Photodissociation of gaseous NCl3 at 193 and 249 nm

Article abstract of DOI:10.1021/jp025655w

The optical absorption spectrum of gaseous NCl₃ was measured from 170 to 350 nm, and experiments were performed in which this molecule was irradiated with a pulsed laser at 193 nm and 249 nm. Photodissociation channels accessible at these wavelengths can produce fragments in singlet, doublet, or triplet manifolds of states. Photodissociation at 249 nm produces prompt banded emission in the visible and near-IR spectral regions that is tentatively ascribed to transitions in NCl₂, suggesting that the dissociation channel leading to doublet fragments (NCl₂ and Cl) is operative at this photolysis wavelength. The absence of emission from excited singlet states of NCl or excited triplet states of Cl₂ suggests that the singlet and triplet dissociation channels are not important at 249 nm. For photolysis at 193 nm, prompt emission is observed from both the a¹Δ and b¹Σ⁺ excited states of NCl. No emission is observed from either NCl₂ or excited triplet states of Cl₂. A measure of the quantum yield for NCl(a¹Δ) production by the 193 nm photodissociation of NCl₃ was determined by observing the transfer of energy from excited NC1 to iodine atoms in the pulsed photolysis system. This method suggests that the yield has a value of 0.8 ± 0.2.

Full text of DOI:10.1021/jp025655w

8422
J. Phys. Chem. A 2002, 106, 8422-8426
Photodissociation of Gaseous NCl3 at 193 and 249 nm†
R. D. Coombe,* J. V. Gilbert, S. S. Beaton, and N. Mateljevic
Department of Chemistry and Biochemistry, UniVersity of DenVer, 2190 E. Iliff AVenue,
DenVer, Colorado 80208
ReceiVed: February 17, 2002; In Final Form: June 4, 2002
The optical absorption spectrum of gaseous NCl3 was measured from 170 to 350 nm, and experiments were
performed in which this molecule was irradiated with a pulsed laser at 193 nm and 249 nm. Photodissociation
channels accessible at these wavelengths can produce fragments in singlet, doublet, or triplet manifolds of
states. Photodissociation at 249 nm produces prompt banded emission in the visible and near-IR spectral
regions that is tentatively ascribed to transitions in NCl2, suggesting that the dissociation channel leading to
doublet fragments (NCl2 and Cl) is operative at this photolysis wavelength. The absence of emission from
excited singlet states of NCl or excited triplet states of Cl2 suggests that the singlet and triplet dissociation
channels are not important at 249 nm. For photolysis at 193 nm, prompt emission is observed from both the
a1∆ and b1Σ+ excited states of NCl. No emission is observed from either NCl2 or excited triplet states of Cl2.
A measure of the quantum yield for NCl(a1∆) production by the 193 nm photodissociation of NCl3 was
determined by observing the transfer of energy from excited NCl to iodine atoms in the pulsed photolysis
system. This method suggests that the yield has a value of 0.8 ( 0.2.
NCl3* (singlet) f NCl(a1∆, b1Σ+) + Cl2(X1Σg )
f NCl2(doublet states) + Cl (2P)
(1)
(2)
(3)
+
Introduction
Nitrogen trichloride has been studied by many researchers
since its identification by Davy1 in the early nineteenth century,
and much of this work has been focused on reactions in solution
or in the gas phase2-4 and on the mechanism of the auto-
decomposition of this highly energetic species.5 Very little is
known about the photochemistry of NCl3 apart from the UV
absorption spectrum reported by Clark and Clyne,2 and results
of exploratory work by Gilbert and co-workers6 on the photo-
dissociation of NCl3 at 249 and 308 nm. Recently, NCl3 has
once again become a subject of interest, this time as a chemical
source of electronically excited NCl(a1∆). Excited NCl(a1∆) has
been shown to effectively transfer energy to iodine atoms,7,8
and both pulsed and cw lasers operating on the 52P1/2 f 52P3/2
transition in iodine have been demonstrated based on this
mechanism.9,10 Gaseous NCl(a1∆) might be generated from NCl3
by either reactions or photodissociation. Exton and co-workers3
have shown that excited NCl(a1∆) is produced when NCl3 reacts
with either H or D atoms, and Schwenz, et al.11 have
demonstrated the efficient production of excited I(52P1/2) when
HI is added to an NCl3/H reactive flow. Work on the auto-
decomposition of NCl3 has produced no evidence of the
production of excited NCl, nor was this species found in the
work by Gilbert and co-workers6 on photolysis at 249 and 308
nm.
f NCl(X3Σ-) + Cl2(3Π states)
In the previous work by Gilbert, et al.,6 for photolysis at 249
and 308 nm, the principal result was the observation of strongly
banded emission in the visible region that was thought to arise
from excited states of either NCl2 (channel (2) above) or the
parent NCl3. In the present work, we report additional, more
extended observations of NCl3 photolysis at 249 nm, and new
results from photolysis of the molecule at 193 nm. The data
indicate the operation of both channels (1) and (2) above, and
we present a measurement of the quantum yield for channel
(1) for photolysis at 193 nm, based on the observation of
emission from excited I(52P1/2) when an iodine atom precursor
is added to the system.
Experimental Details
NCl3 was generated in a shielded apparatus in a fume hood
by passing gaseous Cl2 through an acidified solution of (NH4)2-
SO4, using techniques that have been described in detail
previously.3,6 The NCl3 was collected in a trap at 193 K, and
subsequently allowed to warm to room temperature to allow
any Cl2 in the trap to evaporate into a stream of Ar. At room
temperature, the residual NCl3 is a golden yellow liquid. Once
the NCl3 was generated, it was entrained in a flowing stream
of Ar, and the proportion of gaseous NCl3 in the flow was
determined by passage of the flow through a 1 cm absorption
cell and measurement of the absorbance of 254 nm light from
a low-pressure Hg lamp.
The NCl3 absorption spectrum published by Clark and Clyne2
exhibits a strong absorption with a maximum near 220 nm and
a long, lower absorbance tail extending into the visible. The
latter feature is responsible for the golden yellow appearance
of pure liquid NCl3. Optical excitation of ground state (singlet)
NCl3 produces an excited singlet that can dissociate to two
singlet, two doublet, or two triplet fragments as follows
A small portion of the NCl3/Ar flow was admitted to a
photolysis cell through a passivated stainless steel metering
valve. The remainder of the flow was returned to the fume hood.
† Part of the special issue “Donald Setser Festschrift”.
* To whom correspondence should be addressed. Phone: (303)-871-
2966. E-mail: rcoombe@du.edu.
10.1021/jp025655w CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/20/2002

Photodissociation of Gaseous NCl3
J. Phys. Chem. A, Vol. 106, No. 36, 2002 8423
The photolysis cell was constructed from a stainless steel block
with 2.54 cm diameter channels for the flow of gases, the
passage of the photoylsis laser beam, and the observation of
emissions from the photolysis zone. The photolysis laser beam
passed through the cell via 15 cm long sidearms that were
baffled to reduce the effects of light scattering. The windows
attached to the ends of these sidearms were UV grade fused
silica. Near-IR emission from photolysis products was viewed
at 90° to the axis of the laser beam via a polished KCl window.
IR emissions were observed either directly through the window
by a cooled intrinsic Ge detector equipped with appropriate
filters, or by transmitting the emission via an optical fiber to a
0.25 m monochromator that dispersed the emission before its
detection. Emission in the visible region was observed through
a second fused silica window, also at 90° to the axis of the
laser beam, with an optical fiber/monochromator/photomultiplier
apparatus. The photolysis cell was pumped with a mechanical
pump, and flows were such that the total pressure in the cell
was varied from 0.1 to 1.5 Torr. The pressure in the cell was
measured with a capacitance manometer. For most measure-
ments, the pump was throttled such that the residence time of
the flowing gases in the observation zone of the cell was at
least 8 ms.
Figure 1. Absorption spectrum of gaseous NCl3. Broken line, data
from ref 2; ([) and solid line, present data normalized to ref 2 data at
254 nm.
The gases flowing through the cell were irradiated by the
pulsed emission from an excimer laser (Questek) operated at 1
Hz on either the ArF or KrF transitions at 193 and 249 nm,
respectively. The fluence of the laser pulses was varied between
8 and 116 mJ cm-2. Spectra of emissions produced by the
photolysis were obtained by scanning the monochromator noted
above as the photolysis laser was repetitively pulsed, with data
being acquired using a gated integrator (Stanford Research
Systems) set for various gate widths and delays. Time profiles
of individual emission features were recorded with a digital
oscilloscope (LeCroy). Both the gated integrator and digital
oscilloscope were interfaced to a computer for data display and
analysis.
UV/VUV absorption spectra of NCl3 were recorded using a
VUV absorption spectrometer (Acton Research) equipped with
a deuterium lamp and a photomultiplier tube with a sodium
salicylate phosphor. A flowing stream of NCl3 diluted in Ar
(generated as described above) was passed through a 1.0 cm
path length cell mounted between the exit slit of the mono-
chromator and the photomultiplier tube. Absorption spectra were
determined by manual, point-by-point analysis of background
and sample spectra.
Figure 2. Spectra of emissions in the visible and near-IR regions
produced by 249 nm photolysis of an NCl3/Ar mixture (approximately
2% NCl3) at a total pressure of 0.88 Torr. Spectrum A (500-800 nm)
is from ref 6. Spectrum B (from 750 to 1400 nm) is from present data.
For spectrum B, the delay time from the laser pulse was 15 µs and the
gate width was 15 µs.
2. Photodissociation at 249 nm. Experiments with photolysis
at 249 nm were performed in an effort to reproduce and extend
the data previously reported by Gilbert and co-workers.6
Previous data consisted of spectra of banded emission extending
from roughly 500 to 775 nm. The features of this spectrum
(thought6 to be associated with excited states of NCl2 or NCl3)
were found to have emission lifetimes on the order of 3 µs (for
NCl2) and 34 µs (for NCl3). In the present experiments, we
recorded spectra from 700 to 1500 nm using the 0.25 m
monochromator (equipped with a grating blazed at 2.0 µm and
used in second order) in combination with the cooled Ge
detector, for a variety of delays from the photolysis pulse and
a gate width of 15 µs. Figure 2 shows the spectrum recorded
for a delay of 15 µs, along with the spectrum reported by Gilbert
and co-workers. Clearly, the banded spectrum of the present
data is the extension of that reported by Gilbert, et al., with the
features at 752 and 768 nm common to both data sets.
Results
1. Absorption Spectrum of NCl3. The absorption spectrum
of NCl3 reported by Clark and Clyne2 indicates a major feature
with a maximum near 220 nm, but does not include data for
wavelengths below about 210 nm. Because the present experi-
ments employed photolysis radiation at 193 nm, it was first
necessary to measure the absorption spectrum of the molecule
at shorter wavelengths. Figure 1 shows the spectrum recorded
using the method described above, along with the spectrum
measured by Clark and Clyne. The two spectra are in good
agreement, as expected since the densities of NCl3 in the present
experiments were determined by optical absorption at 254 nm,
using the extinction coefficient at that wavelength from the Clark
and Clyne spectrum.
Time profiles for the banded emission features of the present
spectrum were observed using the digital oscilloscope. The
features were observed to have rapid rise and decay times limited
by the time response of the Ge detector (13 µs), consistent with
the shorter lifetime reported by Gilbert, et al.6 Further, the
intensities of individual features (recorded with the Ge detector
across the full wavelength range of the emission) were all found
to increase linearly with the incident laser fluence. These results
indicate that the emitter of the banded spectrum is a direct
fragment of one-photon photodissociation of NCl3 at 249 nm.
From the present data, it is clear that the very strong
absorption with a maximum at 220 nm extends well below 200
nm. The absorption cross section at 193 nm is 4.5 × 10-18 cm2.
Further, a second maximum is observed at 210 nm, suggesting
a second optical transition. This feature was very reproducible.
The emission spectra recorded in these experiments exhibited
no evidence of the presence of either excited NCl or excited
Cl2, corresponding to photodissociation channels 1 and 3 above.

8424 J. Phys. Chem. A, Vol. 106, No. 36, 2002
Coombe et al.
Figure 3. Spectrum of emission in the visible region produced by
193 nm photolysis of an NCl3/Ar mixture (approximately 2% NCl3) at
a total pressure of 0.40 Torr. For the data shown, the delay time from
the laser pulse was 500 µs, and the gate width was 15 µs. Vibrational
sequences of the b1Σ+ f X3Σ- transition in NCl are indicated.
In scans of the near-IR and visible regions with a variety of
delays up to 1.0 ms, there was no evidence of emission from
either the a1∆ or b1Σ+ states of NCl, which emit at 1076 and
665 nm, respectively. Emission from NCl(a1∆) was sought by
using the Ge detector coupled with a narrow band-pass filter
centered at 1080 nm. Emission from NCl(b1Σ+) was sought by
recording spectra in the visible region with the GaAs photo-
multiplier tube and monochromator (in this case used in first
order with a grating blazed at 500 nm).
Excited Cl2(3Π) has a characteristic banded emission spectrum
in the red visible and near-IR, and is produced5,6 by the auto-
decomposition of NCl3. This emission is easily distinguished
from the emission shown in Figure 2, or from the red emission
of the NCl(b1Σ+ fX3Σ-) transition by either visual observation
or spectral scans, as was noted by Gilbert and co-workers.6 There
is no evidence of emission from excited Cl2 in the spectrum
shown in Figure 2. In the present experiments, we were able to
induce the auto-decomposition flame by throttling the pump such
that the photoysis cell was not fully purged between laser pulses.
In this case, the spectrum in the red visible (recorded with the
photomultiplier tube) clearly revealed features attributable to
Cl2(3Π), with a detector-limited rise and a decay over a few
µs.
3. Photodissociation at 193 nm. Photolysis of NCl3/Ar
mixtures at 193 nm with the same experimental conditions as
employed for the experiments at 249 nm gave different results.
Gated spectra recorded in the visible and near-IR with the GaAs
photomultiplier tube and Ge detector, respectively, exhibited
none of the bands shown in Figure 2. Further, there were no
bands attributable to emissions from Cl2(3Π) states. Instead, the
spectrum of emission in the visible region exhibited the features
shown in Figure 3. These features are clearly identified as the
sequences of the b1Σ+ f X3Σ- transition12 in NCl, with the
most intense band at 665 nm corresponding to the ∆V ) 0
sequence of this transition. Features corresponding to the
∆V ) ( 1, (2, and +3 sequences are also evident, and their
intensities indicate that the excited NCl(b1Σ+) is formed with
considerable vibrational excitation.12,13 The intensity of the
emission varied linearly with the fluence of the photolysis laser,
suggesting that NCl(b1Σ+) is produced directly by the single
photon photodissociation of NCl3.
Figure 4. Time profile of the emission from NCl(a1∆, V ) 0) produced
by 193 nm photolysis of an NCl3/Ar mixture (2.0% NCl3) at a total
pressure of 0.16 Torr.
exhibits a rise limited by the bandwidth of the detector (77 kHz)
followed by a slow decay over several ms. The intensity of this
emission from NCl(a1∆) also varied linearly with the fluence
of the photolysis laser. The decay time of the emission and the
density of NCl3 in the system (1.0 × 1014 cm-3) suggest that
collisional quenching of the excited NCl(a1∆) by the residual
parent NCl3 has a rate constant no greater than 2.5 × 10-12
cm3s-1. In fact, we believe the rate constant to be substantially
smaller than this value, since for the conditions of Figure 4 (total
pressure 0.16 Torr) the loss of NCl(a1∆) is likely to be
dominated by collisions with the vessel wall.
4. Excitation of Iodine Atoms. In previous experiments
performed in our laboratory,7 emission from I(52P1/2) produced
by excitation in collisions with NCl(a1∆) was observed from
the pulsed photolysis of mixtures of ClN3 and CH2I2 at 193
nm. Photodissociation of ClN3 at this wavelength produces
NCl(a1∆) with a branching fraction14 of at least 0.7, and
photodissociation of CH2I2 produces CH2I and I atoms. Of the
I atoms produced from the dissociation of CH2I2, 95% are in
the ground 52P3/2 state and 5% are in the excited 52P1/2 state.15
The collisional excitation of the ground-state iodine atoms by
the NCl(a1∆) present is easily observed in the time profile of
the iodine emission at 1315 nm. The excited iodine produced
by direct photolysis of CH2I2 has a detector limited rise followed
by a slow exponential decay over several ms. The excited iodine
produced by collisional excitation appears as an emission signal
rising over a few hundred µs to a ms after the laser pulse
(depending on the densities of the collision partners), followed
once again by a slow decay over a few ms. Because the rate
constant for excitation of iodine atoms to the 52P1/2 state by
collisions with NCl(a1∆) is known,7,8 and the enhancement in
the absolute number density of I(52P1/2) caused by such
collisional excitation can be experimentally determined from
the I(52P1/2) emission time profile (by using the time profile of
excited iodine emission from phototdissociation of CH2I2 as a
standard), observation of the production of excited iodine atoms
by 193 nm photolysis of mixtures of CH2I2 and NCl3 offers a
means of determining the density of NCl(a1∆) present, and
hence the quantum yield for its production by photodissociation
of NCl3.
Emission from NCl(a1∆) produced by the photolysis at 193
nm was observed by using the cooled Ge detector combined
with a narrow band filter centered at 1080 nm. The transmission
band of the filter was sufficiently narrow such that only the
∆V ) 1 band of the emission was transmitted. Figure 4 shows
the time profile of the NCl (a1∆, ∆V ) 0) emission. The profile
Figures 5A, B, and C show time profiles of 1315 nm iodine
emission produced by the 193 nm photodissociation of CH2I2

Photodissociation of Gaseous NCl3
J. Phys. Chem. A, Vol. 106, No. 36, 2002 8425
of the emission signal to a maximum at 440 µs after the laser
pulse, followed by a decay over several ms. The maximum
intensity of the emission produced from the NCl3/CH2I2/Ar
mixture is a factor of 2.8 greater than the initial intensity of the
emission from CH2I2/Ar, shown in the middle trace, indicating
that the densities of excited and ground-state iodine at the time
of the maximum are 5.6 × 1011 and 3.5 × 1012 cm-3
,
respectively.
At the maximum, the steady-state expression for the density
of excited iodine atoms is
[I(52P1/2)] ) ke[I(52P3/2)][NCl(a1∆)]/kq
(4)
where ke is the second-order rate constant for excitation of iodine
atoms in collisions with NCl(a1∆) and kq is a first-order rate
constant for the overall quenching of the excited iodine atoms.
This expression can be rearranged to give the density of excited
NCl at the steady state, as follows
[NCl(a1∆)] ) (kq/ke)([I(52P1/2)]/[I(52P3/2)]
(5)
For the conditions of Figure 5A, [I(52P1/2)]/[I(52P3/2)] ) 0.16
at the steady state. The value of ke is known7,8 to be 1.6 ×
10-11cm3s-1 at room temperature, and kq is determined from
either the rise or the decay of the iodine emission signal shown
in the figure, depending on which corresponds to the quenching
rate. If, for example, the signal decay (360 s-1) were to
correspond to the quenching rate, then the density of NCl(a1∆)
at the steady state would be determined to be 3.6 × 1012 cm-3
from eq 5. This cannot be the case, though, because for such a
density the rate of excitation of the iodine atoms (using ke )
1.6 × 10-11 cm3s-1) would be only 58 s-1, far slower than the
rate implied by the rise of the signal shown in the figure. On
the other hand, if the rise of the signal (at a rate of 5000 ( 500
s-1, determined by a fit to the data) were to correspond to the
decay of the excited iodine, then eq 5 would suggest the density
of NCl(a1∆) at the steady state to be 5.0 × 1013 cm-3. For this
density the pseudo-first-order rate of excitation of the iodine
atoms would be 800 s-1, significantly faster than the observed
decay of the emission signal (360 s-1). Because the density of
NCl(a1∆) is far greater than the density of ground-state iodine
atoms, and the rate of excitation of the iodine atoms is much
slower than the rate of quenching of the excited iodine, we infer
that the observed decay in fact corresponds to the rate of removal
of NCl(a1∆) from the system. Hence, the rise of the signal would
indeed appear to correspond to the rate of decay of the excited
iodine atoms, and the density of excited NCl at the steady state
is determined to be 5.0 × 1013 cm-3. This value may be
compared to the density of NCl3 dissociated by the 193 nm
laser pulse, 5.1 × 1013 cm-3, to give an apparent quantum yield
for the production of NCl(a1∆) of 0.98.
Figure 5. Time profiles of emission from excited iodine atoms
produced from the pulsed photolysis of either CH2I2/Ar or CH2I2/NCl3-
Ar mixtures, for three different conditions. For each case, a baseline is
shown for photolysis of Ar diluent alone. (A) Lower trace, emission
from I(52P1/2) produced by 193 nm photolysis of a CH2I2/Ar mixture;
initial density of excited iodine atoms is 2.0 × 1011 cm-3. Upper trace,
emission from I(52P1/2) produced by 193 nm photolysis of a CH2I2/
NCl3/Ar mixture. The initial density of excited iodine atoms is 2.0 ×
1011 cm-3, and the density of NCl3 removed by the photolysis is 5.1 ×
1013 cm-3. (B) Lower trace, emission from I(52P1/2) produced by 193
nm photolysis of a CH2I2/Ar mixture; initial density of excited iodine
atoms is 3.3 × 1011 cm-3. Upper trace, emission from I(52P1/2) produced
by 193 nm photolysis of a CH2I2/NCl3/Ar mixture; initial density of
excited iodine is 3.3 × 1011 cm-3, and the density of NCl3 removed by
the photolysis is 1.6 × 1014 cm-3. (C) Lower trace, emission from
I(52P1/2) produced by 193 nm photolysis of a CH2I2/Ar mixture; initial
density of excited iodine atoms is 3.1 × 1011 cm-3. Upper trace,
emission from I(52P1/2) produced by 193 nm photolysis of a CH2I2/
NCl3/Ar mixture; initial density of excited iodine atoms is 3.1 ×
1011 cm-3, and the density of NCl3 removed by the photolysis is 2.8 ×
1014 cm-3
.
(diluted in Ar) and mixtures of CH2I2 and NCl3 (also diluted in
Ar) under three different experimental conditions. As noted in
the figure caption, the densities of photofragments are smallest
for Figure 5A and largest for 5C. In each figure, the bottom
trace is a baseline produced by photolysis of Ar alone. The
middle trace corresponds to emission from excited iodine
produced by photodissociation of CH2I2/Ar. In the middle trace
of Figure 5A, for example, the laser fluence, the initial density
of CH2I2, the CH2I2 absorption cross section, and the known
I(52P1/2)/I(52P3/2) ratio15 are such that the initial densities of
excited state and ground-state iodine atoms for these condi-
The data shown in Figure 5B,C correspond to substantially
higher densities of NCl3 photofragments. A treatment analogous
to that described above results in calculated NCl(a1∆) quantum
yields of 0.56 and 0.46, respectively, for these data. These values
are substantially lower than that calculated from the data of
Figure 5A, and the difference suggests that quenching of
NCl(a1∆) by photofragments in the time between the laser pulse
and the maximum of the iodine emission signal may have an
effect. Taken together, the three results suggest that the intrinsic
quantum yield for production of NCl(a1∆) at 193 nm is
0.8 ( 0.2.
tions are calculated to be 2.0 × 1011 and 3.8 × 1012 cm-3
,
respectively. The short initial “spike” observed in the time
profile corresponds to emission from I2 present in the CH2I2 as
an impurity. The longer decay over several ms corresponds to
quenching of the excited iodine atoms by collisions with other
species present in the mixture or the cell walls.
The upper trace in each figure corresponds to emission from
excited iodine atoms with NCl3 added to the CH2I2/Ar mixture.
In the upper trace of Figure 5A, the excitation of the iodine
atoms by collisions with NCl(a1∆) is clearly evident in the rise
Discussion
The branching fractions among the singlet, doublet, and triplet
photodissociation channels are quite different for photolysis at

8426 J. Phys. Chem. A, Vol. 106, No. 36, 2002
Coombe et al.
249 and 193 nm. The channel leading to singlet photofragments
(eq 1) is clearly open at 193 nm, with a high quantum yield for
the production of NCl(a1∆). Because emission from excited
Cl2(3Π) states was not observed for either 193 or 249 nm
photolysis, it would appear that dissociation of NCl3 to triplet
fragments (eq 3) is not open at these wavelengths. As noted
above, emission from excited Cl2 is readily observed when the
NCl3 auto-decomposition flame is ignited by throttling the pump.
It is interesting to note that emission from excited NCl(a1∆,
bΣ+) is not observed in the auto-decomposition flame, and
indeed emission from excited I(52P1/2) produced by the photo-
dissociation of CH2I2/NCl3/Ar mixtures at 193 nm is quenched
when the auto-decomposition flame is ignited. These observa-
tions are consistent with the auto-decomposition mechanism
proposed by Rubtsov,5 in which Cl atoms are among the
principal products. These atoms are efficient quenchers16,17 of
both NCl(a1∆) and I(52P1/2).
0.8), then NCl(a1∆) self-quenching, with a rate constant19 of
7.2 × 10-12 cm3s-1, would account for another 290 s-1. Hence
the total quenching rate should be about 350 s-1, in very good
agreement with the observed result. For the conditions of Figure
5, this self-quenching process should be dominant in the removal
of excited NCl and the observed decays should in fact be second
order.
The rapid loss of excited I(52P1/2) from the system, exhibited
by the rise of the signals in Figure 5, is more difficult to account
for. The most likely quencher would appear to be the residual
NCl3 in the system. To account for the observed rates, I(52P1/2
)
quenching by NCl3 would have to have a rate constant of
roughly 3 × 10-11 cm3s-1. This value is similar to that reported
by Ray and Coombe7 for I(52P1/2) quenching by ClN3. Current
experiments in our laboratory are directed toward exploring the
rates of energy transfer processes in the I(52P1/2)/NCl(a1∆)/NCl3
system.
While the present data offer a much extended spectrum of
the banded emission produced by photodissociation of NCl3 at
249 nm (Figure 2), few conclusions can be drawn beyond those
reported by Gilbert and co-workers.6 The exponential decay
times for the new bands observed between 750 and 1400 nm
appear to be short, in accord with those expected for excited
states of NCl2. The bands observed by Gilbert et al.6 at shorter
wavelengths had longer lifetimes and were thought to correspond
to excited states of the parent NCl3. Gilbert and Smith18 have
shown that the near UV broadband photolysis of NCl3 trapped
in a low-temperature Ar matrix generates NCl and Cl2. These
authors speculate, though, that the initial photofragments are
NCl2 and Cl and that the NCl arises from either further
photolysis or reaction of the NCl2 in the matrix cage.
Acknowledgment. This work was supported by the U.S.
Air Force Office of Scientific Research under Grant No. F49620-
00-1-0062.
References and Notes
(1) Davy, H. Philos. Trans. R. Soc. 1813, 103, 1.
(2) Clark, T. C.; Clyne, M. A. A. Trans. Faraday Soc. 1969, 65, 2994.
(3) Exton, D. B.; Gilbert, J. V.; Coombe, R. D. J. Phys. Chem. 1991,
95, 2692.
(4) See for example Janda, J. DeVelopments in Inorganic Nitrogen
Chemistry; Colburn, C. B., Ed, Elsevier: New York, 1973, ch. 3.
(5) Rubtsov, N. M. MendeleeV Comm. 1998, 173.
(6) Gilbert, J. V.; Wu, X. L.; Stedman, D. H.; Coombe, R. D. J. Phys.
Chem. 1987, 91, 4265.
(7) Ray, A. J.; Coombe, R. D. J. Phys. Chem. 1993, 97, 3475.
(8) Henshaw, T. L.; Herrera, S. D.; Schlie, L. A. J. Phys. Chem. A
1998, 102, 6239.
(9) Ray, A. J.; Coombe, R. D. J. Phys. Chem. A 1995, 99, 7849.
(10) Henshaw, T. L.; Manke, G. C.; Madden, T. J.; Berman, M. R.;
Hager, G. D. Chem. Phys. Lett. 2000, 325, 537.
(11) Schwenz, R. W.; Gilbert, J. V.; Coombe, R. D. Chem. Phys. Lett.
1993, 207, 526.
The measurements described above suggest that the channel
producing NCl(a1∆) is dominant in NCl3 photodissociation at
193 nm. While production of Cl2(X1Σg+) as a cofragment (eq
1) seems reasonable, molecular chlorine would not make a
significant contribution to the quenching processes observed in
the system, in particular the rapid quenching of excited I(52P1/2
)
(12) Pritt, A. T., Jr.; Patel, D.; Coombe, R. D. J. Mol. Spectrosc. 1981,
observed in the rise of the signals in Figure 5 or the quenching
of NCl(a1∆) reflected in the decay of those signals. It seems
likely that the excited NCl is quenched primarily by collisions
with ground-state iodine atoms and by collisions with other
NCl(a1∆) molecules. For example, the data of Figure 5A exhibit
a decay rate of 360 s-1. From the value of ke and the density of
ground-state I(52P3/2) in the system, NCl(a1∆) quenching by
iodine atoms should contribute about 60 s-1 to this observed
rate. If the NCl(a1∆) density in the system were constant at its
87, 401.
(13) Coombe, R. D.; Patel, D.; Pritt, A. T., Jr.; Wodarczyk, F. J. J. Chem.
Phys. 1981, 75, 2177.
(14) Komissarov, A. V.; Manke, G. C., II; Davis, S. J.; Heaven, M. C.
Proc. SPIE-Int. Soc. Opt. Eng. 2000, 138, 3931.
(15) Pence, W. H.; Baughcum, S. L.; Leone, S. R. J. Phys. Chem. 1981,
85, 3844.
(16) Manke, G. C., II; Setser, D. W. J. Phys. Chem. 1998, 102, 7257.
(17) Burrows, M. D. J. Chem. Phys. 1984, 81, 3546.
(18) Gilbert, J. V.; Smith, L. J. J. Phys. Chem. 1991, 95, 7278.
(19) Henshaw, T. L.; Herrera, S. D.; Haggquist, G. W.; Schlie, L. A. J.
Phys. Chem. A 1997, 101, 4048.
initial value (4.0 × 1013 cm-3 assuming a quantum yield of
,