Furthermore, we can assert, that the highest order of symmetry of
this tetrahedral SF3+
3 complex is C3v , not Td . Making further inferences about the bond

angles with this data would be speculative, and would necessitate making oversimplified
assumptions about the repulsive forces, and the vibrational couplings.

Coplanar triple F+ ejection

In contrast to the orthogonal ejection case, correlated angular distribution of the three
ions for coplanar ejection shows well separated multiple lobes (Fig. 6.14). The intensi-
ties in all lobes except the one at χ12 ∼ 180◦ , χ13 ∼ 90◦ is quite low. The most intense

lobe is entirely consistent with the tee-shaped planar geometry of the SF3+
3 precursor

shown in Figure 6.11. It corresponds to the case when the F+ ion forming the side arm
is the least energetic, while the pair of F+ ions along the long arm have greater energy.
The situation with exchange of energy ordering of the ions along the longer arm is sim-
Dissociation dynamics of sulphur hexafluoride (SF6 ) 183

180
300 o
=90 , =90
o

12 13

o o

160 25 12
=180 ,
13
=90

250 12
o
=90 ,
13
=180
o

counts (arbitrary unit)

140
20 200
120
(degree)
[+150]
100 150
15
80
100
13

[+75]
χ

60 10
50
40
5
20 0
0 30 60 90 120 150 180
0 23
(degree)
0 20 40 60 80 100 120 140 160 180
χ12 (degree)

(a) (b)

Figure 6.14: (a) Correlation map of the distribution of free angles between the three F+
ions for coplanar ejection. (b) Free angle between F+[2] and F+[3] corresponding to the
three main lobes demarcated by (150◦ ≤ χ12 ≤ 180◦ , 60◦ ≤ χ13 ≤ 120◦ ), (60◦ ≤ χ12
≤ 120◦ , 150◦ ≤ χ13 ≤ 180◦ ) and (60◦ ≤ χ12 ≤ 120◦ , 60◦ ≤ χ13 ≤ 120◦), respectively.
To improve the readability of the graph, the baseline of the upper two distributions is
shifted by the amount shown in the brackets.

ilar, and corresponds to the lobe at χ12 ∼ 90◦ , χ13 ∼ 180◦ . The distribution of the angle

χ23 , corresponding to these two fragmentation patterns, is shown in the adjacent panels
of Figure 6.14. The χ23 distributions are very similar and sharply peaked around 90◦ in
both the cases, consistent with the proposed precursor geometry.

The case when the most energetic F+ ion is the one along the side arm explains

the lobe at χ12 ∼ 90◦ , χ13 ∼ 90◦ , but the corresponding distribution of χ23 has peak at
170◦ , which can be understood as follows. When the most energetic F+ ion is along
the side arm , the other F+ ions (and/or the S atom) get repelled from the first F+ . Thus,
even if the less energetic F+ ions start off along the longer arm, the angle χ23 will not
be 180◦ , but will be reduced.

However, the small, broad peak around 20◦ , cannot be explained within this frame-
work. Likewise, there is no consistent explanation for the other less intense, but clear,
lobes in the χ12 , χ13 correlation map.
Dissociation dynamics of sulphur hexafluoride (SF6 ) 184

6.7 Summary

In the present chapter we have studied possible dissociation sequences of SFq+

6 (where

q = 2, 3) formed in the collision of neutral SF6 molecules with 5 MeV u−1 C5+ ion
beam. We observed that almost all the dissociation channels, yielding molecular frag-
ments at the end, start with the emission of one or few neutral fluorine atoms from
SFq+
6 . This reflects the extraordinarily high electronegativity of fluorine which makes

the positive charge to be strongly localized at the sulphur atom. Fragmentation of SF2+
6

is found to be either concerted or sequential in nature. Concerted dissociation leads to

F+ :SF+
5 correlated ion pair whereas sequential dissociation results into the emission of

one neutral F from SF2+ 2+ + + + +

6 followed by the break–up of SF5 into F +SF4 and F2 +SF3 .

The F+ :F+ :SF+

n coincidence islands, observed in the triple ion coincidence spectrum

of SF6 , are found to be originated from dissociation of SF3+

6 where emission of one

neutral F from the parent molecular ion is followed by the violent ejection of two F+
ions from SF3+ +
5 . Slow separation of neutral fluorine atoms from SF3 in subsequent

stages brings it back to stable S+ ion. We have also investigated in detail the ejection

pattern of three F+ ions from the explosion of SF3+

6 . Based on the strong momentum

correlation between the three F+ ions, the explosion scheme has been proposed to oc-
cur in two steps: separation of neutrals (3F atoms), followed by Coulomb explosion
of the residual (SF3+
3 ). Geometrical and combinatorial arguments and supporting data

on angular correlation between the three F+ ions suggests that there are two dominant
precursor geometries for SF3+
3 : tetrahedral and planar (tee–shaped). It is not possible

under the present scheme to give a robust estimate of the bond angles. The prominent
features in the molecular frame angular distributions of the F+ ions are well explained

with the proposed geometric conformation of the precursor. A few less prominent fea-
tures remained unexplained. Whether they can be ascribed to a different fragmentation
sequence or a different precursor state is a matter of speculation at this juncture.
Bibliography

[1] M. Lange, O. Pfaff, U. Müller and R. Brenn, Chemical Physics 230, 117
(1998)

[2] http://www.epa.gov/climatechange/Downloads/ghgemissions/US-GHG-

Inventory-2012-Annex-6-Additional-Information.pdf

[3] J. J. Corr, M. A. Khakoo and J. W. McConkey, J. Phys. B: At. Mol. Phys. 20,
2597 (1987)

[4] R. K. Singh, R. Hippler and R. Shanker, Phys. Rev. A 67, 022704 (2003)

[5] A. V. Snegursky, F. F. Chipev, A. N. Zavilopulo and O. B. Shpenik, Radiat.

Phys. Chem. 76, 604 (2007)

[6] S. Sunil Kumar, P. C. Deshmukh, R. K. Kushawaha, V. Sharma, I. A. Prajap-

ati, K. P. Subramanian and B. Bapat, Phys. Rev. A 78, 062706 (2008)

[7] A. P. Hitchcock and M. J. van der Wiel, J. Phys. B: At. Mol. Phys. 12, 2153
(1979)

[8] T. Masuoka and J. A. R. Samson, J. Chem. Phys. 75, 4946 (1981)

[9] A. H. Al–Nasir, M. A. Chaudhry, A. J. Duncan, R. Hippler, D. M. Campbell

and H. Kleinpoppen, J. Phys. B: At. Mol. Opt. Phys. 29, 1849 (1996)

[10] J. H. D. Eland, Laser. Chem. 11, 259 (1991)

185
Dissociation dynamics of sulphur hexafluoride (SF6 ) 186

[11] B. Bapat, V. Sharma and S. V. K. Kumar, Phys. Rev. A 78, 042503 (2008)

[12] M. B. Miletić, D. D. Goloboćanin, K. F. Zmbov, Bulletin of the Chemists and

Technologists of Macedonia, 23, 139 (2004)

[13] J. H. D. Eland, Mol. Phys. 61, 725 (1987)

[14] R. K. Kushawaha, S. Sunil Kumar, M. R. Jana, I. A. Prajapati, C. P. Safvan

and B. Bapat, J. Phys. B: At. Mol. Opt. Phys. 43, 205204 (2010).
Chapter 7

Summary and Future outlook

In this thesis, we have presented a detailed experimental study of fragmentation dy-

namics of two linear triatomic molecules: carbon dioxide (CO2 ) and carbonyl sulphide

(OCS) along with non–linear polyatomic molecule, sulphur hexafluoride (SF6 ). The
experiments were conducted with low as well as high energy projectile ion beams pro-
vided by the Low Energy Ion Beam Facility (LEIBF) and Pelletron of IUAC, New
Delhi. When projectile ions collide with the neutral molecules several electrons are

stripped of and multiply charged transient molecular ions are formed in the excited
states of the molecule at the equilibrium inter-nuclear distance (Re ). The molecular
ions, thus formed, are usually very unstable and dissociate instantly into copious quan-
tities of charged energetic fragments. The interaction time, “tuned” by the energy of the

ion beam, governs the dissociation mechanisms of molecules under investigation.

In order to probe the dynamics and kinetics of ion impact dissociation of molecu-
les, one needs a dedicated and versatile setup capable of simultaneous measurement of
all multiple-ionization, fragmentation and capture channels. Detection of various frag-

ments combined with complete kinematic information is extremely useful for studying
unstable polyatomic molecular ions. In order to have kinematically complete set of ex-
perimental data we have used a setup that consists of mainly a Time–Of–Flight–Mass–
Spectrometer (TOFMS) equipped with a position sensitive multi–coincidence detection

187
Summary and Future outlook 188

system. Different components of the setup and the relevant electronics for data acqui-
sition are described in detail with their working principles. The important challenge of

molecular dissociation studies lies in the optimization of the TOFMS. When molecular
ions dissociate there are two opposing constraints placed on the experimental condi-
tions namely either (a) angle and energy resolved measurements or (b) determination
of branching ratios, which decides what extraction fields to be used for the TOFMS.
In the former case small extraction voltages are required to ensure good angle and en-

ergy resolution. For the latter case large extraction fields have to be applied to ensure
reasonable collection efficiencies for energetic fragments. Thus an optimum value of
extraction voltage is necessary for the TOFMS. To obtain proper position and TOF
spectra, we tuned the ion beam for proper overlap with the target molecular gas jet. We

also introduced a compensating plate at the entrance of the ion beam to locate the over-
lap volume correctly with respect to the spectrometer axis. In order to test the efficiency
of the spectrometer, we have recorded the collision–induced ionic spectra of Ar and N2
molecules by the impact of various energy ion beam and found that the extraction field

of 100–200 V cm−1 can be a good choice for our spectrometer. The methodology used
for data acquisition and analysis is explained in detail with the help of flow charts and
raw spectra. The measured positions and TOF of fragment ions are converted into the
corresponding momentum vectors from which we can determine the geometry of the
precursor molecular ion and the kinetic energy released in the dissociation process.

On studying the dissociation dynamics of CO2 by 1.2 MeV Ar8+ ion impact, we
came across various fragmentation pathways of multiply charged CO2 molecular ions.
From the geometrical properties and measured slopes of the coincidence islands, we
deduced whether the dissociation process is concerted or sequential in nature. While

the slope is exactly -q1 /q2 for two–body events due to equal distribution of momentum,
it is quite indefinite for three–body events due to the unaccounted momenta carried
away by the undetected fragments. In the present experiment we observed formation
and dissociation of COq+
2 (q = 1 – 4) molecular ions via various channels, with sev-
Summary and Future outlook 189

eral fragment combinations being possible: (i) CO+ :O+ (ii) C+ :Oq+ :O (where q = 1
– 3), (iii) C2+ :Oq+ :O (where q = 1, 2), (iv) C:Om+ :On+ (where m, n = 1, 2) and (v)

C+ :O+ :O+ . By analyzing the data we conclude that dissociation of COq+

2 (where q =

2–4) proceeds mainly via concerted manner. The sequential dissociation has only been
observed in CO2+ + +
2 , which dissociates into CO :O in the first step and fragmentation

of CO+ into C+ +O takes place in the next step. Most of the kinetic energies, released
in the above dissociation channels, are carried away by the terminal O fragments. For

all channels, except C+ +O+ +O and C2+ +O+ +O, the measured KERs are higher than
the KER values predicted by the Coulomb explosion model (CEM).

In the above experiment we have not utilized the imaging (position encoding)
properties of our MCP–based position sensitive delay line detector to its full poten-

tial. Therefore we performed another experiment where triple–ion fragmentation of

CO3+
2 , formed in the fast, heavy-ion-induced ionization of CO2 , using 5 MeV u
−1
Si12+
ions as projectiles, has been studied in detail by the technique of multi–ion coincidence
momentum imaging. This allows the measurement of momentum even in the case of
undetected neutral fragments. From momentum correlation between ejected ions, we

conclude that dissociation of CO3+ + + +

2 into C +O +O is a concerted process involving

linear as well as bent conformations of CO3+

2 . The KER spectrum of this particu-

lar channel has multiple peaks. The range of measured KER values is well explained
from our ab initio quantum chemical calculations, though more detailed calculations

are clearly called for to reproduce the entire KER distribution. The range of angles
between the momentum vectors of the two O+ ions is greater than that governed by
vibrational modes of the linear state, implying participation of the bent electronic states
in dissociation of CO3+
2 . The contribution from bent electronic states is evident from

the small peak around 135◦ and a pronounced tail below 140◦ in the O+ :O+ angular
distribution. Comparison of the break-up pattern of CO3+
2 induced by various pertur-

bations shows differences: the swift perturbations almost exclusively lead to concerted
triple-ion break-up, but significant probability for stepwise break–up is seen with slow
Summary and Future outlook 190

perturbations.

Most of the ion impact dissociation studies, done till now, are on symmetric tri-
atomic molecules like CO2 and CS2 . For this thesis we have investigated the dissoci-
ation dynamics of asymmetric, triatomic molecule OCS by 150 keV Ar+ ion impact.
From the coincidence map we got the evidence of formation and fragmentation of OCS

molecular ions having charge states up to 4. OCS2+ is found to have several dissociation
channels. The two–body dissociation of OCS2+ leading to CO+ +S+ is more favorable
than the O+ +CS+ channel in this experiment. The KER distributions for these two
channels have sharp peaks at 7.8 eV and 9.8 eV respectively. The only two–body dis-

sociation channel of OCS3+ , found in the present experiment, is CO+ +S2+ where the
most probable KER is 13 eV. The sequential dissociation of OCS2+ and OCS3+ has
been observed in O+C+ +S+ , O+ +C+ +S, O+C2+ +S+ and O2+ +C+ +S channels. For
O+ +C+S+ , O+ +C+S2+ , O2+ +C+S+ and C+ +O+ +S+ channels, the oxygen and sulphur

fragments are emitted with momenta making an angle of about 158◦ in the molecular
frame. This indicates simultaneous break–up of O–C and C–S bonds (concerted disso-
ciation) of OCS2+ and OCS3+ having bent geometries. The geometrical contribution
observed is that the central C fragment also has a momentum, as expected, forming
the bond angle. The KER distributions for all the above dissociation channels have

been evaluated from the experimental data and compared with those calculated from
the CE model. It is found that CEM underestimates the experimental KER values due
to its oversimplified point–charge assumption. The measured kinetic energy release
distribution for O+ +C+S+ channel shows a single broad peak at 13 eV. For the three

concerted fragmentation channels of OCS3+ , the most probable KER is same and equal
to 23 eV. We have also found the evidence of formation of OCS4+ which dissociates
into O2+ +C+S2+ , C2+ +S2+ +O and O2+ +C+ +S+ via concerted manner. But due to poor
statistics we could not estimate the bond angles and kinetic energy releases in the above

dissociation channels of OCS4+ .

Keeping the projectile velocity fixed at 1 a.u. we measured the mean kinetic en-
Summary and Future outlook 191

ergy of a fragment ion from dissociative OCS and found that it decreases gradually with
the increasing charge state of the projectile. Thus significant projectile charge state de-

pendence of the fragment ion kinetic energy distributions have been observed in the
dissociation of OCS molecular ions which can be a consequence of formation of multi-
ply charged molecular ion in lower excited states or dissociation of the same parent ion
into highly excited fragment ions.

CO2 is a symmetric molecule having a centre of symmetry through the central

C atom but being heteronuclear, OCS does not have any centre of symmetry. In the
present thesis we have compared the results of ion induced fragmentation of OCS with
CO2 in order to know the effect of replacing one oxygen atom by one sulphur atom.
The experiment has been performed with high energy highly charged ion beam (5 MeV

u−1 Si12+ ) from Pelletron. The number of fragmentation channels observed is more
in case of OCS fragmentation in view of its heteronuclear basis. All the dissociation
channels of doubly and triply ionized CO2 and OCS molecular ions have been analyzed
and we conclude that, for the present experiment, three–body dissociation of OCX2+
proceeds via concerted as well as sequential manner but dissociation of OCX3+ always

takes place in a concerted way (where X = O for CO2 and S for OCS). The maximum
kinetic energy release is same for similar three–body dissociation channels of CO2 and
OCS but the most probable KER is same only for C2+ +X+ +O and C+ +O2+ +X chan-
nels. For others, the most probable KER is generally lower for dissociation channels

of OCS. In triple ion fragmentation of OCS3+ the kinetic energy carried away by the
O+ ion is much larger than the kinetic energy associated with the S+ ion. This is in
contrast with the case of CO3+ +
2 where the two O ions carry equal amount of energies.

The KER distribution of CO3+ + + +

2 → C +O +O has four peaks at 20.7, 26.8, 30.8 and

42.8 eV; out of which only 26.8 and 42.8 eV peaks are present in the KER distribution
of OCS3+ → C+ +O+ +S+ along with two other peaks (22.8 eV and 33 eV). The angular
distribution between the momenta of O+ and S+ ions show a small peak around 135◦
which indicates participation of bent excited state of OCS3+ in the triple ion break–up.
Summary and Future outlook 192

The same angle has also been observed in case of CO3+

2 . However the most probable

angle is not same for both of them. The asymptotic angle between the momentum vec-

tors of two peripheral ions are same for concerted fragmentation channels of CO2+
2 and

OCS2+ leading to C+O+ +O+ and C+O+ +S+ respectively. But, except for C+O2+ +X+
and C+O+ +X2+ channels, the above angle is always smaller for dissociation of OCS3+
compared to CO3+
2 .

After studying the dissociation dynamics of two linear triatomic molecules, one

with a centre of symmetry (CO2 ) and the other with no symmetry (OCS), we decided
to study the fragmentation dynamics of a non–linear polyatomic molecule having high
symmetry. The third molecule of our interest is SF6 , the smallest molecule of Oh sym-
metry. We have investigated the possible dissociation sequences of SF2+ 3+
6 and SF6

formed in the ion impact ionization of SF6 by 5 MeV u−1 C5+ ion beam. We observed
complete two–body dissociation of SF2+ + +
6 into F +SF5 channel. The KER distribution

of this channel has a single sharp peak at 5.5 eV. Three–body dissociation of SF2+
6

proceeds in two steps. In the first step one neutral F is emitted from SF2+
6 with zero

momentum, followed by the break–up of SF2+ + + + + + +

5 into F +SF4 and F2 +SF3 . The F2 :SF3

coincidence island is of particular interest in the present investigation because formation

of F+
2 ion requires bond–association process during the fragmentation of its precursor.

SF3+ 3+
6 dissociates in a sequential manner. In the first step an F atom is ejected and SF5

is formed. This is followed by the violent ejection of two F+ ions either opposite or

perpendicular to each other. Slow separation of neutral F atoms brings the residual SF+
3

to the stable S+ ion, and at each F atom loss there is an increasing likelihood of the SF+
n

(n = 0...2) ion being stable. We have also investigated the ejection pattern of the three
F+ ions from many–body explosion of SF3+
6 . The proposed explosion scheme in this

case is: separation of three neutrals F atoms, followed by Coulomb explosion of the
residual SF3+ +
3 . The angular correlation between the three F ions suggests two domi-

nant precursor geometries for SF3+

3 : tetrahedral and planar (tee–shaped). However we

failed to give an estimation of the bond angle in SF3+

3 at this point.
Summary and Future outlook 193

The ab initio calculation, reported in this thesis, evolves around only one dissocia-
tion channel of CO3+
2 . Detailed multi–dimensional calculations for CO2 , OCS and SF6

will explore all possible molecular states of these species and their dissociation limits.
The experimental setup used in the present experiments, can also be used to address
many of the open questions in the field of ion-molecule collisions. The field is open to
application of novel techniques and methodologies for the study of atoms and molecu-
les. The fundamental question of electron transfer can be addressed for a wide range of

projectile energies by setting up coincidences with the post collision projectile ions. In
future, we would like to investigate the effect of electron capture by the projectile ion
on the dissociation channels of the target molecule. The provision of energy analysis
of the electrons emitted during the interaction can also be added to get an insight to the

origin of these electrons. The dynamics of ion–molecule collisions can be studied to

understand the role of rearrangement processes in the breakup of polyatomic molecular
ions, the significance of initial alignment on the dissociation of molecules and the pos-
sibilities of metastability of diatomic and polyatomic molecular ions. We are interested

in studying the bond rearrangement phenomena in alcohols and aldehydes by the use
of low and as well as high energy highly charged ion impact. In collision experiments,
usually randomly oriented molecular targets are perturbed by projectiles. When the pro-
jectile is highly charged ion, the perturbation is very strong and can change the entire
structure of the target molecule. Our interest is to see whether ionization cross-section

depends on the orientation of the target molecules with respect to the direction and
charge state of the projectile ion beam. For this, coincident detection of both, fragment
ions and post collision projectiles, is required.