Experimental and theoretical studies on the synthesis, spectroscopic data, and reactions of formyl azide

Article abstract of DOI:10.1002/anie.201200029

Small is beautiful: spectroscopic proof or any other indication for the existence of formyl azide (HC(O)N₃) has until now been lacking. Although it liberates dinitrogen much more rapidly than homologous acyl azides, it has been prepared for the first time by four different methods (see scheme). Copyright

Full text of DOI:10.1002/anie.201200029

.
Angewandte
Communications
DOI: 10.1002/anie.201200029
Acyl Azides
Experimental and Theoretical Studies on the Synthesis, Spectroscopic
Data, and Reactions of Formyl Azide**
Klaus Banert,* Christian Berndt, Manfred Hagedorn, Hailiang Liu, Tony Anacker,
Joachim Friedrich,* and Guntram Rauhut*
Dedicated to Professor Adalbert Maercker on the occasion of his 80th birthday
Acyl azides 3b and their Curtius rearrangement to produce
isocyanates 4b have been well-known for more than hundred
years (Scheme 1).[1] These azides are also of great importance
Furthermore, formyl chloride cannot be utilized as a conven-
ient precursor of 3a because this acyl chloride is a short-lived
species, which excludes its simple synthesis and ease of
handling.[4]
Recently, we discovered that nearly all types of aldehydes
1 establish an equilibrium with a-azido alcohols 2 when
treated with hydrazoic acid. The products 2b can be oxidized
under very mild conditions to afford acyl azides 3b.[5] Herein,
we show for the first time that the title compound 3a can be
prepared by oxidation of azidomethanol (2a) and by several
other methods. The spectroscopic data and some reactions of
3a are now investigated, with strong support from high-
quality quantum chemical calculations.
Scheme 1. Curtius rearrangement of acyl azides 3, which can be
prepared from aldehydes 1 via a-azido alcohols 2.
Our attempts to generate 3a from formic acid (5) by using
the Weinstock route[6] were unsuccessful (Scheme 2). This
classical method possibly failed because of insufficient
in modern organic synthesis.[2] The parent compound 3a was
investigated in about a dozen theoretically studies dealing
with its structure, spectroscopic data, and various reactions.[3]
To the best of our knowledge, however, spectroscopic proof or
any other indication for the existence of formyl azide (3a) is
still missing. There may be two reasons why 3a has not been
generated experimentally. Quantum chemical calculations
predicted activation energies for the process 3a!4a that
were significantly lower than those of the Curtius rearrange-
ment reactions of acetyl or benzoyl azide (differences of 3.4–
4.9 and 6.5–10.5 kcalmolꢀ1, respectively).[3e–j] Thus, 3a should
be highly unstable at ambient or even lower temperature.
[*] Prof. Dr. K. Banert, C. Berndt, Dr. M. Hagedorn, H. Liu
Technische Universitꢀt Chemnitz, Organische Chemie
Strasse der Nationen 62, 09111 Chemnitz (Germany)
E-mail: klaus.banert@chemie.tu-chemnitz.de
Homepage: http://www.tu-chemnitz.de/chemie/org/indexe.html
T. Anacker, Prof. Dr. J. Friedrich
Scheme 2. Syntheses of formyl azide (3a).
Technische Universitꢀt Chemnitz, Theoretische Chemie
Strasse der Nationen 62, 09111 Chemnitz (Germany)
Prof. Dr. G. Rauhut
Institut fꢁr Theoretische Chemie, Universitꢀt Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart (Germany)
stability or inferior reactivity of the intermediate anhydride[7]
or because of the presence of an aqueous solution, which
excluded low temperatures. However, nitrosation of formyl-
hydrazine (6) at ꢀ108C led to formyl azide (3a) in moderate
yield (36%). Oxidation of azidomethanol (2a),[5] prepared
from 1a and hydrazoic acid, was successfully performed at
ꢀ358C in the presence of pyridinium chlorochromate (PCC)
or pyridinium dichromate (PDC) to furnish 3a in yields of 36
and 17%, respectively.[8] Moreover, cleavage of the diol 2c,[5]
which is accessible from glyoxal (1c) and hydrazoic acid, by
treatment with periodic acid at ꢀ358C gave 3a in 29%
yield.[8] Formylation of azide salts can also lead to 3a if carried
[**] We gratefully acknowledge financial support from the Deutsche
Forschungsgemeinschaft (BA 903/12-1). We thank Dr. O. Jaurich
(Mettler Toledo) for his help in the measurement of online low-
temperature IR spectra, and Dr. A. Ihle for assistance with preparing
the manuscript. CPU time at the Chemnitzer Hochleistungs-Linux-
Cluster CHiC is gratefully acknowledged. Reactions of Unsaturated
Azides, Part 29; for Part 28, see: K. Banert, F. Kçhler, A. Melzer, I.
Scharf, G. Rheinwald, T. Rꢁffer, H. Lang, R. Herges, K. Heß, N.
Ghavtadze, E.-U. Wꢁrthwein, Chem. Eur. J. 2011, 17, 10071–10080.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201200029.
4718
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4718 –4721

Angewandte
Chemie
Table 2: Experimental and computed fundamental frequencies (cmꢀ1) of
formyl azide (3a) at the CCSD(T)-F12a and B3LYP level of theory.[a]
out at low temperature (ꢀ30 to ꢀ158C) with a highly soluble
[9]
azide source such as QN3 (hexadecyltributylphosphonium
azide). Reagent 7 was found to be superior to 8[10] in
producing 3a in good yield. Solutions of pure 3a could be
conveniently obtained in 50% yield when such reaction
mixtures were recondensed at ꢀ50 to ꢀ158C and 10ꢀ6 bar.[11]
Basis mode Symmetry Expt. CCSD(T)-F12a B3LYP
spectrum
cc-pVTZ-F12
surface
6-311+ +G**
in CCl4 at
RT
scaled harm.
(0.989)
15N-Labeled formyl azide ([15N3]-3a) was also synthesized by
n12
n11
n10
n9
2n4
n8
n7
n6
n5
n4
A’
A’
A’
A’
A’
A’
A’
A’
A’
A’’
A’
A’’
A’
2932 (w)
2156 (s, N3) 2172
1699 (s)
–
1169 (s)
1146 (s)
–
942 (m)
827 (w)
–
–
–
–
2938
3023
2252
1744
1386
–
1241
1008
945
826
583
497
256
using Q15N3
and 7.
[9,12]
1717
1361
1175
1151
993
947
829
584
491
We characterized 3a by its UV, IR (Figure 1, Table 2),
1H NMR, 13C NMR, and 15N NMR spectra (Table 1). The 15
N
NMR spectrum of [15N3]-3a indicated that Ng resonated at
n3
n2
n1
252
171
173
[a] The harmonic B3LYP frequencies were taken from Ref. [3d].
1169 cmꢀ1 is found. Since all the other fundamental frequen-
cies are too far away from these two signals, there must be
a strong coupling between two or several modes. Therefore,
we calculated the potential energy surface at the CCSD(T)-
F12a/cc-pVTZ-F12 level by making use of a multimode
expansion up to the third order.[17] Vibrational configuration
interaction (VCI) calculations were subsequently performed,
which yielded a multitude of vibrational overtones and
combination bands. As a result, we found a strong coupling
of n8 with the first overtone of n4. This Fermi resonance
results in the intensities of both bands being quite similar,
which is in nice agreement with the experimental spectrum. A
comparison of the calculated vibrational frequencies with the
experimental ones led to mean absolute deviations of 55 cmꢀ1
for the scaled B3LYP harmonic frequencies and 8 cmꢀ1 for
the anharmonic CCSD(T)-F12a results. The remaining differ-
ences between the experimental and the theoretical data most
likely arise from truncations within the calculations, as well as
from the fact that the experimental spectrum was determined
in solution rather than the gas phase. Thus, we conclude that
our new calculations are in excellent agreement with the
experimental spectrum and explain all the qualitative aspects
of the vibrational spectrum of 3a (Figure 1).
Even at 08C, solutions of 3a liberated dinitrogen to form
isocyanic acid (4a)[18] by Curtius rearrangement (Scheme 3).
We investigated this process kinetically with the help of
1H NMR spectroscopy by analyzing solutions of 3a in CDCl3
between 0 and 258C. This study led to Ea = (20.3 ꢁ 1.1) kcal
molꢀ1, lnA = 27.0 ꢁ 2.0, DH° = (19.7 ꢁ 1.1) kcalmolꢀ1, DS° =
(ꢀ6.86 ꢁ 2.96) calmolꢀ1 Kꢀ1, and DG°285.5 = (21.7 ꢁ 1.4) kcal
molꢀ1.[19] When the decay of 3a was analyzed at 218C in
CD3CN as the solvent instead of the less-polar CDCl3, the
k value increased only slightly by a factor of 1.07, whereas the
same reaction in the nonpolar solvent [D12]cyclohexane
resulted in a significantly lower (factor of 0.55) k value. On
the other hand, the Curtius rearrangement of 3a in CDCl3 at
258C was much faster (factor of 106) compared with the
analogous reaction of acetyl azide[20,21] under the same
Figure 1. Top: IR spectrum of formyl azide (3a) in CCl4; bottom:
calculated spectrum.
Table 1: Selected spectroscopic data of formyl azide (3a) and [15N3]-3a.[a]
3a
UV
lmax =233 nm
(MeCN):
1H NMR: d=8.10 (brs)
13C NMR: d=167.34 (d, 1JCH =223 Hz)
[15N3]- 1H NMR: d=8.10 (dd, br 2JHN =30.7 Hz, 3JHN =7.3 Hz)
3a
13C NMR: d=167.34 (dd, 1JCN =11.5 Hz, 2JCN =7.3 Hz)
2
15N NMR: d=ꢀ241.8 (ddd, JNH =30.7 Hz, 1JNaNß =17.5 Hz,
2JNN =1.6 Hz, Na), ꢀ144.8 (ddd, 1JNßNa =17.5 Hz,
1
3JNH =7.3 Hz, JNßNg =6.0 Hz, Nß), ꢀ132.7 (dt, br,
2
4
1JNgNß =6.0 Hz, JNN ꢂ JNH =1.6 Hz, Ng)
[a] 1H, 13C, and 15N NMR spectra were recorded in CDCl3 at ꢀ208C and
400, 100.6, and 40.5 MHz, respectively. The UV spectrum was measured
at room temperature.
lower field than Nb, which is typical for electron-poor
2
azides.[13] Furthermore, we observed a remarkable J(15N,1H)
coupling constant of 30.7 Hz in both the 1H NMR and
15N NMR spectra.[14]
The experimental IR spectrum of formyl azide (3a; upper
spectrum in Figure 1) indicates two strong bands at 1146 and
1169 cmꢀ1. These bands cannot be explained on the basis of
the harmonic frequencies obtained from the B3LYP/6-311 +
+ G** calculations of Badawi[3d] (Table 2). Our high-level
CCSD(T)-F12a[15–17] investigationsalso do not explain these
nearly degenerate bands. In both of the calculated spectra,
only the fundamental frequency of n8 close to 1146 or
Angew. Chem. Int. Ed. 2012, 51, 4718 –4721
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4719

.
Angewandte
Communications
rearrangement. A Staudinger reaction of 3a with triphenyl-
phosphine or triisopropyl phosphite resulted in the quantita-
tive formation of the phosphazenes 9a and 9b, respectively.
These products could possibly have applications in aza-Wittig
reactions.[27] Treatment of 3a with cyclooctyne[28] initially gave
the cycloadduct 10a. At room temperature, however, this 1H-
1,2,3-triazole slowly rearranged to the 2H-triazole 10b (40%
yield based on the precursor 7).
Since the products 10a and 10b are significantly larger
than the species in the Curtius rearrangement, we optimized
the geometries at the RI-BP86/TZVP[29] level of theory and
performed single-point calculations with B3LYP-d/def2-
QZVPP using COSMO (e = 4.81). The 1,3-dipolar cyclo-
addition of 3a and cyclooctyne can take place by either a syn
(8.2 kcalmolꢀ1) or an anti (6.0 kcalmolꢀ1) pathway. Both
barriers are significantly smaller than the barrier of the
Curtius rearrangement, and the cycloaddition is possible. The
lowest barrier for the conversion of 10a into 10b is relatively
high at 34.7 kcalmolꢀ1, which explains the relatively slow
rearrangement (Figure 3). Further possible transition states
have similar energy.[19] The reaction energies starting from the
syn and anti product of 10a are ꢀ8.5 and ꢀ3.2 kcalmolꢀ1,
respectively.
Scheme 3. Reactions of formyl azide (3a).
conditions. Our experimental results correspond well with
high-level quantum chemical calculations (see below).
The geometries of the syn and anti conformers of 3a, and
the corresponding transition states were optimized at the
DFT level by using the B3LYP functional[22–24] and the def2-
QZVPP basis set including the dispersion correction of
Grimme[25] (Figure 2). The effect of the solvent on the
Figure 2. Relative CCSD(F12)(T) energies (kcalmolꢀ1) including a cor-
rection for solvation effects and zero-point energy of 3a conformers
and their transition states for interconversion and Curtius rearrange-
ments. Values in brackets correspond to B3LYP/def2-QZVPP.
Figure 3. Transition state of lowest energy for the rearrangement of
10a to 10b.
activation energy was estimated by the COSMO[26] model
using e = 4.81. The computation of the electronic energy was
performed with the CCSD(F12)(T)[24] method at the
B3LYP[22] geometry using the cc-pVTZ-F12 basis set. The
calculated DG°285.5 value of 21.7 kcalmolꢀ1 fits very well with
the experimental value. An increase in the e value from 4.81
to 1 resulted—in complete agreement with the experimental
observations—in a lowering of the DG° value. Considering
the calculated enthalpy DH° = 22.8 kcalmolꢀ1 and entropy
DS° = 3.8 calmolꢀ1 Kꢀ1, we find that the numbers deviate
slightly from the experimental ones, possibly because of the
B3LYP geometry. The error in the single-point approximation
was evaluated for 3a to be ꢀ0.35 kcalmolꢀ1, thus indicating
that the perfect agreement of the DG° value is based on
a fortuitous error cancellation. However, this agreement is
good enough to conclude that the Curtius rearrangement is
a one-step mechanism via TS1 and not a two-step mechanism
via the nitrene, since the energy of TS3 is 6.5 kcalmolꢀ1 higher
than the energy of TS1 (Figure 2). This result does not change,
even with the errors mentioned above.
In summary, we have presented four different methods to
prepare formyl azide (3a). Two of these utilized the unusual
starting compounds 2a and 2c, another route proceeded from
the special salt QN3 and formylation reagents 7 or 8, while the
last method made use of the nitrosation of a hydrazide, which
is the oldest procedure to synthesize acyl azides.[1a] Further-
more, we were able to assign all the strong IR signals of 3a
with the help of high-level quantum chemical calculations.
Although 3a undergoes a rapid Curtius rearrangement
through a concerted mechanism, it can also be applied to
other reactions. Currently, we are investigating whether the
irradiation of 3a can lead to formylnitrene. This short-lived
species has already been studied intensively by theoretical
methods.[30]
Received: January 2, 2012
Published online: March 29, 2012
Intermolecular reactions of formyl azide (3a) can only be
successful if they are able to compete with the rapid Curtius
Keywords: acyl azides · formylation · oxidation ·
quantum chemistry · rearrangement
.
4720 www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 4718 –4721

Angewandte
Chemie
[16] a) T. B. Adler, G. Knizia, H.-J. Werner, J. Chem. Phys. 2007, 127,
[1] a) T. Curtius, Ber. Dtsch. Chem. Ges. 1890, 23, 3023 – 3033; b) T.
Curtius, J. Prakt. Chem. 1894, 50, 275 – 294.
221106; b) K. A. Peterson, T. Adler, H.-J. Werner, J. Chem. Phys.
2008, 128, 084102.
[2] a) P. A. S. Smith, Org. React. 1946, 3, 337 – 449; b) W. Lwowski in
The chemistry of the azido group (Ed.: S. Patai), Wiley
Interscience, London, 1971, pp. 503 – 554; c) W. Lwowski in
Azides and Nitrenes: Reactivity and Utility (Ed.: E. F. V.
Scriven), Academic Press, Orlando, 1984, pp. 205 – 246;
d) Organic Azides: Syntheses and Applications (Eds.: S. Brꢀse,
K. Banert, Wiley, Chichester, 2010.
[17] a) G. Rauhut, A. El Azhary, F. Eckert, U. Schumann, H.-J.
Werner, Spectrochim. Acta 1999, 55, 647 – 658; b) G. Rauhut, J.
Chem. Phys. 2004, 121, 9313 – 9322; c) T. Hrenar, H.-J. Werner,
G. Rauhut, J. Chem. Phys. 2007, 126, 134108; d) G. Rauhut, T.
Hrenar, Chem. Phys. 2008, 346, 160 – 166; e) M. Neff, G. Rauhut,
J. Chem. Phys. 2009, 131, 124129; f) T. Hrenar, G. Rauhut, H.-J.
Werner, J. Phys. Chem. A 2006, 110, 2060 – 2064.
[18] Alternative generation of 4a: G. Fischer, J. Geith, T. M.
Klapçtke, B. Krumm, Z. Naturforsch. B 2002, 57, 19 – 24.
[19] For details, see the Supporting Information.
[20] The Curtius rearrangement of acetyl azide has already been
investigated kinetically (see Ref. [21]). However, the listed
k values are incompatible with the shown activation parameters.
[21] V. A. Zlobin, A. K. Tarasov, Russ. J. Phys. Chem. 1985, 59, 140 –
141.
[22] a) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785 – 789;
b) A. D. Becke, J. Chem. Phys. 1993, 98, 5648 – 5652.
[23] A. Schꢀfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100,
5829 – 5835.
[24] a) TURBOMOLE V6.3 2011, a development of University of
Karlsruhe and Forschungszentrum Karlsruhe GmbH, 1989 –
2007, TURBOMOLE GmbH since 2007, available from http://
www.turbomole.com; b) M. Hꢀser, R. Ahlrichs, J. Comput.
Chem. 1989, 10, 104 – 111; c) O. Treutler, R. Ahlrichs, J. Chem.
Phys. 1995, 102, 346 – 354; d) P. Deglmann, F. Furche, R.
Ahlrichs, Chem. Phys. Lett. 2002, 362, 511 – 518; e) D. P. Tew,
W. Klopper, J. Chem. Phys. 2005, 123, 074101; f) W. Klopper,
C. C. M. Samson, J. Chem. Phys. 2002, 116, 6397 – 6410.
[25] a) S. Grimme, J. Comput. Chem. 2004, 25, 1463 – 1473; b) S.
Grimme, J. Comput. Chem. 2006, 27, 1787 – 1799.
[26] A. Klamt, G. Schꢂꢂrmann, J. Chem. Soc. Perkin Trans. 2 1993,
799 – 805.
[27] Review on aza-Wittig reactions: F. Palacios, C. Alonso, D.
Aparicio, G. Rubiales, J. M. de Los Santos in Organic Azides:
Syntheses and Applications (Eds.: S. Brꢀse, K. Banert), Wiley,
Chichester, 2010, pp. 439 – 467.
[3] a) S. A. Glover, A. Rauk, J. Chem. Soc. Perkin Trans. 2 2002,
1740 – 1746; b) R. H. Abu-Eittah, H. Moustafa, A. M. Al-Omar,
Chem. Phys. Lett. 2000, 318, 276 – 288; c) C. J. Nielsen, C. E.
Sjøgren, J. Mol. Struct. THEOCHEM 1987, 150, 361 – 379;
d) H. M. Badawi, J. Mol. Struct. THEOCHEM 2002, 583, 89 – 97;
e) R. Kakkar, S. Zaidi, R. Grover, Int. J. Quantum Chem. 2009,
109, 1058 – 1069; f) R. H. Abu-Eittah, A. A. Mohamed, A. M.
Al-Omar, Int. J. Quantum Chem. 2006, 106, 863 – 875; g) M. V.
Zabalov, R. P. Tiger, Russ. Chem. Bull. Int. Ed. 2005, 54, 2270 –
2280; h) X. Wenyuan, L. Gousheng, P. Yiyuan, H. Sanguo, Wuli
Huaxue Xuebao (Acta Phys. Chim. Sinica) 1998, 14, 669 – 672;
i) A. Rauk, P. F. Alewood, Can. J. Chem. 1977, 55, 1498 – 1510;
j) V. I. Faustov, E. G. Baskir, A. A. Biryukov, Russ. Chem. Bull.
Int. Ed. 2003, 52, 2328 – 2333; k) P. Pꢁrez, L. R. Domingo, M. J.
Aurell, R. Contreras, Tetrahedron 2003, 59, 3117 – 3125; l) C.
Wentrup, H. Bornemann, Eur. J. Org. Chem. 2005, 4521 – 4524;
m) G. O. Jones, K. N. Houk, J. Org. Chem. 2008, 73, 1333 – 1342.
[4] a) P. Dowideit, R. Mertens, C. von Sonntag, J. Am. Chem. Soc.
1996, 118, 11288 – 11292; b) J. Tyrrell, W. Lewis-Bevan, J. Phys.
Chem. 1992, 96, 1691 – 1696; c) R. W. Davis, M. C. L. Gerry, J.
Mol. Spectrosc. 1983, 97, 117 – 138; d) H. G. Libuda, F. Zabel,
E. H. Fink, K. H. Becker, J. Phys. Chem. 1990, 94, 5860 – 5865;
e) H. Takeo, C. Matsumura, J. Chem. Phys. 1976, 64, 4536 – 4540.
[5] K. Banert, C. Berndt, S. Firdous, M. Hagedorn, Y.-H. Joo, T.
Rꢂffer, H. Lang, Angew. Chem. 2010, 122, 10404 – 10407; Angew.
Chem. Int. Ed. 2010, 49, 10206 – 10209.
[6] J. Weinstock, J. Org. Chem. 1961, 26, 3511.
[7] T. Parasaran, D. S. Tarbell, J. Org. Chem. 1964, 29, 3422 – 3423.
[8] The yields of 3a, generated via 2a or 2c, are based on the
(substoichiometric) amounts of HN3 treated with an excess of
gaseous 1a or 1c, respectively.
[9] K. Banert, Synthesis 2007, 3431 – 3446.
[10] E. J. Vlietstra, J. W. Zwikker, R. J. M. Nolte, W. Drenth, Recl.
Trav. Chim. Pays-Bas 1982, 101, 460 – 462.
[28] L. F. Tietze, T. Eicher, Reaktionen und Synthesen im Organisch-
Chemischen Praktikum und Forschungslaboratorium, 2nd ed.,
Thieme, Stuttgart, 1991, p. 40.
[29] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098 – 3100; b) J. P.
Perdew, Phys. Rev. B 1986, 33, 8822 – 8824.
[30] a) E. A. Pritchina, N. P. Gritsan, A. Maltsev, T. Bally, T. Autrey,
Y. Liu, Y. Wang, J. P. Toscano, Phys. Chem. Chem. Phys. 2003, 5,
1010 – 1018; b) S. A. Stoffregen, R. D. McCulla, R. Wilson, S.
Cercone, J. Miller, W. S. Jenks, J. Org. Chem. 2007, 72, 8235 –
8242; c) D. A. Pantazis, A. C. Tsipis, C. A. Tsipis, J. Phys. Chem.
A 2002, 106, 1425 – 1440; d) N. P. Gritsan, E. A. Pritchina,
Mendeleev Commun. 2001, 11, 94 – 95; e) A. M. Mebel, A.
Luna, M. C. Lin, K. Morokuma, J. Chem. Phys. 1996, 105, 6439 –
6454; f) G. Frenking, J. Schmidt, Tetrahedron 1984, 40, 2123 –
2132; g) A. Mavridis, J. F. Harrison, J. Am. Chem. Soc. 1980, 102,
7651 – 7655; h) D. Poppinger, L. Radom, J. Am. Chem. Soc. 1978,
100, 3674 – 3685; i) D. Poppinger, L. Radom, J. A. Pople, J. Am.
Chem. Soc. 1977, 99, 7806 – 7816; j) J. F. Harrison, G. Shalhoub,
J. Am. Chem. Soc. 1975, 97, 4172 – 4176; k) P. F. Alewood, P. M.
Kazmaier, A. Rauk, J. Am. Chem. Soc. 1973, 95, 5466 – 5475;
l) E. A. Pritchina, N. P. Gritsan, T. Bally, Russ. Chem. Bull. Int.
Ed. 2005, 54, 525 – 532.
[11] Caution! We assume that 3a is a highly volatile and explosive
compound although we never observed any incident. For
potential hazards in handling hydrazoic acid and organic
azides, see T. Keicher, S. Lçbbecke in Organic Azides: Syntheses
and Applications (Eds.: S. Brꢀse, K. Banert), Wiley, Chichester,
2010, pp. 3 – 27.
[12] K. Banert, J. Lehmann, H. Quast, G. Meichsner, D. Regnat, B.
Seiferling, J. Chem. Soc. Perkin Trans. 2 2002, 126 – 134.
[13] D. M. Kanjia, J. Mason, I. A. Stenhouse, R. E. Banks, N. D.
Venayak, J. Chem. Soc. Perkin Trans. 2 1981, 975 – 979.
[14] We did not find any similar values for 2J(15N,1H), e.g., in the
following book: S. Berger, S. Braun, H.-O. Kalinowski, NMR-
Spektroskopie von Nichtmetallen, 15N-NMR-Spektroskopie,
Vol. 2, Thieme, Stuttgart, 1992.
[15] MOLPRO, version 2010.1, a package of ab initio programs, H.-J.
Werner, P. J. Knowles, G. Knizia, F. R. Manby, M. Schꢂtz, et al.
Angew. Chem. Int. Ed. 2012, 51, 4718 –4721
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4721