View Online
actually be formed in the solution photochemistry of triazinyl
azides. Experiments directed towards the isolation of deriva-
tives of these heterocycles therefore may prove useful and are in
progress.
G. B. thanks W. Sander for supporting this work. J. J. W.
acknowledges financial support by the VW Stiftung and the
Deutsche Forschungsgemeinschaft (Heisenberg stipend for
J. J. W., Graduiertenkolleg stipend for F. S.).
Notes and references
1
2
M. S. Platz, Acc. Chem. Res., 1995, 28, 487.
G. B. Schuster and M. S. Platz, Advances in Photochemistry, Wiley,
New York, 1992, vol. 17, p. 69.
3
4
5
6
7
N. P. Gritsan, T. Yuzawa and M. S. Platz, J. Am. Chem. Soc., 1997, 119,
5
059.
R. Born, C. Burda, P. Senn and J. Wirz, J. Am. Chem. Soc., 1997, 119,
061.
W. L. Karney and W. T. Borden, J. Am. Chem. Soc., 1997, 119,
378.
R. A. Evans, M. W. Wong and C. Wentrup, J. Am. Chem. Soc., 1996,
18, 4009.
5
Fig. 1 (a) UV–VIS spectrum obtained after photolysis (3 h, l = 320 nm) of
in Ar, 10 K. (b) UV–VIS spectrum obtained by photolyzing the same
sample subsequently with l = 475–620 nm, 2 h.
1
1
1
C. Wentrup, C. Thétaz, E. Tagliaferri, H. J. Lindner, B. Klitschke,
H.-W. Winter and H. P. Reisenauer, Angew. Chem., Int. Ed. Engl., 1980,
1
9, 566.
M. Kuzaj, H. Lüerssen and C. Wentrup, Angew. Chem., Int. Ed. Engl.,
986, 25, 480.
R. Kayama, S. Hasunuma, S. Sekiguchi and K. Matsui, Bull. Chem. Soc.
Jpn., 1974, 47, 2825.
8
9
1
1
0 R. Kayama, H. Shizuka, S. Sekiguchi and K. Matsui, Bull. Chem. Soc.
Jpn., 1975, 48, 3309.
1
1 W. Sander, J. Org. Chem., 1989, 54, 333.
1
2 Synthesis of 1: The preparation given before (C. V. Hart, J. Am. Chem.
Soc., 1928, 50, 1928) gave only mixtures of products that could not be
separated by crystallization, and the melting point stated to be 85 °C is
most likely incorrect. The following method was successful: To a
solution of cyanuric chloride (3.000 g, 16.27 mmol) in acetone (20 ml)
3
in a small separation funnel, a solution of NaN (1.000 g, 15.38 mmol)
in water (10 ml) was added and the mixture shaken for 5 min. The
organic phase was then separated and allowed to evaporate at room
temperature. The crystals were chromatographed (silica gel, CH
light petroleum = 1+1) to give pure 1 (1.13 g, 5.92 mmol, 36%) as a
colorless powder, mp 59.5–61 °C; d (300 MHz, CDCl ) no signal;
(75.5 MHz, CDCl ) 171.39, 172.60; m/z (FAB) 190.99 (100, M +
H ); n˜ max/cm (Ar, 10 K) 2168.9 (s), 1531.0 (vs), 1512.0 (vs), 1396.8
(s), 1295.8 (s), 1268.0 (s), 1187.6 (m), 1159.9 (m), 947.9 (w), 857.2 (m),
803.8 (w), 730.5 (vw).
2 2
Cl –
H
3
Fig. 2 (a) Difference IR spectrum obtained by subtracting the spectrum of
d
C
+
3
21
1
(Ar, 10 K) from the spectrum obtained after 8 min photolysis (l = 254
nm). The bands pointing downward belong to 1. Bands pointing upward are
assigned to 2 and 3, with the exception of the bands labelled W (water) and
R (ring-opened secondary photoproduct). (b) Calculated IR spectrum
13 The characterization of these final products is in progress; results will be
published at a later time.
(UB3LYP 6-31G*) of triplet nitrene 2. (c) Calculated IR spectrum of 3
(
B3LYP 6-31G*). The correlation with the experimental spectrum is
14 GAUSSIAN 98 software was used in the calculations.
15 Selected data for 2: n˜ max/cm (Ar, 10 K) 1467.6 (s), 1446.4 (vs),
2
1
21
highlighted by grey shading. The intensity of the band at n˜ = 1446.4 cm
is ca. 0.2 absorbance units.
1254.5 (s), 1246.8 (m), 852.9 (w), 848.7 (w), 787.9 (m), 580.0 (w). For
2
1
3
(
1
: n˜ max/cm (Ar, 10 K) 1957.0 (vs), 1951.2 (vs), 1550.0 (m), 1546.2
m), 1110.2 (m), 1050.0 (vs), 814.4. An NNCNN stretching band at n˜ =
957.0 cm–1 for 3 is in the range expected for such a compound. For
the experimental data shows 2 and 3 to be indeed formed as
primary products (Fig. 2).15 This assignment is also consistent
with the UV–VIS data, as the characteristics of the product
initially formed upon photolysis at 320 nm (lmax = 330 and 356
comparison, simple open-chain N-dialkylamino substituted carbodii-
21
mides absorb at n = 2090 cm (ref. 18). The decrease in frequency in
3 reflects the decrease in bond energy induced by strain. A similar trend
has been observed in the case of other cyclic carbodiimides (ref. 6).
16 H. Yamada, H. Shizuka and K. Matsui, J. Org. Chem., 1975, 40,
nm, broad band between 380 and 490 nm) are in line with the
expected behavior of a triplet nitrene.2
1
351.
The UV–VIS-spectrum attributed to 3 (lmax = 328 and 352
nm) is very similar to a spectrum obtained by Yamada et al.16 in
a glassy EPA matrix at 77 K that was assigned to arise from
triplet 4,6-dimethoxy-s-triazinyl nitrene. It thus appears likely
that the product observed by them has to be reassigned to
17 There is a complete lack of absorption beyond l = 420 nm in the spectra
given by Yamada et al. However, a broad band would be expected for
a triplet nitrene. In an EPA matrix, trapping of the didehydrotetrazepine
by EtOH has to be considered.
1
8 W. S. Wadsworth, Jr. and W. D. Emmons, J. Org. Chem., 1964, 29,
816.
17
didehydrotetrazepine 9 or to its trapping products.
2
The observation of didehydrotetrazepine 3 upon photolysis of
azide 1 in an argon matrix could indicate that tetrazepines may
Communication 9/05971G
2114
Chem. Commun., 1999, 2113–2114