A new synthetic path to peri-naphthisoxazoles by oxidative intramolecular cyclization of some β,γ-unsaturated oximes

Article abstract of DOI:10.1039/a809610d

Some β,γ-unsaturated oximes undergo an unprecedented intramolecular cycloaddition to an aryl group to afford peri-naphthisoxazoles when they are irradiated in the presence of PhI(OAc)₂ and I₂ or HgO and I₂ in benzene or by

Full text of DOI:10.1039/a809610d

A new synthetic path to peri-naphthisoxazoles by oxidative intramolecular
cyclization of some b,g-unsaturated oximes
John C. Barnes, William M. Horspool* and George Hynd
Department of Chemistry, The University of Dundee, Dundee, Scotland, UK DD1 4HN.
E-mail: w.m.horspool@dundee.ac.uk
Received (in Liverpool, UK) 7th December 1998, Accepted 27th January 1999
Some b,g-unsaturated oximes undergo an unprecedented
intramolecular cycloaddition to an aryl group to afford peri-
naphthisoxazoles when they are irradiated in the presence of
PhI(OAc)2 and I2 or HgO and I2 in benzene or by thermal
reaction with NaOCl and Et3N in CH2Cl2.
The formation of the peri-naphthisoxazole products from
these irradiations suggests that they might arise from a
conventional intramolecular 1,3-dipolar addition of a nitrile
oxide to an arene. However, there appears to be no literature
precedent for the addition of such a 1,3-dipole to a benzene
moiety, although there are reports of additions to heteroarenes.6
Neither are we aware of a precedent for the formation of a nitrile
oxide via an iminoxyl radical. Thus it was important to verify
whether or not the conventional route to nitrile oxides would
also yield peri-naphthisoxazoles.
The method selected was reaction of the oxime 1b (250 mg,
0.94 mmol) with NaOCl (0.21 ml of an 8% solution, 1.68 mmol)
and Et3N (10 mg, 0.1 mmol) in CH2Cl2 (10 ml).7 The mixture
was stirred at room temperature for 4 days. This indeed yields
the same peri-naphthisoxazole 3b obtained by the photo-
chemical route. However, in this instance the yield is only 44%
and the reaction time is much longer (4 days). Similar thermal
reactivity was observed with the oxime 1d, which gave the peri-
naphthisoxazole 3d in 47%. Interestingly the irradiation of
oxime 1d using either HgO–I2 or PhI(OAc)2–I2 failed to yield a
peri-naphthisoxazole and gave only the corresponding
hydroxydihydroisoxazole 2b in yields varying from 24–45%.
Thus there appears to be a difference between the photo-
chemical and the thermal cyclization.
As part of a general study of the reactivity of b,g-unsaturated
oximes1 we have examined their reactions in the presence of
reagents known to yield free radicals. A survey of the literature
on iminoxyl radicals suggests that such radicals have not been
generated photochemically in solution previously. Therefore,
the methodology for the generation of alkoxyl radicals was used
as a model. Two methods, PhI(OAc)2–I2 in 1,2-dichloroethane2
and HgO–I2 in benzene3 have been used extensively by several
groups and these methods were utilised in this study. Thus we
report the photochemical reactions of the b,g-unsaturated
oximes 1 with both these reagent systems.
It is most likely that both of the photochemical reactions
involve the homolytic fission of OI bonds. There are many
literature examples of hypoiodites formed from alcohols using
both of the methods utilised here.2,3 It is reasonable to assume
that oximes will also yield hypoiodites and, therefore, these
should also undergo OI bond fission with the formation of, in
this instance, iminoxyl radicals such as 4. The cycloaddition
could involve this species with the formation of an intermediate
such as 5.
Final cycloaddition and loss of a hydrogen would yield the
product. Alternatively the iminoxyl radical could lose a
hydrogen to afford the nitrile oxide 6 that would then undergo
cycloaddition. We have not resolved this dilemma since it is
clear that the conventional path to nitrile oxides also affords the
cycloadducts. The one thing that is obvious is that the paths
appear to be subtly different. This is seen by the failure of 1d to
undergo the photochemical cyclization while the thermal
reaction of 1d affords 3d. Currently experiments are in hand to
try to resolve this difference.
The oximes 1 are readily prepared from the corresponding
aldehydes. These are synthesised following the method origi-
nally reported by Zimmerman and co-workers4 and more
recently extended by Armesto et al.1 Typically irradiation (l >
280 nm) of the oxime 1a in benzene in the presence of HgO–I2
(3 equiv. of each) resulted in the formation of two products. The
13C and 1H NMR spectra of the more polar component showed
it to be the known hydroxydihydroisoxazole 2a (13%) pre-
viously reported by Armesto et al.5
The 1H NMR spectrum of the less polar but major component
of the mixture showed this to be considerably different from the
hydroxydihydroisoxazole 2a. The identification of this product
was not easy by conventional means although the integral and
the number of vinylic signals in the 1H NMR spectrum
suggested that addition to one of the aromatic rings had taken
place. X-Ray diffraction† showed this product to be a
peri-naphthisoxazole, 6-phenyl-8,8-dimethyl-10-aza-11-oxatri-
cyclo[7.2.11,9.05,12]dodeca-2,4,6,9-tetraene 3a (55%) (Fig. 1).
The irradiation of the oxime 1a in 1,2-dichloroethane in the
presence of PhI(OAc)2–I2 (1.2 equiv.), best carried out with a
low wattage tungsten lamp, was more efficient and cleaner than
in the presence of HgO–I2, yielding only the peri-naphthisox-
azole 3a in 77% yield and recovered starting material. Other
derivatives of the oxime behaved similarly and 1b and 1c were
converted into the corresponding peri-naphthisoxazoles 3b and
3c in yields of 67 and 68%, respectively.
Fig. 1 X-Ray structure of 3a.
Chem. Commun., 1999, 425–426
425

and E. P. Mayoral, Tetrahedron Lett., 1994, 35, 3785; D. Armesto, A.
Ramos, M. J. Ortiz, M. J. Mancheno and E. P. Mayoral, Recl. Trav. Chim.
Pays-Bas, 1995, 114, 514; D. Armesto, M. G. Gallego, W. M. Horspool
and A. Ramos, J. Chem. Soc., Perkin Trans. 1, 1996, 107; D. Armesto, A.
Ramos, M. J. Ortiz, W. M. Horspool, M. J. Mancheno, O. Caballero and
E. P. Mayoral, J. Chem. Soc., Perkin Trans. 1, 1997, 1535.
2 J. I. Concepcion, C. G. Francisco, R. Hernandez, J. A. Salazar and E.
Suarez, Tetrahedron Lett., 1984, 25, 1553; J. I. Concepcion, C. G.
Francisco, R. Freire, R. Hernandez, J. A. Salazar and E. Suarez, J. Org.
Chem., 1986, 51, 402; M. Angeles, J. A. Salazar and E. Suarez, J. Org.
Chem., 1996, 61, 3999.
3 H. Suginome and S. Yamada, J. Org. Chem., 1984, 49, 3753; H.
Suginome and Y. Nakayama, J. Chem. Soc., Perkin. Trans. 1, 1992,
1843; K. Orito, K. Yorita and H. Suginome, Tetrahedron Lett., 1991, 32,
5999; K. Kobayashi, Y. Kanno, S. Seko and H. Suginome, J. Chem. Soc.,
Perkin. Trans. 1, 1992, 3111; H. Suginome, Y. Nakayama, H. Harada, H.
Hachiro and K. Orito, J. Chem. Soc., Chem. Commun., 1994, 451; H.
Suginome, T. Kondoh, C. Gogonea, V. Singh, H. Goto and E. Osawa,
J. Chem. Soc., Perkin. Trans. 1, 1995, 69; H. Suginome, K. Orito, K.
Yorita, M. Ishikawa, N. Shimoyama and T. Sasaki, J. Org. Chem., 1995,
60, 3052.
We thank the University of Dundee and the EC
(ERBCHRXCT930151) for financial support and the EPSRC
National Crystallographic Service (University of Wales, Car-
diff) for data collection.
Notes and references
† Crystal data for 3a: C18H17NO, M = 263.33, orthorhombic, P212121, a =
8.6306(7), b = 10.610(2), c = 15.119(4) Å, V = 384.4(5) Å3, Z = 4, Dc
= dimensions 0.18 3 0.14 3 0.14 mm was used. Reflections collected /
unique 5768 / 2125 [R(int) = 0.0686] collected on the EPSRC FAST
system at University of Wales, Cardiff. Structure solved by direct methods
(SHELXS-97) (ref. 8) and refined anisotropically using full-matrix least-
squares based on F2 (SHELXL-97/2). Hydrogen atoms of the tricyclic
portion located and refined, other hydrogen atoms placed on calculated
positions, all isotropic. Largest difference peak and hole 0.225 and 20.159
e Å23, R1 = 0.030, wR2 = 0.068 for 1792 independent observed reflections
4 H. E. Zimmerman and A. C. Pratt, J. Am. Chem. Soc., 1970, 92, 6259.
5 D. Armesto, F. Langa, R. P. Ossorio and W. M. Horspool, J. Chem. Soc.,
Chem. Commun., 1990, 123.
6 P. Caramella, G. Cellerino, C. A. Corsico, G. Invernizzi, P. Grunanger,
K. N. Houk and F. M. Albini, J. Org. Chem., 1976, 41, 3349; P.
Caramella, G. Cellerino, P. Grunanger, F. M. Albini and M. R. R.
Cellerino, Tetrahedron, 1978, 34, 3545; P. Caramella, G. Cellerino, K. N.
Houk, F. M. Albini and C. Santiago, J. Org. Chem., 1978, 43, 3006; P.
Caramella, G. Cellerino, P. Grunanger, F. M. Albini and M. R. R.
Cellerino, Tetrahedron, 1978, 34, 3545; V. Librando, U. Chiacchio, A.
Corsaro and G. Gumina, Polycyclic Aromat. Comp., 1996, 11, 313.
7 G. A. Lee, Synthesis, 1982, 508.
[|Fo|
> 4sF, 2q ≤ 50°. The absolute chirality of 3a could not be
unambiguously determined. CCDC 182/1156. Crystal data are available in
CIF format from the RSC web site, see: http://www.rsc.org/suppdata/cc/
1999/425
8 G. M. Sheldrick, in preparation for J. Appl. Crystallogr.
1 See for example: D. Armesto, M. J. Ortiz, A. Ramos, E. P. Mayoral and
W. M. Horspool, J. Org. Chem., 1994, 59, 8115; D. Armesto, A. Ramos
Communication 8/09610D
426
Chem. Commun., 1999, 425–426