Aryl nitrene rearrangements: Spectroscopic observation of a benzazirine and its ring expansion to a ketenimine by heavy-atom tunneling

Article abstract of DOI:10.1021/ja404172s

In the photodecompositions of 4-methoxyphenyl azide (1) and 4-methylthiophenyl azide (5) in argon matrixes at cryogenic temperatures, benzazirine intermediates were identified on the basis of IR spectra. As expected, the benzazirines photochemically rearr

Full text of DOI:10.1021/ja404172s

Communication
pubs.acs.org/JACS
Aryl Nitrene Rearrangements: Spectroscopic Observation of a
Benzazirine and Its Ring Expansion to a Ketenimine by Heavy-Atom
Tunneling
Hiroshi Inui,*,†,‡ Kazuhiro Sawada,† Shigero Oishi,† Kiminori Ushida,† and Robert J. McMahon*,‡
†Department of Chemistry, School of Science, Kitasato University, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa 252-0373,
Japan
‡Department of Chemistry, University of Wisconsin-Madison, 1101 University Avenue, Madison, Wisconsin 53706-1322, United
States
S
* Supporting Information
photolysis in solution at ambient temperature, Tsao and Platz3
ABSTRACT: In the photodecompositions of 4-methox-
yphenyl azide (1) and 4-methylthiophenyl azide (5) in
argon matrixes at cryogenic temperatures, benzazirine
intermediates were identified on the basis of IR spectra. As
expected, the benzazirines photochemically rearranged to
the corresponding ketenimines and triplet nitrenes.
Interestingly, with the methylthio substituent, the
rearrangement of benzazirine 8 to ketenimine 7 occurred
at 1.49 × 10−5 s−1 even in the dark at 10 K, despite a
computed activation barrier of 3.4 kcal mol−1. Because this
rate is 1057 times higher than that calculated for passing
over the barrier and because it shows no temperature
dependence, the rearrangement mechanism is interpreted
in terms of heavy-atom tunneling.
detected tert-butyl-substituted benzazirine, which was identified
by an absorption at 285 nm, as a transient precursor of the
ketenimine, although the singlet nitrene was not observed in this
case. Reports of the observation of benzazirines in matrixes at
cryogenic temperatures have also been limited.6−9 In many
studies on the photodecomposition of phenyl azide and its
derivatives, only triplet nitrenes and ketenimines were
observed.4,5 However, Morawietz and Sander6 found that
when polyfluorinated phenyl azides such as 2,6-difluorophenyl
azide were irradiated, the corresponding benzazirines (2F-type)
were stabilized and observed. Moreover, Carra et al.7 monitored
the reversible photoisomerization between this benzazirine and
the corresponding ketenimine.
Herein we report an investigation of the photodecompositions
of 4-methoxyphenyl azide (1) and 4-methylthiophenyl azide (5)
via low-temperature matrix-isolation spectroscopy together with
density functional theory (DFT) calculations.10 The direct
observation of benzazirines (2H-type) was achieved on the basis
of their clear-cut IR spectra, and new insights into the reactivity of
benzazirines were obtained (Scheme 2).
he photochemistry of aromatic azides has attracted wide
Tinterest for practical applications such as photoresists and
biochemical affinity labeling.1 To achieve the high efficiency
needed for practical reactions, a detailed understanding of the
reaction mechanism is imperative. Therefore, the mechanism has
been extensively investigated by several groups using laser flash
photolysis2,3 and matrix-isolation techniques.4−9 It is well-known
that the irradiation of aromatic azides, including phenyl azide, in
solution results in photodenitrogenation to give singlet nitrenes,
which then undergo intersystem crossing to triplet nitrenes and
ring expansion to ketenimines (Scheme 1).
Scheme 2
Although the intermediates in the ring expansion are thought
to be benzazirines, which can be trapped by ethanethiol to yield
2-ethylthioanilines, the direct observation of these species has
been extremely rare.2 For example, by means of laser flash
Scheme 1
Phenyl azides 1 and 5 were synthesized according to literature
procedures11 and purified using silica gel column chromatog-
raphy with dichloromethane. The azides were stored at −20 °C
until immediately prior to use. When 1 was matrix-isolated in Ar
at 10 K and irradiated with 365 nm light from a 250 W deep-UV
lamp through a bandpass filter, a species having intense bands at
Received: April 26, 2013
Published: June 24, 2013
© 2013 American Chemical Society
10246
dx.doi.org/10.1021/ja404172s | J. Am. Chem. Soc. 2013, 135, 10246−10249

Journal of the American Chemical Society
Communication
1575, 1301, and 1272 cm−1 accumulated as the main product,
accompanied by the disappearance of the bands for 1 [Figure S3
in the Supporting Information (SI)]. The IR peaks of the product
were readily assigned to triplet nitrene 2 by comparison with an
IR spectrum calculated using DFT at the B3LYP/6-31G(d) level.
Subsequent irradiation at 313 or >415 nm for a few minutes
resulted in a decrease in the intensities of the bands due to 2 and
the appearance of a characteristically intense peak at 1894 cm−1,
indicating the formation of a product having a cumulenic double
bond, which was identified as ketenimine 3 (Figure S4).
Geometry optimization of 3 using DFT showed that two
conformers, 3-a and 3-b, exist as true energy-minimum structures
and that 3-a is more stable than 3-b by 2.2 kcal mol−1. These
conformers arise from the orientation of the methyl moiety of the
methoxy substituent (Figure 1). The vibrational frequencies of 3-
Figure 2. (a) Difference IR spectrum recorded after irradiation of
ketenimine 3-a (down bands) matrix-isolated in Ar at 10 K with >415
nm light. The bands labeled N were assigned to nitrene 2. (b−e)
B3LYP/6-31G(d)-calculated IR spectra (scaled by 0.961412) of (b, c)
benzazirines 4-a and 4-b and (d, e) ketenimines 3-a and 3-b,
respectively.
to the following three features: (i) the peak observed at 1519
cm−1 was better reproduced by the peak at 1500 cm−1 calculated
for 4-b than that at 1474 cm−1 for 4-a; (ii) the peak pattern
observed at 900−1200 cm−1 completely agreed with that for 4-b
but not that for 4-a; and (iii) the separation (Δ) of 73 cm−1
between the two peaks observed at 805 and 732 cm−1 in the
region including the out-of-plane deformation of the ring C−H
was well-reproduced (Δ = 77 cm−1) for the peaks at 797 and 720
cm−1 calculated for 4-b, compared with Δ = 109 cm−1 for the
peaks at 831 and 722 cm−1 for 4-a (Figure 2a, up bands). The
rearrangement of 3-a to 4-b is at a structural disadvantage
because of the orientation of the methoxy substituent. However,
the calculated activation barrier for the reaction of 4-a to give 4-b
is extremely low (0.1 kcal mol−1) and should be overcome easily
even at 10 K.13 Thus, it is concluded that irradiation of 3-a at
>415 nm produces 4-a, which then undergoes thermal
isomerization to give 4-b via rotation of the methoxy group.
Few reports of the direct observation of benzazirines, especially
2H-type ones, have been presented. Although Pritchina et al.9
observed an amino-substituted benzazirine in the photoreaction
of 4-aminophenyl azide, no IR peaks were clearly observed
except for a CN stretching band at 1710 cm−1.
Figure 1. Optimized geometries and relative energies for (left)
ketenimine and (right) benzazirine systems (blue, MeO; red, MeS)
obtained from B3LYP/6-31G(d) calculations.10 Zero-point energies are
included. TS = transition state; DA1−2−3−4 = dihedral angle defined by
atoms 1, 2, 3, and 4.
a and 3-b were calculated, scaled by a factor of 0.9614,12 and
compared with the experimental values. Although the two
conformers showed similar spectra, the spectrum calculated for
the more stable conformer 3-a reproduced the observed
spectrum better than that for 3-b in terms of the entire peak
pattern, particularly in the region from 1100 to 1250 cm−1, which
is associated with C−O stretching vibrations (Figure 2a, down
bands).
Ketenimine 3-a could be changed back to nitrene 2 via
irradiation at 365 nm. However, irradiation of a matrix containing
almost exclusively 3-a at >415 nm slowly yielded a different
product exhibiting a weak but characteristic peak at 1717 cm−1,
which is in the region of CN stretching vibrations reported in
the literature (1680−1730 cm−1) (Figures 2 and S16).6−9
Calculations on benzazirine 4 as a possible candidate with a C
N double bond showed that 4 could also exist as two conformers,
4-a and 4-b, with the latter being more stable by 3.0 kcal mol−1.
Similar to 3, the dihedral angle formed by atoms 1−4, which
specifies the orientation of the methoxy group with respect to the
six-membered ring, is very small for 4-b (DA1−2−3−4 = 1.3°),
while that for 4-a is 116.9° (Figure 1). In addition to 4-a and 4-b,
a conformer with DA1−2−3−4 = 173.4° was also obtained;
however, this conformer can isomerize to 4-a without an
activation barrier (Figure S21). The vibrational frequencies
calculated for 4-b, unlike those for 4-a, were in excellent
agreement with the experimental values, particularly with respect
Irradiation of benzazirine 4-b at 365 nm gave mainly the triplet
nitrene 2, while irradiation at 313 nm resulted in a difference
spectrum that showed a decrease in 4-b and increases in 3-a and 2
(Figure S6). The direct formation of 3-a from 4-b is thought to
be difficult because the orientation of the methoxy group makes it
structurally unfavorable, as in the case of the reaction above
discussed. Furthermore, conformer 3-b, which has a favorable
geometry to be the reaction product from 4-b, could not be
detected, in spite of the modest activation barrier (1.4 kcal
mol−1) for the isomerization of 3-b to 3-a (Figure 1). It is highly
probable that the ring-expansion of 4-b to 3-a proceeds by a
consecutive mechanism via 2 or a mechanism involving rotation
of the methoxy substituent utilizing the energy of the 313 nm
light. At present, however, it was not determined whether one or
both mechanisms participated in the reaction. Scheme 2
10247
dx.doi.org/10.1021/ja404172s | J. Am. Chem. Soc. 2013, 135, 10246−10249

Journal of the American Chemical Society
Communication
summarizes the observed interconversions among triplet nitrene
2, ketenimine 3, and benzazirine 4.
As in the photoreaction of 4-b, irradiation of 8 at 313 nm also
gave an IR spectrum that showed an increase in the bands for 7.
Interestingly, however, the rearrangement from 8 to 7 proceeded
even in the dark at 10 K (Scheme 3 and Figure 3). Such a reaction
Next, the photolysis of 4-methylthiophenyl azide (5) was
carried out in an Ar matrix at 10 K. Similar to the photolysis of 1,
three important reactive species, triplet nitrene 6 (1551, 1434,
1313, 1080, 976, and 805 cm−1), ketenimine 7 (νCCN = 1889
cm−1), and benzazirine 8 (νCN = 1716 cm−1), were observed,
depending on the irradiation wavelength (Figures S9−S15).
That is, irradiation of nitrene 6 at 313 or >415 nm led to its ring
expansion to ketenimine 7, and then subsequent irradiation of 7
at >415 nm slowly gave benzazirine 8. In addition, 7 and 8 were
converted back to 6 upon irradiation at 365 nm. Ketenimine 7
also has conformational isomers because of the orientation of the
methylthio substituent (Figure S22). In contrast to the
conformers of methoxy-substituted ketenimine 3, however,
conformers 7-a and 7-b have nearly the same energy (Figure
1) and similar IR spectra according to the DFT calculations
(Figure 3a, up bands). Therefore, it cannot be determined
Scheme 3. Ring Expansion of 8 to 7 Observed in the Dark at
10 K
was not observed between the methoxy-substituted compounds
3 and 4. The B3LYP/6-31G(d)-calculated activation barrier for
the reaction from 8 to 7 is 3.4 kcal mol−1,14 and the rate constant
to surmount this barrier at 10 K is estimated to be 2.00 × 10−63
s−1 using the Eyring equation, indicating that this reaction will
not occur as a thermally activated process at 10 K. However, the
rate constants measured at 10 K for the decrease of 8 and the
increase of 7 were 1.49 × 10−5 and 1.32 × 10−5 s−1, respectively
(Figure 3 bottom).15 In addition, ∼50% of 8 was converted to 7
during 10 h in the dark at 10 K. After 8 was regenerated from 7
with >415 nm light, the rate constants for the decrease of 8 were
also determined at other matrix temperatures and found to be
1.46 × 10−5 s−1 at 15 K, 1.55 × 10−5 s−1 at 20 K, and 1.84 × 10−5
s−1 at 25 K (see section S-8 in the SI). Thus, this rearrangement
barely showed any temperature dependence of the rate constant,
although a small rate acceleration upon warming was observed
because of matrix softening. These results suggest that the ring
expansion reaction of 8 to 7 at cryogenic temperatures occurs via
heavy-atom tunneling, as opposed to a thermally activated (‘over
the barrier’) process.
There is considerable current interest in reactions that proceed
via tunneling mechanisms,15−18 although there are few well-
documented examples of organic reactions that proceed via
heavy-atom tunneling.15−17 The ring expansions of 1H-bicyclo-
[3.1.0]hexa-3,5-dien-2-one to triplet 4-oxocyclohexa-2,5-dien-
ylidene (Scheme 4 top)15a and methylcyclobutylfluoro-carbene
Scheme 4. Previously Reported Reactions Involving Carbon
Tunneling15a,17a
Figure 3. Top: (a) Difference IR spectrum recorded after an Ar matrix
containing benzazirine 8 (down bands) was allowed to stand in the dark
at 10 K for 10 h. (b−e) B3LYP/6-31G(d)-calculated IR spectra for (b, c)
ketenimines 7-b and 7-a and (d, e) benzazirines 8-b and 8-a,
to 1-fluoro-2-methylcyclopentene17 (Scheme 4 bottom) involve
carbon atom tunneling. In the latter reaction, for instance, Zuev
et al.17a measured a rate constant of 4 × 10−5 s−1 in Ar at 8 K,
which is 10152 times greater than the rate would be if the system
had to surmount the computed barrier of ca. 6.5 kcal mol−1.
Furthermore, they reported that the longest distance traveled by
any of the carbon atoms during the tunneling is 0.44 Å. Intrinsic
reaction coordinate (IRC) calculations at the B3LYP/6-31G(d)
level for the transition state connecting the reactant 8-b and the
product 7-b were used to compare the geometry of 8-b with that
of the product having the same energy as 8 along the reaction
pathway (Figure 4), and the maximum distance moved by any of
the atoms was found to be only 0.33 Å for the carbon at the 2-
position of the azirine ring. The fact that this value is smaller than
respectively. Bottom: Plot showing the disappearance of the 1041
−1
cm−1 peak of 8 ( ) and the appearance of the 1889 cm peak of 7 ( )
●
○
in the dark at 10 K.
whether 7 exists only as the marginally more stable conformer 7-
b. Also, the energy difference between the benzazirine con-
formers 8-a and 8-b is only 0.3 kcal mol−1 (Figures 1 and S22).
Although the observed IR spectrum seems to be similar to that
calculated for the marginally more stable conformer 8-b (Figure
3a, down bands), we again cannot conclude whether 8-b is the
sole product.
10248
dx.doi.org/10.1021/ja404172s | J. Am. Chem. Soc. 2013, 135, 10246−10249

Journal of the American Chemical Society
Communication
Education, Culture, Sports, Science and Technology of Japan, a
Sasakawa Scientific Research Grant (15-107) from The Japan
Science Society, a Kitasato University Research Grant for Young
Researchers, and the U.S. National Science Foundation.
REFERENCES
■
(1) (a) Morishita, H.; Ito, M.; Hayashi, N.; Nonogaki, S. J. Photopolym.
Sci. Technol. 1994, 7, 59. (b) Platz, M. S. Photochem. Photobiol. 1997, 65,
193 and references cited therein.
(2) (a) Borden, W. T.; Gritsan, N. P.; Hadad, C. M.; Karney, W. L.;
Kemnitz, C. R.; Platz, M. S. Acc. Chem. Res. 2000, 33, 765. (b) Platz, M.
S.; Gritsan, N. P. Chem. Rev. 2006, 106, 3844.
(3) Tsao, M.-L.; Platz, M. S. J. Am. Chem. Soc. 2003, 125, 12014.
(4) Chapman, O. L.; LeRoux, J.-P. J. Am. Chem. Soc. 1978, 100, 282.
(5) Hayes, J. C.; Sheridan, R. S. J. Am. Chem. Soc. 1990, 112, 5879.
(6) Morawietz, J.; Sander, W. J. Org. Chem. 1996, 61, 4351.
(7) Carra, C.; Nussbaum, R.; Bally, T. ChemPhysChem 2006, 7, 1268.
(8) Grote, D.; Sander, W. J. Org. Chem. 2009, 74, 7370.
(9) Pritchina, E. A.; Gritsan, N. P.; Bally, T. Phys. Chem. Chem. Phys.
2006, 8, 719.
(10) The geometry optimizations, energy calculations, and calculations
of vibrational frequencies and intensities were performed using DFT at
the restricted or unrestricted B3LYP level with the 6-31G(d) basis set.
The energies include zero-point energy scaled by 0.9806. The DFT
calculations were carried out using Gaussian 03. See: Frisch, M. J.;
Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman,
J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi,
M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.;
Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.;
Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision A.1;
Gaussian, Inc.: Pittsburgh, PA, 2003.
Figure 4. IRC calculations at the B3LYP/6-31G(d) level for the
reactions of benzazirine 4-b to give ketenimine 3-b (red) and
benzazirine 8-b to give ketenimine 7-b (blue). Comparisons of the
geometries of the reactants (green circle) and the products having the
same energies as the reactants along the reaction pathways (purple
circles) are shown in section S-9 in the SI.
the previously reported results17a also supports the conclusion
that the ring expansion of 8 to 7 at cryogenic temperatures
proceeds via carbon tunneling. As mentioned above, in the case
of the reaction of methoxy-substituted 4 to give 3, no evidence
for tunneling was obtained. This difference might be interpreted
in terms of the calculated activation barrier and tunneling
distance, which were found to be 6.9 kcal mol−1 and 0.43 Å,
respectively.
In this communication, we have described the direct
observation of 2H-type benzazirines and details of their
reactivity. In the case of methoxy substitution, the most stable
benzazirine conformer, 4-b, was observed on the basis of a clear-
cut IR spectrum, and the photoconversions to ketenimine 3-a
and triplet nitrene 2 were monitored. By contrast, methylthio-
substituted benzazirine 8 afforded ketenimine 7 even in the dark
at 10 K. The measured rate constant was 1057 times greater than
the rate constant would have been if the system had to pass over
the computed barrier of 3.4 kcal mol−1 and showed little
temperature dependence. On the basis of these results, we
conclud that the ring expansion of 8 to give 7 proceeds via heavy-
atom (carbon) tunneling.
(11) Hu, M.; Li, J.; Yao, S. Q. Org. Lett. 2008, 10, 5529.
(12) Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502.
(13) Matrix Isolation Spectroscopy; Barnes, A. J., Orville-Thomas, W. J.,
́
Muller, A., Gaufres, R., Eds.; D. Reidel: Dordrecht, The Netherlands,
̈
1981; Chapter 23, pp 531−549.
(14) This activation barrier was calculated to be 3.4 kcal mol−1 at the
MP2/6-31+G(d,p) level and 4.8 kcal mol−1 at the CCSD(T)/6-
311+G(d,p) level using the MP2/6-31+G(d,p)-optimized geometries
(Figure S18).
(15) (a) Sander, W.; Bucher, G.; Reichel, F.; Cremer, D. J. Am. Chem.
Soc. 1991, 113, 5311. (b) McMahon, R. J.; Chapman, O. L. J. Am. Chem.
Soc. 1987, 109, 683. According to the above papers, in glassy matrixes
with multiple sites, the concentration of a reactant [A] normally
depends on t1/2. In this study, x = 0.50−0.57 was used for [A] = [A]0
exp[−(kt)x].
ASSOCIATED CONTENT
* Supporting Information
Methods and additional results. This material is available free of
charge via the Internet at http://pubs.acs.org.
■
S
(16) McMahon, R. J. Science 2003, 299, 833−834.
(17) (a) Zuev, P. S.; Sheridan, R. S.; Albu, T. V.; Truhlar, D. G.; Hrovat,
D. A.; Borden, W. T. Science 2003, 299, 867. (b) Moss, R. A.; Sauers, R.
R.; Sheridan, R. S.; Tian, J.; Zuev, P. S. J. Am. Chem. Soc. 2004, 126,
10196.
AUTHOR INFORMATION
Corresponding Author
h-inui@kitasato-u.ac.jp; mcmahon@chem.wisc.edu
■
(18) (a) Ley, D.; Gerbig, D.; Schreiner, P. R. Chem. Sci. 2013, 4, 677.
(b) Ley, D.; Gerbig, D.; Wagner, J. P.; Reisenauer, H. P.; Schreiner, P. R.
J. Am. Chem. Soc. 2011, 133, 13614. (c) Schreiner, P. R.; Reisenauer, H.
Notes
The authors declare no competing financial interest.
P.; Pickard, F. C., IV; Simmonett, A. C.; Allen, W. D.; Mat
́
yus, E.;
ACKNOWLEDGMENTS
We gratefully acknowledge financial support through a Grant-in-
Aid for Scientific Research (15750040) from the Ministry of
■
Csaszar, A. G. Nature 2008, 453, 906−909.
́
́
10249
dx.doi.org/10.1021/ja404172s | J. Am. Chem. Soc. 2013, 135, 10246−10249