University of Southern California Dissertations and Theses

Spectroscopy and photodissociation dynamics of hydroxyethyl radicals

(USC Thesis Other)

Spectroscopy and photodissociation dynamics of hydroxyethyl radicals

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content SPECTROSCOPY AND PHOTODISSOCIATION DYNAMICS OF
HYDROXYETHYL RADICALS

by

Laura Wyman Edwards

_______________________________________________________________________

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)

August 2010

Copyright 2010 Laura Wyman Edwards

ii
Table of Contents

List of Tables iv

List of Figures v

Abstract vii

Chapter 1: Introduction
1.1 Overview 1
1.2 1-Hydroxyethyl Radical (CH
3
CHOH) 5
1.3 2-Hydroxyethyl Radical (CH
2
CH
2
OH) Dissociation 7
Pathways
1.4 Theoretical Models 9
1.5 References 11

Chapter 2: Experimental Details
2.1 Overview 15
2.2 Heated Nozzle Design for Radical Production 18
2.3 Diffusion Pump Safety Interlock Design 27
2.4 Sum Frequency Generation 28
2.5 References 32

Chapter 3: Electronic Spectroscopy and Photodissociation
Dynamics of the 1-Hydroxyethyl Radical CH
3
CHOH
3.1 Introduction 33
3.2 Experimental Details 35
3.3 REMPI and Photofragment Yield Spectra 37
3.4 Time of Flight and Energy Spectra of H and D 41
Fragments
3.5 Conclusions 50
3.6 References 58

Chapter 4: D-Atom Products in Predissociation of CD
2
CD
2
OH
From the 202-215 nm Photodissociation of
2-Bromoethanol
4.1 Introduction 64
4.2 Experimental Details 67
4.3 BrCD2CD2OH Absorption Spectrum and Mass 71
Spectrum
4.4 Analysis of D Fragment Velocity and Energy 75
Distributions
4.5 Maximum Velocity of D Fragments and Dissociation 80
Pathways

iii
4.6 Theoretical Estimates of Product Branching Ratios 84
4.7 Conclusion 97
4.8 References 91

Chapter 5: Future Work: Spectroscopy of CH
2
CH2OH
5.1 Introduction 93
5.2 Experimental 95
5.3 Electronic States 98
5.4 References 101

Alphabetized Bibliography 102

Appendix 108
A1 References 112

iv
List of Tables

Table 2.1: Vapor pressure of ICH
2
CH
2
OH vs. temperature 26

Table 3.1: Comparison of quantum defects (δ) and energy of 39
onset (T
0
E
) of Rydberg states in CH
2
OH and CH
3
CHOH

Table 4.1: Values of minimum kinetic energy release in photolysis 83
step

Table 4.2: Maximum velocities expected for reactions (4.6) 83
and (4.7)

Table A1: Thermochemical values used in evaluating eq A1 111

v
List of Figures

Figure 2.1: Schematic of the molecular beam apparatus 16

Figure 2.2: Cross-section schematic of the heated nozzle for 22
experiments on low vapor pressure haloethanol
species, XCH
2
CH
2
OH

Figure 2.3: Heated nozzle side view 23

Figure 2.4: Heated nozzle front view 24

Figure 2.5: Receptacle design for heated nozzle 25

Figure 2.6: System of interlocks to protect diffusion pump from 29
overheating and explosion

Figure 3.1: D photofragment yield spectrum in the region 42
19,000 cm
-1
- 21,500 cm
-1

Figure 3.2: 2 + 1 REMPI of CH
2
OH and CH
3
CHOH in the region 43
of absorption to the 3p
z
state

Figure 3.3: Background subtracted H fragment time-of- flight 47
spectrum from excitation at 21,276 cm
-1

Figure 3.4: The c.m. E
T
distribution of H photofragments 52
following the 1
2
A(3s)←1
2
A transition at
21,276 cm
-1
(2.63 eV)

Figure 3.5: The c.m E
T
distribution of H fragments following 53
31,250 cm
-1
(3.87 eV) excitation

Figure 3.6: The c.m E
T
distribution of D fragments following 54
31,250 cm
-1
(3.87 eV) excitation

Figure 3.7: The c.m E
T
distribution of H fragments following 55
31,250 cm
-1
(3.87 eV) excitation

Figure 3.8: The c.m E
T
distribution of D fragments following 56
31,250 cm
-1
(3.87 eV) excitation

Figure 3.9: Recoil anisotropy parameter β as a function of 57
excitation energy

vi

Figure 4.1: Schematic energy diagram for the decomposition of 65
CH
2
CH
2
OH

Figure 4.2: Absorption spectrum of gas phase 2-bromoethanol 72

Figure 4.3: Mass spectrum of BrCD
2
CD
2
OH at selected 74
wavelengths

Figure 4.4: Experimental TOF distributions, converted to velocity 77
scale, for D atoms from secondary dissociation of
CD
2
CD
2
OH radicals produced by photolysis of
BrCD
2
CD
2
OH at 202, 205, 210 and 215 nm

Figure 4.5: Maximum allowed laboratory frame velocity of D 81
atoms as a function of kinetic energy release in the
photolysis reaction (4.4) at 205 nm

Figure 4.6: Dissociation rates of the CD
2
CD
2
OH radical to 88
different channels

Figure 4.7: Branching ratios for the CD
2
CDOH + D and 89
CD
2
HCDO + D channels for different rotational levels
of the CD
2
CD
2
OH radical

Figure 5.1: Non-classical bridged structure of the CH
2
CH
2
OH
+
ion 95

Figure 5.2: Electronic absorption spectrum of iodoethanol 97

vii
Abstract
The electronic spectroscopy and dissociation dynamics of 1- and 2- hydroxyethyl
radicals were investigated using a combination of Resonance Enhanced Multiphoton
Ionization (REMPI) and time-of-flight (TOF) detection of H and D photofragments.
The 3s and 3p
z
Rydberg states of CH
3
CHOH, CH
3
CHOD, and CD
3
CHOH in the
region 19,400-37,000 cm
-1
were studied in a molecular beam. Absorption to the
structureless 3s Rydberg state begins at 19,600 cm
-1
, though no REMPI spectrum is
observed due to the fast predissociative nature of the state. Instead, the onset is detected
using photofragment spectroscopy of H and D atoms. The electronic transition has a
perpendicular transition moment, analogous to CH
2
OH. The onset of the predissociative
3p
z
state can be detected via REMPI, and is observed at 32360 cm
-1
. Photofragments
detected from dissociation of this state show no recoil anisotropy. However, the β
parameter becomes more negative (indicative of a perpendicular electronic transition
moment) as the excitation energy is decreased. The transition becomes more anisotropic
as the energy decreases through the predicted region of the 3p
x
Rydberg state towards the
onset of the 3s state. The onset of the transition to the 3p
x
state cannot be identified due
to the large background signal from the 3s state. The translational energy distributions of
H and D atims indicates that both the O-H(D) and C
(2)
-H(D) bonds can be broken at
sufficiently high energies.
The ground state photodissociation dynamics of CD
2
CD
2
OH radical were studied
to determine product branching ratios and look for evidence of vinyl alcohol
(CD
2
CDOH) product. Deuterated 2-hydroxyethyl radical (CD
2
CD
2
OH) was generated

viii
from photodissociation of the BrCD
2
CD
2
OH precursor by cleaving the Br-C bond at the
wavelengths 202, 205, 210, and 215 nm. The energy left in the CD
2
CD
2
OH radical after
breaking this bond led to the spontaneous dissociation of the radical. The presence of
acetaldehyde and vinyl alcohol products was inferred by monitoring the velocity
distribution of the D atom co-fragment in the dissociation, and using conservation of
energy to determine the identity of the molecular fragment. Different photolysis
wavelengths were used to determine whether an obvious change in branching ratio
between D and OH dissociation products would become evident as less energy for
dissociation became available to the system. No precipitous drop in D products was seen
toward longer photolysis wavelengths. However, a high velocity tail became apparent at
longer wavelengths, indicating the onset of 2-photon processes at these energies. These
processes are proposed to be owed to two phenomena: an absorption by the radical
beginning at 210 nm, and lengthening of the lifetime of the excited radical as the
available energy above the centrifugal barrier becomes smaller.
Both acetaldehyde and vinyl alcohol products were seen in the dissociation of
CD
2
CD
2
OH. In agreement with RRKM calculations, vinyl alcohol products are expected
to constitute the majority of the dissociations to D + molecular co-fragment.

1
Chapter 1
Introduction
1.1.1 Overview
The discipline of physical chemistry concerns itself with topics of both practical
relevance and the more existential and intriguing questions addressing the quantum
mechanical underpinnings of the universe. For this reason, systems such as the
hydroxyalkyl radicals studied in our laboratory are rewarding in that they are both
socially and scientifically relevant. Practically, they are ubiquitous in controlled
combustion reactions such as those in automobiles as well as in natural and uncontrolled
occurances like fires. The two lowest molecular-weight hydroxyalkyl radicals,
hydroxyethyl and hydroxymethyl, are particularly pervasive and have been identified as
intermediates in the combustion of hydrocarbons
1-3
, in atmospheric chemistry
4,5
and in
isomerization of alkoxy species.
6-12
The 1- and 2- hydroxyethyl structural isomers are
possible products of the combustion of ethanol, which is increasingly used as a fuel and
fuel-additive.
1-3
They can act as intermediates in the production of reactive species such
as OH, CH
3
CH
2
O, and CH
3
;
8-12
leading to common products of hydrocarbon reactions
such as CH
3
CHO and H
2
O;
2,3,8-16
and even CH
2
CHOH
1-3,8-12,14-16
, which eluded detection
until its recent identification in flames of hydrocarbon fuels.
1,3,5

Hydroxyalkyl species are equally valuable as probes into the microscopic world
of molecular dynamics. The electronic states of CH
2
OH, for instance, were at first
largely misunderstood.
17-22
Theoretical treatments of hydroxymethyl in the C
1
symmetry
point group, which is the symmetry of its ground-state geometry, led to the mislabeling of

2
its excited electronic states.
20,22
More careful theoretical studies, together with
experimental studies performed in our laboratory, recognized that the radical's low-
frequency vibrational modes give its wavefunction an average planar C
s
symmetry.
17-19,21

This symmetry is manifest when measuring the anisotropy is the dissociation to CH
2
O +
H from the excited electronic states, which led to the proper characterization and labeling
of these states.
23-25
Our recent work on CH
3
CHOH, with its radical center also on the
C
(1)
-O bond as in CH
2
OH, has demonstrated that the 1-hydroxyethyl radical displays
analogous behavior.
16,26

Of interest in these radical species is also the very low-lying ionization potentials
(I.P.s). Radicals typically have low I.P.s as compared to their closed-shell counterparts,
but those of hydroxyalkyl radicals are as low as 6.67 eV in CH
3
CHOH.
13,15,16,26-29
The
I.P.s are only moderately higher in CH
2
OH and CH
2
CH
2
OH at 7.56 eV
25,30,31
and ≤8.18
eV
13,26,28,29
, respectively. In the case of CH
2
OH and CH
3
CHOH, the low IP gives rise to
the presence of low-energy Rydberg electronic states.
15-19,21,23-25
The possibility of the
existence of both Rydberg and valence states in close proximity creates an additional
challenge in the study of these radicals. In addition, it is difficult to isolate these radicals
because of their fast rate of reactivity with themselves and because the C
(1)
-H bond must
be selectively cleaved using Cl radicals from the photolysis of Cl
2
gas, a highly corrosive
substance.
15,16,23,25
The low-lying electronic states couple to the ground states along
dissociative coordinates of the O-H and C-H bonds in both radicals. Upon electronic
excitation, the radicals dissociate rapidly along these coordinates, on a timescale of less
than half a picosecond.
16,17
These short lifetimes complicate the Resonance Enhanced

3
Multiphoton Ionization (REMPI) technique used in the spectroscopy of these radicals,
which ionizes the radical through an intermediate electronic state, and in the case of the
lowest-lying electronic states makes detection of the radical ion impossible. In order to
probe the spectroscopy of these states, the H products of the dissociation were
monitored.
16,23

The low energies of the electronic states in CH
3
CHOH are of particular
importance in the combustion of ethanol fuel; in the case of the lowest-lying electronic
state, as will be shown in Chapter 3, absorption can occur in a region of the visible
spectrum.
15,16
This means that visible light can electronically excite the radical and
potentially initiate unimolecular reactions. Because of the facility of this absorption, it is
desirable to determine which reactions take place upon excitation of the 1-hydroxyethyl
radical. Given the reactivity of the ethoxy isomer with oxygen,
7
the possible
isomerization of the excited species was of interest. Our studies on the electronic states
of this radical, however, did not show evidence of isomerization. This isomerization was
also searched for extensively in the CH
2
OH radical, given that the CH
3
O radical reacts up
to 3,000 times faster with oxygen in the atmosphere than its hydroxyalkyl isomer.
32-35

As in CH
3
CHOH, no evidence for this isomerization was seen in the electronic
spectroscopy.
23-25
Since the theoretical energy required to surmount the barrier to
isomerization to CH
3
O (13,469 cm
-1
) is lower than that necessary for O-H bond
dissociation (14,000-16,000 cm
-1
), the occurance of isomerization in the excited radical
would not be surprising.
32-34,36
However, the couplings of the electronic states to the
dissociative ground-state coordinates take place at large C-H or O-H separation and do

4
not promote the shift of H from the O to the C atom. Rather, prompt O-H or C-H bond
breakage is observed. In order to elucidate the dynamics on the ground-state surface of
the radical, the O-H stretch coordinate was probed directly through infrared excitation.
37

Again, no evidence for the isomerization reaction was seen, but tunneling through the O-
H dissociation barrier was detected which led to CH
2
O + H products.
The large body of work performed on the CH
2
OH radical by our group informed
the work on CH
3
CHOH reported here. Both hydroxyemethyl and 1-hydroxyethyl have a
lone electron in an anti-bonding orbital centered above the C
(1)
-O bond (π*
C-O
).
21,23

Electronic excitation in both species promotes this electron to non-bonding orbitals,
which activates the C-O stretch vibrational mode.
16,17
C-O vibrational progressions at
~1625 cm
-1
intervals in both radicals are observed in the electronic spectroscopy. The
similarities in the geometries and electronic structure of these radicals make them natural
homologs and allow much to be inferred about the properties of CH
3
CHOH from what
we know about CH
2
OH. This stands in contrast to the 2-hydroxyethyl (CH
2
CH
2
OH)
isomer, where the lone electron lies above C
(2)
in a p-type orbital and the structure of the
cation is strikingly different from the CH
3
CHOH
+
and CH
2
OH
+
ions.
13,26,28,29
The
location of the radical electron in this species relative to the electron-donating OH group
place the 2-hydroxyethyl radicals more logically in the CH
2
CH
2
X series of radicals
(where X= F, Br, I).
Given all of their similarities, the surprisingly low I.P. in CH
3
CHOH with respect
to CH
2
OH was a subject for exploration in a theoretical study.
38
One common
contributor to the depression of I.P.s was reported by Koziol, et al. in an electronic

5
structure study on homologous groups of molecules and their ionization potentials.
39

This study reports that increased stability in the ion structure in combination with
instability of the neutral species logically leads to a low I.P. in that species. An analysis
of the electronic structure of CH
3
CHOH showed that the cation form of CH
3
CHOH is
stabilized by electron donation from the OH group. These stabilizing effects of electron
donation are called hyperconjugation and are often exploited by organic chemists. In
order to test this assumption, an electronic structure study was performed by our group in
collaboration with the group of Dr. Anna Krylov to determine the contribution of this
hyperconjugation effect to the stabilization of the cation and the destabilization of the
neutral CH
3
CHOH relative to CH
2
OH, and consequent lowering of the I.P. in the
CH
3
CHOH radical.
38
Thus, the combined effect of a stable CH
3
CHOH
+
ion and less
stable neutral, is the main contributor in the lowering of the IP of CH
3
CHOH relative to
CH
2
OH.

1.2 1-Hydroxyethyl Radical (CH
3
CHOH)
In light of the current interest in ethanol as a fuel source, there is a concomitant
desire to understand radicals formed from ethanol precursors, including the hydroxyethyl
and ethoxy isomers, which are formed upon loss of an H atom from carbon or oxygen. A
large body of work exists on the ethoxy isomer.
40-51
The 1-and 2-hydroxytheyl isomers,
however, have only recently appeared as subjects of a number of experimental
16,29,52-58
and theoretical
8,13,15,26,28,38,45,49,50
studies. The 1-hydroxyethyl isomer, with the radical
center on the C
(1)
carbon, is the most stable of the three isomers and was a natural target

6
for investigation in our laboratory given its geometric similarity to the hydroxymethyl
radical. As described above, the hydroxymethyl and 1-hydroxyethyl homologs have
similar geometries and symmetry, and therefore parallels were expected in their
electronic transitions and dissociation dynamics. Indeed, with empirical values taken
from the investigation of CH
2
OH, the electronic states of CH
3
CHOH were quickly
identified and even demonstrated familiar dissociation dynamics.
16
The minimum geometry of the CH
2
OH radical is completely asymmetric and
belongs to the C
1
symmetry point group.
18-22
However, owing to the existence of low-
frequency out-of-plane vibrational modes, the radical in fact has an average symmetry
about the plane described by the C-O-H set of bonds, placing it in the C
s
symmetry point
group, as far as the electronic transitions are concerned.
16-19,21,23
In our work with
CH
3
CHOH, which also has a minimum-energy geometry of C
1
symmetry, the same
symmetry of the electronic wavefunction was inferred from the anisotropy in hydrogen
photofragments that predissociate from the electronically excited states.
15

The lowest excited electronic states accessible in CH
3
CHOH were found to be
Rydberg states as in CH
2
OH, and the energy of the transitions to these states was
estimated using the ionization potential of CH
3
CHOH and applying the Rydberg formula
using the quantum defects found in CH
2
OH. The Rydberg formula,

) (
. .
2
 
 
n
Ryd
P I E
Rydberg

7
where E
Rydberg
is the energy of the transition to the Rydberg state, I.P. is ionization
potential, Ryd is the Rydberg Constant (the ionization potential of hydrogen atom), and δ
is the quantum defect, whose purpose is described below. This formula identifies
electronic transitions in atomic hydrogen, and can also be used to successfully predict the
energies of transitions to the diffuse electronic Rydberg states, which depend little on the
nuclear geometry. In neutral molecules, these states are usually higher in energy than
valence electronic states. Because most of the probability density of these states is
concentrated at a large distance from the nuclei, they interact with the nuclei as if the
positive charge density were a point charge, and thus behave like the electronic states of
hydrogen. The quantum defect δ describes the way in which the observed energy of
transition differs from that predicted by the hydrogen model; in other words, the effect of
the nuclear geometry on the energy location of the electronic state. Since the geometries
of the CH
2
OH and CH
3
CHOH homologs are similar, the quantum defects describing the
location of the electronic transitions to the Rydberg states are expected to be similar in
both species. The electronic states of molecules with C
s
symmetry can be either
symmetric (A') or antisymmetric (A") with respect to the plane of symmetry. Using time-
of-flight and core sampling techniques, the symmetries in the observable Rydberg states
of CH
3
CHOH were found to be analogous to those in CH
2
OH.

1.3 2-Hydroxyethyl Radical (CH
2
CH
2
OH) Dissociation Pathways
In the study of hydrocarbon radicals, the 2-hydroxyethyl isomer provides some of
the most intriguing fodder for interrogation. A few investigators have studied it as a

8
product of the photolysis of a haloethanol precursor or as product of the association
reaction between ethene and OH radical. These studies typically address the dissociation
of activated CH
2
CH
2
OH to form OH and CH
2
CH
2
products. Very little data have been
collected, however, on the properties of CH
2
CH
2
OH itself. Theoretical works have
calculated the ionization potential of this radical and the geometry of the CH
2
CH
2
OH
+

ion, which predict that the cation geometry will have a non-linear oxirane structure. A
photoionization study performed by Ruscic and Berkowitz corroborates the assertion that
the cation geometry is significantly different from that of the neutral species. In this
study, a large increase in ion signal from CH
2
CH
2
OH
+
is observed at ~10.25 eV;
however, there is still signal above the background that decreases monotonically down to
~8.18 eV. This indicates that the adiabatic ionization involves a large change in
geometry between the molecule and ion. Using thermochemical considerations, the
authors estimate that the ionization potential of CH
2
CH
2
OH should be close to 7.7 eV.
This is further evidence that the adiabatic ionization potential is lower than can be
spectroscopically observed due to poor Franck-Condon overlap between the neutral
radical and the ion.
In addition to the desire to discover more about the electronic properties of 2-
hydroxyethyl, there has been a recent awakening of interest in the dissociation of
activated CH
2
CH
2
OH to form vinyl alcohol (CH
2
CHOH) and H products. As mentioned
above, many groups have observed hydroxyl radical dissociating from the hydroxyethyl
precursor, but none have attempted to interrogate the minor vinyl alcohol product.
Previously, vinyl alcohol was believed to be an intermediate of minor importance in the

9
combustion of hydrocarbons, as it is unstable compared to its acetaldehyde tautomer.
However, recent experiments have shown significant amounts of vinyl alcohol present in
the flames of hydrocarbons. Chapter 4 describes a study, using a deuterated
bromoethanol precursor to produce activated CD
2
CD
2
OH via photolysis of the C-Br
bond, which probes the D photofragment resulting from the dissociation of the radical to
both vinyl alcohol and acetaldehyde. Conservation of energy considerations are used in
collaboration with theory to identify the channels via which the 2-hydroxyethyl radical
dissociates.

1.4 Theoretical Models
Given the complexity of the TOF data of the D products in CD
2
CD
2
OH
dissociation, modeling was extremely helpful in interpreting the results. Rate
calculations using RRKM theory were used to determine branching ratios between vinyl
alcohol and acetaldehyde products, and thus were very useful for comparison with the D
TOF spectra.
14
The vinyl alcohol and acetaldehyde products are both generated through
loss of D from the C
(1)
carbon, rendering isotopic substitution useless in distinguishing
between the two products. The only method at our disposal to distinguish between these
product channels is the use of energetic considerations, bolstered by the product ratios
predicted by rate calculations and taking into account the energies and geometries of the
transition states. Dr. Stephen Klippenstein of Argonne National Laboratory kindly
assisted us by carrying out calculations of product branching ratios as a function of
excess energy above the dissociation threshold.
14
In addition, since the CD
2
CD
2
OH

10
radical was expected to be highly rotationally excited after dissociation of the
BrCD
2
CD
2
OH precursor, calculations were performed illustrating the effect of the
rotational excitation of CD
2
CD
2
OH on the product branching ratios.
Additionally, we carried out model calculations using impulsive models in order
to estimate the amount of kinetic energy imparted to the CD
2
CD
2
OH radical in the initial
photodissociation step of BrCD
2
CD
2
OH. This energy was used to assess the maximum
amount of energy available to the vinyl alcohol and acetaldehyde products in the
secondary dissociation, although in the final analysis we used the experimental center of
mass (c.m.) kinetic energies taken from 193 and 205 nm dissociations of
BrCH
2
CH
2
OH.
60-62
Since these energies were very close for both photolysis wavelengths
and had very large error bars, it was decided that taking an average of the experimental
numbers would be the most reliable approach. The impulsive model calculations are
given in the Appendix.

11
1.5 References

(1) Cool, T. A.; Nakajima, K.; Mostefaoui, T. A.; Qi, F.; McIlroy, A.; Westmoreland,
P. R.; Law, M. E.; Poisson, L.; Peterka, D. S.; Ahmed, M. J Chem Phys 2003, 119, 8356.

(2) Taatjes, C. A.; Hansen, N.; McIlroy, A.; Miller, J. A.; Senosiain, J. P.;
Klippenstein, S. J.; Qi, F.; Sheng, L.; Zhang, Y.; Cool, T. A.; Wang, J.; Westmoreland, P.
R.; Law, M. E.; Kasper, T.; Kohse-Hoinghaus, K. Science 2005, 308, 1887.

(3) Taatjes, C. A.; Hansen, N.; Miller, J. A.; Cool, T. A.; Wang, J.; Westmoreland, P.
R.; Law, M. E.; Kasper, T.; Kohse-Hoinghaus, K. J Phys Chem A 2006, 110, 3254.

(4) Saito, S. Chem Phys Lett 1976, 42, 399.

(5) Turner, B. E.; Apponi, A. J. Astrophys J 2001, 561, L207.

(6) Choi, H.; Bise, R. T.; Neumark, D. M. J Phys Chem A 2000, 104, 10112.

(7) Faulhaber, A. E.; Szpunar, D. E.; Kautzman, K. E.; Neumark, D. M. J Phys Chem
A 2005, 109, 10239.

(8) Senosiain, J. P.; Klippenstein, S. J.; Miller, J. A. J Phys Chem A 2006, 110, 6960.

(9) Wu, C. W.; Lee, Y. P.; Xu, S. C.; Lin, M. C. J Phys Chem A 2007, 111, 6693.

(10) Xu, S.; Lin, M. C. P Combust Inst 2007, 31, 159.

(11) Xu, Z. F.; Xu, K.; Lin, M. C. Chemphyschem 2009, 10, 972.

(12) Zhu, R. S.; Park, J.; Lin, M. C. Chem Phys Lett 2005, 408, 25.

(13) Curtiss, L. A.; Lucas, D. J.; Pople, J. A. J Chem Phys 1995, 102, 3292.

(14) Edwards, L. W.; Ryazanov, M.; Reisler, H.; Klippenstein, S. J. J Phys Chem A
2010, 114, 5453.

(15) Karpichev, B. Electronic States and Photodissociation Dynamics of Hydroxyalkyl
Radicals, University of Southern California Ph.D. Thesis, 2009.

(16) Karpichev, B.; Edwards, L. W.; Wei, J.; Reisler, H. J Phys Chem A 2008, 112,
412.

(17) Aristov, V.; Conroy, D.; Reisler, H. Chem Phys Lett 2000, 318, 393.

12
(18) Bruna, P. J.; Grein, F. J Phys Chem A 1998, 102, 3141.

(19) Bruna, P. J.; Grein, F. J Phys Chem A 2001, 105, 8599.

(20) Chen, F. W.; Davidson, E. R. J Phys Chem A 2001, 105, 4558.

(21) Johnson, R. D.; Hudgens, J. W. Abstr Pap Am Chem S 1996, 212, 82.

(22) Rettrup, S.; Pagsberg, P.; Anastasi, C. Chem Phys 1988, 122, 45.

(23) Conroy, D.; Aristov, V.; Feng, L.; Reisler, H. J Phys Chem A 2000, 104, 10288.

(24) Conroy, D.; Aristov, V.; Feng, L.; Sanov, A.; Reisler, H. Accounts Chem Res
2001, 34, 625.

(25) Conroy, D. G. Rydberg State of an Open Shell Species: Characterization and
Photophysics of the 3pz State of CH
2
OH, University of Southern California Ph.D. Thesis,
2000.

(26) Feng, L. Spectroscopy and Photodissociation Dynamics of the Hydroxymethyl
Radical (CH2OH), University of Southern California, 2004.

(27) Feng, L.; Demyanenko, A. V.; Reisler, H. J Chem Phys 2004, 120, 6524.

(28) Feng, L.; Demyanenko, A. V.; Reisler, H. J Chem Phys 2003, 118, 9623.

(29) Feng, L.; Huang, X.; Reisler, H. Abstr Pap Am Chem S 2002, 223, C67.

(30) Feng, L.; Huang, X.; Reisler, H. J Chem Phys 2002, 117, 4820.

(31) Feng, L.; Reisler, H. J Phys Chem A 2004, 108, 9847.

(32) Karpichev, B. K., L; Diri, K; Reisler, H.; Krylov, A.I. J Chem Phys 2010,
132, 114308.

(33) Dyke, J. M.; Groves, A. P.; Lee, E. P. F.; Niavaran, M. H. Z. J Phys Chem A
1997, 101, 373.

(34) Nobes, R. H.; Rodwell, W. R.; Bouma, W. J.; Radom, L. J Am Chem Soc 1981,
103, 1913.

(35) Ruscic, B.; Berkowitz, J. J Chem Phys 1994, 101, 10936.

(36) Ruscic, B.; Berkowitz, J. J Chem Phys 1991, 95, 4033.

13

(37) Dyke, J. M.; Ellis, A. R.; Jonathan, N.; Keddar, N.; Morris, A. Chem Phys
Lett 1984, 111, 207.

(38) Adams, G. F.; Bartlett, R. J.; Purvis, G. D. Chem Phys Lett 1982, 87, 311.

(39) Radford, H. E. Chem Phys Lett 1980, 71, 195.

(40) Saebo, S.; Radom, L.; Schaefer, H. F. J Chem Phys 1983, 78, 845.

(41) Wendt, H. R.; Hippler, H.; Hunziker, H. E. J Chem Phys 1979, 70, 4044.

(42) Dobe, S. B., T.; Turanyi, T.; Marta, F.; Grussdorf, J.; Temps, F.; Wagner, H. G. J.
Phys. Chem. 1996, 100, 19864.

(43) Wei, J.; Karpichev, B.; Reisler, H. J Chem Phys 2006, 125.

(44) Karpichev, B.; Reisler, H.; Krylov, A. I.; Diri, K. J Phys Chem A 2008, 112,
9965.

(45) Koziol, L.; Levchenko, S. V.; Krylov, A. I. J Phys Chem A 2006, 110, 2746.

(46) Caralp, F.; Devolder, P.; Fittschen, C.; Gomez, N.; Hippler, H.; Mereau, R.;
Rayez, M. T.; Striebel, F.; Viskolcz, B. Phys Chem Chem Phys 1999, 1, 2935.

(47) Dang, T. T.; Motell, E. L.; Travers, M. J.; Clifford, E. P.; Ellison, G. B.; Depuy,
C. H.; Bierbaum, V. M. Int J Mass Spectrom 1993, 123, 171.

(48) Ebata, T.; Yanagishita, H.; Obi, K.; Tanaka, I. Chem Phys 1982, 69, 27.

(49) Ellison, G. B.; Engelking, P. C.; Lineberger, W. C. J Phys Chem-Us 1982, 86,
4873.

(50) Gonzalez, C.; Sosa, C.; Schlegel, H. B. J Phys Chem-Us 1989, 93, 2435.

(51) Hoyermann, K.; Olzmann, M.; Seeba, J.; Viskolcz, B. J Phys Chem A
1999, 103, 5692.

(52) Inoue, G.; Okuda, M.; Akimoto, H. J Chem Phys 1981, 75, 2060.

(53) Ohbayashi, K.; Akimoto, H.; Tanaka, I. J Phys Chem-Us 1977, 81, 798.

(54) Ramond, T. M.; Davico, G. E.; Schwartz, R. L.; Lineberger, W. C. J Chem Phys
2000, 112, 1158.

14

(55) Sosa, C.; Schlegel, H. B. J Am Chem Soc 1987, 109, 4193.

(56) Sosa, C.; Schlegel, H. B. J Am Chem Soc 1987, 109, 7007.

(57) Zhu, X. K., M. M.; Misra, P. Pure Appl. Opt. 1996, 5, 1021.

(58) Anastasi, C.; Simpson, V.; Munk, J.; Pagsberg, P. J Phys Chem-Us 1990, 94,
6327.

(59) Anastasi, C.; Simpson, V.; Munk, J.; Pagsberg, P. Chem Phys Lett 1989, 164, 18.

(60) Chandler, D. W.; Thoman, J. W.; Hess, W. P. Inst Phys Conf Ser 1991, 355.

(61) Hintsa, E. J.; Zhao, X. S.; Lee, Y. T. J Chem Phys 1990, 92, 22

(62) Ratliff, B. J.; Womack, C. C.; Tang, X. N.; Landau, W. M.; Butler, L. J.; Szpunar,
D. E. Journal of Physical Chemistry A 2010, 114, 4934.

(63) Sapers, S. P.; Hess, W. P. J Chem Phys 1992, 97, 3126.

(64) Shubert, V. A.; Rednic, M.; Pratt, S. T. J Phys Chem A 2009, 113, 9057.

15
Chapter 2
Experimental Details
2.1 Overview
All of the experiments described in this dissertation were performed in the
molecular beam apparatus shown in Figure 2.1. This apparatus consists of a vacuum
chamber divided into two parts by a flange with a skimmer (Beam Dynamics 1.51 mm).
The first is the source region, where the supersonic molecular beam originates and is
cooled. The sample is introduced into the source region via a piezoelectrically-driven
pulsed nozzle operating at 10 Hz. After passing through the skimmer, the beam enters
the detection chamber, where it is crossed with a beam of laser radiation. Windows on
either side of the detection region allow for two parallel, counter-propagating laser beams
to enter the chamber. In the case of CH
3
CHOH, one color (1+1) or (2+2) resonance
enhanced multiphoton ionization (REMPI) can be performed on the radical using only
one laser. In addition, two-laser pump-probe schemes are used to ionize and detect H and
D fragments dissociating from both the 1-and-2-hydroxyethyl radicals. Ions are collected
by a home-built Wiley-McClaren time-of-flight (TOF) apparatus mounted vertically, and
detected by a microchannel plate (MCP) detector (Galileo, 25 mm diameter) installed at
the end of the 18 cm TOF tube.
17,23
The signal from the MCP goes through a fast-current
preamplifier (KOTA A105) to an oscilloscope (Tektronics TDS640A) and is connected
to a computer running LabVIEW software, which collects and stores the data. The laser
and nozzle firing times are controlled by two delay generators (Stanford Research
Systems DG535). When a wavelength scan is required, as when scanning the excited

16

17
electronic states of CH
3
CHOH, LabVIEW is used to control the laser wavelength via a
serial port. LabVIEW can also be used to change the laser delay times, which is useful in
pump/probe experiments where the time between the firing of the first and second lasers
is crucial.
In the CH
3
CHOH experiments, a turbomolecular pump (Leybold, TMP 1000C)
was used to evacuate the source chamber. However, in the experiments using
BrCD
2
CD
2
OH as a precursor, a more robust pump was required that would offer higher
resistance to corrosion by HBr or any other halogen by-products of bromoethanol. Thus,
for these experiments a diffusion pump (Varian, VHS-6) was installed in combination
with an interlock system described in more detail in Section 2.4. This diffusion pump
eliminated the problems with malfunctioning of the turbomolecular pump due to the
corrosive reaction mixture. To protect against contamination in the vacuum chamber by
the pump oil, a liquid-nitrogen-cooled baffles system (Varian, 6") is employed. To
prevent cross-contamination between the roughing pump and diffusion pump oil, a co-
axial foreline trap with a replaceable filter (MDC, KDFT-4150-2) was installed at the
mouth of the roughing pump. The detection region is evacuated by a turbomolecular
pump (Leybold, Turbovac 361). Due to the small fraction of the molecular beam that
passes through the skimmer, contamination and corrosion do not pose a problem to the
turbopump evacuating the detection chamber. The pressure in both the source and the
etection regions is measured using an ion gauge glass tube (Duniway I-75-N) connected
with Phillips-style cables to a controller (Granville Phillips 270).

18
2.2 Heated nozzle design for radical production
Since the experimental schemes for work on CH
3
CHOH and CH
2
CH
2
OH differ,
they will be described in full in the respective chapters of this thesis. In this section I
describe the design of a new heated pulsed nozzle built for use with halogenated ethanol
precursors, which have low vapor pressures. The vapor pressure of ICH
2
CH
2
OH, which
will be used in future work on the CH
2
CH
2
OH radical, is less than one torr at room
temperature. Therefore, in order to increase the partial pressure of this species in the gas
mixture, it must be heated. The nozzle was designed such that the container holding the
liquid haloethanol and the heating apparatus are connected directly to the nozzle head
inside the vacuum chamber. This way, once in the gas phase, the haloethanol will not
have to travel through a long stretch of tubing in order to arrive at the nozzle exit, and
will not risk re-condensing on the inside surface of the tubes.
The nozzle itself consists of a small reactant chamber (with a capacity of
approximately 1.5 cm
3
) for the gas mixture (either BrCH
2
CH
2
OH/He or
CH
3
CH
2
OH/Cl
2
/He for the purposes of the experiments described here) with a quartz
tube attached to a small opening on the vacuum side, as seen in Figure 2.2. This opening
is sealed with a kevlar o-ring seal attached to a plunger. As the plunger pulls back, a
small amount of gas mixture is released into the quartz tube. The amount of gas allowed
through can be controlled by tightening or loosening the plunger against the nozzle exit.
The plunger is controlled by a piezoelectric driver (Physik Instrumente, 100μm)
operating at 10 Hz. The amount of gas in each pulse is measured by monitoring the

19
pressure in the vacuum chamber, which is 2 x 10
-7
Torr in the detection region with the
nozzle operating. The base pressure is ~6 x 10
-8
Torr.
A small stainless steel receptacle, shown in Figure 2.5, is attached directly to the
nozzle head via a 1/8” male NPT nipple, which connects to female NPT joints in both the
receptacle and nozzle head. This receptacle contains ~1 mL of haloethanol in the liquid
phase. A ¼" OD teflon tube is threaded from the outside of the chamber through the
stainless steel pole on which the nozzle is mounted. The Teflon tubing exits the mount
through a vacuum-tight feedthrough on the inside of the chamber, and connects to the
haloethanol receptacle via a 1/8" NPT-1/4" Swagelok tube fitting adapter that screws into
the female 1/8” NPT opening on the receptacle. Helium enters the receptacle via this
tube, is bubbled through the liquid haloethanol, and the haloethanol vapor/He mixture
enters the nozzle chamber directly from the 1/8" female NPT mouth of the receptacle.
Iodoethanol vapor is generated by heating to 40° C of the liquid, at which a vapor
pressure of 7 torr is obtained, using heating wire (Thermocoax, 1.0 mm) connected by
electrically-insulated connectors (Thermocoax, CP10) to electrical wire. This wire is
connected via a vacuum-tight 8-pin electrical feed-through to a power supply (TENMA
72-7245). Type K thermocouple wire is used to monitor the vial temperature, which has
been calibrated to the vapor pressure of the haloethanol.
It is important to mention that both the nozzle head and receptacle are constructed
of stainless steel in this design. In the first design, a plastic nozzle head was used.
23

Halogenated hydrocarbon liquids contain a small fraction of X
2
and HX. In earlier
experiments on CH
2
OH, where Cl
2
gas is introduced directly into the nozzle, metals

20
could not be used in the nozzle design because of the corrosion that resulted from the
interaction of the halogen gas and the metal surface. In order to prevent this corrosion,
which poisoned the CH
2
OH radical,
25
a special nozzle head made of Ultem plastic was
designed, and was used with kevlar o-rings. Anticipating a similar problem with
BrCH
2
CH
2
OH resulting from trace amounts of HBr and Br
2
, the Ultem nozzle was used
in the original design. Upon heating, however, the BrCH
2
CH
2
OH ate through the plastic
nozzle head and polymerized to form a dark-purple solid. Counter to our initial
assumptions, it did not appear to react strongly with the stainless steel model.
The original haloethanol receptacle was made of copper, but over time the copper
was found to corrode and form a green residue. The receptacle was connected via 1/8"
copper tubing to the He inlet and the nozzle head, which created unnecessarily large
amounts of surface area on which the vapor could condense and ran the risk of clogging.
Additionally, it was heavy and caused the nozzle head to sag on its stainless steel mount,
which eventually became permanently deformed. In the new design, the receptacle is
smaller and made of stainless steel. The need for the 1/8" metal tubing was eliminated by
connecting the receptacle directly to the nozzle head, and the He inlet tube directly to the
receptacle. Keeping the precursor liquid inside the vacuum chamber precludes the need
for opaque tubing to protect the photosensitive haloethanol from ambient light. Upon
prolonged exposure to light, the clear bromoethanol decomposes into a brown, murky
liquid, probably composed of Br
2
. The new design is much lighter weight and simple;
there are no small tubes which can clog with material, fewer joints that can leak, and

21
stainless steel has proven to withstand the destructive effects of the heated bromoethanol
on the Ultem design.
The thermocouple wire is calibrated using a small amount of water in the
vial with an alcohol thermometer placed carefully in the water. A Tenma 72-7245 power
supply supplies the current to the heating wire, which is wrapped around the vial and
fixed with an aluminum clamp. The thermocouple readout temperature is calibrated to
the reading on the thermometer. Since the pressure of the ICH
2
CH
2
OH cannot be
measured directly in the chamber, the vapor pressure was measured as a function of
temperature outside the chamber to determine the percentage of ICH
2
CH
2
OH vapor in the
reactant chamber of the nozzle. The vapor pressure to temperature dependence is shown
in Table 2.1. The key to preventing the kind of destructive polymerization that we saw
with BrCH
2
CH
2
OH is to heat very carefully and slowly, avoiding high temperatures. A
temperature of 30°C should suffice for ICH
2
CH
2
OH, since a vapor pressure of 3.0 Torr in
1 atm He was sufficient for our BrCH
2
CH
2
OH experiments.

22

Figure 2.2 Cross-section schematic of the heated nozzle for experiments on low vapor
pressure haloethanol species, XCH
2
CH
2
OH.

23

Figure 2.3 Heated nozzle side view schematic.

24

Figure 2.4 Heated nozzle front view.

25

Figure 2.5 Vial containing liquid XCH
2
CH
2
OH.

26

Temperature, ◦C Pressure, Torr
20 0
25 3.2
30 3.6
35 4.6
40 5.8
45 7.6
50 9.1
55 11.1

Table 2.1 Vapor pressure of ICH
2
CH
2
OH vs. temperature.

27
2.3 Diffusion Pump Safety Interlock Design
The turbomolecular pump used to evacuate the source region of the vacuum
chamber in the CH
3
CHOH experiments was replaced by a diffusion pump in the
CH
2
CH
2
OH experiments because of the possible damage to the turbo pump from
corrosion due to the presence of halogens. Diffusion pumps are very robust and are not
prone to the problems associated with contaminants and misalignment that affect
turbomolecular pumps. There are hazards associated with diffusion pumps that need to
be addressed, however, before they can be used safely and reliably. To begin, they must
be uninterruptedly water-cooled so as not to overheat. Overheating can lead to
contamination of the clean vacuum chamber with the very low-vapor-pressure diffusion
pump oil, which is then difficult to remove from the system. Even at normal operating
temperatures there is a risk of oil-vapor leakage onto clean surfaces, which is addressed
with the use of a liquid nitrogen-cooled baffles system placed between the mouth of the
pump and the entrance to the chamber. There are specific procedures that need to be
followed to minimize the transfer of oil vapor to the vacuum chamber. There is also
some danger of damage to vacuum components in the case of overheating.
Another concern is the backing pressure generated on the diffusion pump by the
roughing pump. The inside of the pump must be evacuated to a pressure below 500
millitorr before the heating apparatus can be engaged. This is to ensure that there is not
enough oxygen present in the pump to ignite at the high temperatures generated by the
heater, which would lead to an explosion in the pump and a potential hazard from the

28
rapid ejection of metal parts. For this reason the pressure in the roughing line of the
diffusion pump must be monitored at all times.
To address these issues, an electrical interlock system was designed to monitor
temperature, cooling water flow, and roughing-line pressure that breaks the electrical
connection in the case of malfunction of any of the systems. This relay system is
depicted schematically in Figure 2.6. A 0.25 GPM brass flow meter switch is attached to
the cooling water line just before it connects to the diffusion pump inlet. A cessation of
water flow within the line opens a switch that breaks the flow of electricity to the heater.
A vacuum relay apparatus (Terranova 924A) monitors the pressure in the roughing line
and cuts the electrical connection if the pressure rises above a value set by the user. The
diffusion pump is protected from back-flow of the roughing pump oil by a filter attached
to the roughing inlet. On our relay system the maximum value was set at 300 millitorr to
provide an additional margin of safety. Finally, the temperature of the heater itself is
monitored using a bi-metal switch. A failure in any of these systems that leads to high
temperature or pressure immediately opens the circuit providing electricity to the
resistive heater mounted at the bottom of the pump and switches the pump off.
Reinstatement of the proper working conditions automatically closes the circuit and turns
the pump back on.

2.4 Sum Frequency Generation
In order to achieve the very short wavelengths required to photolyze bromoethanol, sum
frequency generation (SFG) of the pump laser radiation was

29

208 V
2-phase
~
~
120 V
1-phase
Rel 1
Rel 2
Diffusion
Pump
1 2
3
1. Flow-meter interlock switch..
2. Pressure interlock switch.
3. Temperature interlock switch.
208 V
2-phase
~
208 V
2-phase
~
~
120 V
1-phase
Rel 1
Rel 2
Diffusion
Pump
1 2
3
1. Flow-meter interlock switch..
2. Pressure interlock switch.
3. Temperature interlock switch.

Figure 2.6 System of interlocks to protect diffusion pump from overheating and
explosion. 120 V applied across the two relays closes the circuit between the 208 V, 2-
phase source and the resistive heating element on the pump. All three interlock switches
must be closed in order for the circuit to provide the 120 V drop across the relays.

30
used. To achieve SFG, the radiation generated by the dye laser is first doubled in a non-
linear crystal of the appropriate cut and material, then passed through a polarizer to
polarize the doubled radiation parallel to the fundamental radiation. The fundamental
and doubled radiation are then mixed in another non-linear crystal to create the desired
output wavelength.
Fundamental radiation between 606-645 nm is generated by a Nd:YAG-pumped
dye laser system (Continuum, PL8000/ND6000) using Rhodamine 640 or DCM dye,
which generates an output at the desired wavelength between 20-30 mJ pulse energy.
Next, radiation at wavelengths 303-323 nm is generated by doubling the fundamental
output in a KDP “C” (54.9º) crystal (Inrad) using second-harmonic generation (SHG).
One problem encountered in generating the second harmonic frequency with substantial
efficiency is the difference in phase velocity between the fundamental radiation ω and the
second harmonic 2ω, which can lead to interference effects that diminish the output of
doubled radiation.
59
In KDP crystals, if the fundamental light is a linearly polarized
ordinary wave, the doubled light will exit in the form of an extraordinary wave. The
crystal is rotated to adjust the angle of incidence of the laser beam with respect to the
optic axis so that the index of refraction for the ordinary wave will precisely equal that
for the extraordinary wave, which leads to constructive interference in the waves of
doubled light and a large conversion efficiency. However, after the light is doubled in
the KDP crystal, the fundamental and doubled frequencies must then be mixed in another
non-linear medium to generate the 202-215 nm light. This requires that both frequencies
be vertically polarized with respect to the orientation of the BBO “0” (68.5º) crystal

31
(Inrad). To achieve this condition, the light exiting the KDP crystal is passed through a
multiple-order wave plate. The waveplate was not chosen for its suitability at any
particular frequency, but rather is oriented such that the polarization of a large fraction of
the doubled light is changed while retaining enough of the original polarization of the
fundamental in order for mixing to occur. This is done purely experimentally by
monitoring the polarities of both ω and 2ω with an analyzer. Finally, the two frequencies
are passed through the BBO crystal, which is rotated on its mount to obtain the proper
index matching conditions for third-harmonic generation. The tripled light is separated
from the doubled and fundamental light by UV Harmonic Separator (Inrad, 752-104) and
the tripled 202-215 nm light is aligned through the interaction region of the vacuum
chamber. The four prisms maintain the alignment of the ultraviolet light as the
wavelength is changed.

32
2.5 References

(1) Aristov, V.; Conroy, D.; Reisler, H. Chemical Physics Letters 2000, 318, 393.

(2) Conroy, D.; Aristov, V.; Feng, L.; Reisler, H. J Phys Chem A 2000, 104, 10288.

(3) Conroy, D. G. Rydberg State of an Open Shell Species: Characterization and
Photophysics of the 3pz State of CH2OH, University of Southern California, 2000.

(4) Hecht, E. Optics, Fourth ed.; Addison Wesley: San Francisco, 2002.

33
Chapter 3
Electronic Spectroscopy and Photodissociation Dynamics of the 1-
Hydroxyethyl Radical CH
3
CHOH
3.1 Introduction
The 1-hydroxyethyl radical CH
3
CHOH is a common intermediate in atmospheric
and combustion processes.
1-13
It appears in the photochemistry of the ethoxy radical,
3,6
as
a product of the reaction of halogen atoms with ethanol,
1,2,10,14-16
and is involved in the O
+ C
2
H
5
and OH + C
2
H
4
reactions.
4,7,8,11,12
Given the current interest in ethanol as a fuel
source, it is desirable to understand the excited-state chemistry and dissociation dynamics
of the 1-hydroxyethyl species as a possible product of ethanol combustion. Despite its
importance, there have been limited theoretical
4,5,9,11
and experimental
1,10,13,17
studies
performed on this radical. An unstructured electronic absorption spectrum at 300 K has
been reported in the region 230-300 nm with a maximum absorption cross section of 3.6
x 10
-18
cm
2
.
1
In this region the absorption increases strongly toward shorter wavelengths,
but the measurement ends abruptly at 230 nm. Additionally, a REMPI signal from
CH
3
CHOH has been reported in the range 430-460 nm, which likely arises from 2 + 1
ionization of the radical via a Rydberg state.
17
In accordance with the Rydberg formula,
and using the observed ionization potential of 6.64 eV
10,13
this must be one of the 4p
levels. Unfortunately, these studies provide very little characterization of the observed
electronic states.
A photoionization study by Ruscic and Berkowitz found the adiabatic ionization
potential of the radical to be <6.83 eV.
10
This value was later more carefully evaluated

34
by photoelectron spectroscopy and determined to be 6.64 ± 0.3 eV.
13
The photoelectron
spectra revealed a progression of 1600 ± 30 cm
-1
in the C-O vibrational stretch of the ion,
which is consistent with the removal of the radical electron from an anti-bonding orbital
centered on the C-O bond and consequent shortening of the bond.
13
In addition to these
experimental works, a few theoretical studies on CH
3
CHOH and its isomers have
calculated the minimum geometries of the radical and cation species and discuss the
symmetry of the equilibrium structures.
4,5,9,11

CH
3
CHOH is very similar in geometry to the hydroxymethyl radical CH
2
OH,
which has been the subject of numerous experimental and theoretical studies, including
spectroscopy of the electronic states performed in our lab.
18-45
Like CH
2
OH, which also
has a radical electron centered about the C-O bond, the CH
3
CHOH belongs to the C
1

symmetry point group.
20,21,37-39

In CH
2
OH, however, low barriers to CH
2
inversion (140 cm
-1
) and internal O-H
torsion (1643 cm
-1
) allow for two rapidly interconverting mirror image equilibrium
structures, which leads to an average planar symmetry of the vibronic wavefunction.
37

Given the many similarities to hydroxymethyl, it was first assumed that 1-hydroxyethyl
also contains low-frequency wag and torsional modes which create a plane of symmetry.
Planar symmetry places the radical in the C
s
symmetry point group. In CH
2
OH, the anti-
symmetric nature of the ground state wavefunction about the C
s
plane gives this state A"
symmetry.
20-23,37
The wavefunctions of the 3s and 3p
x
Rydberg states are symmetric
about the plane; they possess A' symmetry, which causes electronic transitions between
the ground state and these states to have a perpendicular transition moment.
20,21,29,30,37

35
The 3p
z
state has A" symmetry, and the transition to the ground state has a parallel
moment.
18,20-23,37
An analogous symmetry in both the ground state and the transitions to
Rydberg states was expected to be observed in CH
3
CHOH. The viability of the C
s

symmetry model was confirmed by the anisotropy seen in the electronic transitions
discussed in Section 3.4. The geometry of the hydroxyethyl and -methyl species is
similar enough that the experimentally observed quantum defects in CH
2
OH were
successfully applied to the Rydberg formula to locate the Rydberg states of CH
3
CHOH
using its ionization potential. The accuracy of the predictions based on CH
2
OH can be
seen in Table 3.1, which shows the photon energy necessary to access the onset of the 3n
l

Rydberg states in both radical species. Here the predicted origin of each state using
δ
CH2OH
is compared to the spectroscopically observed onset, and a value of the quantum
defects δ for the Rydberg states of CH
3
CHOH are given.

3.2 Experimental Details
The CH
3
CHOH radical is produced efficiently by the reaction of Cl with ethanol.
1-hydroxyethyl radical is the most stable product isomer, and is produced with a >95%
yield:
14-16

CH
3
CH
2
OH + Cl → HCl + CH
3
CHOH ΔH=0.46 eV (1.23 kJ/mol) (1)

A mixture of 2.6% CH
3
CH
2
OD, CD
3
CH
2
OH (99.5% and 99% respectively,
Aldrich, used without further purification) or CH
3
CH
2
OH (99.98% AAPER Alcohol,

36
used without further purification) and ~1% Cl
2
(Matheson Tri Gas, High Purity) in He at
2 atm total pressure is prepared in a 4-L glass bulb. A piezoelectrically driven pulsed
nozzle operating at 10 Hz introduces this mixture into the source region of the vacuum
chamber. At the tip of the pulsed nozzle is a 4 mm long x 1.0 mm ID quartz tube. Cl
2
is
photodissociated by 355 nm radiation from a Nd:YAG laser (Spectra Physics, GCR-11; 5
mJ, focused by 30 cm f.l. cylindrical lens) directed at the expansion end of the quartz
tube. This is done to assure that the photolysis event occurs just before expansion of the
beam to minimize secondary reactions. Cl atoms then react with CH
3
CH
2
OH (or one of
the deuterated isomers) to create CH
3
CHOH. The radical beam exits the quartz tube and
is cooled in the expansion into the vacuum chamber to a rotational temperature of ~10-15
K.
18

The dissociation (pump) radiation is generated by a Nd:YAG pumped OPO/OPA
laser system (Continuum PL8000/Sunlite/FX-1; 0.5-2 mJ at 250-400 nm, 14-20 mJ at
470-520 nm; 25 cm f.l. lens). H and D fragments are detected by the probe laser via a
1+1' REMPI scheme. The doubled output (~365 nm, 2 mJ) of a Nd:YAG pumped dye
laser system (Continuum, NY81/ND6000, LDS 751) is focused (20 cm f.l. lens) into a
1000 Torr mixture of Kr (25%) and Ar (75%). The tripled ~121.6 nm radiation is
focused in the detection chamber (MgF
2
7.5 cm f.l. lens). In the anisotropy studies on H
and D, the polarization of the pump laser is controlled by a photoelastic modulator (PEM-
80, Hinds International Inc.). All TOF distributions are recorded with two polarizations
of the laser, parallel and perpendicular to the TOF axis, which allows for calculation of
the recoil anisotropy parameter β.
46

37
While it provides a more accurate speed distribution by restricting velocity data to
a small region parallel to the TOF tube axis, employing a core-sampling regime for
detection greatly reduces the surface area of the detector. This lessens the number of ions
collected, and makes the low ion signal from the observed processes in CH
3
CHOH
difficult to detect. The core was removed from the front of the microchannel plate
(MCP) in order to create a larger "core" with the surface area of the MCP itself. The
voltage is lowered as much as possible while preserving a reasonable signal-to-noise ratio
and also maintaining space-focusing conditions.
47
In this regime, the MCP collects all
ions with a perpendicular component of translational energy (E
T
) lower than 0.6 eV, but
provides increasingly better discrimination against ions with off-axis velocities at higher
E
T
.
48
In order to find the position and width of the low- E
T
peak in the photofragment
distributions, the spectra were fitted with a model velocity distribution composed of two
Gaussian distributions.
46
In the CH
3
CHOH experiments, the TOF apparatus was
calibrated using product E
T
distributions obtained in CH
2
OH photodissociation.
29

3.3 REMPI and Photofragment Yield Spectra
While the ionization potentials (IP) of radicals are usually low (<10 eV),
CH
3
CHOH has an unusually low IP of 6.64 eV.
13
This low IP applied to the Rydberg
formula yields low-lying Rydberg states in the radical. These states are in fact the
lowest-lying excited electronic states, and they are located at lower energies than the
valence states. As discussed above, the electronic transitions to the Rydberg states in 1-
hydroxyethyl were identified using the quantum defects taken from hydroxymethyl

38
radical and applying them to the Rydberg formula along with the ionization potential of
CH
3
CHOH. The 3s and 3p
z
states of CH
3
CHOH were located easily by scanning the
pump laser in the region of the predicted origins. The onset of the 3p
x
state, which was
detectable in CH
2
OH, was unidentifiable by hydrogen photofragment spectroscopy due to
significant lifetime broadening and the large background from overlap with the
structureless 3s state.
31
In hydroxymethyl the 3p
y
state was calculated to have very low
oscillator strength and has not been observed in either radical.
37
In these experiments, the
isotopes CD
3
CHOH and CH
3
CHOD were also used to detect isomerization, determine
selectivity of dissociation processes, and in some cases to minimize hydrogen
background signal by detecting deuterium.
3s region. Owing to strong couplings between this lowest-lying Rydberg state and the
ground state, the CH
2
OH radical upon excitation to the 3s state rapidly predissociates
along the O-H(D) bond coordinate.
29-32,36
A similar phenomenon was expected in
CH
3
CHOH. The nature of REMPI spectroscopy requires that the excited electronic state
have a lifetime long enough to absorb an additional photon and be ionized. Due to its
rapid fragmentation, a REMPI spectrum of the radical through the 3s state cannot be
observed. Nonetheless, spectroscopy on this Rydberg state can be performed using the
hydrogen fragment created in the dissociation. The Lyman-α states of hydrogen and
deuterium have large oscillator strengths, and consequently REMPI of H and D through
these states at ~121.6 nm produces a large signal upon the predissociation of the
electronically-excited radical.
Figure 3.1 shows the REMPI spectrum of the 3s state of CH
3
CHOD obtained by

39
0.73 32,360 (4.01) 33,684 (4.18) 0.65 41,053 (5.09) 3p
y
----- ---------- 27,695 (3.43) 0.94 35,004 (4.34) 3p
x
1.20 19,600 (2.43) 18,527 (2.29) 1.23 25,971 (3.22) 3s
δ T
0
E
,cm
-1
(eV)
Experiment
T
0
E
,cm
-1
(eV)
Estimated
δ T
0
E
,cm
-1
(eV)
Experiment
Hydroxyethyl (CH
3
CHOH) Hydroxymethyl (CH
2
OH) State
0.73 32,360 (4.01) 33,684 (4.18) 0.65 41,053 (5.09) 3p
y
----- ---------- 27,695 (3.43) 0.94 35,004 (4.34) 3p
x
1.20 19,600 (2.43) 18,527 (2.29) 1.23 25,971 (3.22) 3s
δ T
0
E
,cm
-1
(eV)
Experiment
T
0
E
,cm
-1
(eV)
Estimated
δ T
0
E
,cm
-1
(eV)
Experiment
Hydroxyethyl (CH
3
CHOH) Hydroxymethyl (CH
2
OH) State
a b

a. Ref. 22, 31, 33. b. This work.

Table 3.1 Comparison of quantum defects (δ) and energy of onset (T
0
E
)
of Rydberg states in CH
2
OH and CH
3
CHOH. δ are experimental values.

40
monitoring the D photofragment. This state has an onset at 19600 ± 100 cm
-1
(2.43 ±0.01
eV), close to the predicted onset of 18,530 cm
-1
, and is broad and unstructured owing to
its short lifetime. A similar spectrum is seen by monitoring H fragments from
CH
3
CHOH, but the signal-to-noise ratio is slightly better in our deuterium spectra due to
a small permanent hydrogen background in the vacuum chamber from trace
hydrocarbons.
3p
z
region. The 3p
z
Rydberg state is also expected to have a short lifetime, on the order
of half a picosecond or less.
18,22,23
Experimentally we find that the lifetime is still long
enough to observe a REMPI signal from the CH
3
CHOH
+
ion. Figure 3.2 shows the onset
of the 3p
z
state at 32,360 ± 70 cm
-1
(4.01 ± 0.01 eV), and a vibrational progression in the
C-O stretch motion at 1560 ± 100 cm
-1
intervals. As discussed earlier, this vibration
results from the promotion of the radical electron in an anti-bonding orbital about the C-
O bond σ*
C-O
to a non-bonding Rydberg orbital.
13
The bond is thus abruptly shortened
and the stretch motion is activated. An analogous progression is also observed in
CH
2
OH.
18,22

Comparing the 3p
z
spectra of hydroxymethyl and 1-hydroxyethyl in Figure 3.2, it
is evident that the spectral lines are much broader in 1-hydroxyethyl. This indicates that
the lifetime of the electronic state in this radical is shorter. Most likely this is a result of
the fact that, not only are the electronic states of CH
3
CHOH closer to the ground state
energetically, but there are also more vibrational degrees of freedom in the larger radical,
and consequently more low-energy out-of-plane modes that can efficiently
couple the A" 3p
z
state to the A' ground state on which it dissociates. This will be

41
remarked upon further in the discussion of the photofragment spectroscopy of the
radical in Section 3.4.
The onset of the 3p
x
state (predicted at around 35,004 cm
-1
) is not observed in our
hydrogen photofragment measurements due to the domination in this region of absorption
to the 3s state. The 3p
x
region in hydroxymethyl was detectable via REMPI of the
radical,
31
but the shorter lifetime of the excited 1-hydroxyethyl precludes resonant
ionization.

3.4 Time of Flight and Energy Spectra of H and D fragments
As in CH
2
OH, it is evident from the broad spectral lines in the 3p
z
spectrum and
the lack of any structure in the 3s state that the electronically excited radical is rapidly
dissociating. This is confirmed by a large H (or D) signal that appears upon resonant
excitation to the Rydberg states. These excited states are bound electronic states and
should dissociate adiabatically to electronically excited products. However, the large
kinetic energy release in the
dissociation indicates that products are formed in the ground state. Our group's previous
work on hydroxymethyl showed that its Rydberg states are coupled to
the ground state via a conical intersection and predissociate along the ground state
potential energy surface.
29-32,36
Similar dissociation dynamics are anticipated in
CH
3
CHOH. There have been numerous calculations of dissociation and isomerization
barriers on the ground state potential energy surface of the radical.
4,5,9,11
The lowest
barrier channels calculated are

42
19000 19500 20000 20500 21000 21500
0.0
0.2
0.4
0.6
0.8
1.0
D
+
Normalized signal
Frequency, cm
-1

Figure 3.1 D photofragment yield spectrum in the region 19,000 cm
-1
- 21,500 cm
-1
.
Background signal is subtracted, and the signal is normalized to the OPO/OPA laser
energy.

43
32000 33000 34000 35000 36000

Frequency, cm
-1
41000 42000 43000 44000

Figure 3.2 (top) 2 + 1 REMPI of CH
2
OH in the region of absorption to the 3p
z
state
(adapted from ref. 18). (bottom) 2 + 2 REMPI of CH
3
CHOH in the region of absorption
to the 3p
z
state. The lowest energy band of each transition is the origin band. Both
spectra show a vibrational progression in the C-O stretch at ~1600 cm
-1
.

44
CH
3
CHOH    CH
3
CHO + H ΔH=1.04 eV (1)
   CH
2
CHOH + H ΔH=1.46 eV (2)
   CH
3
+ CHOH ΔH=3.13 eV (3)

The molecular elimination channels CH
3
CO + H
2
and CH
3
+ CO + H
2
involve
tight transition states and are not expected to compose a large fraction of the products.
29,30

Isomerization to ethoxy radical can lead to additional dissociation pathways.
11
However,
we conducted isotopic substitution studies that suggest that no isomerization takes place
in the excited 1-hydroxyethyl radical.
TOF spectroscopy on the dissociating radical provides useful information both
about the kinetic energy release in the dissociation and the symmetry of the electronic
transition. TOF spectra are obtained by first choosing a wavelength at which the radical
absorbs strongly, as determined from the photofragment yield spectra. The radical is then
excited at that wavelength and the time of arrival of H(D) atoms is monitored. In both
CH
2
OH and CH
3
CHOH, dissociation from the 3s state reveals a strong negative recoil
anisotropy β in the H(D) fragment distribution (β
3s
=-0.7 ± 0.1 in both radicals).
29,30
This
polarization dependence is visible in Figure 3.3. Here a strong signal is obtained with
perpendicular polarization of the laser relative to the TOF tube, while little signal is seen
with parallel polarization. The negative anisotropy parameter indicates a perpendicular
transition moment and agrees with absorption from the ground A" state to the 3s A' state.
The observation of similar recoil anisotropy from the 3s state in both radicals suggests
that the transition dipole moment in both is similar in direction relative to the O-H bond,

45
which is the only bond dissociation channel observed at this energy. With this in mind,
the first two Rydberg states in CH
3
CHOH are assigned as 1
2
A'(3s) and 2
2
A'(3p
x
).
Knowing the length of our TOF tube and the voltage applied to the ion optics, the
temporal traces can be converted easily to translational energy distributions to obtain
information about the threshold energy needed for dissociation.
47
Using conservation of
energy, the maximum kinetic energy release can be subtracted from the photon energy of
the excitation to supply the energy required to break a O-H(D) or C-H(D) bond, thus
yielding the respective D
0
values. Using hydrogen products as an example, the
relationship is

D
0(H-M)
= hν – (KE
H+M
+ E
M
) (4)

KE
H+M
is the kinetic energy release in the dissociation process that produces H fragments,
hν is the photon energy, and E
M
is the internal energy of the molecular co-fragment
product. The H fragment with the highest velocity corresponds to the bond cleavage
process with the highest kinetic energy release KE
H+M
and thus the lowest internal energy
E
M
in the molecular fragment. Assuming that E
M
=0 for this small fraction of products,
we are left with D
0
for the bond cleavage process. This value, in turn, helps to identify
the molecular product of the dissociation by comparison with theory.
Near the onset of the 3s state, dissociation channels (1) and (2) are energetically
accessible. Both channels have similar barrier heights on the ground state potential
energy surface (calculated at 1.52 and 1.58 eV, respectively).
11
Were the available

46
energy distributed statistically among the vibrational modes upon internal conversion to
the ground state, products from both channels would be expected. However, our
translational energy distributions reveal only one channel in the dissociation. The large
kinetic energy release, with most of the available energy deposited into fragment
translation, agrees with a non-statistical process. This is confirmed by isotope studies
using CH
3
CHOD and CD
2
CHOH. Both yield only the H(D) fragment from O-H(D) bond
cleavage, and the H and D distributions from both are identical and display only one
feature, seen in Figure 3.4. The selectivity of this process may be owed to the nature of
the coupling between the 3s state and the ground state. In CH
2
OH, calculations show an
efficient conical intersection between the two states in the region along the potential
energy surface of an elongated O-H(D) bond.
36
As the radical goes through the conical
intersection it predissociates predominantly via cleavage of that bond. Pumping with
2.63 eV at the onset of absorption to the 3s state yields a maximum center-of-mass (c.m.)
translational energy (E
T
) of 1.53 ± 0.1 eV. Using Equation (4), we obtain D
0
= 1.1 ± 0.1
eV for dissociation via channel (1). second conical intersection located at much higher
energies in CH
2
OH leads to C-H(D) bond cleavage.
29,30,36
As we pump CH
3
CHOH with
higher energy photons and reach the region where the 3p
x
state is predicted to lie, a
change begins to appear in the kinetic energy spectra. In addition to the high translational
energy feature, a low E
T
feature becomes apparent. Figure 3.5 shows the c.m. E
T

distribution of H fragments obtained by exciting CH
3
CHOH at 31,250 cm
-1
(3.87 eV).
The low E
T
peak does not appear in the D photofragment distribution of CD
3
CHOD in
Figure 3.6. The anisotropy in the region of the high E
T
peak is still negative but the

47
-40 -20 0 20 40

H
+
signal, arb. units
Polarization:

||
Relative TOF, ns

Figure 3.3 Background subtracted H fragment time-of-flight spectrum from excitation at
21,276 cm
-1
. Spectra are parallel (solid line) and perpendicular (dashed line) to the TOF
axis. Zero time indicates fragments with no recoil.

48
anisotropy parameter β is lower than at the 3s onset. β=-0.4 ± 0.1 in this region for both
CH
3
CHOH and CH
3
CHOD. The small peak at low translational energy exhibits no
anisotropy, with β=0.0 ± 0.1. The appearance of the second peak at shorter
wavelengths suggests that we are accessing the second conical intersection along the C
(2)
-
H(D) stretch coordinate at these energies. Another interesting feature in this spectrum is
the difference between the maximum translational energy of the fragments, E
Tmax
, and the
energy available to the dissociation process, E
avail
. Previously, we assumed that the
fastest fragments correlated with a cofragment with no internal energy, in which case
E
Tmax
= E
avail
. Applying D
0
to the conservation of energy equation at this new energy,
however, gives us E
avail
=E
hv
-D
0
=3.87-1.2=2.67 ± 0.1 eV, while E
Tmax
from the spectrum
is 2.2 ± 0.1 eV. This implies a minimum internal energy in the cofragment of E
avail
-
E
Tmax
=0.47 ± 0.14 eV. In order for the 1-hydroxyethyl radicals in its Rydberg states to
dissociate along the ground state PES, coupling between the two surfaces must occur
effected by the activation of vibrational modes. The corresponding increase in internal
energy in the acetaldehyde cofragment with increasing excitation energy is indicative of
the transfer of energy in the radical vibrational motion to internal modes of the
dissociation products. With decreasing wavelength, there is more energy available to be
deposited in out-of-plane modes. This reduces the anisotropy parameter in the
perpendicular transition.
As the dissociation energy is further increased, the intensity of the low E
T
peak
grows larger. Figures 3.7 and 3.8 show D and H spectra from dissociation of the radical
at 35,460 cm
-1
. This photon energy accesses the region beyond the onset of the 3p
z
state.

49
Figure 3.7 shows the absence of the low-energy peak from the CH
3
CHOD isotope, and
the anisotropy parameter is now β=-0.2 ± 0.1 for the peak at high E
T
. Figure 3.8 shows a
hydrogen spectrum from the non-deuterated isotope. It is clear from the distribution that
not only does the C
(2)
-H dissociation channel become more prominent, but the anisotropy
parameter in the O-H dissociation becomes less negative with increasing excitation
energy. The dependence of β on excitation energy is shown in Figure 3.9. In addition,
the difference between E
T
and E
avail
becomes larger with increasing energy, indicating
that more energy is deposited into molecular vibrations at shorter wavelengths. Beyond
the origin of the 3p
z
onset, a reduction in the anisotropy in the spectra also results from
radicals excited via a parallel transition. It is surprising that at high pump energies the
electronic transition does not show a strong positive β parameter, as would be expected
from the parallel transition from the ground state to the 3p
z
state, but rather exhibits a loss
of polarization. There are a few considerations that can lead to the reduction in
anisotropy at these energies. First, there may still be contributions from the broad 3s and
3p
x
states in the distribution. Second, due to unique properties associated with planar
symmetry, both horizontal and vertical polarizations of the pump laser can lead to a
strong photofragment signal. The plane allows for two different orientations of the
radical that result in absorption parallel to the plane, and thus, counter-intuitively, both
parallel and perpendicular electric fields can lead to a parallel-type distribution of H(D)
fragments. The isotropic nature of the C
(2)
-H dissociation in channel (2) can be similarly
rationalized, and may also be a result of the favored geometries that lead to non-adiabatic
transitions.

50
In order to determine the energy required for C
(2)
-H(D) dissociation to produce
channel (2), deconvolution of the H spectrum was performed assuming two Gaussian
distributions of E
T
. This procedure gives a value for C
(2)
-H bond fission in the vinyl
alcohol channel of D
0
< 2.7 ± 0.1 eV. This is larger than the calculated value of 1.46 eV,
but lower than the energy required for dissociation from C
(1)
, with D
0
=3.37 eV.
11

3.5 Conclusions
Electronic transitions to low-lying Rydberg states and photodissociation dynamics
of the CH
3
CHOH radical and its deuterated isotopes were studied in a molecular beam
in in the region 19,600-37,000 cm
-1
. REMPI studies of the radical cation and
photofragment spectroscopy on H and D atoms from predissociation of the
electronically-excited radical were used to characterize these Rydberg states. The
Rydberg formula was used with an experimental quantum defect, δ, taken from CH
2
OH
radical in order to identify and assign the electronic states, and these assignments were
confirmed with anisotropy studies. The predominant absorption in the observed region
is to the 3s Rydberg state, assigned as 1
2
A'(3s), with an onset at 19,600 ± 100 cm
-1
(2.43
± 0.01 eV). In excitation to the 3s region, only O-H(D) bond cleavage is observed
despite the similar barrier height and energetic accessibility of C-H(D) bond cleavage
from the C
(2)
. The selectivity of the dissociation at these energies is owed to the nature
of the conical intersection between the 3s and ground states in the region on the PES of
an elongated O-H(D) bond.

51
The onset of the 2
2
A'(3p
x
) state cannot be determined directly from the
photofragment spectra, but the appearance of a new low E
T
isotropic peak from channel
(2) dissociation indicates that another bond-breaking channel may be accessed at energies
above the predicted 3p
x
onset of 27,695 cm
-1
(3.43 eV). This channel, C
(2)
-H bond
cleavage, first appears only ~1 eV above its thermochemical threshold, and appears to
coincide with accessing a second conical intersection leading to C
(2)
-H(D) bond cleavage.
The origin band of the transition to the 2
2
A"(3p
z
) state lies at 32,360 ± 70 cm
-1
(4.01 ±
0.01 eV). Excitation to this state is characterized by a vibrational progression in the C-O
stretch with 1560 ± 100 cm
-1
intervals. Photofragment spectroscopy in the region of the
onset of 3p
z
excitation reveals dissociation processes from both channels (1) and (2) to
produce both acetaldehyde and vinyl alcohol molecular co-fragments. D
0
values for
dissociation to these channels are found to be 1.1 ± 0.01 eV and < 2.7 ± 0.01 eV,
respectively. As the photon energy of the excitation becomes larger, more energy is
deposited into internal degrees of freedom in CH
3
CHOH, which allows for more efficient
internal conversion from the excited state to the ground state and leads to vibrationally-
excited molecular co-fragments upon H(D) loss.

52

Figure 3.4 The c.m. E
T
distribution of H photofragments following the 1
2
A(3s)←1
2
A
transition at 21,276 cm
-1
(2.63 eV). The arrow indicates the available energy, E
avail
. The
β parameter is plotted as a function of E
T
in the top panel.

53

Figure 3.5 The c.m E
T
distribution of H fragments following 31,250 cm
-1
(3.87 eV)
excitation. The arrow indicates the available energy, E
avail
. The β parameter is plotted as
a function of E
T
in the top panel.

54

Figure 3.6 The c.m E
T
distribution of D fragments following 31,250 cm
-1
(3.87 eV)
excitation. The available energy, E
avail
differs slightly from the H spectrum due to the
difference in zero-point energies of the deuterated isotopes. The β parameter is plotted as
a function of E
T
in the top panel.

55

Figure 3.7 The c.m E
T
distribution of H fragments following 31,250 cm
-1
(3.87 eV)
excitation. The β parameter is plotted as a function of E
T
in the top panel.

56

Figure 3.8 The c.m E
T
distribution of D fragments following 31,250 cm
-1
(3.87 eV)
excitation. The β parameter is plotted as a function of E
T
in the top panel

57

Figure 3.9 Recoil anisotropy parameter β as a function of excitation energy.

58
3.6 References

(1) Anastasi, C.; Simpson, V.; Munk, J.; Pagsberg, P. Chem Phys Lett 1989, 164, 18.

(2) Anastasi, C.; Simpson, V.; Munk, J.; Pagsberg, P. J Phys Chem-Us 1990, 94,
6327.

(3) Choi, H.; Bise, R. T.; Neumark, D. M. J Phys Chem A 2000, 104, 10112.

(4) Cleary, P. A.; Romero, M. T. B.; Blitz, M. A.; Heard, D. E.; Pilling, M. J.;
Seakins, P. W.; Wang, L. Phys Chem Chem Phys 2006, 8, 5633.

(5) Curtiss, L. A.; Lucas, D. J.; Pople, J. A. J Chem Phys 1995, 102, 3292.

(6) Faulhaber, A. E.; Szpunar, D. E.; Kautzman, K. E.; Neumark, D. M. J Phys Chem
A 2005, 109, 10239.

(7) Hoyermann, K.; Olzmann, M.; Seeba, J.; Viskolcz, B. J Phys Chem A 1999, 103,
5692.

(8) Lindner, J.; Loomis, R. A.; Klaassen, J. J.; Leone, S. R. J Chem Phys 1998, 108,
1944.

(9) Nobes, R. H.; Rodwell, W. R.; Bouma, W. J.; Radom, L. J Am Chem Soc 1981,
103, 1913.

(10) Ruscic, B.; Berkowitz, J. J Chem Phys 1994, 101, 10936.

(11) Senosiain, J. P.; Klippenstein, S. J.; Miller, J. A. J Phys Chem A 2006, 110, 6960.

(12) Tully, F. P. Chem Phys Lett 1988, 143, 510.

(13) Dyke, J. M.; Groves, A. P.; Lee, E. P. F.; Niavaran, M. H. Z. J Phys Chem A
1997, 101, 373.

(14) Ahmed, M.; Peterka, D. S.; Suits, A. G. Phys Chem Chem Phys 2000, 2, 861.

(15) Rudic, S.; Murray, C.; Ascenzi, D.; Anderson, H.; Harvey, J. N.; Orr-Ewing, A. J.
J Chem Phys 2002, 117, 5692.

(16) Taatjes, C. A.; Christensen, L. K.; Hurley, M. D.; Wallington, T. J. J Phys Chem
A 1999, 103, 9805.

59
(17) Edelbuttel-Einhaus, J. H., K.; Rohde, G.; Seeba, J.;. “Proceedings of the 24th
Symposium International on Combustion”; 1992, Pittsburgh, PA.

(18) Aristov, V.; Conroy, D.; Reisler, H. Chem Phys Lett 2000, 318, 393.

(19) Bauschlicher, C. W.; Partridge, H. J Phys Chem-Us 1994, 98, 1826.

(20) Bruna, P. J.; Grein, F. J Phys Chem A 1998, 102, 3141.

(21) Bruna, P. J.; Grein, F. J Phys Chem A 2001, 105, 8599.

(22) Conroy, D.; Aristov, V.; Feng, L.; Reisler, H. J Phys Chem A 2000, 104, 10288.

(23) Conroy, D.; Aristov, V.; Feng, L.; Sanov, A.; Reisler, H. Accounts Chem Res
2001, 34, 625.

(24) Curtiss, L. A.; Kock, L. D.; Pople, J. A. J Chem Phys 1991, 95, 4040.

(25) Dóbé, S. B., T.; Tyrányi, T.; Márta, J.; Grussdorf, J.; Temps, F.; Wagner, H.G. J
Phys Chem-Us 1996, 100, 19864.

(26) Dulcey, C. S.; Hudgens, J. W. J Phys Chem-Us 1983, 87, 2296.

(27) Dulcey, C. S.; Hudgens, J. W. B Soc Chim Belg 1983, 92, 583.

(28) Dulcey, C. S.; Hudgens, J. W. J Chem Phys 1986, 84, 5262.

(29) Feng, L.; Demyanenko, A. V.; Reisler, H. J Chem Phys 2003, 118, 9623.

(30) Feng, L.; Demyanenko, A. V.; Reisler, H. J Chem Phys 2004, 120, 6524.

(31) Feng, L.; Huang, X.; Reisler, H. J Chem Phys 2002, 117, 4820.

(32) Feng, L.; Huang, X.; Reisler, H. Abstr Pap Am Chem S 2002, 223, C67.

(33) Feng, L.; Reisler, H. J Phys Chem A 2004, 108, 9847.

(34) Feng, L.; Wei, J.; Reisler, H. Abstr Pap Am Chem S 2004, 227, U246.

(35) Feng, L.; Wei, J.; Reisler, H. J Phys Chem A 2004, 108, 7903.

(36) Hoffman, B. C.; Yarkony, D. R. J Chem Phys 2002, 116, 8300.

(37) Johnson, R. D.; Hudgens, J. W. Abstr Pap Am Chem S 1996, 212, 82.

60
(38) Marenich, A. V.; Boggs, J. E. J Chem Phys 2003, 119, 10105.

(39) Marenich, A. V.; Boggs, J. E. J Chem Phys 2003, 119, 3098.

(40) Pagsberg, P.; Munk, J.; Sillesen, A.; Anastasi, C. Chem Phys Lett 1988, 146, 375.

(41) Radford, H. E. Chem Phys Lett 1980, 71, 195.

(42) Saebo, S.; Radom, L.; Schaefer, H. F. J Chem Phys 1983, 78, 845.

(43) Tyndall, G. S.; Wallington, T. J.; Hurley, M. D.; Schneider, W. F. J Phys Chem-
Us 1993, 97, 1576.

(44) Walch, S. P. J Chem Phys 1993, 98, 3076.

(45) Wei, J.; Feng, L.; Reisler, H. Abstr Pap Am Chem S 2004, 227, U330.

(46) Syage, J. A. J Chem Phys 1996, 105, 1007.

(47) Conroy, D. Rydberg State of an Open Shell Species: Characterization and
Photophysics of the 3pz State of CH2OH, University of Southern California, 2000.

(48) Karpichev, B.; Edwards, L. W.; Wei, J.; Reisler, H. J Phys Chem A 2008, 112,
412.

61
Chapter 4
D-atom products in predissociation of CD
2
CD
2
OH from the 202-215 nm
photodissociation of 2-bromoethanol
4.1 Introduction
The 2-hydroxyethyl radical, CH
2
CH
2
OH, is a subject of great interest given the
dearth of experimental work on this radical in addition to interesting qualities predicted
theoretically.
1,2
One such quality, the unusual bridged ion structure, appears to explain
the results of a photoionization study that is one of few reported experimental works on
this species.
3
In this study by Ruscic et al., a weak signal from the CH
2
CH
2
OH
+
peak
decreases more-or-less linearly to lower energies, implying that a large change in
geometrical structure accompanies ionization. Because of the lack of a sharp onset in the
ion spectrum, only an upper limit to the adiabatic I.P. can be determined. The authors
assign it as ≤ 8.35 ± 0.06 eV, and possibly ≤ 8.18 ± 0.08 eV. Quantum defects from
previous experiments cannot be employed in an attempt to interrogate the Rydberg states
of 2-hydroxyethyl, which should have geometries similar to the bridged cation. As
described in the introduction, wheras in CH
2
OH and CH
3
CHOH the lone electron is
centered over the C
(1)
-O bond,
4-8
in CH
2
CH
2
OH, the lone electron resides over C
(2)
.
1,2

An electronic structure study was performed by our group together with our theoretical
collaborators to calculate the energies of the valence and Rydberg states in CH
2
CH
2
OH.
9

In future work, we hope to observe these states spectroscopically. An electronic
absorption has been observed at room temperature in the region 210-265 nm, but no
attempt to characterize the observed electronic state has been made.
10

62
We were also drawn to study the CH
2
CH
2
OH radical because of its dissociation
products: acetaldehyde and vinyl alcohol. These products have been identified in the
decomposition of the 1- and 2-hydroxyethyl structural isomers (CH
3
CHOH and
CH
2
CH
2
OH, respectively), in addition to the major decomposition channel in
CH
2
CH
2
OH to CH
2
CH
2
+ OH.
8,11-15
Of particular interest is the vinyl alcohol product,
CH
2
CHOH. Vinyl alcohol tautomerizes readily to the more thermodynamically stable
acetaldehyde isomer upon storage and was previously postulated to exist only as a
transient intermediate in chemical reactions. Despite high barriers to isomerization in the
gas phase, it has been excluded previously from standard models of hydrocarbon
oxidation.
11
Its recent discovery in significant concentrations in flames, however, as well
as its identification as a dissociation product in electronically excited 1-hydroxyethyl
radicals, has underscored the importance of this molecule in combustion.
8,11-15
We
wished to observe it firsthand through detection of D fragments dissociating from
CD
2
CD
2
OH. The presence of the acetaldehyde tautomer in combustion reactions is well
established, but it was desirable to determine the extent of its involvement in the
dissociation of the radical. Theoretical calculations of rate constants appear to confirm
that acetaldehyde only plays a minor role in the dissociation process, especially given the
tight transition state to isomerization to CH
3
CH
2
O that must precede dissociation to
CH
3
CHO.
The addition of OH to C
2
H
4
,
11,12,14,16
which can lead to formation of vinyl alcohol
in flames, has been the subject of both experimental and theoretical studies. The reverse
reaction is dominant in the dissociation of CH
2
CH
2
OH, as has been demonstrated before

63
by using photolysis of haloethanols to generate 2-hydroxyethyl radicals.
17,18
The
aforementioned theoretical predictions assert, however, that other dissociation pathways
of CH
2
CH
2
OH, such as dissociation to vinyl alcohol and acetaldehyde, can also play a
role.
2,13
To date, no report has been published on dissociation pathways leading to H
fragments. The three relevant channels are:

CH
2
CH
2
OH → CH
2
CH
2
+ OH (major channel) (1)
D
M-OH
= 1.14 eV (109.6 kJ/mol)

CH
2
CH
2
OH → CH
2
CHOH + H (vinyl alcohol channel) (2)
D
M-H
= 1.18 eV (113.8 kJ/mol)

CH
2
CH
2
OH → CH
3
CHO + H (acetaldehyde channel) (3)
D
M-H
= 0.76 eV (72.8 kJ/mol)

where D
X-Y
is the dissociation energy of channel XY → X + Y. Relevant energies and
barriers based on the calculations of Senosiain et al.
13
are summarized in figure 4.1.
Channel (1) dominates at energies close to the dissociation threshold and has been
observed in secondary dissociation of rovibrationally excited CH
2
CH
2
OH generated by
UV photolysis of haloalcohols.
17,18
In these studies it was concluded that a large fraction
of the CH
2
CH
2
OH radicals were generated in high rotational states, which are less easily
coupled to the C-H bond cleavage coordinate. According to the rate calculations of

64
Senosiain et al.,
13
other channels, in particular the vinyl alcohol channel, should become
more important with increasing energy above the initial photolysis threshold. For
example, in the 202 nm photolysis of BrCH
2
CH
2
OH where there is ~3.14 eV available to
products after C-Br photolysis, vinyl alcohol can comprise a few percent of the products
as discussed in Section 4.4. CH
2
CH
2
OH radicals can be produced efficiently by
photolysis of haloethanols.
17-21
Hintsa et. al. photolyzed BrCH
2
CH
2
OH at 193 nm and,
using time-of-flight (TOF) spectroscopy, observed intact CH
2
CH
2
OH products with high
kinetic and internal energies.
18
Moreover, they concluded that some of the radicals had
high rotational excitation that was not available for predissociation on the experimental
timescale. Predissociation of “hot” radicals (i.e. those produced with lower product
kinetic energies/higher internal energies) to CH
2
CH
2
+ OH was detected by electron
impact mass spectroscopy.
18
Sapers and Hess dissociated BrCH
2
CH
2
OH and
ICH
2
CH
2
OH and detected OH fragments by laser induced fluorescence (LIF).
17
They
concluded that following photolysis of BrCH
2
CH
2
OH at 202 nm, OH radicals were
produced by predissociation of internally excited CH
2
CH
2
OH, whereas following
ICH
2
CH
2
OH photolysis at 266 nm, absorption of a second 266 nm photon by
CH
2
CH
2
OH was the main source of OH. Chandler et. al. used photofragment imaging of
ground-state Br(
2
P
3/2
) and spin-orbit excited Br*(
2
P
1/2
) (hereafter denoted by Br and Br*,
respectively) and determined the maximum and minimum kinetic energies of fragments
produced in the 205 nm photolysis of bromoethanol.
19
Butler and coworkers carried out
similar imaging experiments at 193 nm with higher resolution and concluded that Br was
the main product.
21
A recent imaging study of ICH
2
CH
2
OH photolyzed at 253 – 298

65

-109.6
(-111.4)
20
0
-50
CH
2
CH
2
+ OH
OCH
2
CH
3
HOCHCH
2
+ H
CH
3
CHO + H
CH
2
O + CH
3
CH
2
CH
2
OH
23.9
(22.2)
-95.4
(-97.0)
28.9
(34.7)
4.2
-36.8
-54.8
Energy/kJ mol
-1
0.0
10
-10
-20
-30
-40
-60
-70
-80
-90
-100
-110
-6.4
(0.0)
-23.8
(-22.4)

Figure. 4.1 Schematic energy diagram for the decomposition of CH
2
CH
2
OH. Energies
(in kJ/mol) are from ref 13. The corresponding barriers for CD
2
CD
2
OH, calculated in
this work, are given in parentheses.

66
nm
20
has shown that CH
2
CH
2
OH from the major dissociation channel to I*(
2
P
1/2
) +
CH
2
CH
2
OH does not undergo significant secondary dissociation. However, the authors
conclude that 45–60% of the CH
2
CH
2
OH radicals correlated with I(
2
P
3/2
) do dissociate as
a result of second photon absorption or dissociative photoionization. Based on the
published room temperature absorption spectrum of CH
2
CH
2
OH between 210 and 265,
10

it was suggested that the radical could absorb an additional photon at the 266 nm
photolysis wavelength and dissociate.
In this chapter, I report the detection of D photofragments following 202–215 nm
photolysis of BrCD
2
CD
2
OH and conclude that they are generated by predissociation of
internally excited CD
2
CD
2
OH. The partially deuterated product was used in order to
distinguish between O- and C-bound hydrogens, and also to improve the signal-to-noise
ratio in the experiment. D atoms were detected by 1+1’ resonance enhanced multiphoton
ionization (REMPI) via its strong Lyman-α transition (~121.6 nm), and their TOF
distributions were determined. Bromoethanol was chosen as the precursor because it
absorbs strongly around 202 nm, and at this wavelength even after cleavage of the C-Br
bond a fraction of the product 2-hydroxyethyl radicals should contain enough internal
energy to predissociate via both the OH and D product channels.
13,15
D fragment
maximum velocities and kinetic energies are used here to infer dissociation pathways.
The theoretical calculations predict the energy dependence of the branching between
channels (1) to (3) in CD
2
CD
2
OH dissociation via RRKM calculations based on the
model of ref 4.

67
4.2 Experimental Details
Unlike the CH
3
CHOH experiments, the primary purpose of the CH
2
CH
2
OH
(CD
2
CD
2
OH) experiments was to detect product channels of the vibrationally and
rotationally excited radical as it dissociates from its ground electronic state. The
BrCD
2
CD
2
OH precursor is first photolyzed with 202-215 nm light. The photolysis step
occurs in the interaction region, which is in contrast to CH
3
CHOH, where radical
production occurred in the source region of the vacuum chamber. After photolysis, D
atoms are ionized by the probe laser radiation and detected by the MCP. BrCD
2
CD
2
OH
(Aldrich, 98 atom % D, used without further purification) is evaporated to its vapor
pressure of about 3 Torr at 25°C into a 4 L glass bulb filled with He to a final pressure of
1 atm. The pulsed nozzle described in detail in Chapter 2 introduces this mixture to the
source region of the vacuum chamber. As noted above, in these experiments the source
chamber is pumped by a diffusion pump equipped with a liquid nitrogen cooled trap and
the detection chamber is evacuated by a turbopump. The molecular beam is cooled in a
supersonic expansion to an estimated rotational temperature of 10–15 K
22
and passes
through the skimmer into the detection chamber. After photolysis of BrCD
2
CD
2
OH, D
photofragments are probed by a laser beam counterpropagating with the photolysis laser,
both of which are aligned perpendicular to the molecular beam. Detection of D atoms
eliminates many of the problems associated with H signals from photolysis of
background hydrocarbons. This is especially important in these experiments because of
the additional hydrocarbon background from the diffusion pump.

68
The photolysis laser radiation between 202-215 nm created in the SFG process
described in Chapter 2 is focused into the detection chamber using a 25 cm focal length
(f.l.) lens. The final laser output is polarized perpendicular to the TOF detector axis.
Previous work indicated that photolysis of bromoethanol leads to a nearly isotropic
distribution of Br fragments,
18,19,21
and the angular distributions of the D atoms, which
are generated by slow secondary predissociation, are unlikely to depend on pump laser
polarization.
The D fragments are detected by 1 + 1' REMPI via the Lyman-α transition. The
output of a Nd:YAG pumped dye laser system (Continuum, NY81/ND6000) pumping
LDS 751 dye is doubled in a KDP "D" (46.6º) crystal (Inrad) to produce 6 mJ of 365 nm
light, then focused (20 cm f.l. lens) into a Kr/Ar tripling cell in which phase matching
conditions are met by slow addition of argon to 200 Torr of krypton, until the scattered
tripled light and the deuterium signal are observed at total pressures 900-1000 Torr. The
tripled 121.6 nm radiation, along with the residual 365 nm light, is focused in the
detection chamber using a MgF
2
7.5 cm f.l. lens.
The D
+
ions are accelerated towards the detector, perpendicularly to the laser and
molecular beams, by the static electric field
8
toward the MCP. The TOF apparatus was
calibrated by photolyzing HBr at 202, 205, 210, and 215 nm and monitoring the TOF
distribution of H photofragments using the aforementioned 1 + 1' REMPI detection
scheme. The experimental calibration constant for H was scaled according to mass ratio
to obtain the calibration constant 340
ns
s ms /
for D. The experimental broadening of the
TOF spectra was 1.5-3 ns (mainly due to detection laser pulse width), corresponding to

69
750±250 m/s in D velocity scale. This experimental arrangement does not give good
velocity resolution for D ions with velocities below ~8000 m/s.
8
However, because the D
fragments in these experiments arise from secondary fragmentation, they have a wide
range of energy distributions, and little information is gained from increasing the
resolutions for slow D fragments. In order to determine the bond dissociation energy
involved in each fragmentation process, only the D fragments with highest kinetic energy
are used. The maximum observed kinetic energy at each photolysis wavelength can,
thus, help identify the dissociation channel(s) leading to D atoms. No isotopic
substitution experiments were performed to help identify the observed channels given
that dissociation to both acetaldehyde and vinyl alcohol proceeds via C
(1)
bond cleavage.
Nor can the cofragments be identified using their ionization potentials, since under our
experimental conditions they cannot be ionized with enough efficiency to detect.
It is important to establish that the D fragments are generated following one-
photon absorption in bromoethanol, since this information, along with thermochemical
considerations, is used to determine the maximum velocity of the detected D
+
ions. It is
therefore mandatory to suppress as much as possible multiphoton processes by reducing
the pump pulse energy. This was achieved by moving the photolysis focal point away
from the D detection region. Under these conditions the D
+
signal intensities increased
linearly with increasing pump pulse energy. The optimal focusing condition was
determined by monitoring several ion peaks in the TOF spectra, and moving the position
of the photolysis lens until the ion peaks were still visible, but with intensities well below
the maximum. When monitoring the D
+
velocity distribution at the focal region of the

70
photolysis lens, a tail at high velocities, presumably from generation of D products by
multiphoton processes, begins to grow in at photolysis laser pulse energies exceeding 100
μJ. Under our defocused conditions, the high-velocity tail is almost absent when laser
pulse energies are kept at <50 μJ before the 25 cm lens. In the reported experiments, the
pump laser pulse energies are always limited to 20–25 μJ, which is the lowest laser power
at which an adequate signal-to-noise ratio can still be obtained. Thus loose focusing
conditions were used in combination with low laser power in order to assure that one-
photon photolysis of BrCD
2
CD
2
OH was achieved with the pump laser radiation.
Other sources of deuterium production were also considered and ruled out. No
detectable background signal was observable from the photoysis laser alone, and the
small deuterium background from the probe laser was subtracted from the spectra,
elminating the possibility of contributions from absorption of the Lyman-α radiation by
BrCD
2
CD
2
OH leading to primary D loss. D fragments from dissociation via the probe
would exhibit much higher kinetic energies than those observed here. Direct D
dissociation in the photolysis step was also considered. Absorption in the wavelength
range 202-215 nm is to an anti-bonding orbital on the C-Br bond, which is also the
weakest in the molecule, and the principal photolysis process is expected to involve
cleavage of the C-Br bond. Finally, variation of the delay time between the pump and
probe lasers demonstrated that the probe ionized D created after the pump pulse and did
not itself produce any additional deuterium.

71
4.3 BrCD
2
CD
2
OH Absorption Spectrum and Mass Spectrum
No published spectra of BrCH
2
CH
2
OH or BrCD
2
CD
2
OH were found to determine
at which wavelengths the molecule would efficiently absorb, although photolysis studies
have been performed at selected wavelengths.
18,19,21
The room temperature gas phase
absorption spectrum of BrCD
2
CD
2
OH, shown in Figure 4.2, is broad and structureless.
The absorption starts around 250 nm and increases monotonically to 200 nm. In halogen-
containing hydrocarbons, the most weakly bound electrons are located in non-bonding
orbitals on the halogen atoms and the lowest-energy electronic excitation is to an
antibonding σ* orbital on the carbon-halogen bond.
23-26
In the halogens and hydrogen
halides, the first accessible σ* electronic excitation has been termed the "Q-complex" by
Mulliken, and is comprised of three components labeled
3
Q
1
,
3
Q
0
, and
1
Q
1
.
24,27
For alkyl
bromides, both singlet and triplet states are energetically accessible and absorption to
both is allowed due to spin-orbit coupling.
25
In the dissociation both the ground state Br
and electronically excited Br* can be generated.
18,19,21,25
Br* is correlated with the
3
Q
0

state, which is accessible via a parallel transition, whereas Br is correlated with the
1
Q
1

and
3
Q
1
states, excited via a perpendicular transition.
25
In the alkyl iodides, due to the
strong spin-orbit coupling in iodine, the
3
Q
0
state carries most of the oscillator strength at
the center of the absorption band, and a recent study on ICH
2
CH
2
OH photolysis at 266
nm showed a 75% yield of I* product.
20
However, in molecules for which spin-orbit
coupling is weaker, the parallel transition becomes less allowed. Some Br* product was
observed in BrCH
2
CH
2
OH photolysis at 193 and 205 nm, but the major product was
ground state Br.
19,21
It is possible, of course, that at different wavelengths the three

72
190 200 210 220 230 240 250 260 270 280 290
0.0
0.1
0.2
0.3
0.4
0.5
Absorption
W avelength (nm )
2-brom oethanol
Absorption C urve

Figure 4.2 Absorption spectrum of gas phase 2-bromoethanol. The dashed lines indicate
the 202-215 nm wavelength range of the present experiments.

73
excited states contribute in different proportions to the absorption. However, based on
the 193 and 205 nm results, it was assumed that Br atoms in their
2
P
3/2
ground state are
the major dissociation products. It is interesting to note that almost no angular anisotropy
has been observed in the Br product distribution.
18,19,21
It has been suggested that this
may be due to a large geometry change upon dissociation and not to the nature of the
electronic absorption or a long lifetime.
18

Before attempting to perform spectroscopy on D atoms from the secondary
fragmentation process, we first performed an analysis of the mass spectrum. Mass
spectra were taken at higher photolysis laser power than was used in the D-detection
experiments, typically 50-100 μJ. To be able to observe species other than D atoms, the
ionization wavelength was shifted slightly off-resonance from the 121.6 nm D peak. An
example of such a spectrum can be seen in Figure 4.3. The photolysis energies used for
these experiments were much higher than those used in the D detection experiments,
~100-200 μJ. The main ion peaks were at m/z = 47 and 49, corresponding to the masses
of CD
2
CDOH (or CD
2
HCDO) and CD
2
CD
2
OH. Additionally, smaller peaks
corresponding to the masses of
79
BrCD
2
CD
2
OH
+
and
81
BrCD
2
CD
2
OH
+
(m/z = 128 and
130) were observed. There can be more than one source of the m/z = 47 and 49 peaks,
which may derive either from ionization of dissociation products of BrCD
2
CD
2
OH or
from dissociative ionization of BrCD
2
CD
2
OH (i.e., fragmentation of BrCD
2
CD
2
OH
+
ions
generated by multiphoton absorption).
3
Unfortunately, the signal levels for CD
2
CD
2
OH
+

or CD
2
CDOH
+
were small, and D was the only fragment ion detected with signal to noise
sufficient to obtain TOF distributions.

74
0 10 20 30 40 120 130
81
BrCD
2
CD
2
OH
79
BrCD
2
CD
2
OH
Ion Signal
Mass (am u)
CD
2
CD
2
OH
CD
2
CDOH

Figure 4.3 Mass spectrum of BrCD
2
CD
2
OH at selected wavelengths.

75
.4.4 Analysis of D fragment velocity and energy distributions
The main dissociation sequence responsible for production of D fragments from
secondary predissociation of “hot” CD
2
CD
2
OH radicals following BrCD
2
CD
2
OH
photolysis is:

BrCD
2
CD
2
OH   
 h
CD
2
CD
2
OH* + Br (4)
CD
2
CD
2
OH*    CD
2
CDOH + D (5)
   CD
2
HCDO + D (6)

where CD
2
CD
2
OH* denotes radicals with sufficient internal energy to dissociate.
Our TOF results can be converted straightforwardly to laboratory frame velocity
distributions of D fragments. The velocity distributions obtained from the TOF
distributions at 202–215 nm photolysis are displayed in Figure 4.4. In order to identify
the dissociation channel(s) responsible for D products, we use energy conservation:

KE
D+M
= hν – (D
Br-R
+ KE
Br+R
+ E
Br
+ D
D-M
+ E
M
) (7)

Here CD
2
CD
2
OH is denoted by R, and M denotes either of the two molecular co-
fragments: CD
2
CDOH or CD
2
HCDO (reactions 5 and 6). KE
X+Y
is the kinetic energy
release in the XY → X + Y dissociation process, hν is the photon energy, and E
X
is the
internal energy of co-fragment X. Because we are interested only in the maximum

76
velocity and correspondingly maximum kinetic energy, we set E
M
= 0 and E
Br
= 0 (see
below).
For the sequence of steps (4) and (5+6), the final D-atom velocity v
D
is
determined by the velocities v
R
of the radicals generated in reaction (4) (in the laboratory
frame) and the velocities v
D
(R)
of D atoms from reactions (5) and/or (6) (in the radical
frame of reference):

) (R
D R D
v v v
  
  (8)

Using momentum conservation, the velocities of the radical in the laboratory
frame and of the D atom in the radical frame, obtained from the corresponding kinetic
energy release of each process, are given by:












Br
R
R
R Br
R
m
m
m
KE
v
1
2
,











M
D
D
M D R
D
m
m
m
KE
v
1
2
) (
(9)

where m
X
denotes the mass of each fragment. The kinetic energy release depends on the
partitioning of the initial photon energy hν among the available degrees of freedom in
reaction (4):

hν = D
Br-R
+ KE
Br+R
+ E
R
(10)

77

Figure 4.4 Experimental TOF distributions, converted to velocity scale, for D atoms
from secondary dissociation of CD
2
CD
2
OH radicals produced by photolysis of BrCD
-
2
CD
2
OH at 202, 205, 210 and 215 nm. Predicted maximum velocity limits for reactions
(4.5) and (4.6) (lower and higher values, respectively) are indicated by vertical dotted
lines and solid circles with error bars.

78

E
R
= D
D-M
+ KE
D+M
(11)

Recognizing that CD
2
CD
2
OH is created with a broad range of internal energies, the
velocity distribution of product D atoms is not unique but derives from a superposition of
these distributions for the various internal energies of the radicals. Therefore, it is
impossible to reconstruct an exact velocity distribution in reactions (4.5) and/or (4.6)
from the TOF measurements.
However, we can easily estimate the maximum velocity of the D atoms for each
dissociation channel. From eqs 8–11 this velocity is:

 









  











   
M
D
D
R Br M D R Br
Br
R
R
R Br
D
m
m
m
KE D D h
m
m
m
KE
v
1
) ( 2
1
2 
(12)

This velocity depends on the kinetic energy release in the initial photolysis process (4).
Figure 4.5 shows an example of such dependence for 205 nm photolysis, which is the
only wavelength in this study for which the limits of KE
Br+R
have been published.
19
The
figure shows that lower KE in step (4) (i.e. higher E
R
) correlates with higher D velocity
(due to higher KE in eq 11).
However, data obtained from photolysis at 205 and 193 nm
18,19,21
indicate that the kinetic
energy release for products of reaction (5) is generally high, peaking around 1.5 eV for
either 
phot
wavelength. The lower limits of the observed KE’s for Br and Br* products

79
are summarized in Table 4.1. The difference in the minimum observed values of KE for
different wavelengths apparently does not exceed the experimental errors.
Results on ICH
2
CH
2
OH photolysis also show that the lower limit of KE is nearly constant
for a broad range of photolysis wavelengths (258–298 nm).
20
Thus a value of KE
Br+R
=
0.95 ± 0.10 eV was used to establish limits on the maximum velocity of D fragments
produced by reactions (5) and (6) for all wavelengths in the present study. This is
indicated in Figure 4.4 with horizontal error bars for each reaction. The energy
considerations given by eqs 10 and 11 require knowledge of the dissociation energies for
steps (4)-(6). However, the final equation, eq 12, contains only the sum of D
Br-R
and D
D-
M
, i.e. the energy 
r
E of the overall reaction

BrR → Br + D + M (13)

which can be estimated from the enthalpies of formation of the atoms and the stable
molecules: bromoethanol, vinyl alcohol, and acetaldehyde. The results,
derived from a combination of experimental and theoretical data (see Appendix 1), are


r
E ≡ D
Br R
+ D
D M
=428  12 kJ/mol = 4.44  0.13 eV (vinyl alcohol channel)
= 387
3
16


kJ/mol = 4.01
03 . 0
17 . 0


eV (acetaldehyde channel)

These values are used in eq 4.12 to calculate the curves shown in Figure 4.5,
where the shaded areas correspond to the associated uncertainties.

80
The total uncertainties in the determination of the maximum velocities of the D
fragments, derived from the combination of uncertainties in the dissociation energies and
in the lower limits of KE observed in the photolysis reaction (4), are indicated in Figure
4.5 by the vertical error bars. Table 4.2 summarizes the maximum D-fragment velocities
expected for each dissociation channel at the investigated photolysis wavelengths. These
values are also indicated in Figure 4.5.
Finally, we explain why we use only E
Br
= 0 in eq 4.7. Experimental results show
that Br* products are also generated with high kinetic energies, as shown in Table
4.1,
19,21
and thus the maximum velocity of D atoms for reactions (4.4) - (4.6) associated
with Br* is lower than for the corresponding Br channel.

4.5 Maximum Velocities of D Fragments and Dissociation Pathways.
Figure 4.3, which displays the velocity distributions of D atoms obtained at 202,
205, 210, and 215 nm photolysis wavelengths, also displays the maximum velocities for
reactions (4.5) and (4.6), obtained as described in section 4.4 and corrected for
experimental broadening. In each plot the values at higher velocities correspond to
reaction (4.6). The error bars include those in Table 4.2 as well as the experimental
uncertainties. The high velocity limit of the distribution shifts slightly to lower velocities
at longer photolysis wavelengths, as expected. No dramatic changes in the spectra are
expected with wavelength variation because of the wide distribution of product kinetic

81

Figure 4.5 Maximum allowed laboratory frame velocity of D atoms as a function of
kinetic energy release in the photolysis reaction (4.4) at 205 nm.. Lower and upper curves
correspond to reactions (4.5) and (4.6), respectively. See the text for details.

82
and internal energies generated in the primary and secondary dissociation steps.
Nevertheless, the agreement in all the plots between the maximum observed velocity
values and those predicted by eq 12 reinforces our conclusion that the observed D
fragments are products of reactions (4.5) and (4.6); i.e. they derive from secondary
dissociation of “hot” CD
2
CD
2
OH radicals. Further examination of the curves in Figure
4.5 identifies a small region at high velocities with a low fragment population whose
maximum value corresponds to reaction (4.6), and a region displaying a large increase in
fragment populations that begins at velocities estimated for the onset of reaction (4.5). It
is thus reasonable that the favored process that produces D atoms is the vinyl alcohol
channel (see also below).
We notice, however, that in the velocity distributions obtained with 210 and 215
nm photolysis, a small tail persists that extends beyond the maximum velocity allowed
for reaction (4.6). This tail, which does not seem to vary any more with reduction in laser
power, may be the result of absorption of a second 210 or 215 nm photon by CD
2
CD
2
OH.
Second photon absorption is likely to be more efficient at longer wavelengths because at
low photolysis energies a larger fraction of long-lived CD
2
CD
2
OH radicals (i.e. those
with energies either below the dissociation barrier or above it but with long dissociation
lifetimes) are produced. As discussed previously,
17,18,20
a significant fraction of the
radicals’ internal energy is tied up in rotational motion, which does not couple well to the
dissociation coordinates. These “rotationally metastable” radicals have a higher
probability of absorbing a second photon and then dissociating to D products with very
high velocities.
17,20
At shorter wavelengths, the “hotter” radicals dissociate faster and

83
reference 
phot
, nm Min KE
Br+R
, eV min KE
Br*+R
, eV
19 205 1.08 ± 0.06 1.17 ± 0.06
18 193 0.85 –
21 193 1.00 1.05

Table 4.1 Values of minimum kinetic energy release in photolysis step.

λ
phot
, nm hν, eV
max v
D
, m/s
vinyl alcohol acetaldehyde
202 6.138
9900
1100
1300


12000
1100
500



205 6.048
9400
1200
1400


11600
1100
500



210 5.904
8500
1300
1700


10900
1200
600



215 5.767
7500
1500
2100


10200
1300
600



Table 4.2 Maximum velocities expected for reactions (4.6) and (4.7).

84
have a reduced probability for absorption of a second photon. Recent theoretical
calculations of the electronic states of CH
2
CH
2
OH indicate that absorption to both
Rydberg and valence states is possible at this wavelength range.
9
The existence of
“rotationally metastable” radicals also reduces the branching ratio of the D-producing
channels (4.5) and (4.6) relative to the lowest energy channel that gives OH + C
2
H
4
(see
below). Thus, the longer lifetime of the radical produced from 210 and 215 nm
photolysis, in combination with the onset of an electronic absorption in CD
2
CD
2
OH leads
to the increased efficiency of two-photon processes at these wavelengths and the
appearance of a high energy tail in the D fragment velocities that persists down to very
low pump power.

4. 6 Theoretical estimates of product branching ratios.
According to the calculations of Senosiain et al.
13
the dissociation energy of the
CH
2
CH
2
OH radical to the lowest channel, reaction (4.1), is 1.14 eV and the reaction
proceeds without a reverse barrier. The theoretical work of Senosiain et al. couples high
level electronic structure calculations and transition state theory with a master equation
analysis for the kinetics of the CH
2
CH
2
+ OH reaction. The branching ratios arising from
the collisionless limit of this analysis are at least qualitatively related to the present
observation of the production of D atoms in the dissociation of CD
2
CD
2
OH. Figure 11 of
ref 13 suggests that the vinyl alcohol + H channel dominates over the acetaldehyde + H
channel and at high energies it also dominates over the formaldehyde + methyl channel.
13

That plot also appears to suggest the importance of a water + vinyl channel, but this

85
channel arises largely from direct abstraction from ethylene by OH, which is not relevant
to the present analysis. The consideration of ethylene + OH as the reactants in that work
obscures the branching to that channel. However, examination of the underlying RRKM
analysis suggests that the branching to that channel dominates the CH
2
CH
2
OH
dissociation kinetics, with only a few percent branching to the H atom producing
channels.
The underlying microcanonical RRKM analysis of ref 13 could be compared
directly to the present observations. However, the partial deuteration may have some
effect on the predicted rate constants and branching ratios. Thus, to properly compare
with the present experiments, we have evaluated the rovibrational properties of the
stationary points for the relevant partially deuterated species at the B3LYP/6-
311++G(d,p) level, and the relevant reaction barriers for these species are included in
parentheses in Figure 4.1. These rovibrational properties were then incorporated in a
revised RRKM analysis for the partially deuterated species of interest here. Note that, in
predicting the dissociation rates to D
2
CO + CD
2
H and to CD
2
HCDO + D, we have
simply multiplied the isomerization rate from CD
2
CD
2
OH to CD
2
HCD
2
O by the ratio of
the decomposition rates to these channels (from CD
2
HCD
2
O) relative to the total
decomposition rate from CD
2
HCD
2
O. This approach is accurate since the back
isomerization from CD
2
HCD
2
O to CD
2
CD
2
OH is negligible.
As per the discussion in section 3.4 and given the known reaction energies, the
excess energies (relative to CD
2
CD
2
+ OH) in the 2-hydroxyethyl radical decomposition
following photolysis at 202-215 nm will extend up to about 5000 to 8000 cm
-1
. The

86
present RRKM predicted microcanonical dissociation rates for the partially deuterated
analogs of reactions (4.1)-(4.3) at these excess energies are shown in Figure 4.6 and these
rates reflect mainly dissociation via reaction (4.1). These microcanonical rates are for
radicals with a total angular momentum J of 4, and similar results are found for modestly
larger total angular momentum. The reaction channels terminating in D products are
minor, constituting no more than a few percent of all products. Nevertheless, we have no
problem detecting these channels via the very sensitive 1+1’ REMPI scheme for H and D
atoms.
The acetaldehyde product correlated with H(D) co-fragments could not be
discernible from vinyl alcohol in our study because dissociation to both species proceeds
via C-D bond cleavage of the α-carbon and thus isotopic substitution would be of no
help. Although the barrier to isomerization to the ethoxy radical that precedes
dissociation to the acetaldehyde channel is slightly lower than that for direct dissociation
to vinyl alcohol, the isomerization, which involves a tight transition state where the H
atom is transferred from O directly to the β-carbon, is inefficient.
13,15,28
Additionally, the
dissociation of CH
3
CH
2
O to the acetaldehyde product competes with the reaction that
yields H
2
CO + CH
3
, which has a lower barrier (see Figure 4.1), and thus the latter should
comprise the major dissociation channel of the ethoxy radical, in both CH
2
CH
2
OH and
CD
2
CD
2
OH.
13,15

Our calculations show that vinyl alcohol should be the major dissociation product
in channels leading to H (or D) fragments. The energy dependent branching ratios for the
two D atom loss channels are illustrated in Figure 4.7 for radicals in rotational states J of

87
4, 20, 52, and 100. The branching to both these channels decreases with increasing J as
the rotational barriers become more significant. Thus, the contributions of the D
channels are largest for those “hot” 2-hydroxyethyl radicals that are born with modest
rotational excitation. These are also the radicals that dissociate the fastest and whose
products are likely to be observed in our detection region.
As shown in Figure 4.6, the branching to channel (4.6) (acetaldehyde) relative to
channel (4.5) (vinyl alcohol) decreases with increasing energy. At an excess energy of
8000 cm
-1
, which roughly corresponds to D fragments with maximum velocities, the
branching to the acetaldehyde channel is predicted to be a few percent of that to the vinyl
alcohol channel. This modest branching to reaction (6) is in reasonable agreement with
the experimental TOF distributions in Figure 4.4, and with the conclusions of ref 13 and
ref 15 of calculations for the fully hydrogenated radical. The increase in this relative
branching with decreasing energy also matches well with the changes in the TOF
distributions with increasing photolysis wavelength.

4.7 Conclusions
Experimental observations of D fragments from the predissociation of
rovibrationally excited 2-hydroxyethyl radicals, CD
2
CD
2
OH, are reported and possible
dissociation channels are analyzed by theory. The radicals are produced by photolysis of
2-bromoethanol at 202-215 nm. Previous results established that some of these radicals
are generated with sufficient internal energy to undergo secondary dissociation, and the
lowest dissociation channel to OH + ethylene was indeed observed previously.
17,18

88

Figure 4.6 Dissociation rates of the CD
2
CD
2
OH radical to the different channels as
indicated in the figure.

89

Figure 4.7 Branching ratios for the CD
2
CDOH + D and CD
2
HCDO + D channels for
different rotational levels of the CD
2
CD
2
OH radical.
90
Calculations have predicted that while the main dissociation channel is OH +
ethylene, two minor channels may generate H or D atoms, one associated with vinyl
alcohol and the other with acetaldehyde co-fragments.
13,15
In the present experiments, D
fragments are detected by 1+1’ REMPI and their TOF distributions are determined.
From analysis of the maximum velocities and kinetic energies of the observed D
fragments it is concluded that they originate from the decomposition of CD
2
CD
2
OH, but
the experimental resolution is insufficient to distinguish between the two possible
channels leading to D products. Theoretical analysis and RRKM calculations of
microcanonical dissociation rates and branching ratios for the range of available excess
energies (up to 5000-8000 cm
-1
above the OH + C
2
D
4
threshold) indicate that: (i) the D-
producing channels are minor compared to the OH + C
2
D
4
channel, constituting at most
a few percent; (ii) the branching ratio is more favorable towards these products when the
reactant radicals have low rotational energy; and (iii) the vinyl alcohol channel is strongly
favored over the acetaldehyde channel at all but the lowest excess energies.
91
4.8 References
(1) Curtiss, L. A.; Lucas, D. J.; Pople, J. A. J. Chem. Phys. 1995, 102, 3292.

(2) Sosa, C.; Schlegel, H. B. Journal of the American Chemical Society 1987, 109,
4193.

(3) Ruscic, B.; Berkowitz, J. J. Chem. Phys. 1994, 101, 10936.

(4) Dyke, J. M.; Groves, A. P.; Lee, E. P. F.; Niavaran, M. H. Z. Journal of Physical
Chemistry A 1997, 101, 373.

(5) Feng, L. Spectroscopy and Photodissociation Dynamics of the Hydroxymethyl
Radical (CH2OH), University of Southern California, 2004.

(6) Johnson, R. D.; Hudgens, J. W. Abstracts of Papers of the American Chemical
Society 1996, 212, 82.

(7) Karpichev, B. Electronic States and Photodissociation Dynamics of Hydroxyalkyl
Radicals, University of Southern California, 2009.

(8) Karpichev, B.; Edwards, L. W.; Wei, J.; Reisler, H. J. Phys. Chem. A 2008, 112,
412.

(9) Karpichev, B. K., L; Diri, K; Reisler, H.; Krylov, A.I. J. Chem. Phys. 2010, 132,
114308

(10) Anastasi, C.; Simpson, V.; Munk, J.; Pagsberg, P. Journal of Physical Chemistry
1990, 94, 6327.

(11) Taatjes, C. A.; Hansen, N.; McIlroy, A.; Miller, J. A.; Senosiain, J. P.;
Klippenstein, S. J.; Qi, F.; Sheng, L.; Zhang, Y.; Cool, T. A.; Wang, J.; Westmoreland, P.
R.; Law, M. E.; Kasper, T.; Kohse-Hoinghaus, K. Science 2005, 308, 1887.

(12) Taatjes, C. A.; Hansen, N.; Miller, J. A.; Cool, T. A.; Wang, J.; Westmoreland, P.
R.; Law, M. E.; Kasper, T.; Kohse-Hoinghaus, K. Journal of Physical Chemistry A 2006,
110, 3254.

(13) Senosiain, J. P.; Klippenstein, S. J.; Miller, J. A. Journal of Physical Chemistry A
2006, 110, 6960.

(14) Cool, T. A.; Nakajima, K.; Mostefaoui, T. A.; Qi, F.; McIlroy, A.; Westmoreland,
P. R.; Law, M. E.; Poisson, L.; Peterka, D. S.; Ahmed, M. Journal of Chemical Physics
2003, 119, 8356.

92
(15) Xu, Z. F.; Xu, K.; Lin, M. C. Chemphyschem 2009, 10, 972.
(16) Tully, F. P. Chemical Physics Letters 1983, 96, 148.

(17) Sapers, S. P.; Hess, W. P. Journal of Chemical Physics 1992, 97, 3126.

(18) Hintsa, E. J.; Zhao, X. S.; Lee, Y. T. Journal of Chemical Physics 1990, 92, 2280.
(19) Chandler, D. W.; Thoman, J. W.; Hess, W. P. Institute of Physics Conference
Series 1991, 355.

(20) Shubert, V. A.; Rednic, M.; Pratt, S. T. Journal of Physical Chemistry A 2009,
113, 9057.

(21) Ratliff, B. J.; Womack, C. C.; Tang, X. N.; Landau, W. M.; Butler, L. J.; Szpunar,
D. E. Journal of Physical Chemistry A 2010, 114, 4934.

(22) Aristov, V.; Conroy, D.; Reisler, H. Chemical Physics Letters 2000, 318, 393.

(23) Krajnovich, D.; Butler, L. J.; Lee, Y. T. Journal of Chemical Physics 1984, 81,
3031.

(24) Mulliken, R. S. Physical Review 1936, 50, 1017.

(25) Pence, W. H.; Baughcum, S. L.; Leone, S. R. Journal of Physical Chemistry
1981, 85, 3844.

(26) Mullikan, R. S. Physical Review 1934, 46, 549.

(27) Mullikan, R. S. Physical Review 1940, 57, 500.

(28) Zhu, R. S.; Park, J.; Lin, M. C. Chemical Physics Letters 2005, 408, 25.

93
Chapter 5
Future Work: Spectroscopy of CH
2
CH
2
OH
5.1 Introduction
The ultimate goal of our work with halogenated ethanol precursors is to create a
cold CH
2
CH
2
OH radical upon which spectroscopy can be performed. Very little is
known about this radical species, although there have been intriguing experimental and
theoretical reports that suggest that the stable ion has a non-classical bridged structure.
This structure is shown in Figure 5.1. The large geometry change from the ground state
structure to the bridged structure upon ionization has been proposed by Ruscic and
Berkowitz to explain the unusual ionization spectrum of the radical, which demonstrates
a large increase in ion collection at the onset of the vertical ionization potential, but
shows a long tail that persists to much lower wavelengths.
1
This tail is believed to be due
to the approach towards the adiabatic ionization, which has low oscillator strength and an
uncertain onset. The large difference between the neutral and ion geometries makes the
Franck-Condon overlap very small at the origin of this transition.
1,2
Sapers and Hess
report in their work on CH
2
CH
2
OH that the radical appears to absorb well at 266 nm, a
fact which they used to explain the presence of high rotational states of the OH
dissociation product at high laser fluences.
3
Shubert et al. dissociated ICH
2
CH
2
OH at
266 nm, and explained the discrepancy between the CH
2
CH
2
OH
+
ion signal and the I
+
ion
signal in their photoionization spectra by arguing that the hydroxyethyl radical was
absorbing a second photon and

94

Figure 5.1 Non-classical bridged structure of the CH
2
CH
2
OH
+
ion. The bond lengths
and angles used in this drawing are those calculated by Curtis, et al.
2

95
dissociating.
4
In addition, our group together with the theoretical group of Dr. Anna
Krylov has performed calculations on the electronic states of this radical that suggest that
an excited electronic state might indeed be accessible via a 266 nm absorption.
5
The
fourth harmonic of our Nd:YAG laser is 266 nm, which provides an excellent high-power
source for pumping this electronic state.
The ICH
2
CH
2
OH precursor has a strong absorption that peaks at around this
wavelength, as seen in Figure 5.2. Therefore, photolysis of this molecule should be easy
at 266 nm, also using an Nd:YAG source. However, the fact that the radical also absorbs
at the same wavelength means that care must be taken to ensure that not more than one
photon is absorbed in the photolysis step. This can be achieved with low laser fluences,
and by assuring that the photolysis laser hits the molecular beam at the tip of the quartz
tube just before the expansion, as with the CH
3
CHOH experiments.

5.2 Experimental
ICH
2
CH
2
OH is an ideal precursor to create CH
2
CH
2
OH radicals for several
reasons, which have been outlined in part above. The peak of one of the electronic
absorptions by ICH
2
CH
2
OH lies in the region of the fourth harmonic of an Nd:YAG
laser, giving us a high-power source with which to perform this photolysis. Unlike in
BrCH
2
CH
2
OH, the strong spin-orbit coupling in ICH
2
CH
2
OH makes the parallel
transition to the Q
0
state strongly allowed, and ICH
2
CH
2
OH has been shown to dissociate
to primarily I(
2
P
1/2
) (I*) products.
4,6
Dissociating the molecule at 266 nm

96

200 250 300 350
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Absorption
Wavelength (nm)
2-Iodoethanol

Figure 5.2 Electronic absorption spectrum of iodoethanol. This spectrum was taken in
solution in hexanes at room temperature on a Varian Cary 300 UV-Vis
Spectrophotometer.

97
(~4.66 eV) requires ~2.4 eV to break the C-I bond.
7
The spin-orbit splitting of I is 0.94
eV.
4
If most of the I products are created in the spin-orbit excited state, then a little more
than one eV is left in the internal modes of the radical. This amount of energy is barely
enough for further dissociation, making it unlikely that a large fraction of radicals would
dissociate. Some of the remaining energy will go into translation of the dissociating
fragments, which can leave the radical even colder. In this respect, ICH
2
CH
2
OH is the
ideal precursor for the creation of the CH
2
CH
2
OH radical. The problem, however, is that
this molecule has a vapor pressure at room temperature so low as to be undetectable by
our instruments. A small amount of it must be evaporated into the gas phase in order for
it to be experimentally viable as a precursor. For this reason, we had to make
modifications to our experimental design in order to accommodate its unique properties.
The heated pulsed nozzle in Chapter 2 was designed specifically with
ICH
2
CH
2
OH in mind. The vapor pressure of this species is negligible at room
temperature, but upon heating to 40
◦
C has a vapor pressure of ~ 7 Torr, enough to
comprise a 1% mixture in 1 atm of He gas. The heated precursor will enter the quartz
tube at the tip of the nozzle and be photolyzed in the source region of the vacuum
chamber by the fourth harmonic output of an Nd:YAG. This is the same experimental
design as in the CH
3
CHOH experiments, in which the third Nd:YAG harmonic was used
instead of the fourth. The radical produced in this photolysis will then be cooled in the
adiabatic expansion, and spectroscopy will be performed by a tunable laser source such
as the ND6000 dye laser (Continuum) used in the experiments in Chapter 4 or the Sunlite
OPO/OPA (Continuum), pumped by the third harmonic of our Powerlite Nd:YAG
98
(Continuuem) source.

5.3 Electronic States
Drs. Boris Karpichev, Kadir Diri, and Lukasz Koziol, in a joint study between our
group and the group of Dr. Anna Krylov at USC, performed calculations on the two
lowest-energy electronic excitations in CH
2
CH
2
OH and determined the minimum
geometries of the neutral ground-state and cation as well as the electronically-excited
species.
5
As in previous theoretical works, it was found that there is no minimum on the
cation surface in the open-chain geometry.
2,8
The authors found that the vertical
ionization of the open-chain neutral CH
2
CH
2
OH leads to immediate collapse to either the
ring-structure non-classical geometry, or to isomerization to CH
3
CHOH. In addition, the
Rydberg-state energy curves with respect to change in the geometrical conformers run
almost parallel to those of the cation, which suggests that the dynamics on the Rydberg
state potential energy surfaces (PESs) will be similar to those on the cation PES. The
isomerization pathways of the Rydberg states should be very similar to those in the
cation.
There are two minima and several low-lying saddle points on the potential energy
surface of the neutral CH
2
CH
2
OH ground state, all lying within 900 cm
-1
of the global
minimum. The lowest-energy electronic excitations from all of these states occur either
from the singly occupied molecular orbital (SOMO) on the C
(2)
atom to the 3s Rydberg
state, or the valence excitation from the HOMO
-1
to the SOMO. Since the Rydberg states
behave like the cation’s ground states, excitation to the 3s state will cause the radical to
99
isomerize, either to the CH
3
CHOH radical or to the bridged structure. If the radical
isomerizes to CH
3
CHOH, the 3s state of this radical is expected to couple strongly to the
ground state surface, and dissociation will occur rapidly along the O-H bond coordinate.
9

A different set of dissociation products is expected to result from excitation to the
valence state. The HOMO
-1
is bonding along the C-C bond, and in the valence transition
it is excited to an orbital with anti-bonding character along this bond leading to a possible
dissociation. In fact, three dissociation products of this excitation have been found,
producing CH
2
+ CH
2
OH, CH
2
CH
2
+ OH, and CH
2
CHOH + H. By performing pump-
probe experiments and monitoring the dissociation products, it may be possible to tell
which excitation is taking place. Again, we see that hydrogen can come from many
possible sources (C
(2)
-H and O-H bond cleavage in CH
3
CHOH, C
(1)
-H bond cleavage in
CH
2
CH
2
OH). This time, however, the dissociation will be a primary process, and it will
be much easier to determine the source of the H photoproducts using energetics. The
three processes have three distinct dissociation energies
9,10
:

CH
3
CHOH → CH
3
CHO + H ΔH = 1.04 eV
→ CH
2
CHOH + H ΔH = 1.46 eV

CH
2
CH
2
OH → CH
2
CHOH + H ΔH = 1.17 eV

The maximum kinetic energy of the H fragment from each process will be known
with much greater accuracy because of the lack of the primary dissociation process taking
100
place in our earlier experiments in the cleavage of the C-Br bond. In these experiments,
the C-I bond will be broken before the expansion and cooling of the radical, and the
energy associated with this process will not affect the energy of the radical dissociation
process occurring in the detection chamber. Excitation from the lowest-lying neutral
electronic state is primarily to the 3s Rydberg orbital, which should produce H atoms
from O-H dissociation along the ground state coordinate. However, another conformer of
the ground state lies only 325 cm
-1
above the global minimum, and has oscillator strength
to several excited states. A useful experiment might be to determine the difference in
branching ratio of H to CH
2
OH as the laser frequency is increased above the threshold for
3s excitation. The lowest energy excitation possible in the radical is from the global
minimum structure to the 3s Rydberg state. CH
2
OH has well-established REMPI
schemes, and will be detectable via ionization through its 3p
z
Rydberg state.
11

101
5.4 References
(1) Ruscic, B.; Berkowitz, J. Journal of Chemical Physics 1994, 101, 10936.

(2) Curtiss, L. A.; Lucas, D. J.; Pople, J. A. Journal of Chemical Physics 1995, 102,
3292.

(3) Sapers, S. P.; Hess, W. P. Journal of Chemical Physics 1992, 97, 3126.

(4) Shubert, V. A.; Rednic, M.; Pratt, S. T. Journal of Physical Chemistry A 2009,
113, 9057.

(5) Karpichev, B. K., L; Diri, K; Reisler, H.; Krylov, A.I. Journal of Chemical
Physics 2010, 132, 114308.

(6) Pence, W. H.; Baughcum, S. L.; Leone, S. R. Journal of Physical Chemistry
1981, 85, 3844.

(7) Bernardes, C. E. S.; da Piedade, M. E. M.; Amaral, L. M. P. F.; Ferreira, A. I. M.
C. L.; da Silva, M. A. V. R.; Diogo, H. P.; Cabral, B. J. C. Journal of Physical Chemistry
A 2007, 111, 1713.

(8) Nobes, R. H.; Rodwell, W. R.; Bouma, W. J.; Radom, L. Journal of the American
Chemical Society 1981, 103, 1913.

(9) Karpichev, B.; Edwards, L. W.; Wei, J.; Reisler, H. Journal of Physical
Chemistry A 2008, 112, 412.

(10) Edwards, L. W.; Ryazanov, M.; Reisler, H.; Klippenstein, S. J. Journal of
Physical Chemistry A 2010, 114, 5453.

(11) Aristov, V.; Conroy, D.; Reisler, H. Chemical Physics Letters 2000, 318, 393.

102
Alphabetized Bibliography

(1) Adams, G. F.; Bartlett, R. J.; Purvis, G. D. Chem Phys Lett 1982, 87, 311.

(2) Ahmed, M.; Peterka, D. S.; Suits, A. G. Phys Chem Chem Phys 2000, 2, 861.

(3) Amaral, G.; Xu, K. S.; Zhang, J. S. Journal of Chemical Physics 2001, 114, 5164.

(4) Anastasi, C.; Simpson, V.; Munk, J.; Pagsberg, P. Chem Phys Lett 1989, 164, 18.

(5) Anastasi, C.; Simpson, V.; Munk, J.; Pagsberg, P. J Phys Chem-Us 1990, 94,
6327.

(6) Aristov, V.; Conroy, D.; Reisler, H. Chem Phys Lett 2000, 318, 393.

(7) Batt, L.; Burrows, J. P.; Robinson, G. N. Chem Phys Lett 1981, 78, 467.

(8) Bauschlicher, C. W.; Partridge, H. J Phys Chem-Us 1994, 98, 1826.

(9) Bernardes, C. E. S.; da Piedade, M. E. M.; Amaral, L. M. P. F.; Ferreira, A. I. M.
C. L.; da Silva, M. A. V. R.; Diogo, H. P.; Cabral, B. J. C. Journal of Physical Chemistry
A 2007, 111, 1713.

(10) Bruna, P. J.; Grein, F. Journal of Physical Chemistry A 1998, 102, 3141.

(11) Bruna, P. J.; Grein, F. Journal of Physical Chemistry A 2001, 105, 8599.

(12) Caralp, F.; Devolder, P.; Fittschen, C.; Gomez, N.; Hippler, H.; Mereau, R.;
Rayez, M. T.; Striebel, F.; Viskolcz, B. Phys Chem Chem Phys 1999, 1, 2935.

(13) Chandler, D. W.; Thoman, J. W.; Hess, W. P. Inst Phys Conf Ser 1991, 355.

(14) Choi, H.; Bise, R. T.; Neumark, D. M. Journal of Physical Chemistry A 2000,
104, 10112.

(15) Cleary, P. A.; Romero, M. T. B.; Blitz, M. A.; Heard, D. E.; Pilling, M. J.;
Seakins, P. W.; Wang, L. Phys Chem Chem Phys 2006, 8, 5633.

(16) Conroy, D. Rydberg State of an Open Shell Species: Characterization and
Photophysics of the 3pz State of CH2OH, University of Southern California, 2000.

(17) Conroy, D.; Aristov, V.; Feng, L.; Reisler, H. Journal of Physical Chemistry A
2000, 104, 10288.

103
(18) Conroy, D.; Aristov, V.; Feng, L.; Sanov, A.; Reisler, H. Accounts Chem Res
2001, 34, 625.

(19) Cool, T. A.; Nakajima, K.; Mostefaoui, T. A.; Qi, F.; McIlroy, A.; Westmoreland,
P. R.; Law, M. E.; Poisson, L.; Peterka, D. S.; Ahmed, M. Journal of Chemical Physics
2003, 119, 8356.

(20) Cox, J. D.; Wagman, D. D.; Medvedev, V. A. CODATA Key Values for
Thermodynamics; Hemisphere Publishing Corp.: New York, 1989.

(21) Curtiss, L. A.; Kock, L. D.; Pople, J. A. Journal of Chemical Physics 1991, 95,
4040.

(22) Curtiss, L. A.; Lucas, D. J.; Pople, J. A. Journal of Chemical Physics 1995, 102,
3292.

(23) da Silva, G.; Bozzelli, J. W. Journal of Physical Chemistry A 2006, 110, 13058.

(24) da Silva, G.; Kim, C. H.; Bozzelli, J. W. Journal of Physical Chemistry A 2006,
110, 7925.

(25) Dang, T. T.; Motell, E. L.; Travers, M. J.; Clifford, E. P.; Ellison, G. B.; Depuy,
C. H.; Bierbaum, V. M. Int J Mass Spectrom 1993, 123, 171.

(26) Davies, A. G.; Neville, A. G. J Chem Soc Perk T 2 1991, 2021.

(27) Diau, E. W. G.; Lee, Y. P. Journal of Chemical Physics 1992, 96, 377.

(28) Dóbé, S. B., T.; Tyrányi, T.; Márta, J.; Grussdorf, J.; Temps, F.; Wagner, H.G. J
Phys Chem-Us 1996, 100, 19864.

(29) Dulcey, C. S.; Hudgens, J. W. J Phys Chem-Us 1983, 87, 2296.

(30) Dulcey, C. S.; Hudgens, J. W. B Soc Chim Belg 1983, 92, 583.

(31) Dulcey, C. S.; Hudgens, J. W. Journal of Chemical Physics 1986, 84, 5262.

(32) Durig, J. R.; Shen, S.; Guirgis, G. A. J Mol Struct 2001, 560, 295.

(33) Dyke, J. M.; Groves, A. P.; Lee, E. P. F.; Niavaran, M. H. Z. Journal of Physical
Chemistry A 1997, 101, 373.

(34) Ebata, T.; Yanagishita, H.; Obi, K.; Tanaka, I. Chem Phys 1982, 69, 27.

104
(35) Edelbuttel-Einhaus, J. H., K.; Rohde, G.; Seeba, J.;. “Proceedings of the 24th
Symposium International on Combustion”;
, 1992, Pittsburgh, PA.

(36) Edwards, L. W.; Ryazanov, M.; Reisler, H.; Klippenstein, S. J. Journal of
Physical Chemistry A 2010, 114, 5453.

(37) Ellison, G. B.; Engelking, P. C.; Lineberger, W. C. J Phys Chem-Us 1982, 86,
4873.

(38) Faulhaber, A. E.; Szpunar, D. E.; Kautzman, K. E.; Neumark, D. M. Journal of
Physical Chemistry A 2005, 109, 10239.

(39) Feng, L. Spectroscopy and Photodissociation Dynamics of the Hydroxymethyl
Radical (CH2OH), University of Southern California, 2004.

(40) Feng, L.; Demyanenko, A. V.; Reisler, H. Journal of Chemical Physics 2003,
118, 9623.

(41) Feng, L.; Demyanenko, A. V.; Reisler, H. Journal of Chemical Physics 2004,
120, 6524.

(42) Feng, L.; Huang, X.; Reisler, H. Journal of Chemical Physics 2002, 117, 4820.

(43) Feng, L.; Huang, X.; Reisler, H. Abstr Pap Am Chem S 2002, 223, C67.
(44) Feng, L.; Reisler, H. Journal of Physical Chemistry A 2004, 108, 9847.

(45) Feng, L.; Wei, J.; Reisler, H. Abstr Pap Am Chem S 2004, 227, U246.

(46) Gonzalez, C.; Sosa, C.; Schlegel, H. B. J Phys Chem-Us 1989, 93, 2435.

(47) Hecht, E. Optics, Fourth ed.; Addison Wesley: San Francisco, 2002.

(48) Hintsa, E. J.; Zhao, X. S.; Lee, Y. T. Journal of Chemical Physics 1990, 92, 2280.

(49) Hippler, H.; Viskolcz, B. Phys Chem Chem Phys 2000, 2, 3591.

(50) Hoffman, B. C.; Yarkony, D. R. Journal of Chemical Physics 2002, 116, 8300.

(51) Hoyermann, K.; Olzmann, M.; Seeba, J.; Viskolcz, B. Journal of Physical
Chemistry A 1999, 103, 5692.

(52) Inoue, G.; Okuda, M.; Akimoto, H. Journal of Chemical Physics 1981, 75, 2060.

(53) Johnson, R. D.; Hudgens, J. W. Abstr Pap Am Chem S 1996, 212, 82.
105
(54) Joo, D. L.; Merer, A. J.; Clouthier, D. J. J Mol Spectrosc 1999, 197, 68.

(55) Karpichev, B. Electronic States and Photodissociation Dynamics of Hydroxyalkyl
Radicals, University of Southern California, 2009.

(56) Karpichev, B.; Edwards, L. W.; Wei, J.; Reisler, H. Journal of Physical
Chemistry A 2008, 112, 412.

(57) Karpichev, B. K., L; Diri, K; Reisler, H.; Krylov, A.I. Journal of Chemical
Physics 2010, 132, 114308.

(58) Klippenstein, S. J. Private Communication.

(59) Koziol, L.; Levchenko, S. V.; Krylov, A. I. Journal of Physical Chemistry A
2006, 110, 2746.

(60) Krajnovich, D.; Butler, L. J.; Lee, Y. T. Journal of Chemical Physics 1984, 81,
3031.

(61) Lin, J. J.; Shu, J.; Lee, Y. T.; Yang, X. Journal of Chemical Physics 2000, 113,
5287.

(62) Lindner, J.; Loomis, R. A.; Klaassen, J. J.; Leone, S. R. Journal of Chemical
Physics 1998, 108, 1944.

(63) Liu, G. X.; Ding, Y. H.; Li, Z. S.; Fu, Q.; Huang, X. R.; Sun, C. C.; Tang, A. C.
Phys Chem Chem Phys 2002, 4, 1021.

(64) Marenich, A. V.; Boggs, J. E. Journal of Chemical Physics 2003, 119, 3098.

(65) Matsika, S.; Yarkony, D. R. Journal of Chemical Physics 2002, 117, 6907.

(66) Mullikan, R. S. Physical Review 1934, 46, 549.

(67) Mullikan, R. S. Physical Review 1940, 57, 500.

(68) Mulliken, R. S. Physical Review 1936, 50, 1017.

(69) Nobes, R. H.; Rodwell, W. R.; Bouma, W. J.; Radom, L. J Am Chem Soc 1981,
103, 1913.

(70) Oehlers, C.; Wagner, H. G.; Ziemer, H.; Temps, F.; Dobe, S. Journal of Physical
Chemistry A 2000, 104, 10500.

(71) Ohbayashi, K.; Akimoto, H.; Tanaka, I. J Phys Chem-Us 1977, 81, 798.
106
(72) Osborn, D. L.; Leahy, D. J.; Neumark, D. M. Journal of Physical Chemistry A
1997, 101, 6583.

(73) Pagsberg, P.; Munk, J.; Sillesen, A.; Anastasi, C. Chem Phys Lett 1988, 146, 375.

(74) Park, J.; Zhu, R. S.; Lin, M. C. Journal of Chemical Physics 2002, 117, 3224.

(75) Pence, W. H.; Baughcum, S. L.; Leone, S. R. J Phys Chem-Us 1981, 85, 3844.
(76) Radford, H. E. Chem Phys Lett 1980, 71, 195.

(77) Ramond, T. M.; Davico, G. E.; Schwartz, R. L.; Lineberger, W. C. Journal of
Chemical Physics 2000, 112, 1158.

(78) Ratliff, B. J.; Womack, C. C.; Tang, X. N.; Landau, W. M.; Butler, L. J.; Szpunar,
D. E. Journal of Physical Chemistry A 2010, 114, 4934.

(79) Rettrup, S.; Pagsberg, P.; Anastasi, C. Chem Phys 1988, 122, 45.

(80) Riley, S. J. W., K.R. Faraday Discussions of the Chemical Society 1972, 53, 132.

(81) Rodler, M.; Blom, C. E.; Bauder, A. J Am Chem Soc 1984, 106, 4029.

(82) Rudic, S.; Murray, C.; Ascenzi, D.; Anderson, H.; Harvey, J. N.; Orr-Ewing, A. J.
Journal of Chemical Physics 2002, 117, 5692.

(83) Ruscic, B.; Berkowitz, J. Journal of Chemical Physics 1994, 101, 10936.

(84) Saebo, S.; Radom, L.; Schaefer, H. F. Journal of Chemical Physics 1983, 78, 845.

(85) Saito, S. Chem Phys Lett 1976, 42, 399.

(86) Sapers, S. P.; Hess, W. P. Journal of Chemical Physics 1992, 97, 3126.

(87) Senosiain, J. P.; Klippenstein, S. J.; Miller, J. A. Journal of Physical Chemistry A
2006, 110, 6960.

107
(88) Shao, Y.; Molnar, L. F.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S. T.;
Gilbert, A. T. B.; Slipchenko, L. V.; Levchenko, S. V.; O'Neill, D. P.; DiStasio, R. A.;
Lochan, R. C.; Wang, T.; Beran, G. J. O.; Besley, N. A.; Herbert, J. M.; Lin, C. Y.; Van
Voorhis, T.; Chien, S. H.; Sodt, A.; Steele, R. P.; Rassolov, V. A.; Maslen, P. E.;
Korambath, P. P.; Adamson, R. D.; Austin, B.; Baker, J.; Byrd, E. F. C.; Dachsel, H.;
Doerksen, R. J.; Dreuw, A.; Dunietz, B. D.; Dutoi, A. D.; Furlani, T. R.; Gwaltney, S. R.;
Heyden, A.; Hirata, S.; Hsu, C. P.; Kedziora, G.; Khalliulin, R. Z.; Klunzinger, P.; Lee,
A. M.; Lee, M. S.; Liang, W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E. I.; Pieniazek, P.
A.; Rhee, Y. M.; Ritchie, J.; Rosta, E.; Sherrill, C. D.; Simmonett, A. C.; Subotnik, J. E.;
Woodcock, H. L.; Zhang, W.; Bell, A. T.; Chakraborty, A. K.; Chipman, D. M.; Keil, F.
J.; Warshel, A.; Hehre, W. J.; Schaefer, H. F.; Kong, J.; Krylov, A. I.; Gill, P. M. W.;
Head-Gordon, M. Phys Chem Chem Phys 2006, 8, 3172.

(89) Shu, J.; Lin, J. J.; Lee, Y. T.; Yang, X. M. Journal of Chemical Physics 2001,
115, 849.

(90) Shubert, V. A.; Rednic, M.; Pratt, S. T. Journal of Physical Chemistry A 2009,
113, 9057.

(91) Sinha, P.; Boesch, S. E.; Gu, C. M.; Wheeler, R. A.; Wilson, A. K. Journal of
Physical Chemistry A 2004, 108, 9213.

(92) Sosa, C.; Schlegel, H. B. J Am Chem Soc 1987, 109, 7007.

(93) Sosa, C.; Schlegel, H. B. J Am Chem Soc 1987, 109, 4193.

(94) Syage, J. A. Journal of Chemical Physics 1996, 105, 1007.

(95) Taatjes, C. A.; Christensen, L. K.; Hurley, M. D.; Wallington, T. J. Journal of
Physical Chemistry A 1999, 103, 9805.

(96) Taatjes, C. A.; Hansen, N.; McIlroy, A.; Miller, J. A.; Senosiain, J. P.;
Klippenstein, S. J.; Qi, F.; Sheng, L.; Zhang, Y.; Cool, T. A.; Wang, J.; Westmoreland, P.
R.; Law, M. E.; Kasper, T.; Kohse-Hoinghaus, K. Science 2005, 308, 1887.

108
Appendix
In order to apply eq 4.12, the values of the two dissociation energies D
Br-R
and
D
D-M
are required. D
Br-R
was estimated from C-Br bond energies in similar molecules to
be 289  8 kJ/mol.
1
In more recent work the standard enthalpy of dissociation was
estimated at 305  8 kJ/mol from a combination of experimental data and theoretical
calculations.
2
No experimental data for the dissociation energy of 2-hydroxyethyl radical
are available. However, the sum of the dissociation energies that is required in eq 12 is
equal to the energy of the net reaction (13), and the standard enthalpy of reaction (13) can
be calculated straightforwardly.
The standard enthalpy of reaction needs to be corrected to obtain the energy of
reaction at 0 K, and the isotopic differences in energies between the partially deuterated
and the non-deuterated molecules should also be taken into account. The thermal
corrections are composed of translational (3/2 RT for all species), rotational and
vibrational (3/2 RT and H
v
for molecules) contributions, and the ΔPV = ΔnRT term.
Isotopic corrections for the molecules are taken as differences in zero point vibrational
energies (ZPE) of corresponding isotopologs (Δ
D
ZPE). The expression used for the
energy of reaction is:


r
E = 
f
H
o
(Br) + 
f
H
o
(H) + 
f
H
o
(M)  
f
H
o
(BrR)
  5RT  H
v
(M) + H
v
(BrR) + 
D
ZPE(M)  
D
ZPE(BrR). (A1)

The vibrational enthalpies H
v
and ZPE were calculated in the harmonic oscillator
109
approximation from frequencies obtained by quantum chemical calculations at the
B3LYP/aug-cc-pVTZ level using the Q-Chem software package.
3
The calculated
frequencies were used because the experimental information is incomplete even for the
non-deuterated species, and is entirely missing for the deuterated analogs of interest.
However, for CD
2
CDOH, 9 of 15 fundamental frequencies were measured in an argon
matrix.
4
The most complete set of spectroscopic constants (harmonic frequencies and
anharmonic constants) derived from experimental data is available for CH
3
CHO, and
comparison of the experimental harmonic frequencies with the theoretical calculations
shows an RMS deviation of about 20 cm
–1
and a best fit scaling factor of 0.999  0.003.
For BrCH
2
CH
2
OH and CH
2
CHOH, only the fundamental experimental frequencies are
available.
5,6
The best fit of the theoretical harmonic frequencies to these data gives an
RMS deviation of 20 cm
–1
, which is comparable to the discrepancies between different
experimental determinations. The derived scaling factor (experimental
fundamental/theoretical harmonic) is 0.964  0.003, which is very close to the
recommended scaling factor 0.9676 for calculations of fundamental frequencies from
B3LYP/aug-cc-pVTZ harmonic frequencies. This agreement suggests that the theoretical
frequencies can be reliably used in calculations of the thermal and isotopic corrections.
For this purpose the frequencies were scaled by 0.9867, the factor recommended for
calculations of H
v
and ZPE.
7

The experimental values of the standard enthalpies of formation and the
theoretically estimated H
v
and ZPE are listed in Table A1. Note that the enthalpies of
formation of the molecular species have asymmetric error bars. The reason for this is that
110
the values listed in the table are the most recent results, believed to be the most reliable,
and the smaller errors correspond to the reported error bars for those values. However,
the discrepancies between the listed values and the earlier (but still apparently reliable)
values
1,6,8
are larger than the estimated errors, so these differences are shown as larger
error bars in one direction. Additionally, the enthalpy of formation of 2-bromoethanol
refers to Gg, the most stable conformer, but at room temperature in the gas phase a few
percent of the molecules are in the Tt and Tg conformations, which have 4-6 kJ/mol
higher energies.
5
These contributions are also included in the error bars.
The error bars for the final results of the energy of the overall reaction (13) were
obtained by assuming that the errors in the enthalpies of formation of molecules are
additive and all other errors are uncorrelated. The estimated errors in Table A1 (one
standard deviation) are: 0.055 kJ/mol (5 cm
–1
) for H
v
(298.15 K) and 1.2 kJ/mol (100 cm
–
1
) for ZPEs according to ref 5.
111

Species 
f
H
0
, kJ/mol H
v
(298.15 K)
,
kJ/mol (cm
–1
)
ZPE,
cm
–1

ZPE
D
,
cm
–1


D
ZPE=ZPE
D
–
ZPE,
kJ/mol (cm
–1
)
BrCH
2
CH
2
OH
–221.7
10
7 . 0


[a]
5.89 (492) 15353 12451 –34.7 (2902)
Br +111.81  0.12 [b]    
H +217.998  0.006 [b]    
CH
2
CHOH
–125.4
11
2


[c]
2.15 (180) 12211 10178 –24.3 (2033)
CH
3
CHO
–166.2
7 . 0
5 . 4


[d]
2.82 (236) 11952 9950 –23.9 (2002)
[a] =Ref.
9
, [b] =Ref.
8
, [c] = Ref.
10
, [d] = Ref.
11

Table A1: Thermochemical values used in evaluating eq A1.
112
A1 References
(1) Ruscic, B.; Berkowitz, J. J Chem Phys 1994, 101, 10936.

(2) Bernardes, C. E. S.; da Piedade, M. E. M.; Amaral, L. M. P. F.; Ferreira, A. I. M.
C. L.; da Silva, M. A. V. R.; Diogo, H. P.; Cabral, B. J. C. J Phys Chem A 2007, 111,
1713.

(3) Shao, Y.; Molnar, L. F.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S. T.;
Gilbert, A. T. B.; Slipchenko, L. V.; Levchenko, S. V.; O'Neill, D. P.; DiStasio, R. A.;
Lochan, R. C.; Wang, T.; Beran, G. J. O.; Besley, N. A.; Herbert, J. M.; Lin, C. Y.; Van
Voorhis, T.; Chien, S. H.; Sodt, A.; Steele, R. P.; Rassolov, V. A.; Maslen, P. E.;
Korambath, P. P.; Adamson, R. D.; Austin, B.; Baker, J.; Byrd, E. F. C.; Dachsel, H.;
Doerksen, R. J.; Dreuw, A.; Dunietz, B. D.; Dutoi, A. D.; Furlani, T. R.; Gwaltney, S. R.;
Heyden, A.; Hirata, S.; Hsu, C. P.; Kedziora, G.; Khalliulin, R. Z.; Klunzinger, P.; Lee,
A. M.; Lee, M. S.; Liang, W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E. I.; Pieniazek, P.
A.; Rhee, Y. M.; Ritchie, J.; Rosta, E.; Sherrill, C. D.; Simmonett, A. C.; Subotnik, J. E.;
Woodcock, H. L.; Zhang, W.; Bell, A. T.; Chakraborty, A. K.; Chipman, D. M.; Keil, F.
J.; Warshel, A.; Hehre, W. J.; Schaefer, H. F.; Kong, J.; Krylov, A. I.; Gill, P. M. W.;
Head-Gordon, M. Physical Chemistry Chemical Physics 2006, 8, 3172.

(4) Rodler, M.; Blom, C. E.; Bauder, A. Journal of the American Chemical Society
1984, 106, 4029.

(5) Durig, J. R.; Shen, S.; Guirgis, G. A. Journal of Molecular Structure 2001, 560,
295.

(6) Joo, D. L.; Merer, A. J.; Clouthier, D. J. Journal of Molecular Spectroscopy 1999,
197, 68.

(7) Sinha, P.; Boesch, S. E.; Gu, C. M.; Wheeler, R. A.; Wilson, A. K. Journal of
Physical Chemistry A 2004, 108, 9213.

(8) Traeger, J. C.; Djordjevic, M. Eur. J. Mass Spec. 1999, 5, 319.

(9) Karpichev, B. K., L; Diri, K; Reisler, H.; Krylov, A.I. J Chem Phys 2010, 132,
114308.

(10) Cox, J. D.; Wagman, D. D.; Medvedev, V. A. CODATA Key Values for
Thermodynamics; Hemisphere Publishing Corp.: New York, 1989.

(11) da Silva, G.; Bozzelli, J. W. J Phys Chem A 2006, 110, 13058.

Abstract (if available)

Abstract The electronic spectroscopy and dissociation dynamics of 1- and 2- hydroxyethyl radicals were investigated using a combination of Resonance Enhanced Multiphoton Ionization (REMPI) and time-of-flight (TOF) detection of H and D photofragments.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Electronic states and photodissociation dynamics of hydroxyalkyl radicals

PDF

Photoelectron and ion imaging investigations of spectroscopy, photoionization, and photodissociation dynamics of diazomethane and diazirine

PDF

Photodissociation dynamics of atmospherically relevant small molecules in molecular beams

PDF

Dissociation energy and dynamics of HCl and aromatic-water clusters

PDF

Photodissociation dynamics of group V hydrides

PDF

Electronic structure and spectroscopy of excited and open-shell species

PDF

Photodissociation dynamics and energetics of HCl-(H₂O)₃

PDF

Electronic structure and spectroscopy of radicals in the gas and condensed phases

PDF

Ultrafast spectroscopic interrogation and simulation of excited states and reactive surfaces

PDF

Following redox chemistry and excited state dynamics in solution using liquid jet photoelectron spectroscopy

PDF

Development and implementation of methods for sliced velocity map imaging. Studies of overtone-induced dissociation and isomerization dynamics of hydroxymethyl radical (CH₂OH and CD₂OH)

PDF

Spectroscopy of hydrogen and water clusters in helium droplets

PDF

Unraveling photodissociation pathways in pyruvic acid and the role of methylhydroxycarbene

PDF

Imaging the state-specific vibrational predissociation of hydrogen bonded complexes

PDF

Ion imaging studies of the spectroscopy and photodissociation dynamics of chloromethyl radical and nitric oxide dimer

PDF

Fourier transform infrared studies of guest-host interactions in ice

PDF

Understanding organic and inorganic semiconductor surfaces by coherent nonlinear spectroscopy

PDF

Probing charge transfer and electric fields at the bulk and interface using vibrational spectroscopy

PDF

Vibrational spectroscopy and molecular orientation at interfaces: correlating the structure of a material's surface to its properties

PDF

Dissociation energy and dynamics of water clusters

Asset Metadata

Creator Edwards, Laura Wyman (author)

Core Title Spectroscopy and photodissociation dynamics of hydroxyethyl radicals

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry (Chemical Physics)

Publication Date 08/10/2010

Defense Date 05/21/2010

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag hydroxyethyl,OAI-PMH Harvest,REMPI,vinyl alcohol

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Reisler, Hanna (committee chair), Benderskii, Alexander (committee member), Kresin, Vitaly V. (committee member)

Creator Email ledwards@aaas.org,lwedward@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m3380

Unique identifier UC168448

Identifier etd-Edwards-3924 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-382914 (legacy record id),usctheses-m3380 (legacy record id)

Legacy Identifier etd-Edwards-3924.pdf

Dmrecord 382914

Document Type Dissertation

Rights Edwards, Laura Wyman

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu