Dihydro[c]benzazepin-3-ones via conjugated nitrone-allene precursors

Article abstract of DOI:10.1021/ol0056832

Treatment of o-propargylaryl nitrones with base provided 1,2-dihydro[c]benzazepin-3-ones in good yields. The straightforward transformation is explained on the basis of a multistep rearrangement involving conjugated allene-nitrones as precursors of a 1,7-dipolar electrocyclization process that is followed by further bond reorganizations.

Full text of DOI:10.1021/ol0056832

ORGANIC
LETTERS
2000
Vol. 2, No. 8
1117-1120
Dihydro[c]benzazepin-3-ones via
Conjugated Nitrone−Allene Precursors
Karin Knobloch and Wolfgang Eberbach*
Institute of Organic Chemistry and Biochemistry, UniVersity of Freiburg,
Albertstrasse 21, D-79104 Freiburg, Germany
eberbach@organik.chemie.uni-freiburg.de
Received February 17, 2000
ABSTRACT
Treatment of o-propargylaryl nitrones with base provided 1,2-dihydro[c]benzazepin-3-ones in good yields. The straightforward transformation
is explained on the basis of a multistep rearrangement involving conjugated allene−nitrones as precursors of a 1,7-dipolar electrocyclization
process that is followed by further bond reorganizations.
The participation of an allene unit in pericyclic processes is
amply documented in many [2 + 2] cycloadditions,1 Diels-
Scheme 1
Alder reactions,2 and the corresponding 1,3-dipolar meth-
odology.3 Much less common are examples involving
electrocyclic ring closures of allene systems.4 We have been
engaged for some years in studies directed toward the
application of 1,7-dipolar cyclization reactions5 in order to
develop new methods in heterocyclic synthesis. After the
extensive use of dipoles bearing butadienyl and butenynyl
groups as 4π-moieties, leading to a variety of new five-, six-,
and seven-membered heterocycles,6 we now report on results
with nitrones derivatives of type A (Scheme 1).7
It was the intention of this work to generate the allene
unit of A by base-catalyzed tautomerization of the corre-
sponding propargyl derivative B. The experiments were
performed with the benzannulated nitrone system 6, which
is available by the route sketched in Scheme 2.
Starting with the bromo aldehydes 1, carbonyl protection
and subsequent Grignard reaction gave compounds 2, which
after liberation of the aldehyde function, afforded the nitrones
6d,e,n by treatment of the aldehydes 5 with methyl and
phenyl hydroxylamine, respectively. The nitrones 6f-k,o
were made available from 2 by sequential hydrodesilylation
(f 3), introduction of a terminal substituent (f 4) by the
Sonogashira method8 or n-butyllithium/methyl iodide treat-
ment, deprotection (f 5), and nitrone formation. For the
(1) (a) Hopf, H. In The Chemistry of Allenes; Landor, S. R., Ed.;
Academic Press: New York, 1982; Vol. 2, pp 525-562. (b) Pasto, D. J.
Tetrahedron 1984, 40, 2805-2827.
(2) (a) Carruthers, W. Cycloaddition Reactions in Organic Synthesis;
Pergamon Press: Oxford, 1990. (b) Larock, R. C. ComprehensiVe Organic
Transformations; Wiley-VCH: Weinheim, 1999; p 541. See also ref 1a.
(3) For a recent review, see: Broggini, G.; Zecchi, G. Gazz. Chim. Ital.
1996, 126, 479-488.
(4) Hopf, H. In The Chemistry of Allenes; Landor, S. R., Ed.; Academic
Press: New York, 1982; Vol. 2, pp 563-577.
(5) Reviews on 1,7-dipolar cyclizations: (a) Zecchi, G. Synthesis 1991,
181-188. (b) Groundwater, P. W.; Nyerges, M. AdV. Heterocycl. Chem.
1999, 73, 97-129.
(6) (a) Eberbach, W.; Trostmann, U. Chem. Ber. 1985, 118, 4035-4058.
(b) Eberbach, W.; Roser, J. Tetrahedron 1986, 42, 2221-2234. (c) Maier,
W.; Eberbach, W.; Fritz, H. HelV. Chim. Acta 1991, 74, 1095-1101. (d)
Bussenius, J.; Laber, N.; Mu¨ller, T.; Eberbach, W. Chem. Ber. 1994, 127,
247-259. (e) Lopez-Calle, E.; Ho¨fler, J.; Eberbach, W. Liebigs Ann. Chem.
1996, 1855-1866. (f) Marx, K.; Eberbach, W. Chem. Eur. J. 2000, in press.
(7) Knobloch, K. Part of the forthcoming Dissertation, University
Freiburg i. Br.
(8) Sonogashira, K. In Metal-Catalyzed Cross-Coupling Reactions;
Diederich, F., Stang, P. J., Eds.; Wiley-VCH: Weinheim, 1998; pp 203-
229.
10.1021/ol0056832 CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/29/2000

Scheme 2
synthesis of 6a-c,l,m direct deprotection of 3 (f 5) was
followed by reaction with the corresponding hydroxylamines
(see Scheme 2).
In some cases the 1H NMR spectra of the crude products
containing the nitrones 6 as major component showed
additional signals of a product that turned out to belong to
the epimino indenes 7. The formation of 7 is explained by
a known pathway involving a 1,3-dipolar cycloaddition
reaction affording the isoxazolines 8, which are subsequently
transformed into the aziridines 7 (Scheme 3).9,10 It must be
methanol in the presence of 0.5-1 equiv of sodium methylate
or potassium hydroxide. In the case of the N-methyl
derivative 6a total conversion was reached after 0.5 h,
resulting in a reaction mixture which, after aqueous workup
and flash chromatographic purification, gave a single mon-
omeric product in 84% yield. The structural identification
of the compound as the dihydrobenzazepin-3-one 10a
(Scheme 4) is based on the elemental analysis and the MS
Scheme 4
Scheme 3
and other spectroscopic data. In the IR spectrum there is a
strong band at 1650 cm-1 indicating a carbonyl group; the1H
due to the particular annulation of the isoxazolines that no
direct evidence for 8 was obtained. However, control
experiments with 6a,d,e,i revealed that on standing or on
heating in boiling benzene the nitrones are indeed trans-
formed into the bicyclic systems 7.
(9) The ring transformation of type 8 f 7 can be brought about either
by thermal or photochemical activation of 4-isoxazolines; in the latter case
the isolation of intermediate azomethine ylides was successfully ac-
complished: Lopez-Calle, E.; Eberbach, W. J. Chem. Soc., Chem. Commun.
1994, 301-302.
(10) For thermal tranformations of type 8 f 7, see: (a) Huisgen, R.;
Niklas, K. Heterocycles 1984, 22, 21-26. (b) Padwa, A.; Wong, G. S. K.
J. Org. Chem. 1986, 51, 3125-3133. (c) Mullen, G. B.; Bennet, G. A.;
Georgiev, V. St. Liebigs Ann. Chem. 1990, 109-110.
The cyclization experiments were typically carried out at
room temperature by stirring solutions of the nitrones 6 in
1118
Org. Lett., Vol. 2, No. 8, 2000

NMR spectrum shows, besides signals in the aromatic region
and the singlet for the N-methyl group, a singlet for the
methylene protons at C-1 and an AB pattern ascribed to the
olefinic protons H-4 and H-5 (see Table 1). The final proof
of structure 6a came from the results of an X-ray analysis.11
Scheme 5
1
Table 1. Yields and Selected H NMR Data for Azepinones
10a
1H NMR data
yield (%)
δ1-H
4.24
δ4-H
δ5-H
J 4,5 (Hz)
a
b
c
d
e
f
g
h
i
84b
6.41
6.55
6.36
7.08
7.20
6.98
7.20
∼7.2f
6.89
7.06
∼7.4f
∼7.4f
7.20
7.28
6.99
7.09
7.13
7.21
12.2
12.2
12.1
84c
4.66
4.27
4.12
4.55
4.10
4.57
4.50
4.72
4.24/4.53h
4.28
4.18
4.58
4.08
4.22
85c
24b,d
68c,e
86b,g
46b,g
75b
40b
j
k
l
77b
64c
77b
6.33
6.44
12.2
12.2
m
n
o
93c
25c,i
76c
(ii) 8π-cyclization 9 f 11,5,6 (iii) N-O cleavage (11 f
12),6d,e,13 (iv) diradical recombination (12 f 13), (v) one-
or two-step cyclization of the azadienyl cyclopropanone 13
to 14, and finally, (vi) 1,5-H shift 14 f 10. There is
precedent at least for the first three steps of this pathway
(see references).
a Reaction conditions: 0.2 molar in MeOH, 0.5-1 equiv of base, 0.5-5
b
c
d
e
h (f, 24 h; g, 20 h), rt. Base: NaOMe. Base: KOH. +29% a.
f
g
h
+26% b. Partially covered by signals of Ar-H. Solvent: CH2Cl2.
i
J1a,1b) 15.3 Hz. +43% l.
The transformation 6 f 10 is not restricted to benzannu-
lated derivatives. Preliminary work has shown that the same
reaction takes place with alkenoannulated compounds and
also with simple linear derivatives. For instance, treatment
of the diphenyl nitrone 15 with NaOMe/MeOH produces the
corresponding azepinone 16 in 94% yield; in this case there
is no final H-migration (Scheme 6).
The conversion of type 6 f 10 turned out to be quite
general and was successfully accomplished for many dif-
ferent derivatives. Thus, under analogous conditions the
nitrones 6b-o afforded the corresponding benzazepinones
10b-o with yields up to 93%. As a result of partial
hydrodesilylation during the reaction of 6d,e,n the additional
amount of 10a, 10b and 10l, respectively, has to be added
to the yields given in the table.
With regard to the conformational mobility of the annu-
lated azepinone system, only the N-tert-butyl 4-phenyl
compound 10j has some rigidity under ambient conditions.
According to the 1H NMR spectra the methylene protons at
C-1 are not equivalent and therefore give rise to a pair of
doublets (J ) 15.3 Hz).
Scheme 6
From the inspection of the structural differences of the
starting and final products, 6 and 10, it becomes clear that
there is no simple connection between the nitrone and
azepinone compounds. The most striking feature concerns
the interchange of the terminal and central positions of the
allene unit of 9, i.e., the carbon atom connected to R1 ends
up at C-4 of 10, which is now flanked by both former allene
carbons. A possible but still tentative mechanism for the
observed transformation includes the sequence outlined in
Scheme 5: (i) propargyl-allene tautomerization (6 f 9),12
Despite the structural simplicity of 1,2-dihydro[c]benza-
zepin-3-ones 10, there are only a few synthetic methods
available, mostly used for the synthesis of further annulated
compounds,14 which frequently possess biological activity.15
The parent compound of 6 (6a, NH instead of NMe) has
(13) (a) Eberbach, W.; Roser, J. Tetrahedron Lett. 1987, 28, 2689-2692.
(b) Eberbach, W.; Laber, N. Tetrahedron Lett. 1992, 33, 61-64.
(14) (a) Gschwend, H. W.; Hamdan, A. J. Org. Chem. 1982, 47, 3652-
3657. (b) Ro¨hrkasten, R.; Kreher, R. P. Chem. Ber. 1991, 124, 2085-2090.
(c) Kai, H.; Yamauchi, M.; Murai, S. Tetrahedron Lett. 1997, 38, 9027-
9030.
(11) Knobloch, K.; Keller, M.; Eberbach, W. paper in preparation.
(12) (a) Taylor, D. R. Chem. ReV. 1967, 67, 317-359. Grimaldi, J.,
Bertrand, M. Bull. Soc. Chim. Fr. 1971, 947-957; 4316-4320.
Org. Lett., Vol. 2, No. 8, 2000
1119

been synthesized in modest yield by irradiation of â-azi-
donaphthalene in methanol and subsequent hydrolysis of the
iminoether.16 The novel access to 10 described in this paper
is a useful and widely applicable method for simple and more
complex derivatives of this heterocyclic system.17
Acknowledgment. This work was supported by the Fonds
der Chemischen Industrie.
Supporting Information Available: Full characterization
data of compounds 7a,d,e,i, 10a-o, and 16. This material
is available free of charge via the Internet at http://pubs.acs.org.
(15) (a) Hawthorne, J. O.; Mihelic, E. L. U.S. Patent 3,668,232, 1972;
Chem. Abstr. 1972, 77, P101199s; U.S. Patent 3,551,414v, 1970; Chem.
Abstr. 1971, 74, P125484. (b) Gorshkova, V. K.; Saratikov, A. S.;
Tignibidina, L. G. Pharm. Chem. J. (Engl. Transl.) 1994, 28, 158-162.
(16) Rigaudy, J.; Igier, C.; Barcelo, J. Tetrahedron Lett. 1975, 16, 3845-
3848.
(17) Typical procedure, exemplified for 10a, is as follows. The solution
of 6a (100 mg, 0.58 mmol) and sodium methylate (16 mg, 0.29 mmol) in
4 mL of dry methanol was stirred at room temp. After complete conversion
(0.5 h, monitored by TLC) the solution was diluted with water (10 mL)
OL0056832
and then extracted with CH2Cl2 (3 × 10 mL). The combined organic phase
was washed successively with saturated NH4Cl and NaCl solutions, dried,
and concentrated. Flash chromatography of the residue (SiO2, cyclohexane/
ethyl acetate 1:1, 1:2) gave 10a (84 mg, 84%) as a white solid; mp 132-
133 °C (ethanol).
1120
Org. Lett., Vol. 2, No. 8, 2000