Time- and product-resolved photodissociations of bromotoluene radical cations

Article abstract of DOI:10.1063/1.473289

Photodissociations of o-, m-, and p-bromotoluene radical cations have been studied in the wavelength range 575-475 nm using Fourier transform-ion cyclotron resonance (FT-ICR) mass spectrometry. The parent ions were prepared by charge-transfer reactions of bromotoluenes with toluene-d₈ radical cations produced by two-photon ionization of toluene-d₈ at 266 nm. Bromotoluene radical cations dissociate to C₇H₇⁺ by loss of Br. The dissociation rates were measured by time-resolved photodissociation spectroscopy. Structures of C₇H₇⁺ from one-photon dissociation were identified by their bimolecular reactivities with toluene-d₈. The C₇H₇⁺ products from all three isomers were identified as the benzyl cation. No unreactive tropylium ions were detected within experimental limits. The rate constants measured in this work were combined with the previous photoelcctron-photoion-coincidence results to refine activation parameters for the Rice-Ramsperger-Kassel-Marcus rate-energy curves, k(E), for the low barrier rearrangement process. The activation barriers are estimated to be 1.66, 1.80, and 1.78 eV for the o-, m-, and p-bromotoluene radical cations, respectively, whereas the entropy changes for the activation, ΔS^?(1000 K), are -9.6, -7.2, and -5.6 eu., respectively. The mechanism of the rearrangement process is presented to account for the predominant formation of the benzyl cation.

Full text of DOI:10.1063/1.473289

Time- and product-resolved photodissociations of bromotoluene radical cations
Byungjoo Kim and Seung Koo Shin
Citation: The Journal of Chemical Physics 106, 1411 (1997); doi: 10.1063/1.473289
View online: http://dx.doi.org/10.1063/1.473289
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/106/4?ver=pdfcov
Published by the AIP Publishing
Articles you may be interested in
Gas-phase electronic spectroscopy of the indene cation (C9H8 +)
J. Chem. Phys. 138, 224307 (2013); 10.1063/1.4808380
Photodissociation of gaseous CH3COSH at 248 nm by time-resolved Fourier-transform infrared emission
spectroscopy: Observation of three dissociation channels
J. Chem. Phys. 138, 014302 (2013); 10.1063/1.4768872
Product branching ratios in photodissociation of phenyl radical: A theoretical ab
initio/Rice–Ramsperger–Kassel–Marcus study
J. Chem. Phys. 136, 234305 (2012); 10.1063/1.4726455
Two-dimensional cross-spectral correlation analysis and its application to time-resolved Fourier transform
emission spectra of transient radicals
J. Chem. Phys. 123, 184104 (2005); 10.1063/1.2074147
Resolved high Rydberg spectroscopy of benzenerare gas van der Waals clusters: Enhancement of spin–orbit
coupling in the radical cation by an external heavy atom
J. Chem. Phys. 110, 5589 (1999); 10.1063/1.478457
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
131.94.16.10 On: Sun, 21 Dec 2014 20:38:07

Time- and product-resolved photodissociations of bromotoluene
radical cations
Byungjoo Kim and Seung Koo Shin
Department of Chemistry, University of California, Santa Barbara, Santa Barbara, California 93106
͑Received 29 August 1996; accepted 14 October 1996͒
Photodissociations of o-, m-, and p-bromotoluene radical cations have been studied in the
wavelength range 575–475 nm using Fourier transform-ion cyclotron resonance ͑FT-ICR͒ mass
spectrometry. The parent ions were prepared by charge-transfer reactions of bromotoluenes with
toluene-d8 radical cations produced by two-photon ionization of toluene-d8 at 266 nm.
Bromotoluene radical cations dissociate to C7Hϩ7 by loss of Br. The dissociation rates were
measured by time-resolved photodissociation spectroscopy. Structures of C7Hϩ7 from one-photon
dissociation were identified by their bimolecular reactivities with toluene-d8. The C7Hϩ7 products
from all three isomers were identified as the benzyl cation. No unreactive tropylium ions were
detected within experimental limits. The rate constants measured in this work were combined with
the previous photoelectron-photoion-coincidence results to refine activation parameters for the
Rice–Ramsperger–Kassel–Marcus rate-energy curves, k(E), for the low barrier rearrangement
process. The activation barriers are estimated to be 1.66, 1.80, and 1.78 eV for the o-, m-, and
p-bromotoluene radical cations, respectively, whereas the entropy changes for the activation,
⌬S‡͑1000 K͒, are Ϫ9.6, Ϫ7.2, and Ϫ5.6 eu., respectively. The mechanism of the rearrangement
process is presented to account for the predominant formation of the benzyl cation. © 1997
American Institute of Physics. ͓S0021-9606͑97͒00604-1͔
I. INTRODUCTION
the dissociation threshold to channel I are needed. In this
work, photodissociations of o-, m-, and p-bromotoluene
radical cations were studied in the internal energy range 2.2–
2.7 eV using Fourier transform ion cyclotron resonance ͑FT-
ICR͒ spectrometry. Bromotoluene radical cations are chosen
as they afford energy-selective access to channel II without
opening channel I owing to a large gap in dissociation
threshold between the two channels. For instance, thermo-
chemical thresholds for channel I are 2.68, 2.58, and 2.90 eV
for o-, m-, and p-isomers, respectively,1,3,9,13 whereas acti-
vation barriers for channel II determined from PEPICO ex-
periments in the internal energy range 2.7–3.5 eV are 1.80
eV for both o- and p-bromotoluene radical cations and 1.84
eV for the m isomer.1 A number of structural studies were
reported to date.1,4–7 Jackson et al. studied the structures of
C7Hϩ7 from EI of bromotoluenes by observing their ion-
molecule reactions with toluene using ICR spectrometry.7
They reported the branching ratio favoring the benzyl struc-
ture, ͓benzylϩ͔/͓tropyliumϩ͔, ranging from 99/1 at 14 eV to
93/7 at 25 eV, which was nearly identical among all three
isomeric radical cations. In contrast to the ICR results,
McLafferty and co-workers reported the opposite trend,
͓benzylϩ͔/͓tropyliumϩ͔ϭ18/82 at 70 eV, from the CA mass
spectrometric studies.5,6 Olesik et al. reexamined the struc-
tures by CA methods and reported the branching ratio favor-
ing the tropylium ion.1
Unimolecular dissociations of halotoluene radical cat-
ions have attracted much interest in recent years for the elu-
cidation of the mechanisms of the C7Hϩ7 formation.1–11 The
halotoluene radical cations ͑Cl, Br, and I͒ with a few eV
internal energy dissociate to C7Hϩ7 by loss of halogen. Olesik
et al. measured rates of energy-selective dissociations of ha-
lotoluene radical cations ͑Cl, Br, and I͒ in the internal energy
range 2.7–3.5 eV by using the photoelectron-photoion coin-
cidence ͑PEPICO͒ technique.1 Dunbar and co-workers mea-
sured rates of one-photon dissociation ͑1PD͒ of iodotoluene
radical cations in the internal energy range 2.1–2.7 eV by
time-resolved photodissociation ͑TRPD͒ spectroscopy.2,3
The structures of C7Hϩ7 products were studied in the separate
experiments employing collisional activation ͑CA͒ mass
spectrometry1,4–6 or ion cyclotron resonance ͑ICR͒
spectrometry.7 These studies provided a two-channel picture
as shown in Fig. 1, following Dunbar’s notation.3 The direct
C–X ͑Xϭhalogen͒ bond cleavage, denoted channel I, yields
tolyl cations, whereas the rearrangement process, denoted
channel II, produces either the benzyl or tropylium cation or
both. The identity of C7Hϩ7 from channel II has not been well
established, however, most workers have concluded that both
isomers are formed with a preference to the tropylium
ion.1–3,8 In structural studies,4–7 the C7Hϩ7 ions were pro-
duced mostly by electron impact ͑EI͒ that imparts a broad
internal energy distribution to the parent ion.12 Conse-
quently, a number of structures identified from there do not
correlate directly to the selected dissociation channels.
To help establish the mechanism of the C7Hϩ7 formation
from channel II, the energy-selective studies of both the rates
of dissociation and the structures of C7Hϩ7 products below
In order to overcome the deficiency of EI that lacks the
energy selectivity, the parent ions were prepared by the
photoionization-charge-transfer ͑PICT͒ method.12 The PICT
involves a sequence of ionization and charge-transfer pro-
cesses 1–3.
C D CD ϩ2h␯ 266 nm͒→C D CDϩ•ϩeϪ,
͑1͒
͑
6
5
3
6
5
3
J. Chem. Phys. 106 (4), 22 January 1997
0021-9606/97/106(4)/1411/7/$10.00
© 1997 American Institute of Physics
1411
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
131.94.16.10 On: Sun, 21 Dec 2014 20:38:07

1412
B. Kim and S. K. Shin: Photodissociation of bromotoluene cations
All chemicals were purchased from Aldrich Inc. and
used without further purification except several freeze-pump-
thaw cycles. Gaseous bromotoluene and toluene-d8 were
separately guided into the ICR cell by two electropolished
stainless steel tubes connected to either a leak valve or a
home-built pulsed valve. The purity of each sample was
tested by taking EI mass spectra. Sample inlet lines were
baked out and pumped out overnight to prevent any cross
contaminations between isomers. EI mass spectra of residual
background in the ICR chamber and the inlet lines were
checked for contamination before introducing a new sample.
The sample pressures were measured by observing the kinet-
ics of ion-molecule reactions with known rate constants.12,19
Partial pressures of bromotoluenes and toluene-d8 were kept
constant at 4ϫ10Ϫ8 and 2ϫ10Ϫ8 Torr, respectively.
Two-photon ionization of toluene-d8 was achieved by
the 266 nm laser pulse from the frequency quadrupled output
of a Nd:YAG laser ͑Spectra-Physics, GCR-150͒. A typical
pulse width was ϳ10 ns. The unfocused UV laser beam of a
3 mm diameter was passed through the center of the ICR cell
along the magnetic field axis. A typical laser power mea-
sured before entering a vacuum window was ϳ2 mJ/pulse.
Photodissociation of bromotoluene radical cations was ac-
complished in the wavelength range 575–475 nm by a coun-
terpropagating second laser beam that overlapped with the
UV laser beam. A 532 nm laser pulse was the frequency
doubled output of a Nd:YAG laser ͑Continuum, NY-8010͒.
The longer wavelength range 555–575 nm was covered by a
dye laser ͑Continuum, ND-6000͒ with Rhodamine-590
pumped by the 532 nm output of the Continuum Nd:YAG
laser. The shorter wavelength range 495–475 nm was pro-
vided by the same dye laser with Coumarine-480 pumped by
the 355 nm output of the Continuum Nd:YAG laser. The
photodissociation laser beam was adjusted to a 2-mm diam-
eter at a typical laser power of 1–3 mJ/pulse.
FIG. 1. A schematic illustration of the two-channel picture for dissociations
of halotoluene radical cations.
C6D5CDϩ3 •ϩCH3C6H4Br→C6D5CD3ϩCH3C6H4Brϩ•,
͑2͒
CH3C6H4Brϩ•ϩCH3C6H4Br→CH3C6H4BrϩCH3C6H4Brϩ•.
͑3͒
The two-photon ionization of toluene-d8 at 266 nm generated
its radical cation. Because the ionization potential ͑IP͒ of
toluene-d8, 8.83 eV,13 is higher than those of bromotoluenes,
ϳ8.7 eV,1 toluene-d8 radical cations underwent the charge-
transfer process 2 with bromotoluenes to yield bromotoluene
radical cations. The vibrationally excited bromotoluene radi-
cal cations then relaxed by both radiative relaxation14–16 and
thermalization by self-charge-exchange reactions with the
parent neutrals at room temperature.12 Unlike EI, the PICT
ensures a thermal internal energy distribution of the parent
ion in the absence of resistive heating by hot filament.12
The parent ions were then optically excited to the ex-
cited electronic state by a second laser in the wavelength
range 575–475 nm. The rapid internal conversion to the
ground electronic state prepares vibrationally hot molecular
ions.17 The wavelengths were chosen so as to excite the par-
ent ions above activation barriers for channel II but below or
near thermochemical thresholds for channel I. The rates of
1PD of bromotoluene radical cations were measured by
TRPD spectroscopy.2,3,12,18 The rate constants measured in
the internal energy range 2.2–2.7 eV were combined with
the previous PEPICO data1 in the internal energy range 2.7–
3.5 eV for the refinement of RRKM parameters. Structures
of C7Hϩ7 products were identified at each wavelength by ob-
serving their ion-molecule reactions with toluene-d8. The
identity of the C7Hϩ7 product combined with the activation
parameters are used to establish the mechanism of the rear-
rangement process.
The photodissociation laser was delayed 2 s after two-
photon ionization of toluene-d8. It ensures the completion of
charge-transfer reactions of toluene-d8 radical cations with
bromotoluenes and the subsequent thermalization by self-
charge-exchange reactions with bromotoluene neutrals.12 Be-
cause there is no resistive heating of the ICR cell by hot
filament, the temperature of molecule is considered the same
as the vacuum housing temperature ͑20 °C͒.12
For time-resolved detection of photodissociation pro-
cesses, the temporal buildup of C7Hϩ7 was monitored as a
function of time delay between the photodissociation laser
and the beginning of radio-frequency ͑rf͒ detection pulse.
The detection rf is in resonance with the ICR frequency of
C7Hϩ7 . The duration of the rf burst was set to 20 ␮s for the
slow kinetics measurements in the wavelength range 575–
532 nm. As the dissociation rate increases in shorter wave-
lengths, the rf duration was reduced to 10 ␮s. A precise
timing of the photodissociation laser pulse was monitored by
using a fast photodiode.
II. EXPERIMENT
Experimental setups were previously described in
detail.12 The FT-ICR spectrometer used in this work consists
of a 5.0 T superconducting magnet ͑Oxford Magnet Inc.͒, a
home-built vacuum chamber with a 2-in. cubic trapping ICR
cell, and an IonSpec Omega/486 FT-data system. There were
0.325-in. holes in the centers of both trapping plates to pro-
vide optical access along the axial direction. The background
pressure in the ICR chamber was typically below 9.0ϫ10Ϫ10
Torr.
The bimolecular reactivities of C7Hϩ7 products toward
toluene-d8 were monitored at each wavelength by taking
mass spectra at 20 s after photodissociation. The completion
of ion-molecule reactions was checked by observing tempo-
J. Chem. Phys., Vol. 106, No. 4, 22 January 1997
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
131.94.16.10 On: Sun, 21 Dec 2014 20:38:07

B. Kim and S. K. Shin: Photodissociation of bromotoluene cations
1413
FIG. 3. TRPD spectra of the m-bromotoluene radical cation at 570, 555,
532, and 495 nm.
FIG. 2. TRPD spectra of the o-bromotoluene radical cation at 570, 555,
532, and 495 nm. The solid curves are the calculated TRPD spectra using
the convoluted signal Eq. ͑5͒ with the RRKM parameters reported in Table
III.
the initial rate-energy curve parameters for RRKM calcula-
tions. Second, the convoluted signal Eq. ͑5͒ was used to
calculate the TRPD data at each wavelength to compare with
the experiments. Last, the RRKM parameters were refined
iteratively until the TRPD curves calculated using the Eq. ͑5͒
best fitted experiments at all wavelengths.
ral variations of ion intensities. The benzyl cation reacts with
toluene-d8 to yield CD3C6D4CHϩ2 , while the tropylium cat-
ion remains unreactive.7,19–21
III. TIME-RESOLVED DATA ANALYSIS
The ICR signal equation for the product ion is given in
Eq. ͑4͒ for the photodissociation of the parent ion with a
given internal energy E:12,18
IV. RESULTS
A. Time-resolved photodissociation (TRPD) spectra
and RRKM parameters
1Ϫexp Ϫk E͒⌬
͔
͓
͑
S k E͒,t ϭC⌬ 1Ϫexp Ϫk E͒t
,
͓ ͑
͔
͓
͑
͔
ͩ
ͪ
k E͒⌬
͑
TRPD spectra of o-, mϪ, and p-bromotoluene radical
cations are shown in Figs. 2, 3, and 4, respectively. TRPD
spectra for all three isomers show nonzero product yields at
tϭ0 owing to two-photon processes as reported previously.12
The rate of two-photon dissociation is much faster than the
time scale of the rf burst duration. The relative intensity of
time-zero intercept increases as the laser power increases,
confirming that the initial fast rising components during rf
burst excitation result from at least two-photon process.
͑4͒
where S[k(E),t] is the product ion intensity, C is the pro-
portionality constant, ⌬ is the duration of the detection rf
burst, k(E) is the dissociation rate constant for a given in-
ternal energy E, and t is the time delay between the photo-
dissociation laser pulse and the beginning of the detection rf
burst. As the dephasing rates for the parent and the daughter
ions in a 5 T magnetic field are much faster than the disso-
ciation rate, the parent ion correction term due to dephasing
is not included in the signal Eq. ͑4͒. Because the parent ions
produced by charge-transfer reactions were thermalized at
the room temperature, the signal equation was convoluted
with a thermal internal energy distribution of the parent
ions:22,23
E
0
S t͒ ϭ
P E ͒S k E͒,t ,
͓ ͑
͑5͒
͑
͑
͔
͗
͘
͚
T
E
ϭ0
T
where S(t) is the convoluted signal equation and P(E ) is
͗
͘
T
a truncated Boltzmann distribution of the parent ion with a
dissociation threshold, E0. E is the sum of a thermal internal
energy and a photon energy, EϭETϩh␯.
A typical procedure for the rate-energy curve fitting is as
follows.12 First, the TRPD data were fitted to a single expo-
nential curve to give an approximate rate constant at each
wavelength. The approximate rate constants as a function of
FIG. 4. TRPD spectra of the p-bromotoluene radical cation at 575, 555,
average internal energy, Eϭ E ϩh␯, were then used to set
532, and 495 nm.
͗
͘
T
J. Chem. Phys., Vol. 106, No. 4, 22 January 1997
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
131.94.16.10 On: Sun, 21 Dec 2014 20:38:07

1414
B. Kim and S. K. Shin: Photodissociation of bromotoluene cations
TABLE I. Average dissociation rate constants of bromotoluene radical cat-
ions.
a
k ͑sϪ1
͒
Wavelength
ET ϩh␯
͗
͘
͑nm͒
eV
ortho
meta
para
575
570
555
532
495
475
2.258
2.277
2.336
2.430
2.606
2.712
2000Ϯ 100
6900Ϯ 600
11900Ϯ1000
23000Ϯ1500
1100Ϯ 100
1800Ϯ 200
3700Ϯ 200
4700Ϯ 400
8700Ϯ 500
50000Ϯ3600 20000Ϯ 700 31000Ϯ1400
45000Ϯ5000
aErrors are estimated from the uncertainty of fitting TRPD spectra.
Hence, the photodissociation laser power was adjusted in
such a way to make 1PD become dominant in the TRPD
spectra.
FIG. 6. RRKM rate-energy curves for the m-bromotoluene radical cation.
The solid lines drawn in Figs. 2–4 along with experi-
mental data are the best-fit curves using the convoluted sig-
nal Eq. ͑5͒ based on the refined RRKM parameters. The
calculated curves fit the experimental data very well at all
wavelengths. The average dissociation rate constant obtained
at each wavelength is summarized in Table I. In the common
internal energy range 2.34–2.61 eV, the average rate con-
stants are 11.9–50ϫ103 sϪ1 for o isomer, 1.8–20ϫ103 sϪ1
for m isomer, and 4.7–31ϫ103 sϪ1 for the p isomer. Since
the rates of 1PD are much faster than the infrared-radiative
relaxation rate of 160 sϪ1 reported for the p-iodotoluene
ion,2 the radiative relaxation processes are not included in
data analyses. The dissociation rate decreases in order; ortho,
para-, and meta-isomers. The same trend was previously ob-
served from 1 PD of iodotoluene radical cations at 532 nm in
our laboratory.24
C–Br dissociation channel I are also included in Figs. 5–7 as
dotted lines. They are calculated by using thermochemical
thresholds summarized in Table II. The entropy changes for
channel I are assumed to be the same for all three isomers as
those of iodotoluene radical cations, ⌬S‡͑1000 K͒ϭ7.6 eu.3,8
In the present wavelength range, average internal energies of
o and p isomers are less than thermochemical thresholds for
channel I, whereas those of the m isomer in the 495–475 nm
range lie above the threshold for channel I. Since the disso-
ciation rate to channel I is several orders of magnitude
slower than that to channel II, the relative yield of channel I
must be less than 0.1% even with the highest energy studied
in the PEPICO experiments. Hence, the previous PEPICO
data in the internal energy range 2.7–3.5 eV are combined
with the present results in the internal energy range 2.2–2.7
eV to derive the RRKM parameters for channel II. Table III
lists the RRKM parameters that fit both experiments. The
activation barriers, E0, are 1.66, 1.80, and 1.78 eV for o-,
m-, and p-bromotoluene radical cations, respectively. The
entropy changes for activation, ⌬S‡͑1000 K͒, are Ϫ9.0 eu
for ortho, Ϫ7.2 eu for meta, and Ϫ5.6 eu for para isomers,
which indicates the tight transition states for channel II.
The average rate constants are plotted as a function of
average internal energy in Figs. 5, 6, and 7 for o, m, and p
species, respectively. The solid symbols represent this work
and the open symbols denote the PEPICO results from the
Ref. 1. For comparisons, the rate-energy curves for the direct
FIG. 5. RRKM rate-energy curve for the o-bromotoluene radical cation.
Solid symbols denote the data from this work and open symbols from the
previous PEPICO studies. The solid line represents the RRKM rate-energy
curve for channel II and the dotted line is for channel I.
FIG. 7. RRKM rate energy curves for the p-bromotoluene radical cation.
J. Chem. Phys., Vol. 106, No. 4, 22 January 1997
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
131.94.16.10 On: Sun, 21 Dec 2014 20:38:07

B. Kim and S. K. Shin: Photodissociation of bromotoluene cations
1415
TABLE II. Thermochemical data for chemical species related to this work.
TABLE III. RRKM kinetic parameters E0 ͑eV͒ and ⌬S‡͑1000 K͒ ͑eu͒ for
the bromotoluene radical cations.a
a
a
M
IP͑M͒ ͑eV͒
⌬H0f,0(Mϩ) ͑kJ molϪ1
͒
ortho
meta
para
o-CH3C6H4Br
m-CH3C6H4Br
p-CH3C6H4Br
C6D5CD3
C6H5CH2 ͑benzyl͒
c-C7H7 ͑tropylium͒
o-C6H4CH3 ͑o-tolyl͒
m-C6H4CH3 ͑m-tolyl͒
p-C6H4CH3 ͑p-tolyl͒
⌬Hf,0͑Br͒aϭ117.9 kJ molϪ1
8.64Ϯ0.01
8.73Ϯ0.01
8.67Ϯ0.01
8.83b,c
937
943
͑a͒ Channel Ia
a
E0
1.66
Ϫ9.0
1.80
Ϫ7.2
1.78
Ϫ5.6
932
⌬S‡
924c
919
͑b͒ Channel IIb
7.20c
6.24c
8.71a,b
E0
2.68
7.6
2.58
7.6
2.90
7.6
⌬S‡
d
͑1078͒
1074
1094
aThe estimated errors are Ϯ0.02 eV.
bE0’s are estimated from thermochemical thresholds for the direct dissocia-
tion and ⌬S‡’s are from the iodotoluene ion system ͑Refs. 3 and 8͒.
aReferences 1, 25, 26.
bReference 27.
cReference 13.
tivities with toluene-d8. All C7Hϩ7 ions underwent reaction 8
to completion and yielded CD3C6D4CHϩ2 , indicating the ben-
zyl structure.7,19–21,28
dEstimated from the activation barrier, 2.11 eV, for the direct dissociation of
o-iodotoluene radical cation ͑Ref. 3͒.
C7Hϩ7 ϩC6D5CD3→CD3C6D4CHϩ2 ϩC6H5D.
͑8͒
Figure 8 combines the rate-energy curves for all three
isomeric ions for direct comparisons. The meta-
bromotoluene radical cation shows the slowest dissociation
in the entire energy range. The o isomer dissociates faster
than the para isomer in the low energy regime, while the
PEPICO results show comparable dissociation rates for both
isomers in the high energy regime starting from ϳ2.7 eV.
The slope of the rate-energy curve that indicates the tightness
of the transition state becomes less negative in going from
ortho to meta to para isomers. In the case of ortho isomer,
the PEPICO data deviates significantly from the RRKM fit in
the intermediate energy regime.
No unreactive components were observed even after 20 s,
suggesting the absence of the tropylium ion from 1PD in the
present wavelength range. It is concluded that the rearrange-
ment channel II leads to the formation of the benzyl ion in all
three isomeric ions.
Meanwhile, when all C7H7ϩ products from both one- and
two-photon processes were analyzed, there remained a small
fraction of C7Hϩ7 unreactive. They are most likely from a
two-photon process. The previous studies in our laboratory
revealed that the tolyl cations from 2PD of iodotouene radi-
cal cations at 640 nm isomerize mostly to the benzyl ion with
a small fraction of the tropylium ion.24 It is suggested that
the unreactive tropylium ions result from the isomerization
of tolyl cations from two-photon process. Further quantita-
tive analyses for the two-photon products were not per-
formed.
B. The structures of C7H؉7 products
Because the products ions formed within the first 20 ␮s
were originated from more than one-photon processes, they
were ejected out of the ICR cell by applying an ejection rf
pulse after 20 ␮s from photodissociation. The rest of C7H7ϩ
ions, resulted solely from 1PD, were analyzed for their reac-
V. DISCUSSIONS
One of the most important results from this study is the
identification of the structures of C7Hϩ7 products from chan-
nel II in conjunction with the rates of energy-selective dis-
sociations of bromotoluene radical cations. It was previously
suggested from the CA mass spectrometric studies that the
rearrangement channel II led to both the benzyl and tropy-
lium cations.1,2,5 To the contrary, herein, we provide evi-
dence against the formation of the tropylium ion in channel
II from all three isomeric ions. Similar results were previ-
ously observed from iodotoluene radical cations.24 The chlo-
rotoluene radical cations are expected to yield similar results
to those of both bromo- and iodotoluene radical cations. The
discrepancy between the present study and the earlier CA
studies may be due to the difference in preparation of C7Hϩ7 .
In CA studies, the ionizing electron energy was 70 eV and
the source temperature was 150 °C, whereas, in the present
study, the bromotoluene radical cations were prepared at
room temperature by the PICT and C7Hϩ7 was produced by
energy-selective photodissociations. The observation of the
tolyl structure in CA studies indicates that the internal energy
of the precursor ion from EI may have been significantly
FIG. 8. RRKM rate-energy curves for channel II of all three isomers. The
solid symbols are the experimental data from this work and the open sym-
bols from the previous PEPICO studies.
J. Chem. Phys., Vol. 106, No. 4, 22 January 1997
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
131.94.16.10 On: Sun, 21 Dec 2014 20:38:07

1416
B. Kim and S. K. Shin: Photodissociation of bromotoluene cations
␣-H migration than ͓1,3͔ ␣-H migration.29 With halotolu-
enes, however, the positive charge may be localized near
halogens by forming partial double-bond resonance struc-
tures that perturb electronic configurations of the ring car-
bons. Thus, the barrier height as well as the direction of
H-atom migration can be significantly modified from the
toluene system. The positive charge character delocalized on
the halogen atom is evident in photoelectron spectra of ha-
lobenzenes and halotoluenes.34–36
A. The [1,2] ␣-H migration
The ͓1,2͔ ␣-H migration is a well-known rearrangement
process in carbonium ions.30,33 Once formed, intermediate
radical cations can subsequently undergo either ͓1,2͔ or ͓1,4͔
CH2 migrations and/or ͓1,2͔ H migrations. The benzyl ion
precursors can be formed by the ͓1,2͔ H migration in the
ortho-isomer, or by the sequential ͓1,4͔ CH2 and ͓1,2͔ H
migrations in the meta-isomer, or by the ͓1,4͔ CH2 migration
in the para-isomer. The ͓1,2͔ CH2 migration may lead to an
insertion of CH2 into a C–C bond to form the cycloheptatrie-
nyl halide radical cation as a precursor for the tropylium ion.
The ͓1,4͔ CH2 migration may involve bicyclo-͓2,2,1͔-hepta-
2,5-dienyl halide radical cations as local minima along the
reaction coordinate. The energetics of both ͓1,2͔ and ͓1,4͔
CH2 migrations are not well established. The barrier for the
͓1,2͔ H migration along the benzene ring is estimated to be
ϳ0.8 eV.31
FIG. 9. The proposed mechanism of dissociations of bromotoluene radical
cations to C7Hϩ7 with a loss of Br.
higher than the channel I threshold. The EI also permits dif-
ferent isomers of bromotoluene to be formed, thereby lead-
ing to different product ion structures. The present results
certainly raise questions about the validity of many EI/CA
results on the structure of the lowest barrier product.
It is interesting to see that activation parameters do not
differ significantly among o, m, and p isomers. In addition,
the activation parameters are not affected very much by the
halogen substitution ͑Cl, Br, and I͒. For chlorotoluene radical
cations, Olesik et al. reported activation barriers of 1.70 eV
for the ortho isomer and 1.77 eV for both meta- and para
isomers.1 For iodotoluene radical cations, Dunbar and co-
worker determined activation barriers of ϳ1.88 eV for all
three isomers.2,3 These similarities in the product structures
and kinetics among different halotoluenes provide a newer
insight into the intramolecular rearrangement process in ha-
lotoluene radical cations. First, the halogen atom is not con-
sidered to be an active species that migrates in the rearrange-
ment process. Second, the C–X ͑Xϭhalogen͒ bond cleavage
occurs at the final step after forming precursors for the ben-
zyl cation. Last, the halogen atom may affect the rearrange-
ment process by forming resonance structures. McLafferty
and Bockhoff have previously proposed the isomerization of
halotoluene radical cations to methylene–cyclohexadienyl
halide radical cations to account for the benzyl ion
formation.4 Our observations lend experimental supports to
the McLafferty rearrangement, although the existence of
such reaction intermediates needs to be verified by experi-
ments.
Figure 9 illustrates the mechanism modified from that of
McLafferty et al.4 to account for the present results in con-
junction with other relevant experimental1–11 and theoretical
studies.27,29–33 The first step is the ͓1,2͔ or ͓1,3͔ sigmatropic
migration of an ␣-H atom to the ipso- or ortho-position, re-
spectively. The second and subsequent steps include the
͓1,2͔ or ͓1,4͔ CH2 migration and ͓1,2͔ H-atom migration. No
theoretical transition state calculations are available for ha-
lotoluene radical cations. In the case of toluene radical cat-
ion, ab initio calculations predict a lower barrier for ͓1,2͔
B. The [1,3] ␣-H migration
The ͓1,3͔ ␣-H migration is not a favored process in neu-
tral systems.30 For the radical cations, however, it may be
comparable in energetics to the ͓1,2͔ ␣-H migration.33 The
͓1,3͔ ␣-H migration forms the methylene–cyclohexadienyl
halide radical cations in o-isomer by one step. For meta- and
para-isomers, the formation of benzyl ion precursors in-
volves the subsequent ͓1,2͔ H migrations. The difference be-
tween ͓1,2͔ and ͓1,3͔ ␣-H migrations is in the CH2 migra-
tion. The former process involves both CH2- and H-atom
migrations, whereas the latter process relies only on the
H-atom migration. Therefore, the ͓1,3͔ ␣-H migration can
lead to the exclusive formation of the benzyl ion without
branching to the tropylium ion.
Though there is no direct evidence for or against the
͓1,2͔ ␣-H migration mechanism, the absence of tropylium
ion favors the ͓1,3͔ ␣-H migration as the primary step in the
rearrangement processes. This model predicts a lower activa-
tion barrier for the o-bromotoluene radical cation than m-
and p-isomers, as the bromine in the ortho position helps
develop a more positive charge character on the ortho car-
bon. It is well known that the positive charge character helps
the H atom to migrate in radical cations.30 In cases of chlo-
rotoluene radical cations, the ortho-isomer has a lower bar-
rier than other isomers.1 For iodotoluene radical cations, the
rates of 1PD at 532 nm decreases in order of ortho-, para,
and meta-isomers, similar to those observed in bromotoluene
radical cations.24 On the other hand, if the ͓1,2͔ ␣-H migra-
tion is the primary step, the subsequent H- or CH2-migration
J. Chem. Phys., Vol. 106, No. 4, 22 January 1997
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
131.94.16.10 On: Sun, 21 Dec 2014 20:38:07

B. Kim and S. K. Shin: Photodissociation of bromotoluene cations
1417
must be the rate-determining step. The absence of the tropy-
lium ion suggests the ͓1,2͔ CH2 migration as the highest
activation barrier process. The difference in barrier height
among three isomers indicates the ͓1,2͔ H migration as the
lowest activation process. The proposed mechanism calls for
further experimental studies of the identification of reaction
intermediates and theoretical transition state calculations
along the reaction coordinate.
7 J.-A. A. Jackson, S. G. Lias, and P. Ausloos, J. Am. Chem. Soc. 99, 7515
͑1977͒.
8 Y. S. Cho, M. S. Kim, and J. C. Choe, Int. J. Mass. Spectrom. Ion Pro-
cesses 145, 187 ͑1995͒.
9 R. C. Dunbar, J. P. Honovich, and B. Asamoto, J. Phys. Chem. 92, 6935
͑1988͒.
10 R. C. Dunbar and G. H. Weddle, J. Phys. Chem. 92, 5706 ͑1988͒.
11 C. Lifshitz, I. Levin, S. Kababia, and R. C. Dunbar, J. Phys. Chem. 95,
1667 ͑1991͒.
12 S. K. Shin, S. J. Han, and B. Kim, Int. J. Mass. Spectrom. Ion Processes
͑in press͒.
13 S. G. Lias, J. E. Bartmess, J. F. Liebman, J. L. Holmes, R. D. Levin, and
W. G. Mallard, J. Phys. Chem. Ref. Data 17, Suppl. 1 ͑1988͒.
14 R. C. Dunbar, J. Phys. Chem. 91, 2801 ͑1987͒.
15 R. C. Dunbar, J. Chem. Phys. 90, 7369 ͑1989͒.
16 R. C. Dunbar, Mass Spectrom. Rev. 11, 309 ͑1992͒.
17 R. C. Dunbar, in Gas Phase Ion Chemistry, edited by M. T. Bowers
͑Academic, New York, 1979͒, Vol. 3, Chap. 20.
18 R. C. Dunbar, J. Phys. Chem. 91, 2801 ͑1987͒.
19 P. Ausloos, J.-A. A. Jackson, and S. G. Lias, Int. J. Mass. Spectrom. Ion
Processes 33, 269 ͑1980͒.
VI. CONCLUSION
The rates of energy-selective dissociations of bromotolu-
ene radical cations as well as the structures of the corre-
sponding C7Hϩ7 products were determined using FT-ICR
spectrometry. In contrast to the widely accepted presumption
that the lowest activation barrier rearrangement process leads
to both the benzyl and tropylium ions with a preference to
the tropylium ion, the present study reveals the benzyl cation
as the predominant rearrangement product. The primary
source of the tropylium ion in unimolecular dissociations of
halotoluene radical cations is considered to be the isomeriza-
tion of tolyl cations produced via the direct C–X bond cleav-
age. The rate measurement below thresholds of the direct
C–Br bond cleavage allows the accurate determination of
activation parameters for the rearrangement process. The
proposed mechanism calls for theoretical investigations.
20 S. Wexler and R. P. Clow, J. Am. Chem. Soc. 90, 3940 ͑1968͒.
21 P. Ausloos, J. Am. Chem. Soc. 104, 5259 ͑1982͒.
22 F. S. Huang and R. C. Dunbar, Int. J. Mass. Spectrom. Ion Processes 109,
151 ͑1991͒.
23 J. D. Faulk, R. C. Dunbar, and C. Lifshitz, J. Am. Chem. Soc. 112, 7893
͑1990͒.
24 S. K. Shin, B. Kim, R. L. Jarek, and S.-J. Han ͑unpublished͒.
25 T. Baer, J. C. Morrow, J. D. Shao, and S. Olesik, J. Am. Chem. Soc. 110,
5633 ͑1988͒.
26 J. C. Treager and R. G. McLoughlin, Int. J. Mass Spectrom. Ion Phys. 27,
319 ͑1978͒.
27 C. Lifshitz, Acc. Chem. Res. 27, 138 ͑1994͒.
28 J. Shen, R. C. Dunbar, and G. A. Olah, J. Am. Chem. Soc. 96, 6227
͑1974͒.
ACKNOWLEDGMENTS
S.K.S. acknowledges the support from the National Sci-
ence Foundation Grant No. CHE-9302959, the National Sci-
ence Foundation Young Investigator Award CHE-9457668.
This work was also made possible by the Santa Barbara La-
ser Pool under NSF Grant No. CHE-9413030.
29 C. Lifshitz, Y. Gotkis, A. Ioffe, J. Laskin, and S. Shaik, Int. J. Mass.
Spectrom. Ion Processes 125, R7 ͑1993͒.
30 R. A. Woodward and R. Hoffman, The Conservation of Orbital Symmetry
͑Academic, New York, 1971͒.
31 W. J. van der Hart, J. Am. Soc. Mass Spectrom. 6, 513 ͑1995͒.
32 C. Cone, M. J. S. Dewar, and D. Landman, J. Am. Chem. Soc. 99, 372
͑1977͒.
33 M. J. S. Dewar and D. Landman, J. Am. Chem. Soc. 99, 2446 ͑1977͒.
34 K. Kimura, S. Katsumata, Y. Achiba, T. Yamazaki, and S. Iwata, Hand-
book of HeI Photoelectron Spectra of Fundamental Organic Molecules
͑Halsted, New York, 1981͒.
1 S. Olesik, T. Baer, J. C. Morrow, J. J. Ridal, J. Buschek, and J. L. Holmes,
Org. Mass Spectrom. 24, 1008 ͑1989͒.
2 R. C. Dunbar and C. Lifshitz, J. Chem. Phys. 94, 3542 ͑1991͒.
3 C. Y. Lin and R. C. Dunbar, J. Phys. Chem. 98, 1369 ͑1994͒.
4 F. W. McLafferty and F. M. Bockhoff, Org. Mass Spectrom. 14, 183
͑1979͒.
35 D. W. Turner, C. Baker, A. D. Baker, and C. R. Brundle, Molecular
Photoelectron Spectroscopy ͑Wiley-Interscience, London, 1970͒.
36 E. W. Fu, P. P. Dymerski, and R. C. Dunbar, J. Am. Chem. Soc. 98, 337
͑1976͒.
5 F. W. McLafferty and J. Winkler, J. Am. Chem. Soc. 96, 5182 ͑1974͒.
6 C. J. Proctor and F. W. McLafferty, Org. Mass Spectrom. 18, 193 ͑1983͒.
J. Chem. Phys., Vol. 106, No. 4, 22 January 1997
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:
131.94.16.10 On: Sun, 21 Dec 2014 20:38:07