Femtosecond pump-probe studies of dichlorine monoxide in solution

Article abstract of DOI:10.1021/jp030213l

The first femtosecond pump-probe studies of ClOCl photochemistry in solution are presented. Following 266-nm photoexcitation of ClOCl dissolved in perfluorohexane, the resulting evolution in optical density is monitored at seven wavelengths ranging from 266 to 400 nm. A depletion in optical density consistent with ground-state ClOCl photolysis is observed, followed by increases in optical density at 266 and 315 nm assigned to the production of ClO and ClClO, respectively. Kinetic analysis of the temporal evolution in optical density establishes that photoproduct appearance occurs on the ～10-ps time scale. Later time decay of the optical density at 315 nm is mirrored by a corresponding increase in optical density at 266 nm consistent with thermal decomposition of ClClO to produce ClO and Cl on the ～100-ps time scale. The quantum yields for photoproduct formation are determined through analysis of the absolute change in optical density. This analysis establishes that the quantum yields for ClClO and ClO production are 0.4 ±0.1 and 0.6 ± 0.1, respectively. Finally, the observation of ClClO production following ClOCl photoexcitation is similar to the behavior observed for chlorine dioxide (OClO) suggesting that photoisomerization is a general feature of halooxide reactivity in solution.

Full text of DOI:10.1021/jp030213l

5508
J. Phys. Chem. A 2003, 107, 5508-5514
ARTICLES
Femtosecond Pump-Probe Studies of Dichlorine Monoxide in Solution
Catherine C. Cooksey and Philip J. Reid*
Department of Chemistry, UniVersity of Washington, Box 351700, Seattle, Washington 98195
ReceiVed: February 17, 2003; In Final Form: May 1, 2003
The first femtosecond pump-probe studies of ClOCl photochemistry in solution are presented. Following
266-nm photoexcitation of ClOCl dissolved in perfluorohexane, the resulting evolution in optical density is
monitored at seven wavelengths ranging from 266 to 400 nm. A depletion in optical density consistent with
ground-state ClOCl photolysis is observed, followed by increases in optical density at 266 and 315 nm assigned
to the production of ClO and ClClO, respectively. Kinetic analysis of the temporal evolution in optical density
establishes that photoproduct appearance occurs on the ∼10-ps time scale. Later time decay of the optical
density at 315 nm is mirrored by a corresponding increase in optical density at 266 nm consistent with thermal
decomposition of ClClO to produce ClO and Cl on the ∼100-ps time scale. The quantum yields for
photoproduct formation are determined through analysis of the absolute change in optical density. This analysis
establishes that the quantum yields for ClClO and ClO production are 0.4 ( 0.1 and 0.6 ( 0.1, respectively.
Finally, the observation of ClClO production following ClOCl photoexcitation is similar to the behavior
observed for chlorine dioxide (OClO) suggesting that photoisomerization is a general feature of halooxide
reactivity in solution.
Introduction
ments also occurring resulting in re-formation of the parent
molecule.25-35 While geminate recombination and photoisomer-
ization are central to the reactivity of OClO, it is unclear if this
behavior is reflected by other halooxides such as ClOCl.
Halooxide photochemistry is of interest due to the participa-
tion of these compounds in a variety of atmospheric processes.1-4
The environmental impact of these compounds results from their
ability to photochemically produce atomic halogens, with the
extent of halogen production dependent on the environment in
which the chemistry occurs. For example, photolysis of chlorine
dioxide (OClO) promotes dissociation to form ClO and O, or
Cl and O2.5,6 In the gas phase, the quantum yield for atomic
chlorine production is modest with Φ ) 0.04; however, in
condensed environments the quantum yield increases to Φ )
0.1 and unity in aqueous solution and ice, respectively. Current
interest involves elucidating the fundamental reasons behind this
behavior in order to understand the environmental impact of
halooxides in both homogeneous and inhomogeneous settings.
Of interest here is the photochemical reaction dynamics of
dichlorine monoxide (ClOCl). It is believed that this compound
plays only a minor role in stratospheric chemistry. It is the
anhydrous form of hypochlorous acid (HOCl), a chlorine
reservoir species,7 and ClOCl is used as a convenient photolytic
source of ClO in laboratory studies.8 The photochemistry of
ClOCl provides a comparative example with respect to other
halooxides including OClO. Studies of gaseous OClO have
demonstrated that photolysis leads predominately to ClO and
O.9-20 However, in low-temperature matrixes, photochemical
production of the structural isomer, ClOO, occurs to an
appreciable extent.21-24 Solution photochemistry represents an
intermediate case in which both ClO/O and ClOO production
occurs, with geminate recombination of the ClO/O photofrag-
Our knowledge regarding ClOCl photochemistry is limited,
with only a few gas phase and in matrix isolation studies
presented in the literature.22,36-47 Nelson et al. studied the gas-
phase dissociation of ClOCl using photofragment translational
energy spectroscopy with photoexcitation at 308, 248, and 198
nm.40 Dissociation to form ClO and Cl was observed exclusively
with photoexcitation at 308 nm. However, with photoexcitation
at 248 nm, an additional channel was observed and assigned to
vibrationally excited ClO that undergoes secondary dissociation
to form Cl and O(3P). Finally, two distinct photochemical
channels were observed following 198-nm photoexcitation
involving Cl2 formation with atomic oxygen produced as either
O(3P) or O(1D). Broadband flash photolysis studies of Nick-
olaisen and co-workers were performed in which the yield of
ClO as a function of bath gas pressure was determined.41 The
ClO yield was observed to decrease with increasing pressure
for wavelengths greater than 300 nm leading to the suggestion
that photoexcitation of ClOCl at wavelengths greater than 300
nm results in rapid intersystem crossing to two triplet states
from which dissociation to form ClO and Cl or internal
conversion to the ground state occurs. Tanaka et al. have used
photofragment ion imaging to investigate the primary photo-
chemistry of gas-phase ClOCl following excitation at 235 nm
and found that production of the lowest-velocity Cl atoms was
consistent with a three-body fragmentation process.43 Finally,
Molina and co-workers have recently measured the quantum
yields of ClO and Cl production using chemical ionization mass
* To whom correspondence should be addressed. E-mail: preid@
chem.washington.edu.
10.1021/jp030213l CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/26/2003

Dichlorine Monoxide in Solution
J. Phys. Chem. A, Vol. 107, No. 29, 2003 5509
Figure 1. Absorption spectra of ClOCl and its potential photoprod-
ucts: Cl2, ClOOCl, ClO, and ClClO. The molar absorptivity values
are those of ClOCl dissolved in carbon tetrachloride.55 The gas-phase
absorption spectra of Cl2, ClOOCl, and ClO are taken from the
literature.51,54 The absorption spectrum of ClClO is for this species
isolated in an argon matrix.46 The dashed line indicates the excitation
wavelength employed in these studies.
Figure 2. Schematic of the pump-probe spectrometer employed in
this work. The following abbreviations have been used: BBO,
frequency doubling â-barium borate crystal; BS, beam splitter; C,
mechanical chopper; DL, optical delay line; DM, dichroic mirror; DP,
Brewster dispersion prisms; λ/2, zero-order half-wave plate; OPA,
optical parametric amplifier; 3×, frequency tripler; PL, calcite polarizer;
S, sample.
spectrometry following photolysis of gaseous ClOCl at 320, 360,
400, and 440 nm.44 When normalized to the quantum yield for
ClO and Cl production at 255 nm, it was found that the ClO
production quantum yield demonstrates little actinic-wavelength
dependence. In addition, these authors concluded that the
photochemical behavior could be interpreted entirely in terms
of transitions between singlet states, in opposition to the triplet-
state mechanism suggested by Nicolaisen et al.
The first matrix isolation studies of ClOCl were performed
by Rochkind and Pimentel.22 In this work, ClOCl in low-
temperature nitrogen and argon matrixes was photolyzed, and
the resulting reaction mixture was probed by IR and Raman
spectroscopy. It was established that photolysis of ClOCl results
in formation of the structural isomer, ClClO, and the ClO dimer.
Johnsson et al. reported the first UV absorption spectra of ClClO
formed by the photolysis of ClOCl in an argon matrix.46
Subtraction of the UV spectra before and after photolysis was
performed to determine the UV absorption spectrum of ClClO
(λmax ) 260 nm). In addition, using the absorption cross section
of gaseous ClOCl at 260 nm, an estimate for the ClClO
maximum absorption cross section of 1.3 × 10-17 cm2/molecule
was obtained.
spectral range over which optical-density evolution associated
with photoproduct formation is expected to occur. Following
ClOCl photoexcitation, substantial increases in optical density
are observed at 315 and 266 nm and are assigned to ClClO and
ClO, respectively. Kinetic analysis of this evolution reveals that
both photoproducts appear on the ∼10 ps time scale. In addition
to these dynamics, later-time decomposition of ClClO to produce
ClO and Cl is observed on the ∼100 ps time scale. Given these
results, a photochemical model is presented in which ClOCl
dissociates to form ClO and Cl or undergoes photoisomerization
to produce ClClO that undergoes subsequent ground-state
thermal dissociation to produce ClO and Cl. With this model,
quantum yields for ClClO and ClO production of 0.4 ( 0.1
and 0.6 ( 0.1, respectively, are established. Finally, the solution
phase photochemistry of ClOCl is compared to that of OClO.
The production of ClClO observed here mirrors the production
of ClOO following OClO photolysis suggesting that photoi-
somerization is a general feature of halooxide photochemistry
in solution.
There is very little information regarding the reaction
dynamics of ClOCl in solution. The first studies of ClOCl in
solution were performed in the 1930s,48 with further studies not
performed until recently. In this recent work, Esposito et al.49
obtained the resonance Raman spectrum of ClOCl in CCl4.
Using 282.4-nm excitation, they investigated the dependence
of the resonance Raman spectra on incident pulse energy and
observed scattering assignable to both ClClO and ClO. The
power dependence of photoproduct formation was investigated
and found to be consistent with both ClClO and ClO being
primary photoproducts of ClOCl.
In this paper, we present the first femtosecond transient
absorption study of ClOCl photochemistry in solution. The
evolution in optical density following ClOCl photolysis at 266
nm is monitored at seven probe wavelengths ranging from 266
to 400 nm. The absorption spectra of the parent species and all
potential photoproducts observed in studies performed to date
are presented in Figure 1. The figure demonstrates that the probe
wavelengths employed here provide good coverage of the
Experimental Section
A schematic diagram of the pump-probe spectrometer is
presented in Figure 2. An intracavity doubled Nd:VO4 laser
(Spectra Physics Millenia V) was used to pump a home-built
Ti:sapphire oscillator that produced 30-fs pulses (full width at
half-maximum (fwhm)) centered at 780 nm at a repetition rate
of 91 MHz. Amplification of these pulses was performed using
the chirped-pulse technique (Clark-MXR CPA-1000-PS). The
postcompression amplified output consisted of 85-fs fwhm
pulses centered at 780 nm with pulse energy of 850 µJ at a
repetition rate of 1 kHz.
The amplified output was split into two components. Twenty
percent of the output was frequency tripled (Spectra Physics
TP-IA) to generate the actinic field. The remainder of the
amplified output was used to generate the probe field. For probe
wavelengths at 400 and 266 nm, the amplified output was
frequency doubled and tripled, respectively. For probe wave-
lengths from 286 to 350 nm, the fourth harmonic of the signal

5510 J. Phys. Chem. A, Vol. 107, No. 29, 2003
Cooksey and Reid
TABLE 1: Kinetic Parameters Determined from Analysis of the Pump-Probe Data for ClOCl Dissolved in Perfluorohexane
b
λpra (nm)
A1
t1 (ps)
A2
t2 (ps)
A3
t3 (ps)
offsetd
266
286
300
317
350
400
-0.11 ( 0.06c
-0.40 ( 0.15
-0.38 ( 0.12
-0.47 ( 0.06
-0.54 ( 0.07
-0.99 ( 0.0003
9.6 ( 6.1
8.8 ( 2.9
11.4 ( 4.3
9.3 ( 4.3
3.2 ( 3.3
0.14 ( 0.10
-0.35 ( 0.13
0.60 ( 0.15
122.8 ( 78.6
>1000
0.53 ( 0.16
0.33 ( 0.15
41.6 ( 35.5
26.4 ( 6.2
28.3 ( 12.6
7.9 ( 3.4
0.29(0.27
381.6 ( 94.7
199.3 ( 118.3
196.2 ( 56.9
20.1 ( 6.7
0.42 ( 0.06
0.12(0.09
0.33 ( 0.12
0.13(0.05
-0.0001 ( 0.0002
0.0002 ( 0.0002
a Wavelength at which the transient change in optical density was measured. b Amplitudes are normalized such that ∑|Ai| ) 1. c Errors represent
1 standard deviation from the mean of all measurements employed at a given wavelength. d A time constant of 10 000 ps was included to represent
any persistent offset in optical density at long delay times.
field from an optical parametric amplifier (OPA, Quantronix
TOPAS) was employed. Isolation of the signal fourth harmonic
was accomplished using a pair of CaF2 prisms that also provided
precompensation for group-velocity dispersion of the probe field.
Following the prism pair, a beam splitter separated the probe
into the sample and reference beams. A calcite polarizer was
placed in the probe line to define the polarization. The pump
beam was temporally adjusted relative to the probe beam using
a delay stage. The pump polarization was oriented to 54.7°
relative to the probe using a zero-order half-wave plate to
minimize the contribution of rotational dynamics to the data.
Pump and probe pulse energies of 4.5 and 0.2 µJ, respectively,
were employed. The optical-density changes were found to vary
linearly with pump power. The focus of the pump beam was
adjusted so that the probe beam waist was approximately five
times smaller than that of the pump at the sample. The
instrument response was measured by the optical Kerr effect in
water and was 350 ( 50 fs for all probe wavelengths. Samples
were delivered to a fused-silica flow cell with CaF2 windows
having a path length of 0.2 mm. The flow rate was sufficient
to replenish the sample volume between actinic events.
Measurement of the pump-induced change in sample optical
density was performed as follows. The sample and reference
probe signals were detected with UV-enhanced photodiodes
(Advanced Photonix, Inc. SD 200-13-23-242). The photodiode
outputs were processed by gated integration (Stanford Research
Systems SR250) and subtracted on a shot to shot basis. A
mechanical chopper operating at half the amplifier repetition
frequency modulated the pump beam, and the gated-integrator
output was delivered to a computer where the voltage difference
between adjacent laser shots (i.e., pump on versus pump off)
was determined providing a measure of the pump-induced
optical-density change. Up to 1000 laser shots per time point
were acquired for an individual scan, with three or four scans
averaged to produce the data shown here. A convolution of the
instrument response with a sum of exponentials was fit to the
data using the Levenberg-Marquardt algorithm. Goodness of
the fit was judged by ø2 values as well as visual inspection of
the residuals. Errors reported here correspond to one standard
deviation of the mean of multiple measurements taken (five or
greater).
the reaction, the mercuric oxide was allowed to settle to the
bottom of the flask and the solution of ClOCl in PFH was
decanted, resulting in 35-40 mL of stock solution. Stock
solutions were typically 0.2-0.3 M in ClOCl as determined by
UV-vis absorption and would undergo decomposition into
mainly chlorine dioxide within 1-2 days. To minimize the
decomposition rate, stock solutions were further diluted with
PFH until a ClOCl concentration of ∼0.05 M was reached. This
diluted stock solution would last 1 week in the freezer. Further
dilution of these solutions was performed to produce the 7.2
mM solutions used in the pump-probe work. The results
reported here were determined to be identical with ∼2-fold
changes in concentration. The concentration of the sample was
monitored at the start and end of the experiment, and the sample
was kept at ∼0° C during the experiment using an ice bath. In
addition to the kinetic studies described above, the absolute
change in optical density at each probe wavelength was
determined employing a fresh 5.0 mM sample of ClOCl in PFH
to minimize variation in sample concentration in this absolute
comparative study.
Results
Figure 3 depicts the time evolution in optical density observed
following 266-nm photolysis of ClOCl in perfluorohexane.
Representative probe wavelengths at 266, 286, 300, 317, 350,
and 400 nm are presented in the figure. To simplify presentation,
the results are discussed with respect to three wavelength
regions, 266-300 nm, 300-350 nm, and greater than 350 nm.
266 nm e λpr < 300 nm. In this wavelength region, both
photoinduced ground-state ClOCl depletion and photoproduct
formation are observed. At 266 nm, an initial decrease in optical
density is observed consistent with the depletion of ground-
state ClOCl due to photolysis. Following this initial decrease,
an increase in optical density is observed that persists out to
the longest delays investigated (500 ps). The evolution in optical
density observed at 266 nm was fit to a sum of three
exponentials convolved with the instrument response resulting
in appearance time constants of 9.6 ( 6.1 and 122.8 ( 78.6 ps
and a long time offset. The time constants determined for this
and other probe wavelengths are presented in Table 1. At 286
nm, an increase in optical density is observed that appears with
a time constant of 8.8 ( 2.9 ps. Scans out to longest delays
suggest that there may be an extremely modest increase in
optical density occurring at long times, with a time constant
for this evoltion being >1000 ps.
Samples of dichlorine monoxide (ClOCl) dissolved in per-
fluorohexane (PFH) were prepared following a modified version
of the method outlined by Cady.50 Chlorine gas (99%, JCI Jones
Chemicals) was bubbled into a flask containing 50 mL of PFH
(Acros) cooled in a dry ice-acetonitrile bath. Under the
assumption that the molar absorptivity of Cl2 in PFH is the same
as in CCl4, Cl2 concentrations of 0.75 M were generated as
determined by UV-vis absorption. Yellow mercuric oxide
(Acros) was then added to the Cl2 solution at a molar ratio of
1.1 HgO to Cl2, and this mixture was stirred over an ice bath
for at least 45 min. Formation of ClOCl was evidenced by a
change in solution color from green to reddish brown. After
300 nm e λpr < 350 nm. The evolution in optical density
observed in this wavelength region is dominated by photoprod-
uct appearance and decay. At 300 nm, a substantial increase in
optical density occurs with a time constant of 11.4 ( 4.3 ps
and then undergoes biexponential decay with time constants of
41.6 ( 35.5 and 381.6 ( 94.7 ps. Similar behavior is observed
at 317 nm, where the increase in optical density occurs with a

Dichlorine Monoxide in Solution
J. Phys. Chem. A, Vol. 107, No. 29, 2003 5511
Figure 4. The transient absorption spectra of ClOCl dissolved in
perfluorohexane. The time delay for each spectrum is labeled in the
figure.
to the 300 and 317 nm data. At 350 nm, this optical density
increase occurs with a time constant of 3.2 ( 1.3 ps and
undergoes subsequent biphasic decay with time constants of 28.3
( 12.6 and 196.2 ( 56.9 ps. Similar behavior is observed at
400 nm, where an instrument-response-limited increase in
optical density is observed, which undergoes a subsequent
increase with a time constant of 7.9 ( 3.4 ps. The optical density
evolution at 400 nm ends by single-exponential decay with a
time constant of 20.1 ( 6.7 ps.
Transient Absorption Spectra. In addition to the single-
wavelength studies presented above, the temporal evolution of
the absolute optical density change as a function of probe
wavelength was determined as presented in Figure 4. There are
two major features evident in these data, the depletion and
subsequent increase in optical density at 266 nm and the increase
and subsequent decay in optical density at ∼315 nm. Following
photolysis, depletion at 266 nm is evident consistent with ClOCl
ground-state depletion due to photolysis. Studies of the optical
density change at early times employing finer time resolution
(data not shown) established a maximum optical-density deple-
tion of 0.008 ( 0.002 due to photolysis of ClOCl. In addition
to this depletion, an increase in optical density centered at ∼315
nm appears. As the depletion at 266 nm begins to grow steadily
into an absorption, the absorption at 315 nm also increases,
reaching a maximum of 0.030 OD at 17.5 ps. This absorption
then decreases to 0.014 OD by 100 ps. Comparison of the
maximum optical density for this feature to the 0.008 OD
depletion of ClOCl suggests that the species giving rise to the
strong absorption evident in Figure 4 has an extinction coef-
ficient that is 3.75 times greater than the ∼600 M-1 cm-1 value
for ClOCl at 266 nm. This simple analysis assumes that all
photoexcited ClOCl results in the production of the 315-nm
absorbing photoproduct, an assumption that will revisited
below.
Figure 3. Time-resolved pump-probe dynamics for ClOCl dissolved
in perfluorohexane. Probe wavelengths and optical density scale are
given in the figure. Best fit to the data by a sum of exponentials
convolved with the instrument response is given by the solid line, with
corresponding fit parameters presented in Table 1.
time constant of 9.3 ( 4.3 ps and decays with time constants
of 26.4 ( 6.2 and 199.3 ( 118.3 ps. As will be discussed below,
the amplitude of the optical density increase observed in this
wavelength region is significantly greater than the corresponding
depletion due to ClOCl photolysis. Therefore, these data
demonstrate that following ClOCl photolysis, a photoproduct
is formed whose absorption cross section is significantly greater
than that of the parent compound.
In summary, the results presented in Figure 4 demonstrate
that ClOCl photolysis results in the production of two species
that absorb at 315 and 266 nm. In addition to this result, there
are two other features evident in Figure 4 that should be noted.
There is a small optical density change at 400 nm observed
following photolysis, which quickly reaches a maximum of
0.008 OD before undergoing complete decay. This behavior is
consistent with the expected red-edge enhancement of the
absorption cross section accompanying vibrational excitation
for the species responsible for the 315-nm absorption band.
350 nm e λpr. In this wavelength region, an increase in
optical density is observed that is modest in amplitude relative

5512 J. Phys. Chem. A, Vol. 107, No. 29, 2003
Cooksey and Reid
Finally, there is an apparent isosbestic point at 286 nm for times
greater than 17.5 ps. The observation of an isosbestic point
suggests that the evolution in optical density in this wavelength
region is well modeled as a two-component system, with decay
of the species responsible for the 315 nm absorption band
resulting in the production of the species absorbing at 266 nm.
band is assigned to ClClO since it is the only species with the
requisite molar absorptivity to account for the observed mag-
nitude of optical-density evolution. Regarding the necessity of
a 55-nm shift in absorption-band position between the argon
matrix studies and this work for this assignment to hold,
theoretical calculations of the vertical excitation energies for
gaseous ClClO predict that the only electronic transition with
significant oscillator strength occurs at 298 nm, in agreement
with our assignment.53 Therefore, the question becomes why
one would expect any difference between the ClClO absorption
band maximum observed in low-temperature matrixes and that
predicted by theory. A potential answer to this question is found
in the environmental dependence of the ClClO molecular
structure. The IR absorption spectra of ClClO isolated in argon
and nitrogen matrixes indicate that the two chlorine atoms are
not equivalent and that the frequency of the Cl-O stretch is
only slightly shifted from that of isolated ClO.22 These observa-
tions indicate that in the matrix, ClClO is structurally perturbed
with an exceedingly weak Cl-Cl bond. Sensitivity of the Cl-
Cl bond strength to environment will be reflected by environ-
mental dependence of the ClClO potential energy surfaces. The
Cl-Cl bond is expected to be weaker in matrixes relative to
the gas phase due to the polarizability of the surrounding matrix.
Although one would generally expect similar behavior in
solution, the polarizability of perfluorohexane is quite modest.
As such, the Cl-Cl bond should be stronger, and the structure
of ClClO will correspondingly be more proximate to that of
gaseous ClClO. Therefore, a shift of the ClClO absorption-band
maximum between low-temperature matrixes and perfluoro-
hexane is not unreasonable.
Discussion
Discussion of the results presented above is predicted on
assignment of the optical-density evolution to those products
formed following ClOCl photolysis. As discussed in the
Introduction, the potential photolysis products and their corre-
sponding molar absorptivities are presented in Figure 1. The
species presented in the figure represent those photoproducts
observed in previous studies of ClOCl photochemistry in all
phases. Inspection of Figure 1 reveals that the current informa-
tion regarding photoproduct absorption cross sections cannot
be used to make a straightforward assignment of the evolution
at 315 nm. However, the simple estimate molar extinction
coefficient for the 315-nm absorbing species derived above
(2300 M-1 cm-1) is most consistent with the production of
ClClO, which has a molar extinction coefficient of 3400 M-1
cm-1 when isolated in low-temperature matrixes.46 Previous
studies of ClClO production following ClOCl photolysis were
performed in argon matrixes where the formation of ClClO was
evident by the appearance of an absorption band centered at
260 nm. Therefore, assignment of the 315-nm absorption band
observed in this study to ClClO requires that the absorption-
band maximum of this species undergoes a 55-nm shift between
a low-temperature argon matrix and perfluorohexane. We will
revisit this issue below after discussing other possible assign-
ments.
A second possible assignment for the 315-nm absorbing
species is that it corresponds to ClO. However, in both the gas
phase and water, absorption of this species has a maximum of
280 nm (Figure 1). In addition, the molar absorptivity of this
compound is only slightly larger than that of ClOCl such that
assignment of the 315-nm absorption band to ClO is question-
able. Some ClOCl photolysis studies have observed the forma-
tion of ClOOCl;22,45 however, it is doubtful that this species is
responsible for the absorption intensity at 315 nm. In addition
to the modest molar absorptivity of this compound (Figure 1),
this photoproduct is formed by a bimolecular process that should
be reflected by concentration dependence of the observed
kinetics. The evolution in optical density did not change with a
2-fold change in concentration suggesting that bimolecular
processes are not responsible for the observed photoproduct
formation dynamics (data not shown).
The increase in optical density observed at 266 nm (Figures
3 and 4) is assigned to production of ClO. Comparison to the
315-nm absorption band suggests that the absorbing species at
266 nm has a smaller molar absorptivity; however, comparison
to the reduction in optical density at 266 nm corresponding to
ClOCl depletion demonstrates that the molar absorptivity of this
species is greater than that of ClOCl. Given the discussion
presented above and the data presented in Figure 1, ClO is the
only reasonable assignment for optical-density evolution at 266
nm.
Further support for these assignments is found in the
resonance Raman studies of Esposito et al.49 In this work,
resonance Raman spectra of ClOCl in CCl4 were obtained with
excitation at 282.4 nm as a function of pulse energy. Figure 5
presents the power dependence of the ClOCl resonance Raman
spectrum. The spectra represents the scattering intensities
integrated over the temporal duration of the laser pulse (∼5
ns) such that all species formed to an appreciable extent, and
that have a significant scattering cross sections at this excitation
wavelength, will be observed. Figure 5A is the spectrum
obtained with a pulse energy of 57 µJ providing for significant
photolysis. Figure 5B is the spectrum obtained with a pulse
energy of 6.7 µJ corresponding to limited photolysis. By
subtraction of the low-energy spectrum from that obtained at
high pulse energy, scattering assignable to the photoproducts
is observed (Figure 5C). In this difference spectrum, scattering
assignable to ClClO and ClO is evident, demonstrating that these
two species are produced following ClOCl photoexcitation and
that these species have appreciable Raman scattering cross
sections consistent with the presence of oscillator strength in
this wavelength region.
The final photoproducts to consider are molecular and atomic
chlorine. Molecular chlorine does have an absorption maximum
at 330 nm, but the extinction coefficient for this transition is
only 66 M-1 cm-1, far below the estimate for the molar
absorptivity of the 315-nm absorbing species presented above.51
Studies of gaseous ClOCl have observed atomic chlorine
production; however, the absorption bands of atomic chlorine
are located in the far-UV, removed from the wavelength region
of interest. That said, it is possible that atomic chlorine forms
a charge-transfer complex with the solvent; however, the high
ionization potential of perfluorohexane (16.6 eV) should result
in a charge-transfer absorption band maximum of ∼162 nm.52
Therefore, it is doubtful that molecular or atomic chlorine are
responsible for the substantial changes in absorption intensity
observed at 315 nm.
The proposed ClOCl photochemical reaction dynamics that
occur following photoexcitation at 266 nm is presented in Figure
6. Photoexcitation of ground-state ClOCl results in photoi-
Given the above, we propose the following assignments for
the pump-probe results presented here. The 315-nm absorption

Dichlorine Monoxide in Solution
J. Phys. Chem. A, Vol. 107, No. 29, 2003 5513
photoproduct production as that determined by their respective
concentrations at 17.5 ps and write
1 ) φClClO(17.5 ps) + φClO(17.5 ps)
The 0.008 reduction in optical density at 266 nm (see above)
resulting from ground-state ClOCl depletion corresponds to a
concentration of 6.5 × 10-5 M. The total optical density change
of 0.003 OD observed at 17.5 ps at 266 nm contains a -0.008
( 0.002 OD contribution from ClOCl depletion; therefore, the
optical density originating from ClO is 0.011 ( 0.002 OD
(assuming only ClO and ClOCl contribute to the optical density
at 266 nm). Using this optical density and the 0.2-cm cell path
length employed, we can determine the concentration of ClO
by simply using the literature value for the molar extinction
coefficient of this species. Two choices for this constant are
available: ∼800 M-1 cm-1 for aqueous ClO,54 and 1386 M-1
cm-1 for gaseous ClO.51 Considering that the solvent employed
here is devoid of significant intermolecular interactions (see
above), we have employed the gas-phase extinction coefficient
for this analysis. Given this, we calculate a ClO concentration
at 17.5 ps of (4.0 ( 0.7) × 10-5 M such that the quantum yield
for ClO production is 0.6 ( 0.1, and the corresponding quantum
yield for ClClO production is 0.4 ( 0.1. By use of this quantum
yield for ClClO production and the 0.030 OD observed at 17.5
ps, the molar extinction coefficient for this species at 315 nm
is estimated to be 5700 ( 1100 M-1 cm-1, which is the expected
magnitude for this species.
With respect to halooxide reactivity as a whole, the ClOCl
photochemical model presented here has many similarities with
the solution-phase behavior observed for OClO. As mentioned
above, geminate recombination and photoisomerization play
prominent roles in the photochemical reaction dynamics of
OClO in solution.6 The model for ClOCl presented here does
not include geminate recombination for the following reason.
In the case of aqueous OClO, ∼90% of the photoexcited OClO
molecules undergo dissociation in to ClO and O which
subsequently recombine with a quantum yield of 0.9 ( 0.1.30,31
However, in a weakly associating solvent such as fluorotrichlo-
romethane, only 30% of the ClO and O photofragments undergo
geminate recombination.35 Given that the studies described here
are performed in perfluorohexane, it is likely that the extent of
geminate recombination is modest. On comparison of ClOCl
and OClO, photoisomerization occurs for both species, with
appearance and decay of the isomers (ClClO and ClOO,
respectively) occurring on similar time scales.35 Therefore, it
appears that photoisomerization is a general feature of halooxide
photochemistry in solution. However, the optical-density as-
signments discussed above need to be verified in order to
confirm the proposed model and quantum yields provided here.
Figure 5. (A) Resonance Raman spectrum of ClOCl in CCl4 obtained
with 57 µJ/pulse at 282.4 nm. Scattering due to ClOCl, ClClO, and
ClO is observed. (B) Resonance Raman spectrum of ClOCl in CCl4
obtained with 6.7 µJ/pulse at 282.4 nm. Scattering due to ClOCl is
still present, but scattering due to ClClO and ClO has diminished with
the decrease in pulse energy. (C) Difference spectrum generated by
subtracting (B) from (A) until the scattering due to ClOCl has vanished.
This demonstrates that the only photoproducts of ClOCl are ClClO
and ClO.
Figure 6. Proposed reaction scheme for the photolysis of ClOCl in
perfluorohexane. Excitation of ClOCl results in photodissociation to
produce ClO and Cl and photoisomerization to produce ClClO.
Following the initial photoproduct production, thermal dissociation of
ground-state ClClO results in the production of ClO and Cl.
somerization of the parent molecule to form ClClO as evidenced
by the appearance of a strong absorption at 315 nm. In addition,
primary dissociation to form ClO and Cl occurs as evidenced
by an increase in optical density at 266 nm. Both photoproducts
appear on the ∼10-ps time scale. Following these dynamics,
the isosbestic point at 286 nm for times greater than 17.5 ps is
consistent with the thermal decomposition of ClClO resulting
in ClO and Cl production on the ∼100-ps time scale. The
description of the reaction dynamics presented in Figure 6 is
basic, with effects such as vibrational relaxation not included
in the description at present. It is possible that the photoproduct
appearance times represent a combination of photoproduct
production and subsequent vibrational relaxation. However,
overlap of the photoproduct absorption bands (Figure 1) makes
the deconvolution of these effects based on the transient
absorption data rather ambiguous. Experiments better suited to
such a deconvolution are currently underway.
Given the model presented above, analysis of optical density
evolution evident in Figure 4 allows us estimate the quantum
yields for ClO and ClClO production. As noted above, an
isosbestic point is observed for delay times greater than 17.5
ps consistent with the evolution of a two-component system
assigned to the thermal decomposition of ClClO to produce ClO.
Given this observation, we define the quantum yield for
Conclusion
We have presented the first femtosecond transient absorption
studies of ClOCl photochemistry in solution. Photolysis of
ClOCl at 266 nm in perfluorohexane results in both ClClO and
ClO production, with these species appearing on the ∼10-ps
time scale. Following ClClO production, ground-state thermal
decomposition of this species occurs on the ∼100-ps time scale
resulting in the production of ClO and Cl. If one compares the
solution-phase photochemistry of ClOCl to that of OClO,
photolysis of both halooxides results in the production of their
respective structural isomers. Furthermore, the production and
decay kinetics of these isomers occurs on similar time scales
suggesting that photoisomerization is a general feature of
halooxide photochemistry in solution.

5514 J. Phys. Chem. A, Vol. 107, No. 29, 2003
Cooksey and Reid
(28) Thøgersen, J.; Thomsen, C. L.; J. Aa. Poulsen; Keiding, S. R. J.
Phys. Chem. A 1998, 102, 4186.
(29) Philpott, M. J.; Charalambous, S.; Reid, P. J. Chem. Phys. 1998,
236, 207.
(30) Hayes, S. C.; Philpott, M. J.; Reid, P. J. J. Chem. Phys. 1998, 109,
2596.
(31) Hayes, S. C.; Philpott, M. P.; Mayer, S. G.; Reid, P. J. J. Phys.
Chem. A 1999, 103, 5534.
(32) Thomsen, C. L.; Philpott, M. P.; Hayes, S. C.; Reid, P. J. J. Chem.
Phys. 2000, 112, 505.
(33) Thomsen, C. L.; Reid, P. J.; Keiding, S. R. J. Am. Chem. Soc. 2000,
122, 12795.
Acknowledgment. The authors thank Matthew Sulham and
Gordon Mitchell for their assistance in the preparation of ClOCl.
The National Science Foundation is acknowledged for their
support of this work (CHE-0091320). Acknowledgment is also
made to the donors of the Petroleum Research Fund, adminis-
tered by the American Chemical Society (P.J.R.). P.J.P. is an
Alfred P. Sloan Fellow and is a Cottrell Scholar of the Research
Corporation. C.C.C. would like to thank the Ford Foundation
for support through an environmental sciences fellowship.
(34) Philpott, M. J.; Hayes, S. C.; Thomsen, C. L.; Reid, P. J. Chem.
Phys. 2001, 263, 389.
References and Notes
(1) Rowland, F. S. Annu. ReV. Phys. Chem. 1991, 42, 731.
(2) Renard, J. B.; Pirre, M.; Robert, C.; Huguenin, D. J. Geophys. Res.
1998, 103, 25383.
(3) Solomon, S.; Borrmann, S.; Garcia, R. R.; Portmann, R.; Thomason,
L.; Poole, L. R.; Winker, D.; McCormick, M. P. J. Geophys. Res. 1997,
102, 21411.
(4) Donsig, H. A.; Herridge, D.; Vickerman, J. C. J. Phys. Chem. A
1998, 102, 2302.
(5) Vaida, V.; Simon, J. D. Science 1995, 268, 1443.
(6) Reid, P. J. J. Phys. Chem. A 2002, 106, 1473.
(7) Molina, M. J.; Tso, T.-L.; Molina, L. T.; Wang, F. C.-Y. Science
1987, 238, 1253.
(8) Renard, J. J.; Bolker, H. I. Chem. ReV. 1976, 76, 487.
(9) Davis, H. F.; Lee, Y. T. J. Phys. Chem. 1992, 96, 5681.
(10) Davis, H. F.; Lee, Y. T. J. Phys. Chem. 1996, 100, 8142.
(11) Bishenden, E.; Haddock, J.; Donaldson, D. J. J. Phys. Chem. 1991,
95, 2113.
(12) Bishenden, E.; Donaldson, D. J. J. Chem. Phys. 1993, 99, 3129.
(13) Bishenden, E.; Donaldson, D. J. J. Chem. Phys. 1994, 101, 9565.
(14) Baumert, T.; Herek, J. L.; Zewail, A. H. J. Chem. Phys. 1993, 99,
4430.
(15) Delmdahl, R. F.; Bakker, B. L. G.; Parker, D. H. J. Chem. Phys.
2000, 112, 5298.
(16) Delmdahl, R. F.; Ullrich, S.; Gericke, K. H. J. Phys. Chem. A 1998,
102, 7680.
(17) Delmdahl, R. F.; Baumgartel, S.; Gericke, K.-H. J. Chem. Phys.
1996, 194, 2883.
(18) Roth, M.; Maul, C.; Gericke, K. H. J. Chem. Phys. 1997, 107,
10582.
(19) Ruhl, E.; Jefferson, A.; Vaida, V. J. Phys. Chem. 1990, 94, 2990.
(20) Furlan, A.; Scheld, H. A.; Huber, J. R. J. Chem. Phys. 1997, 106,
6538.
(21) Arkell, A.; Schwager, I. J. Am. Chem. Soc. 1967, 89, 5999.
(22) Rochkind, M. M.; Pimentel, G. C. J. Chem. Phys. 1967, 42, 1361.
(23) Peterson, K. A.; Werner, H.-J. J. Chem. Phys. 1992, 96, 8948.
(24) Mueller, H. S. P.; Willner, H. J. Phys. Chem. 1993, 97, 10589.
(25) Thøgersen, J.; Jepsen, P. U.; Thomsen, C. L.; Poulsen, J. A.; Byberg,
J. R.; Keiding, S. R. J. Phys. Chem. A 1997, 101, 3317.
(26) Philpott, M. J.; Charalambous, S.; Reid, P. J. Chem. Phys. Lett.
1997, 281, 1.
(35) Hayes, S. C.; Thomsen, C. L.; Reid, P. J. J. Chem. Phys. 2001,
115, 11228.
(36) Edgecombe, F. H. C.; Norrish, R. G. W.; Thrush, B. A. Proc. R.
Soc. London 1957, A243, 455.
(37) Basco, N.; Dogra, S. K. Proc. R. Soc. London 1971, 323, 401.
(38) Sander, S. P.; Friedl, R. R. J. Phys. Chem. 1989, 93, 4764.
(39) Chichinin, A. I. Chem. Phys. Lett. 1993, 209, 459.
(40) Nelson, C. M.; Moore, T. A.; Okumura, M.; Minton, T. K. J. Chem.
Phys. 1994, 100, 8055. Moore, T. A.; Okumura, M.; Minton, T. K. J. Chem.
Phys. 1997, 107, 3337.
(41) Nickolaisen, S. L.; Miller, C. E.; Sander, S. P.; Hand, M. R.;
Williams, I. H.; Francisco, J. S. J. Chem. Phys. 1996, 104, 2857.
(42) Aures, R.; Gericke, K.-H.; Maul, C.; Trott-Kreigeskorte, G. J. Chem.
Phys. 2002, 117, 2141.
(43) Tanaka, Y.; Kawasaki, M.; Matsumi, Y.; Fujiwara, H.; Ishiwata,
T.; Rogers, L. J.; Dixon, R. N.; Ashfold, M. N. R. J. Chem. Phys. 1998,
109, 1315.
(44) Smith, G. D.; Tablas, F. M. G.; Molina, L. T.; Molina, M. J.J.
Phys. Chem. A 2001, 105, 8658.
(45) Chi, F. K.; Andrews, L. J. Phys. Chem. 1973, 77, 3062.
(46) Johnsson, K.; Engdahl, A.; Nelander, B. J. Phys. Chem. 1995, 99,
3965.
(47) Gane, M. P.; Williams, N. A.; Sodeau, J. R. J. Chem. Soc., Faraday
Trans. 1997, 93, 2747.
(48) Dickinson, R. G.; Jeffreys, C. E. P. J. Am. Chem. Soc. 1930, 52,
4288.
(49) Esposito, A. P.; Reid, P. J.; Rousslang, K. W. J. Photochem.
Photobiol. A 1999, 129, 9.
(50) Cady, G. H. Inorg. Synth. 1957, 5, 156.
(51) Demore, W. B.; Sander, S. P.; Golden, D. M.; Hampson, R. F.;
Kurylo, M. J.; Howard, C. J.; Ravishankara, A. R.; Kolb, C. E.; Molina,
M. J. In Chemical Kinetics and Photochemical Data for Use in Stratospheric
Modeling; NASA Contractor Report CR-198863; NASA Jet Propulson
Laboratory, California Institute ofTechnology, Pasadena, CA, 1994.
(52) Chateauneuf, J. E. Chem. Phys. Lett. 1989, 164, 577.
(53) Delbene, J. E.; Watts, J. D.; Bartlett, R. J. Chem. Phys. Lett. 1995,
246, 541.
(54) Klaning, U. K.; Wolff, T. Ber. Bunsen-Ges. Phys. Chem. 1985,
89, 243.
(55) Anbar, M.; Dostrovsky, I. J. Chem. Soc. 1954, 1105.
(27) Poulsen, J. A.; Thomsen, C. L.; Keiding, S. R.; Thøgersen, J. J.
Chem. Phys. 1998, 108, 8461.