Transient absorption spectra and reaction kinetics of singlet phenylnitrene and its 2,4,6-tribromo derivative in solution [14]

Full text of DOI:10.1021/ja970205g

J. Am. Chem. Soc. 1997, 119, 5061-5062
5061
phenylnitrenes has been accumulated.2,8 Bromination should,
at the same time, accelerate intersystem crossing (isc) due to
the heavy-atom effect. The lifetime of singlet phenylnitrene is
limited by ring expansion, isc competes effectively only at
Transient Absorption Spectra and Reaction Kinetics
of Singlet Phenylnitrene and Its 2,4,6-Tribromo
Derivative in Solution
2
Reto Born, Clemens Burda, Paul Senn, and Jakob Wirz*
temperatures below 200 K. 2,4,6-Tribromophenyl azide (2)
9
is known to react predominantly Via the triplet nitrene that
Institut f u¨ r Physikalische Chemie der
UniVersit a¨ t Basel, Klingelbergstrasse 80
CH-4056 Basel, Switzerland
2
dimerizes to the corresponding azo compound. The transient
spectra obtained by picosecond pump-probe spectroscopy of 2
in dichloromethane clearly required three spectral components
in the factor analysis and a biexponential rate law in the kinetic
fitting procedure (Figure 2). The resulting rate constants k1 )
ReceiVed January 22, 1997
Time-resolved techniques have provided lifetimes for a
representative selection of carbenes in both the singlet and the
1
0
-1
8
-1
6
.0 ( 1.9 × 10
s
and k2 ) 9.4 ( 0.3 × 10 s gave a
good fit to the amplitude coefficients (Figure 2, inset).
1
triplet state, but this is not so for nitrenes. In particular, neither
The two time-resolved processes are attributed to formation
of singlet 2,4,6-tribromophenylnitrene from the excited singlet
state of 2, followed by isc of the singlet nitrene (Figure 2,
spectrum at 50 ps delay) to the triplet nitrene. The characteristic
final spectrum (Figure 2, spectrum at 17 ns delay) is clearly
that of triplet 2,4,6-tribromophenylnitrene which has been
singlet nor triplet phenylnitrene has been detected in solution.
The lifetime of singlet phenylnitrene at room temperature in
solution is limited by ring expansion to 1-azacyclohepta-1,2,4,6-
1
0(1 -1
tetraene and was estimated as 10
s
on the basis of several
assumptions.2 Platz and co-workers have since provided
convincing evidence that denitrogenation and ring expansion
proceed stepwise upon photolysis of phenyl azide (1). The
2
identified previously. The assignment of the first two transient
3
intermediates rests on two observations: (i) Proton transfer in
stage was set to identify singlet phenylnitrene by flash pho-
tolysis. We report transient absorption spectra determined by
pump-probe spectroscopy of phenyl azide (1), 2,4,6-tribro-
1
4
:1 aqueous acetonitrile produced the nitrenium ion 4, λmax )
10,11
20 and 600 nm.
The rate of formation and the yield of 4
-
1
increased upon addition of perchloric acid, (τobsd) ) k0 +
4
mophenyl azide (2), and 2,4,6-tribromophenyl sulfoximine (3).
+
+
+
9
-1 -1
kH [H ] with kH ) 3.5 ( 0.1 × 10 M s , and the decay
rate of the singlet nitrene increased accordingly. (ii) Pump-
probe spectroscopy of sulfoximine 3 gave the same, albeit
weaker, transient absorption changes and the same rate constant
as azide 2 for the second process (isc of singlet nitrene). The
spectra taken immediately after the pump pulse were, however,
different (broad absorption, λmax ) 340 and 400 nm, tailing to
>
500 nm) from those obtained with 2, and the first process
Solutions (ca. 10- M) were circulated in a flow system,
pumped at 248 nm (4 mJ, 0.7 ps half-width, 12 Hz), and probed
by a delayed continuum pulse (310-700 nm) of the same
3
10 -1
had a different rate constant, k1 ) 4.5 ( 0.4 × 10 s . These
observations indicate that the primary transient absorptions
detected by pump-probe spectroscopy of 2 (Figure 2, spectrum
at 2 ps delay) and 3 are due to the excited singlet states of these
compounds.
5
duration. Transient absorption spectra were recorded at ca.
5
0 different time delays ranging from 2 ps to 1.8 ns relative to
the excitation pulse. Delays of up to 17 ns were produced by
passing the probe beam over an extended delay line, but these
spectra were not used to determine kinetics. The transient
absorbance data were subjected to global analysis, i.e., factor
analysis followed by nonlinear least-squares fitting of appropri-
ate kinetic models.5
We now return to parent phenyl azide (1). Factor analysis
of the data matrix obtained by pump-probe spectroscopy
indicated that three components were sufficient to reproduce
the observed spectra within experimental accuracy. Fitting of
a single exponential rate law to the reduced data set gave a
,6
9
-1
first-order rate constant of 4.6 ( 0.6 × 10 s for the overall
1
2
The transient absorbance changes produced by flash pho-
tolysis of phenyl azide (1) in dichloromethane are shown in
Figure 1. Experiments with other solvents were hampered by
the deposition of polymeric tars on the windows of the flow
cell. The final spectrum (1.8 ns delay) is the same as that
spectral changes. In the preceding paper, Gritsan, Yuzawa,
and Platz report kinetic flash photolysis of 1 in pentane at
temperatures below 0 °C and thereby provide a direct measure-
ment of the lifetime of singlet phenylnitrene. At first sight our
results appear to be inconsistent with the data reported by these
authors: we observe time-resolved absorbance decay predomi-
nantly at wavelengths >370 nm (Figure 1), whereas they report
decay at 350 nm and a slight rise at longer wavelengths. This
suggests that, as in the case of 2, the transient spectra observed
with 1 might represent two sequential processes and that only
2
observed by nanosecond flash photolysis of 1 and has been
unambiguously assigned to the product of ring expansion,
1
-azacyclohepta-1,2,4,6-tetraene.7 Prior to the discussion of
these observations the tribromo derivatives 2 and 3 are
considered.
Ortho-substituents retard ring expansion and substantial
evidence for an enhanced lifetime of ortho-disubstituted singlet
(
8) (a) Poe, R.; Grayzar, J.; Young, M. J. T.; Leyva, E.; Schnapp, K.;
Platz, M. S. J. Am. Chem. Soc. 1991, 113, 3209-3211. (b) Poe, R.; Schnapp,
K.; Young, M. J. T.; Grayzar, J.; Platz, M. S. J. Am. Chem. Soc. 1992,
114, 5054-5067. (c) Marcinek, A.; Platz, M. S. J. Phys. Chem. 1993, 97,
12674-12677. (d) Marcinek, A.; Platz, M. S.; Chan, S. Y.; Floresca, R.;
Rajagopalan, K.; Golinski, M.; Watt, D. J. Phys. Chem. 1994, 98, 412-
419. (e) Karney, W. L.; Borden, W. T. J. Am. Chem. Soc. 1997, 119, 3347-
3350. (f) Gritsan, N. P.; Zhai, H. B.; Yuzawa, T.; Karweik, D.; Brooke, J.;
Platz, M. S. J. Phys. Chem. A 1997, 101, 2833-2840.
(
1) Platz, M. S. Acc. Chem. Res. 1995, 28, 487-492.
(
2) Leyva, E.; Platz, M. S.; Persy, G.; Wirz, J. J. Am. Chem. Soc. 1986,
1
08, 3783-3790.
(
3) Marcinek, A.; Leyva, E.; Whitt, D.; Platz, M. S. J. Am. Chem. Soc.
1
993, 115, 8609-8612.
(
4) Azide 2 (0.4 g) was heated to 175 °C for 1.5 h in DMSO. The product
3
7
was purified by chromatography: mp 98-100 °C; H-NMR (CDCl3): δ
(9) A small amount of polymer is formed, presumably via ring expansion
of the singlet nitrene.
(10) The lifetime of 2,4,6-tribromophenylnitrenium ion is 200 ns; 2,6-
dibromo-1,4-benzoquinone was the main product isolated after photolysis
in aqueous acid. Details will be reported elsewhere.
(11) For a review on protonation of singlet aryl nitrenes, see: McClelland,
R. A. Tetrahedron 1996, 52, 6823-6858. See, also: McClelland, R. A.;
Kahley, M. J.; Davidse, P. A.; Hadzialic, G. J. Am. Chem. Soc. 1996, 118,
4794-4803. Michalak, J.; Zhai, H. B.; Platz, M. S. J. Phys. Chem. 1996,
100, 14028-14036.
.68 (s, 2H), 3.22 (s, 6H).
(
5) Details of the experimental set-up and of the data analysis were
described: Hasler, E.; H o¨ rmann, A.; Persy, G.; Platsch, H.; Wirz, J. J. Am.
Chem. Soc. 1993, 115, 5400-5409.
(
6) Bonneau, R.; Wirz, J.; Zuberb u¨ hler, A. D. Pure Appl. Chem. 1997,
in the press.
(
7) Schrock, A. K.; Schuster, G. B. J. Am. Chem. Soc. 1984, 106, 5228-
5
234. Li, Y.-Z.; Kirby, J. P.; George, M. W.; Poliakoff, M.; Schuster, G.
B. J. Am. Chem. Soc. 1988, 110, 8092-8098.
S0002-7863(97)00205-9 CCC: $14.00 © 1997 American Chemical Society

5062 J. Am. Chem. Soc., Vol. 119, No. 21, 1997
Communications to the Editor
Figure 1. Pump-probe absorption spectra of 1 in dichloromethane. The points shown in the inset are the loading coefficients (arbitrary scale) of
the first eigenspectrum determined by factor analysis of the complete set of pump-probe spectra. The full line is the best fit of a biexponential rate
law to the complete set of data (global analysis, ref 5).
Figure 2. Pump-probe absorption spectra of 2 in dichloromethane. Inset: see caption to Figure 1.
Table 1. Rates of Formation (k
Arylnitrenes in Dichloromethane
1
) and Decay (k
2
) of Singlet
measurements at -20 °C. These data confirm the above
interpretation: the two processes are resolved, and the slower
spectral changes are consistent with those reported by Gritsan
et al. Thus, we assign the component exhibiting an absorption
peak at 350 nm to singlet phenylnitrene. Tribromination shifts
the absorption maximum to 395 nm. Similar band shifts of
about 40 nm were found in the spectra of the triplet nitrenes
/(10 s-1)
9
k
/(10 s-1)
9
λ
max/nm
arylnitrene
k
1
2
ca. 1.7a)
0.94 ( 0.03
phenylnitrene
10 ( 3
60 ( 2
350
395
2
,4,6-tribromophenylnitrene
a
The kinetic data determined at temperatures below ambient by
1
2
Gritsan et al. are considered to be more accurate.
2
derived from 1 and 2.
The absorption maxima and the rates of formation and decay
of singlet phenylnitrenes are collected in Table 1. The transient
spectra determined with 1 at room temperature are congested
due to overlap between the formation and decay processes. The
observed rates of formation represent a lower limit for the rate
of nitrogen elimination from singlet excited azide, since the
initial absorbance changes may correspond to the decay of the
excited azide, internal conversion of electronically excited singlet
nitrene, thermalization of a vibrationally excited singlet nitrene,
or a combination of these processes. Ring expansion is the
dominant decay process of phenylnitrene in solution at room
temperature. The lifetime of 2,4,6-tribromophenylnitrene is
mainly limited by isc, but protonation to form the nitrenium
ion is competitive in aqueous solution.
the slower one was resolved by Gritsan et al. Indeed, the peak
at about 350 nm in our spectra decays discernibly more slowly
than the broad absorption above 400 nm. Fitting of a biexpo-
nential function to the data (inset of Figure 1) gave k1 ) 1.0 (
1
0
-1
9 -1
0
.3 × 10
spectral component exhibits a broad absorption in the range of
70-470 nm, the longer-lived one shows a relatively narrow
s
and k2 ) 1.7 ( 1.1 × 10 s . The short-lived
3
peak at 350 nm. While our data are consistent with a
biexponential model function representing two independent
processes, the large standard errors of the rate constants show
that spectral changes are too small and the rate constants of the
two processes too similar to warrant the use of such a model
on purely statistical grounds.
In order to obtain conclusive evidence that the spectra
observed at room temperature represent a congested view of
two or more processes overlapping in time, we repeated the
Acknowledgment. We are grateful to Professors Gritsan and Platz
for delaying publication of their manuscript12 in order to enable joint
publication of these studies. This work was supported by the Swiss
National Science Foundation. Dedicated to Prof. Waldemar Adam on
the occasion of his 60th birthday.
(12) Gritsan, N. P.; Yuzawa, T.; Platz, M. S. J. Am. Chem. Soc. 1997,
1
19, 5059-5060.
JA970205G