Reaction of 2-methoxy-3H-azepine with NBS: Efficient synthesis of 2-substituted 2H-azepines

Article abstract of DOI:10.1021/jo0500232

(Chemical Equation Presented) The reaction of 2-methoxy-3H-azepines, in the presence or absence of a nucleophile, with N-bromosuccinimide (NBS) gave a regioselective 1,4-adduct from which the corresponding 2H-azepine derivatives were formed via base-promoted hydrogen bromide elimination, generally in moderate to quantitative yield. Competitive formation of 4-bromo-2-methoxy-3H- azepine by electrophilic substitutuion or 3H-azepin-2-yl 2H-azepin-2-yl ether by transetherification was minimized at lower reaction temperatures. Quantitative substitution of 2-(2′,4′,6′-trichlorophenoxy)-2H-azepine derivatives, formed in moderate yield from the respective 3H-azepine and NBS in the presence of 2,4,6-trichlorophenol (TCP), by various nucleophiles gave the corresponding 2-substituted 2H-azepine. Among these nucleophiles were alkanethiol and alkylamine that are not tolerated in the reaction of 3H-azepine and NBS.

Full text of DOI:10.1021/jo0500232

Reaction of 2-Methoxy-3H-azepine with NBS: Efficient Synthesis
of 2-Substituted 2H-Azepines
Christopher E. J. Cordonier, Kyosuke Satake,* Mikihiko Atarashi, Yousuke Kawamoto,
Hideki Okamoto, and Masaru Kimura
Department of Chemistry, Faculty of Science, Okayama University,
Tsushima-Naka 3-1-1, Okayama 700-8530, Japan
satake@cc.okayama-u.ac.jp
Received January 4, 2005
The reaction of 2-methoxy-3H-azepines, in the presence or absence of a nucleophile, with
N-bromosuccinimide (NBS) gave a regioselective 1,4-adduct from which the corresponding 2H-
azepine derivatives were formed via base-promoted hydrogen bromide elimination, generally in
moderate to quantitative yield. Competitive formation of 4-bromo-2-methoxy-3H-azepine by
electrophilic substitutuion or 3H-azepin-2-yl 2H-azepin-2-yl ether by transetherification was
minimized at lower reaction temperatures. Quantitative substitution of 2-(2′,4′,6′-trichlorophenoxy)-
2H-azepine derivatives, formed in moderate yield from the respective 3H-azepine and NBS in the
presence of 2,4,6-trichlorophenol (TCP), by various nucleophiles gave the corresponding 2-substituted
2H-azepine. Among these nucleophiles were alkanethiol and alkylamine that are not tolerated in
the reaction of 3H-azepine and NBS.
Introduction
In previous work, from the more thermally stable 3H-
azepine substrates, we reported the first synthesis of 2H-
azepines where 2,7-dimethoxy-2H-azepines 2a-c were
formed specifically by adding methanol to the reaction
mixture of the corresponding 2-methoxy-3H-azepines
1a-c5 and bromine followed by workup with aq alkaline
(Scheme 1).6 Similar treatment of 2,5- and 3,6-di-tert-
Synthesis of 2H-azepine derivatives as naturally oc-
curring alkaloids1 was initiated upon isolation of chalci-
porone and norchalciporyl propionate.2 As well, synthetic
value of 2H-azepine has been generated from the purpose
of studying the characteristic thermal [1,5]-hydrogen
shift to the more thermally stable 3H-azepine,3 which
consequently provides a challenge in synthesis.1 Recently
2-methoxy-2H-azepine derivatives have also been real-
ized as an indispensable precursor in the synthesis of
azatropylium ions in solution by demethoxylation using
TiCl4.4
SCHEME 1. Conversion of 3H-Azepine to
2H-Azepine with Bromine
* Address correspondence to this author. Phone and fax: +81-86-
751-7841.
butyl-3H-azepines (1d,e)3a also gave 2-methoxy-2H-
azepines 2d,e, respectively.7 Formation of 2-methoxy-2H-
azepine can be paralleled to trapping tropylium bromide,
(1) (a) Albini, A.; Bettinetti, G.; Minoli, G. J. Am. Chem. Soc. 1991,
113, 6928-6934. (b) Hamprecht, D.; Josten, J.; Steglich, W. Tetrahe-
dron 1996, 52, 10883-10902. (c) Jones, D. W.; Thornton-Pett, M. J.
Chem. Soc., Perkin Trans. 1 1995, 809-815. (d) Steglich, W.; Bauer,
H.; Grosse-Bley, M.; Jeschke, R.; Josten, J.; Klein, J. J. Heterocycl.
Chem. 1990, 27, 107-110.
(2) (a) Spiteller, P.; Hamprecht, D.; Steglich, W. J. Am. Chem. Soc.
2001, 123, 4837-4838. (b) Sterner, O.; Steffan, B.; Steglich, W.
Tetrahedron 1987, 43, 1075-1082.
(3) (a) Satake, K.; Okuda, R.; Hashimoto, M.; Fujiwara, Y.; Okamoto,
H.; Kimura, M.; Morosawa, S. J. Chem. Soc., Perkin Trans. 1 1994,
1753-1757. (b) Satake, K.; Okuda, R.; Hashimoto, M.; Fujiwara, Y.;
Watadani, I.; Okamoto, H.; Kimura, M.; Morosawa, S. J. Chem. Soc.,
Chem. Commun. 1991, 1154-1156. (c) Satake, K.; Saitoh, H.; Kimura,
M.; Morosawa, S. Heterocycles 1994, 38, 769-778.
(4) Satake, K.; Kubota, Y.; Cordonier, C.; Okamoto, H.; Kimura, M.
Angew. Chem., Int. Ed. 2004, 43, 736-738.
(5) (a) Masaki, M.; Fukui, K.; Kita, J. Bull. Chem. Soc. Jpn. 1977,
50, 2013-2016. (b) Kubota, Y.; Satake, K.; Ikui, R.; Okamoto, H.;
Kimura, M. Bull. Chem. Soc. Jpn. 2003, 76, 805-811.
(6) Satake, K.; Cordonier, C.; Kubota, Y.; Jin, Y.; Kimura, M.
Heterocycles 2003, 60, 2211-2215.
(7) Satake, K.; Tawada, Y.; Okamoto, H.; Kimura, M. J. Chem. Soc.,
Perkin Trans. 1 1997, 2015-2017.
10.1021/jo0500232 CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/15/2005
J. Org. Chem. 2005, 70, 3425-3436
3425

Cordonier et al.
TABLE 1. Conversion of 3H-Azepine to 2-Substituted
2H-Azepine with NBSa
SCHEME 2. Formation and Substitution of
Tropylium
reaction
SCHEME 3. Conversion of 3H-Azepine to
2H-Azepine with NBS
starting temp
yield
product (%)
entry compd
(°C)
additive R1
R3
1
1a
1a
-41
none
H
H
H
H
C4H4O2N
3a
4a
3a
4a
3b
3c
4c
3c
4c
5a
3a
5b
3b
2a
2b
2c
3c
6
81
19
2
25
none
C4H4O2N
27
68
3
4
1b
1c
-98
-98
none
none
tBu C4H4O2N
Me C4H4O2N
Me
100
73
22
5
6
7
1c
1a
25
none
Me C4H4O2N
Me
H
H
97
53
-41
AcNHMe
AcNMe
C4H4O2N
37
1b
1a
-98
-41
AcNHMe tBu AcNMe
77
tBu C4H4O2N
21
8
9
10
MeOH
MeOH
MeOH
H
MeO
99
1b
1c
-98
-98
tBu MeO
100
66
Me MeO
formed from bromine and 1,3,5-cycloheptatriene (CHT),8
with methanol9 (Scheme 2). Regioselective substitution
at the C2 position by methanol has been predicted based
on frontier molecular orbital consideration of the interac-
tion between a nucleophile and the previously hypoth-
esized azepinium ion intermediates, for which the π-LU-
MO coefficients were largest at the C2 position.3,7 While
azatropylium is theoretically predicted to be a relatively
stable aromatic cation,10 azepine is expected to be more
susceptible to oxidation than CHT leading to azatropy-
lium because of an electronegative nitrogen atom in the
triene conjugation.11
Although the reaction conditions are similar to those
of CHT and bromine, even in situ by NMR spectral
measurement of the reaction mixture, evidence of an
azepinium ion intermediate has not been observed. To
explore the reaction mechanism including the reason for
regioselective introduction of the methoxy group at the
C2 position, attempts were made to determine the
structure of any reaction products without adding metha-
nol to the 3H-azepine and NBS reaction mixture prior
to and after treatment with alkaline.6
Me C6H2O2N
33
11
12
13
14
15
1b
1a
1b
1a
1a
-98
-41
-98
-41
-41
tBuOH
TCP
TCP
AcOH
PrSH
tBu tBuO
91
H
C6H2Cl3O
7a
7b
8
1a
9a
1b
1c
47
tBu C6H2Cl3O
64
H
AcO
80
H
74
H
PrS
21
16
17
1b
1c
-98
-98
PrSH
PrSH
tBu
Me
>95
>95
a Reactions of 3H-azepine (1 mmol), an additive (2 mmol), and
NBS (1.1 mmol) were carried out in CH2Cl2 under a nitrogen
atmosphere at the specified temperature for 10 h, then an excess
of triethylamine was introduced, and the solution was stirred at
room temperature for 5 h. Yields are based on the crude isolated
products.
of 1c in the presence of methanol produced both 2-meth-
oxy-2H-azepine 2c in 66% yield and 2-succinimido-2H-
azepine 3c in 33% yield (Table 1, entry 10). In the
absence of methanol (Scheme 3) reaction of 3H-azepines
1a-c efficiently gave 2-succinimido-2H-azepines 3a-c
in 81%, quantitative, and 73% yield, respectively (Table
1, entries 1, 3, and 4). Formation of 3a and 3c was
accompanied by formation of the corresponding regiose-
lectively brominated products, 4-bromo-2-methoxy-3H-
azepine (4a) in 19% yield and 4-bromo-2-methoxy-5-
methyl-3H-azepine (4c) in 22% yield, for which the MS
and NMR spectral data of each supports the structure.
Replacing bromine with NBS led to a new efficient
method for converting 3H-azepine to 2H-azepine.6 In
contrast to the conversion with bromine, the chemose-
lectivity of NBS as well as a significantly lower concen-
tration of HBr in the reaction environment are factors
attributable to improved yields. Results for the conver-
sion of 1a-c in the presence of other nucleophiles are
listed in Table 1.
In the presence of methylamine or diethylamine the
starting material was almost completely recovered with
no evidence of 2-amino-2H-azepine formation. In the
presence of N-methylacetamide both succinimide and
amide substituted products 3a,b and 5a,b were formed
(Table 1, entries 6 and 7). Similar to N,N-dimethylac-
etamide,12 due to restricted rotation about the C-N
bonds, at room temperature the 2-(N-methylacetamido)-
Results and Discussion
Reaction of 3H-Azepine and NBS. An alternative
attempt to understand the mechanism of the previously
reported 3H-azepine to 2H-azepine conversion in Scheme
16,7 by replacing bromine with NBS (Scheme 3) resulted
in quantitative conversion of 3H-azepines 1a,b to 2-meth-
oxy-2H-azepines 2a,b (Table 1, entries 8 and 9). Reaction
(8) Doering, W. Von E.; Knox, L. H. J. Am. Chem. Soc. 1954, 76,
3203-3206.
(9) Doering, W. Von E.; Knox, L. H. J. Am. Chem. Soc. 1957, 79,
352-356.
(10) (a) Di Stefano, M.; Rosi, M.; Sgamellotti, A.; Negri, F. Chem.
Phys. 2004, 302, 295-308. (b) Di Stefano, M.; Rosi, M.; Sgamellotti,
A.; Ascenzi, D.; Bassi, D.; Franceschi, P.; Tosi, P. J. Chem. Phys. 2003,
119, 1978-1985. (c) Smith, D. A.; Bitar, J. J. Org. Chem. 1993, 58,
6-8. (d) Batori, S.; Gompper, R.; Meier, J.; Wagner, H.-U. Tetrahedron
1988, 44, 3309-3318.
(11) (a) Paul, I. C.; Johnson, S. M.; Paquette, L. A.; Barrett, J. H.;
Haluska, R. J. J. Am. Chem. Soc. 1968, 90, 5023-5024. (b) Paquette,
L. A. J. Am. Chem. Soc. 1963, 85, 4053-4054.
3426 J. Org. Chem., Vol. 70, No. 9, 2005

Synthesis of 2-Substituted 2H-Azepines
was also easily substituted by stronger nucleophiles.
Attempts to isolate the acetoxy derivatives from the
reaction mixtures of 1b and 1c in the presence of acetic
1
acid failed but by qualitative analysis of the H NMR
spectra of the crude reaction mixtures, the corresponding
2-acetoxy-2H-azepines were formed in moderate yield.
Although a large loss through decomposition on silica gel
occurred, pure 8 was isolated by MPLC at 0 °C.
In the presence of propane thiol, NBS favorably reacted
with propane thiol to give propyl disulfide as the major
product along with the unreacted starting material.
Despite reaction of 3H-azepine being the minor reaction
pathway, the 2-propylthio-2H-azepine 9a was nonethe-
less formed in low yield from 1a although no 2-propylthio-
2H-azepine was isolated from the reaction of 1b or 1c
(Table 1, entries 15, 16, and 17).
The reaction temperatures were determined by maxi-
mization of monocyclic 2H-azepine formation, which were
also found to be the minimum temperatures at which the
3H-azepines would react within a reasonable amount of
time. Predominance of 4a increased in the reaction of 1a
and NBS as the temperature increased from -41 °C, and
at room temperature 3a and 4a were formed in 27% and
68% yields, respectively (Table 1, entry 2). Reaction of
1c and NBS at room temperature gave 4-bromo-2-
methoxy-5-methyl-3H-azepine (4c) in near quantitative
yield (Table 1, entry 5). Unexpected formation of bisazepi-
nyl ethers 10-12 from the reaction of 1b and NBS
occurred at temperatures above -63 °C and became
predominant above 0 °C. The temperature-dependent
product distribution is summarized in Table 2. Although
not discussed here, it should be noted that the product
distribution is also dependent on the solvent system.
Generally a 0.1-0.3 M solution of 3H-azepine in dichlo-
romethane gave good results.
FIGURE 1. The E and Z structures of 5 observed in the NMR
spectra due to restricted rotation about the amide C-N bond.
2H-azepines 5a and 5b were respectively found to exist
in 55:45 and 58:42 molar ratios of Z and E conformers
(Figure 1) for which chemical exchange was confirmed
by observation of positive signals in the 2D NOESY
spectra between corresponding proton signals. Assign-
ment of the signals to the corresponding isomer in the
1H and 13C NMR spectrum of the isomeric mixture was
accomplished with 1H-1H COSY, HMQC, and HMBC
techniques. Z and E geometry was concluded based on
the difference in peak width and relative chemical shift
of the N-methyl NMR signals (Z-5a: δH 3.18, ω1/2 ) 1.32
( 0.24 Hz; E-5a: δH 3.11, ω1/2 ) 1.90 ( 0.24 Hz; Z-5b:
δH 3.21, ω1/2 ) 1.53 ( 0.24 Hz; E-5b: δH 3.14, ω1/2 ) 1.95
( 0.24 Hz) due to cis and trans long-range coupling and
anisotropic shift due to shielding by the CdO bond in
comparison to dimethylacetamide.12
Similar to the conversion in the presence of methanol,
even in the presence of sterically hindered tert-butyl
alcohol or 2,4,6-trichlorophenol (TCP), which is a com-
paratively weak nucleophile, the corresponding 2-tert-
butoxy-2H-azepine 6 and 2-TCP-2H-azepines 7a,b (Table
1, entries 11, 12, and 13) were formed in good and
moderate yields. Use of TCP as an example of a phenol
derivative in this reaction was 2-fold. First, TCP is
relatively resistant to bromination whereas unsubsti-
tuted phenols react competitively with NBS under the
reaction conditions to give complex mixtures of bromi-
nated phenol derivatives and unreacted 3H-azepine.
Second, based on the results of Table 1, oxygen nucleo-
philes generally react efficiently under these conditions,
thus although the TCP group is a relatively weak
nucleophile TCP should nonetheless react to give the
corresponding 2H-azepine. Unfortunately due to the
exceptional ability of TCP as a leaving group, isolation
of 7a,b by the standard procedure of extracting byprod-
ucts with water (triethylammonium bromide) and aque-
ous base (succinimide) led to hydrolysis of 7a,b and
attempted isolation by medium-pressure liquid chroma-
tography (MPLC) on silica gel or alumina even at 0 °C
led to complete decomposition. Although separation of
7a,b from 3a,b, triethylammonium bromide and succin-
imide was accomplished to a good degree by extraction
of 7a,b from the crude reaction mixture with hexane,
triethylammonium trichlorophenoxide and 7a,b could not
be completely separated thus characterization was lim-
ited to NMR and MS spectroscopy.
Reaction of 1b and NBS in a 2:1 molar ratio at room
temperature (eq 1) gave 5-tert-butyl-3H-azepin-2-yl 4-tert-
butyl-7-methoxy-2H-azepin-2-yl ether (13) in 84% yield.
Peculiar formation of bisazepinyl ethers is an unprec-
edented transesterification of the imidate oxygen in 1b,
replacing the methyl group with a 2H-azepinyl group or
vice versa substitution of a 2H-azepine at C2 by the
oxygen of a 2-methoxy-3H-azepine. There are no parallel
transesterifications or transetherifications of any form
that have been reported in the literature to our knowl-
edge and out of the 3H-azepines examined here, only 1b
led to formation of bisazepinyl ethers.
Structure Determination of Bisazepinyl Ethers.
The structure of bisazepinyl ether 13 was substantiated
by MS (FAB) m/z 343 (M + H)+. The structures of each
ring linked by oxygen were determined to be a 5-tert-
butyl-3H-azepin-2-yl and a 4-tert-butyl-7-methoxy-2H-
azepin-2-yl group based on 1H and 13C NMR spectral
data, which shows nearly superimposable signals to those
of 1b and 2b in the region of olefinic ring protons.
Structures of 10-12 were concluded similarly.
2-Acetoxy-7-methoxy-2H-azepine 8 was formed in good
yield from 1a in the presence of acetic acid (Table 1, entry
14). Similar to the TCP derivatives, the acetoxy group
(12) (a) Reeves, L. W.; Shaddick, R. C.; Shaw, K. N. Can. J. Chem.
1971, 49, 3683-3691. (b) Martin, M. L.; Ricolleau, G.; Poignant, S.;
Martin, G. J. J. Chem. Soc., Perkin Trans. 2: Phys. Org. Chem. (1972-
1999) 1976, 182-186.
J. Org. Chem, Vol. 70, No. 9, 2005 3427

Cordonier et al.
TABLE 2. Temperature-Dependent Product Distribution of the Reaction of 1b and NBSa
crude yield (%)
temp (°C)
time (h)
3b
10
11
12
-98
-63
-23
-13
0
10
8
95
89
66
62
27
18
10
5
2
13
15
22
23
25
58
6
5
5
7
9
15
40
43
28
6
5
6
5
4
7
10
3
rt
40
2
0.5
1
a Reactions of 1b (1 mmol) and NBS (1.1 mmol) were carried out in CH2Cl2. Yields were determined based on the H NMR spectra of
the reaction mixture.
Two isomers of 12 were separated by MPLC on silica
gel. The NMR spectra and incapacity for chemical
exchange suggests they are stereoisomers. Although clear
evidence is insufficient, the cis and trans geometry of the
bromine atom and succinimido group in the two isomers
was estimated by comparing GIAO chemical shift calcu-
lation13 (HF/6-311+G (2d,p)) results based on the opti-
mized cis and trans structures (B3LYP/6-31G*)14 of 14b
(Figure 2a,b), calculated using the Gaussian9815 program,
with the 1H NMR spectra of 12 (Figure 2c,d). The
assignments were made based on the extreme variation
in chemical shift of the C2 proton on the adduct ring,
observed in both the experimental spectral data (7.40 and
6.74 ppm) and theoretically calculated spectral values
(6.94 and 6.18 ppm for trans and cis 14b). The exception-
ally low field shift of the C2 proton resonating at 7.40
ppm and that estimated to resonate at 6.94 ppm in the
calculated trans structure can be expected from contact
deshielding by the bromine atom16 on C5. Close proximity
of the C2 proton and the C5 bromine atom is predicted
for the trans structure (2.770 Å). Thus by consideration
of this theoretical and experimental coherence, assign-
FIGURE 2. A comparison of the calculated chemical shifts
(HF/6-311+G (2d,p)) of the optimized structures (B3LYP/6-
ment of the stereoisomer with the proton resonating at
7.40 ppm to the trans adduct and that resonating at 6.74
ppm to the cis adduct was made. Calculated chemical
shifts for C2 protons of the optimized structures of cis
and trans 14b and observed chemical shifts for the
stereoisomers of 12 are shown in Figure 2.
31G*) of the (a) trans and (b) cis 1,4-adducts of 14b with the
chemical shifts of the stereoisomers of 12 (c and d). Chemical
shift values are expressed in ppm.
1,4-Addition of NBS. The reaction intermediates
responsible for this conversion were elucidated by NMR
observation of the reaction product(s) of 3H-azepine and
NBS in the presence or absence of methanol. Regiose-
lective formation of NBS 1,4-adduct 5-bromo-5,6-dihydro-
2-succinimido-2H-azepines 14a,b was confirmed by sig-
nals corresponding to a single compound with coupling
constants that distinguish the ring structure (14a: J2-3
) 2.7 Hz, J3-4 ) 11.7 Hz, J4-5 ) 2.7 Hz, J5-6 ) 4.9 Hz,
J5-6′ ) 12.5 Hz, J6,gem ) 13.7 Hz; and 14b: J2-3 ) 3.4
Hz, J5-6 ) 3.7 Hz, J5-6′ ) 4.0 Hz, J6,gem ) 18.6 Hz) in the
NMR spectral data upon completion of the reaction of
3H-azepines 1a,b with NBS in CD2Cl2 (Table 3, entries
1 and 2) and observation of the (M + H)+ signal in the
FAB-MS and FAB-HRMS spectra of the crude reaction
mixtures. In the presence of methanol regioselective
formation of bromomethoxy 1,4-adduct 5-bromo-5,6-di-
hydro-2-methoxy-2H-azepines 15a,b was similarly con-
(13) (a) Wolinski, K.; Hinton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990,
112, 8251-8260. (b) Cheeseman, J. R.; Trucks, G. W.; Keith, T. A.;
Frisch, M. J. J. Chem. Phys. 1996, 104, 5497-5509.
(14) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652. (b) Frisch,
M. J.; Pople, J. A.; Binkley, J. S. J. Chem. Phys. 1984, 80, 3265-3269.
(15) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-
Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, Revision A.9;
Gaussian, Inc.: Pittsburgh, PA, 1998.
(16) Yonemoto, T. Can. J. Chem. 1966, 44, 223-231.
3428 J. Org. Chem., Vol. 70, No. 9, 2005

Synthesis of 2-Substituted 2H-Azepines
TABLE 3. 1,4-Addition of Bromine and a Nucleophile to
4a in a 4 to 1 molar ratio (Table 3, entries 1-4), which
is the same molar ratio as 3a to 4a suggesting competi-
tion between 1,4-addition and electrophilic substitution.
Treatment of the 1,4-adducts with triethylamine gave
quantitative conversion to the respective 2H-azepines
and triethylamine hydrobromide via 1,2-elimination of
HBr (eq 2). Although there are examples of ionic 1,4-
3H-Azepinesa
reaction
temp
(°C)
starting
entry compd
yield
product (%)
additive R1
R3
1
1a
-41
H
H
4a
14a
14b
15a
15b
10
20
80
100
99
100
1
C4H4O2N
2
3
4
5
1b
1a
1b
-98
-41
-98
0
tBu C4H4O2N
addition of bromine and a nucleophile in the literature,17
this is the first report of ionic 1,4-addition of NBS to a
conjugated diene to give the 1-bromo-4-succinimido ad-
duct as far as we know. Typical yields of 1,4-addition to
1,3-butadiene derivatives using NBS in the presence of
water, alcohol, or carboxylic acid are around 50%.17 In
comparison to other π-electron-rich conjugated dienes,
the 1,3-butadiene function in the 3H-azepine ring shows
exceptionally high reactivity and reaction regioselectivity.
The reasons for the difference are not yet fully understood
but it is thought to be due to the rigid geometry of the
conjugated imino-diene structure in the azepine ring.
Reaction of Bisazepinyl Ether 13 with NBS. There
are two unique features of bisazepinyl ethers. One, a
molecule possessing both a 2H- and a 3H-azepine ring
provides a system where the relative reactivity of both
isomers can be compared. The other, the steric effect from
close proximity of two azepine rings renders bisazepinyl
ethers thermally stable relative to the monoring ana-
logues. Consequences of the thermal stability are ob-
served in the 1,5-hydrogen isomerization rate of the 2H-
ring of 13 (τ1/2 ) 2150 min at 80 °C) compared to 2b (τ1/2
) 210 min at 80 °C) and in the NMR signals for the C3
methylene protons at room temperature of the 3H-ring
of 13, 2.62 (1H, dd) and 3.16 ppm (1H, dd), show distinct
geminal coupling (Jgem ) 12.2 Hz) compared to 1b, 2.59
ppm, (2H, d). The latter implies the ring inversion of the
3H-azepine moiety of 13 is sterically hindered. The
thermal stability of 13 afforded an opportunity for
isolation and examination of stable 1,4-adducts of the 3H-
azepine ring.
MeOH
H
MeO
MeOH tBu MeO
tBu
13
tBu C4H4O2N
12
88
100
6
13
0
MeOH tBu MeO
16
a Reactions of 3H-azepine (1 mmol), an additive (2 mmol), and
NBS (1 mmol) were carried out in CD2Cl2. Yields were determined
based on the 1H NMR spectra of the reaction mixture.
firmed (Table 3, entries 3 and 4, 15a: J2-3 ) 2.2 Hz, J3-4
) 11.8 Hz, J4-5 ) 4.9 Hz, J5-6 ) 3.4 Hz, J5-6′ ) 12.0 Hz,
J6,gem ) 13.4 Hz; and 15b: J2-3 ) 3.1 Hz, J5-6 ) 3.7 Hz,
J5-6′ ) 4.0 Hz, J6,gem ) 18.9 Hz). Upon precipitation from
acetone 1,4-adduct 14b was stable and pure enough for
complete spectroscopic characterization but unfortu-
nately, due to instability, clear observation of 14a, 15a,
and 15b was limited to the crude reaction mixture within
a few hours of formation. Complete 1H and 13C NMR
spectral data of adducts 14a,b and 15a,b were attained
for which assignment of the signals to the structures was
1
accomplished by using HMQC and H-1H COSY tech-
1
niques (Table 4). Similarity between the calculated H
NMR shifts for the trans structure of 14b opposed to the
1
cis structure (Figure 2a,b) and the H NMR spectra of
14b (Table 4), in particular at the C2 proton (δ 7.00 ppm),
suggests 14b is the trans adduct, which was also pre-
dicted to be more stable than the cis adduct by 0.0049
au (3.1 kcal/mol) by comparison of the B3LYP/6-31G*14
optimized structures. Electrophilic addition of bromine
occurred selectively at C4 with concurrent nucleophilic
addition selectively at C7 of the 3H-azepine ring. Forma-
tion of other stereoisomers has not been observed.
Reaction of 13 and NBS at 0 °C gave a <1% yield of
10 and a 88% yield of a 24:76 molar ratio of trans- and
cis-12. The 1,4-adduct 16 was formed quantitatively in
The 1,4-adducts 14b, 15a, and 15b were quantitatively
formed as sole products but 14a was formed along with
TABLE 4. NMR Resonances of 1,4-Adductsa
compd
R1
R2
C2
C3
C4
C5
C6
H2
H3
H4
H5
H6ax
H6eq
14a
H
C4H4O2N
C4H4O2N
MeO
62.0
60.5
86.4
84.8
62.7
61.0
80.4
82.7
131.3
129.2
134.6
133.4
121.5
121.6
124.8
124.7
131.7
149.1
131.1
147.9
151.8
150.8
152.9
152.1
42.0
41.9
42.6
43.1
39.8
40.1
42.3
41.8
40.8
39.7
40.8
40.3
47.0
47.0
44.2
44.8
6.45
7.00
5.44
5.75
6.74
7.40
5.65
5.59
5.90
5.84
5.76
5.86
5.53
5.37
6.10
6.06
5.75
4.90
4.85
4.81
4.86
4.76
4.77
4.83
4.79
3.05
3.05
3.08
3.03
4.01
4.12
3.95
3.81
3.56
3.16
3.33
3.09
3.21
3.17
2.95
2.89
14b
tBu
H
15a
15b
5.80
tBu
tBu
tBu
tBu
tBu
MeO
syn-12
anti-12
16
C4H4O2N
C4H4O2N
MeO
16
MeO
a Resonance values are displayed in ppm.
J. Org. Chem, Vol. 70, No. 9, 2005 3429

Cordonier et al.
TABLE 5. 1,4-Elimination of HBr from the 1,4-Adducts
12 and 13a
entry
substrate
R
product
yield (%)
1
2
12
16
C4H4O2N
MeO
11
17
92
68
FIGURE 3. Relative stabilities of dihydroazepinium isomers,
representative of the cation intermediate A, arising from
electrophilic attack on the available positions of the 3H-azepine
ring. AM1 energies are in kcal/mol and B3LYP/6-31G* energies
are in au.
a Yields are based on isolated products.
SCHEME 4. Conversion Mechanism
FIGURE 4. LUMO coefficients of the proposed reaction
intermediate A for R1 ) H, R2 ) OMe calculated by AM1.18
philic 1,4-addition followed by 1,2-elimination. Compe-
tition of electrophilic substitution by bromine selectively
at C4 with regioselective 1,4-addition and the observation
that 4-bromo-2-methoxy-3H-azepines are not formed
from 1,4-adducts under the reaction conditions further
suggest the conjugated cation intermediate A that is
common to both reaction pathways, proposed in Scheme
4, as the first reaction intermediate. Similarly, the
observation that methanol does not displace the succin-
imide moiety of 1,4-adducts 14a,b also supports the
reaction pathway via cation intermediate A. Although the
methoxy moiety in the conjugation is expected to enhance
reactivity and regioselectivity by π-electron donation to
the conjugation, 2-alkyl-substituted 3H-azepines (i.e.
1d,e) show similar regioselectivity7 suggesting the reac-
tivity is general.
the presence of methanol again as a mixture of two
stereoisomers in a 1:4 molar ratio (Table 3, entries 5 and
6). Treatment of adduct 12 with DBU and 16 with Bu4-
NOH did not afford the corresponding 2H-azepines by
1,2-elimination of hydrogen bromide but instead the
substituted 3H-azepines 11 and 17 by 1,4-elimination
(Table 5). Elimination occurred at room temperature,
ruling out the likelihood of 1,5-hydrogen shift3 of a 2H-
azepine intermediate. Logically, 1,4-elimination as op-
posed to 1,2-elimination is a consequence of steric inter-
ference at C6 by the azepinoxy substituent at C7 of the
dihydroazepine ring.
Conversion Mechanism of 3H-Azepine to 2-Sub-
stituted-2H-azepine. The reaction system for the con-
version of 3H-azepine to 2H-azepine in Scheme 4 was
concluded by observation of regioselective 1,4-addition of
electrophilic bromine to C4 and a nucleophile to C7 of
the 3H-azepine ring followed by 1,2-elimination of HBr.
The regioselectivity of C4 toward electrophilic attack can
be explained in terms of the resulting intermediate cation
stability. Figure 3 shows the theoretical stabilities cal-
culated by AM118 and ab initio (B3LYP/6-31G*) meth-
ods14 for the cations potentially formed from 3H-azepine.
Concurrent with the experimental result, the cation in
Figure 3c is the most stable, which clearly supports the
proposed mechanism. The largest LUMO coefficient in
the intermediate cation A, shown in Figure 4, is on the
carbon that was C7 of the parent 3H-azepine and
therefore concurs with the experimental result that this
is the most reactive location toward a nucleophile.
Theoretical agreement with unambiguous experimental
results provides strong evidence that the 3H- to 2H-
azepine conversion mechanism is electrophilic + nucleo-
The formation mechanism of bisazepinyl ethers from
1b remains unclear. A likely mechanism could be nu-
cleophilic attack of the oxygen lone pair electrons of one
azepine molecule on the initially formed conjugated
cation intermediate A that was proposed in Scheme 4.
Removal of the methyl group from the new ether link
may occur via bromide substitution to give methyl
1
bromide. Careful observation of the H NMR spectrum
of the reaction mixture arising from 1b and NBS at room
temperature suggests formation of methyl bromide by a
singlet peak at 2.57 ppm. At higher temperatures forma-
tion of bisazepinyl ether from 1b, as opposed to electro-
philic bromination observed with 1a and 1c, can most
reasonably be considered due to steric interaction be-
tween the C4 position and the tert-butyl group in the
conjugated cation intermediate A.
(17) (a) Konya, N.; Seko, S. Eur. Patent 1072589 (20010131), 2001.
(b) Van, T. N.; Kimpe, N. Tetrahedron 2000, 56, 7969-7973. (c)
Deagostino, A.; Tivola, P. B.; Prandi, C.; Venturello, P. Synlett 1999,
1841-1843. (d) Wakamatsu, K.; Kigoshi, H.; Niiyama, K.; Niwa, H.;
Yamada, K. Tetrahedron 1986, 42, 5551-5558. (e) Pearson, A. J.; Ray,
T. Tetrahedron Lett. 1986, 27, 3111-3114.
(18) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P.
J. Am. Chem. Soc. 1985, 107, 3902-3909.
3430 J. Org. Chem., Vol. 70, No. 9, 2005

Synthesis of 2-Substituted 2H-Azepines
TABLE 6. Nucleophilic Substitution of 2-TCP- and 2-AcO-2H-Azepinesa
starting
compd
reaction time
at 30 °C
yield
(%)
entry
half-life
nucleophile
R1
R3
product
1
2
3
4
5
6
7
8
9
7b
7a
8
5 h
30 min
10 h
3 min
45 s
10 min
35 s
36 min
14 min
260 min
90 min
8 min
PrNH2
MeOH
MeOH
MeOH
tBuOH
AcOH
PrSH
tBu
H
PrNH
MeO
MeO
MeO
tBuO
AcO
PrS
18
2a
2a
2b
6
100
100
100
100
100
37
100
100
100
H
7b
7b
7a
7a
8
30 min
25 h0
25 h
tBu
tBu
H
8
25 h
H
9a
9a
9b
25 h
PrSH
H
PrS
7b
25 h
PrSH
tBu
PrS
a Reactions were carried out in CDCl3 with the specified nucleophile in excess of 2H-azepine. Reaction times were determined by the
time required for change in the 1H NMR spectra of the reaction mixture to become unobservable. Half-life values were extrapolated from
the time-dependent 1H NMR spectra. Yields were determined based on the 1H NMR spectra of the reaction mixture.
SCHEME 5. Substitution via 2-TCP-2H-Azepine
azepine derivatives providing an efficient method for
generating thermally less stable 2H-azepine derivatives.
Experimental Section
2-Methoxy-3H-azepines 1a and 1b were prepared according
to the established method5 and structures were confirmed by
1
identical H NMR spectra to those previously reported.
2-Methoxy-5-methyl-3H-azepine (1c). 4-Nitrotoluene (30.0
g, 219 mmol), tri-n-butylphosphine (97.4 g, 481 mmol), and
50 mL of methanol were heated in an autoclave at 150 °C for
25 h. Excess methanol was then removed and the reaction
mixture distilled. The fraction from 24 to 26 °C at 4 Torr
contained 1c (18.3 g, 61%) as a colorless oil. 1H NMR (500
MHz, CDCl3) δ 1.81 (3H, s), 2.51 (2H, d, J ) 6.5 Hz), 3.65 (3H,
s), 5.01 (1H, td, J ) 6.5, 1.0 Hz), 5.82 (1H, dd, J ) 8.5, 1.0
Hz), 6.81 (1H, d, J ) 8.5 Hz); 13C NMR (50 MHz, CDCl3) δ
20.9 (q), 32.6 (t), 54.2 (q), 113.1 (d), 118.0 (d), 135.8 (s), 136.6
(d), 153.5 (s). IR (film) νmax 1626, 1545, 1439, 1251, 1197, 1025
cm-1; UV-vis (EtOH) λmax 257 nm (log ꢀ 3.68); MS (FAB) m/z
138 (M + H)+. Anal. Calcd for C8H11NO: C, 70.04; H, 8.08; N,
10.21. Found: C, 69.58; H, 8.07; N, 10.20.
General Procedure A for Conversion of 3H-Azepines
to 2H-Azepines with Use of NBS. NBS (0-10% molar
excess) was dropped into a solution of 2-methoxy-3H-azepine
1 and an additive (200-500% molar excess) in CH2Cl2 (0.1-
0.3 M solution of 1) stirred under a nitrogen atmosphere at
the specified reaction temperature then the solution was
stirred at the specified reaction temperature for 10 h before
triethylamine or dimethylethylamine (300% molar excess) was
introduced and then the solution was then stirred at room
temperature for 5 h, unless otherwise specified. Volatile
components were then removed under reduced pressure then
the remaining material was redissolved in CH2Cl2, washed
once with water then twice with aq K2CO3, then dried over
MgSO4. Volatile components were then removed under reduced
pressure.
General Procedure B for Conversion of 3H-Azepines
to 2H-Azepines with Use of NBS. NBS (0-10% molar
excess) was dropped into a solution of 2-methoxy-3H-azepine
1 and an additive (200-500% molar excess) in CH2Cl2 (0.1-
0.3 M solution of 1) stirred under a nitrogen atmosphere at
the specified reaction temperature then the solution was
stirred at the specified reaction temperature for 10 h before
triethylamine or dimethylethylamine (300% molar excess) was
introduced with additional stirring at room temperature for 5
h. Volatile components were then removed under reduced
pressure and the soluble 2H-azepine derivatives were ex-
tracted from the remaining material with hexane.
Nucleophilic Substitution of 7a,b and 8. TCP is
best known for its function as a leaving group in peroxy-
oxalate chemiluminescent reactions.19 Similarly, as a
substituent on the azepine ring, TCP also functioned as
a good leaving group. 2-TCP-2H-azepines 7a,b were
quantitatively ipso-substituted by various nucleophiles
to give the corresponding 2H-azepines (Table 6). Among
the nucleophiles that successfully substituted the TCP
group were propylamine and propane thiol, which failed
to react efficiently under the 3H-azepine to 2H-azepine
conversion protocol. An efficient sequential one-pot syn-
thesis of 2-propylthio-2H-azepine 9b using the TCP
derivative as an intermediate was developed (Scheme 5),
which extends the range of nucleophiles that can be
substituted into the 2 position of the 2H-azepine ring.
In this process most of the TCP is recovered and can be
reused.
The kinetic properties, mechanism, and range of the
substitution reaction of, as well as thermal properties of,
7 and 8 will be examined in detail in future papers.
Substitution of 2-TCP-2H-azepines by carbon nucleo-
philes will also be investigated.
Conclusions. Improved efficiency in the conversion
of 3H-azepines to 2H-azepines was accomplished by the
use of NBS for which a variety of nucleophiles were
substituted at the 2 position. For nucleophiles not toler-
ated by NBS, indirect substitution was effected with TCP.
The alcohol, carboxylic acid, phenol, amide, amine, and
sulfide nucleophiles examined here represent general
examples of oxygen, nitrogen, and sulfur nucleophiles,
thus elucidating the flexibility of one aspect of this novel
reaction. This reaction may be general to a variety of 3H-
(19) (a) Stevani, C. V.; Silva, S. M.; Baader, W. J. Eur. J. Org. Chem.
2000, 4037-4046. (b) Rauhut, M. M. Acc. Chem. Res. 1969, 2, 80-87.
J. Org. Chem, Vol. 70, No. 9, 2005 3431

Cordonier et al.
Reaction of 1a at -41 °C. According to procedure A, NBS
(400 mg, 2.25 mmol) was reacted with 1a (250 mg, 2.03 mmol)
at -41 °C, which gave a mixture of 3a and 4a. Hexane
extraction of the residue gave 4a (79 mg, 19%) as a brown oil.
The remaining insoluble 3a (360 mg, 81%) upon recrystalli-
zation from AcOEt and hexane gave colorless prisms. Purifica-
tion of 4a performed by medium-pressure liquid chromatog-
raphy (MPLC) on ICN 32-63 silica gel at 0 °C eluted by 1:9
v/v AcOEt-hexane gave a colorless oil (70 mg, 17%).
d, J ) 9.2 Hz), 6.49 (1H, d, J ) 11.8 Hz), 6.94 (1H, d, J ) 11.8
Hz), 7.05 (1H, d, J ) 9.2 Hz); 13C NMR (75 MHz, CDCl3) δ
29.3 (q), 29.3 (q), 34.7 (s), 36.0 (s), 51.4 (t), 53.8 (q), 65.3 (d),
107.1 (d), 118.3 (d), 124.7 (d), 126.4 (d), 127.2 (d), 139.0 (d),
142.7 (s), 146.9 (s), 160.4 (s), 165.5 (s); IR (KBr) νmax 1676, 1639,
1446, 1390, 1367, 1270, 1249, 1183 cm-1; UV-vis (EtOH) λmax
264 nm (log ꢀ 3.77); MS (FAB) m/z 421 (M + 1)+. Anal. Calcd
for C21H29BrN2O2: C, 59.86; H, 6.94; N, 6.65. Found: C, 59.87;
H, 6.99; N, 6.45.
7-Methoxy-2-succinimido-2H-azepine (3a): mp 169-170
°C; 1H NMR (600 MHz, CDCl3) δ 2.78 (4H, s), 3.60 (3H, s),
5.35 (1H, dd, J ) 4.6, 1.7 Hz), 6.20 (1H, ddd, J ) 10.0, 5.3, 1.7
Hz), 6.23 (1H, dd, J ) 10.0, 4.6 Hz), 6.52 (1H, d, J ) 11.7 Hz),
6.75 (1H, dd, J ) 11.7, 5.3 Hz); 13C NMR (150 MHz, CDCl3) δ
28.2 (t), 54.1 (q), 62.7 (d), 126.2 (d), 126.3 (d), 131.4 (d), 137.6
(d), 161.6 (s), 176.8 (s); IR (KBr) νmax 1711, 1626, 1444, 1396,
1253, 1214, 1176 cm-1; UV-vis (EtOH) λmax 273 nm (log ꢀ
2.82); MS (FAB) m/z 221 (M + H)+. Anal. Calcd for C11H12N2O3:
C, 59.99; H, 5.49; N, 12.72. Found: C, 60.02; H, 5.54; N, 12.54.
4-Bromo-2-methoxy-3H-azepine (4a): colorless liquid; 1H
NMR (600 MHz, CDCl3) δ 3.10 (2H, s), 3.74 (3H, s), 5.76 (1H,
dd, J ) 8.1, 6.1 Hz), 6.50 (1H, d, J ) 6.1 Hz), 6.94 (1H, d, J )
8.1 Hz); 13C NMR (150 MHz, CDCl3) δ 43.8 (t), 55.2 (q), 104.3
(s), 113.9 (d), 128.9 (d), 137.2 (d), 149.4 (s); IR (film) νmax 1624,
1531, 1439, 1249, 1189, 1029 cm-1; UV-vis (EtOH) λmax 266
nm (log ꢀ 3.84); MS (FAB) m/z 202 (M + H)+; HRMS (FAB)
m/z found 201.9883, calcd for C7H9NOBr (M + H)+ 201.9868.
Reaction of 1a at 25 °C. According to procedure A, NBS
(400 mg, 2.25 mmol) was reacted with 1a (250 mg, 2.03 mmol)
at 25 °C, which gave a mixture of 3a and 4a. Hexane extraction
of the nonvolatile component gave 4a (280 mg, 68%) as a brown
oil. The remaining insoluble 3a (120 mg, 27%) upon recrys-
tallization from AcOEt and hexane gave colorless prisms.
Purification of 4a performed by MPLC on ICN 32-63 silica gel
at 0 °C eluted by 1:9 v/v AcOEt-hexane gave a colorless oil
(250 mg, 61%).
5-tert-Butyl-7-succinimido-3H-azepin-2-yl 4-tert-butyl-
7-methoxy-2H-azepin-2-yl ether (11): colorless plates; mp
88-91 °C; 1H NMR (600 MHz, CDCl3) δ 1.04 (9H, s), 1.16 (9H,
s), 2.70 (4H, br s), 2.98 (1H, ddd, J ) 12.5, 5.6, 1.5 Hz), 3.31
(1H, dd, J ) 12.5, 7.8 Hz), 3.57 (3H, s), 5.64 (1H, d, J ) 5.6
Hz), 5.71 (1H, dd, J ) 7.8, 5.6 Hz), 5.83 (1H, d, J ) 5.6 Hz),
6.40 (1H, d, J ) 1.5 Hz), 6.44 (1H, d, J ) 12.0 Hz), 6.89 (1H,
d, J ) 12.0 Hz); 13C NMR (50 MHz, CDCl3) δ 28.0 (t), 28.9 (t),
29.4 (q), 29.6 (q), 34.6 (s), 34.9 (s), 37.4 (t), 53.5 (q), 65.8 (d),
120.2 (d), 124.5 (d), 125.0 (d), 125.6 (d), 128.3 (s), 139.4 (d),
146.6 (s), 146.8 (s), 160.8 (s), 171.5 (s), 174.4 (s), 175.6 (s); IR
(KBr) νmax 1719, 1686, 1630, 1446, 1379, 1321, 1251, 1189 cm-1
;
UV-vis (EtOH) λmax 254 nm (log ꢀ 3.82); MS (FAB) m/z 440
(M + H)+; HRMS (FAB) m/z found 440.2549, calcd for
C25H34N3O4 (M + H)+ 440.2549. Anal. Calcd for C25H33N3O4:
C, 68.31; H, 7.57; N, 9.56. Found: C, 67.92; H, 7.40; N, 10.19.
cis-5-Bromo-4-tert-butyl-5,6-dihydro-2-succinimido-
2H-azepin-7-yl 4-tert-butyl-7-methoxy-2H-azepin-2-yl ether
(cis-12): colorless plates; mp 124-146 °C; 1H NMR (300 MHz,
CDCl3) δ 1.08 (9H, s), 1.20 (9H, s), 2.61-2.67 (2H, m), 2.68-
2.74 (2H, m), 3.21 (1H, dd, J ) 15.0, 7.1 Hz), 3.48 (3H, s), 4.01
(1H, dd, J ) 15.0, 1.3 Hz), 4.76 (1H, d, J ) 7.1 Hz), 5.53 (1H,
dd, J ) 5.8, 1.0 Hz), 5.74 (2H, s), 6.40 (1H, d, J ) 11.9 Hz),
6.74 (1H, d, J ) 5.8 Hz), 6.92 (1H, ddd, J ) 11.9, 1.3, 1.0 Hz);
13C NMR (150 MHz, CDCl3) δ 28.3 (t), 29.6 (q), 29.7 (q), 34.8
(s), 37.0 (s), 39.8 (d), 47.0 (t), 53.2 (q), 62.7 (d), 68.7 (d), 121.5
(d), 124.1 (d), 127.7 (d), 139.1 (d), 146.7 (s), 151.8 (s), 160.2
(s), 172.8 (s), 176.0 (s); IR (KBr) νmax 1702, 1669, 1634, 1452,
1328, 1265, 1201 cm-1; UV-vis (EtOH) λmax 262 nm (log ꢀ
3.41); MS (FAB) m/z 520 (M + H)+. Anal. Calcd for C25H34-
BrN3O4: C, 57.69; H, 6.58; N, 8.07. Found: C, 57.44; H, 6.62;
N, 7.89.
Reaction of 1b at -98 °C. According to procedure A, NBS
(1.00 g, 5.62 mmol) was reacted with 1b (1.00 g, 5.58 mmol)
at -98 °C, which gave 3b (1.54 g, 100%) as a slightly yellow
solid. Recrystallization of 3b from AcOEt and hexane gave
colorless prisms.
trans-5-Bromo-4-tert-butyl-5,6-dihydro-2-succinimido-
2H-azepin-7-yl 4-tert-butyl-7-methoxy-2H-azepin-2-yl ether
(trans-12): colorless plates; mp 145-146 °C; 1H NMR (500
MHz, CDCl3) δ 1.05 (9H, s), 1.17 (9H, s), 2.67 (4H, s), 3.17
(1H, dd, J ) 15.0, 7.0 Hz), 3.72 (3H, s), 4.12 (1H, dd, J ) 15.0,
0.5 Hz), 4.77 (1H, d, J ) 7.0 Hz), 5.23 (1H, d, J ) 4.5 Hz),
5.37 (1H, dd, J ) 5.0, 0.5 Hz), 5.40 (1H, d, J ) 4.5 Hz), 6.42
(1H, d, J ) 12.0 Hz), 6.91 (1H, d, J ) 12.0 Hz), 7.40 (1H, d, J
) 5.0 Hz); 13C NMR (75 MHz, CDCl3) δ 28.3 (t), 29.5 (q), 29.6
(q), 34.5 (s), 37.1 (s), 40.1 (d), 47.0 (t), 54.0 (q), 61.0 (d), 69.6
(d), 121.6 (d), 123.9 (d), 128.4 (d), 138.9 (d), 145.8 (s), 150.8
(s), 159.2 (s), 172.4 (s), 175.7 (s); IR (KBr) νmax 1713, 1655, 1450,
1342, 1323, 1263, 1199 cm-1; UV-vis (EtOH) λmax 268 nm (log
ꢀ 3.29); MS (FAB) m/z 520 (M + H)+. Anal. Calcd for C25H34-
BrN3O4: C, 57.69; H, 6.58; N, 8.07. Found: C, 57.30; H, 6.55;
N, 7.83.
Reaction of 1c at -98 °C. According to procedure A, NBS
(280 mg, 1.57 mmol) was reacted with 1c (200 mg, 1.46 mmol)
at -98 °C for 10 h, which gave a mixture of 3c and 4c.
Separation of the reaction mixture by MPLC on silica gel at 0
°C eluted by 1:1 v/v AcOEt-hexane gave 3c (250 mg, 73%)
and 4c (70 mg, 22%) as a colorless oil. Recrystallization of 3c
from AcOEt and hexane gave colorless prisms.
4-tert-Butyl-7-methoxy-2-succinimido-2H-azepine (3b):
mp 121-122 °C; 1H NMR (200 MHz, CDCl3) δ 1.06 (9H, s),
2.75 (4H, s), 3.58 (3H, s), 5.21 (1H, d, J ) 5.2 Hz), 5.95 (1H, d,
J ) 5.2 Hz), 6.47 (1H, d, J ) 12.4 Hz), 6.93 (1H, d, J ) 12.4
Hz); 13C NMR (75 MHz, CDCl3) δ 28.3 (t), 29.4 (q), 34.8 (s),
54.0 (q), 63.2 (d), 123.3 (d), 124.9 (d), 138.9 (d), 147.6 (s), 161.3
(s), 176.8 (s); IR (KBr) νmax 1705, 1644, 1444, 1404, 1251, 1210,
1187 cm-1; UV-vis (EtOH) λmax 276 nm (log ꢀ 3.18); MS (FAB)
m/z 277 (M + H)+. Anal. Calcd for C15H20N2O3: C, 65.20; H,
7.30; N, 10.14. Found: C, 65.26; H, 7.32; N, 10.02.
Temperature-Dependent Reaction of 1b. NBS (200 mg,
1.12 mmol) was reacted with 1b (200 mg, 1.12 mmol) in CH2-
Cl2 at the temperature and for the length of time specified in
Table 2. Volatile components were then removed. The crude
yields given in Table 2 were measured based on the mass of
the product mixture and component ratios determined by
integration of the 1H NMR spectra of the crude reaction
mixtures. Yields were reconfirmed by comparison to a benzene
internal standard in the 1H NMR spectra. The products were
isolated from the reaction mixtures by MPLC on ICN 32-63
silica gel at 0 °C eluted by 1:3 v/v AcOEt-hexane. Recrystal-
lization of 10 from THF and hexane gave colorless plates.
Recrystallization of 11 from hexane gave colorless plates.
Recrystallization of cis-12 from AcOEt and ether gave colorless
plates. Recrystallization of trans-12 from AcOEt and ether
gave colorless plates.
7-Methoxy-4-methyl-2-succinimido-2H-azepine (3c): col-
orless prisms; mp 134.5-135.5 °C; 1H NMR (600 MHz, CDCl3)
δ 1.87 (3H, s), 2.78 (4H, s), 3.61 (3H, s), 5.33 (1H, d, J ) 5.1
Hz), 5.96 (1H, d, J ) 5.1 Hz), 6.44 (1H, d, J ) 11.7 Hz), 6.50
(1H, d, J ) 11.7 Hz); 13C NMR (150 MHz, CDCl3) δ 20.7 (q),
28.2 (t), 54.0 (q), 62.8 (d), 124.9 (d), 127.1 (d), 135.2 (s), 141.1
(d), 161.3 (s), 176.8 (s); IR (KBr) νmax 1702, 1628, 1446, 1390,
1253, 1205, 1160 cm-1; UV-vis (EtOH) λmax 282 nm (log ꢀ
4-Bromo-5-tert-butyl-3H-azepin-2-yl 4-tert-butyl-7-meth-
oxy-2H-azepin-2-yl ether (10): colorless plates; mp 174-175
°C; 1H NMR (200 MHz, CDCl3) δ 1.06 (9H, s), 1.35 (9H, s),
3.28 (1H, d, J ) 13.0 Hz), 3.53 (1H, d, J ) 13.0 Hz), 3.66 (3H,
s), 5.27 (1H, d, J ) 5.0 Hz), 5.71 (1H, d, J ) 5.0 Hz), 6.17 (1H,
3432 J. Org. Chem., Vol. 70, No. 9, 2005

Synthesis of 2-Substituted 2H-Azepines
3.15); MS (FAB) m/z 315 (M + H)+. Anal. Calcd for C12H14N2O3:
C, 61.53; H, 6.02; N, 11.96. Found: C, 61.55; H, 6.02; N, 11.77.
4-Bromo-2-methoxy-5-methyl-3H-azepine (4c): colorless
liquid; 1H NMR (300 MHz, CDCl3) δ 1.95 (3H, s), 3.12 (2H, s),
3.74 (3H, s), 5.83 (1H, d, J ) 8.2 Hz), 6.84 (1H, d, J ) 8.2 Hz);
13C NMR (75 MHz, CDCl3) δ 22.1 (q), 43.7 (t), 55.0 (q), 102.4
(s), 117.6 (d), 133.9 (s), 136.3 (d), 151.5 (s); IR (film) νmax 1618,
1535, 1439, 1259, 1191, 1027 cm-1; UV-vis (EtOH) λmax 261
nm (log ꢀ 3.79); MS (FAB) m/z 216 (M + H)+. Anal. Calcd for
C8H10BrNO: C, 44.47; H, 4.66; N, 6.48. Found: C, 44.38; H,
4.48; N, 6.44.
Reaction of 1c at 25 °C. According to procedure A, NBS
(280 mg, 1.57 mmol) was reacted with 1c (200 mg, 1.46 mmol)
at 25 °C for 1 h, which gave 4c (305 mg, 97%) as a yellow oil.
Purification of 4c performed by MPLC on ICN 32-63 silica gel
eluted by 1:9 v/v AcOEt-hexane gave a colorless oil (290 mg,
92%).
Reaction of 1a in the Presence of N-Methylacetamide.
According to procedure A, NBS (800 mg, 4.49 mmol) was
reacted with 1a (500 mg, 4.06 mmol) in the presence of
N-methylacetamide (600 mg, 8.21 mmol) at -41 °C, which
gave a mixture of 3a, 4a, and 5a. Separation of the reaction
mixture by MPLC on alumina at 0 °C eluted by 1:1 v/v THF-
hexane gave 3a (330 mg, 37%), 4a (60 mg, 7%), and 5a (420
mg, 53%). Recrystallization of 5a from AcOEt and hexane gave
colorless prisms.
1a (250 mg, 2.03 mmol) in the presence of methanol (165 µL,
4.07 mmol) at -41 °C, which gave 2a (308 mg, 99%) as a yellow
oil. Purification of 2a performed by MPLC on ICN 32-63 silica
gel at 0 °C eluted by 1:9 v/v AcOEt-hexane gave a colorless
oil (250 mg, 80%).
2,7-Dimethoxy-2H-azepine (2a): colorless oil; 1H NMR
(600 MHz, CDCl3) δ 3.47 (3H, s), 3.68 (3H, s), 4.31 (1H, dd, J
) 3.9, 1.7 Hz), 5.96 (1H, dd, J ) 10.0, 3.9 Hz), 6.05 (1H, ddd,
J ) 10.0, 5.4, 1.7 Hz), 6.47 (1H, d, J ) 11.5 Hz), 6.72 (1H, dd,
J ) 11.5, 5.4 Hz); 13C NMR (150 MHz, CDCl3) δ 53.7 (q), 54.7
(q), 87.4 (d), 124.0 (d), 125.6 (d), 135.6 (d), 138.1 (d), 158.9 (s);
IR (film) νmax 1634, 1448, 1251, 1207, 1125 cm-1; UV-vis
(EtOH) λmax 274 nm (log ꢀ 3.28); MS (EI) m/z 153 (M+, 15).
Anal. Calcd for C8H11NO2: C, 62.73; H, 7.24; N, 9.14. Found:
C, 61.76; H, 7.23; N, 9.07. Instability of 2a results in unsat-
isfactory elemental analysis data.
Reaction of 1b in the Presence of Methanol. According
to procedure A, NBS (1.07 g, 6.01 mmol) was reacted with 1b
(1.00 g, 5.58 mmol) in the presence of methanol (1.2 mL, 29.6
mmol) at -98 °C, which gave 2b (1.17 g, 100%) as a slightly
yellow solid. Recrystallization of 2b from hexane gave colorless
prisms.
4-tert-Butyl-2,7-dimethoxy-2H-azepine (2b): colorless
prisms; mp 37-37.5 °C; 1H NMR (200 MHz, CDCl3) δ 1.10 (9H,
s), 3.48 (3H, s), 3.71 (3H, s), 4.23 (1H, d, J ) 3.8 Hz), 5.77
(1H, dd, J ) 3.8, 1.2 Hz), 6.46 (1H, d, J ) 12.4 Hz), 6.94 (1H,
dd, J ) 12.4, 1.2 Hz); 13C NMR (50 MHz, CDCl3) δ 29.5 (q),
34.5 (s), 53.6 (q), 54.8 (q), 87.9 (d), 124.5 (d), 128.1 (d), 139.2
(d), 145.1 (s), 158.6 (s); IR (KBr) νmax 1640, 1448, 1261, 1199,
1112 cm-1; UV-vis (EtOH) λmax 276 nm (log ꢀ 3.19); MS (FAB)
m/z 210 (M + H)+. Anal. Calcd for C12H19NO2: C, 68.87; H,
9.15; N, 6.69. Found: C, 68.87; H, 9.11; N, 6.56.
Reaction of 1c in the Presence of Methanol. According
to procedure A, NBS (2.85 g, 16.0 mmol) was reacted with 1c
(2.00 g, 14.6 mmol) in the presence of methanol (0.94 g, 29.3
mmol) at -98 °C for 3 h, which gave a mixture of 2c and 3c.
Hexane extraction of the residue gave 2c (1.60 g, 66%) as a
yellow solid that upon recrystallization from hexane gave
colorless prisms. The remaining insoluble 3c (1.12 g, 33%)
upon recrystallization from AcOEt and hexane gave colorless
prisms.
2-(N-Methylacetamido)-7-methoxy-2H-azepine (5a): col-
orless prisms; mp 69-70 °C.; E-5a: 1H NMR (600 MHz, CDCl3)
δ 2.02 (3H, s), 3.10 (3H, s), 3.63 (3H, s), 4.84 (1H, d, J ) 4.4
Hz), 5.90 (1H, dd, J ) 9.8, 4.4 Hz), 6.15 (1H, dd, J ) 9.8, 5.6
Hz), 6.54 (1H, d, J ) 11.5 Hz), 6.75 (1H, dd, J ) 11.5, 5.6 Hz);
13C NMR (150 MHz, CDCl3) δ 21.7 (q), 28.6 (q), 53.7 (q), 70.0
(d), 125.9 (d), 126.3 (d), 133.4 (d), 137.6 (d), 160.0 (s), 170.1
(s). Z-5a: 1H NMR (600 MHz, CDCl3) δ 2.12 (3H, s), 3.18 (3H,
s), 3.62 (3H, s), 5.75 (1H, dd, J ) 9.8, 4.4 Hz), 5.78 (1H, d, J
) 4.4 Hz), 6.12 (1H, dd, J ) 9.8, 5.6 Hz), 6.47 (1H, d, J ) 11.5
Hz), 6.70 (1H, dd, J ) 11.5, 5.6 Hz); 13C NMR (150 MHz,
CDCl3) δ 22.3 (q), 31.1 (q), 53.6 (q), 64.8 (d), 126.0 (d), 126.2
(d), 133.9 (d), 137.7 (d), 160.2 (s), 170.9 (s); IR (KBr) νmax 1653,
1630, 1446, 1404, 1265, 1241, 1226, 1201, 1052, 1011 cm-1
;
UV-vis (EtOH) λmax 276 nm (log ꢀ 3.20); MS (FAB) m/z 195
(M + H)+; HRMS (FAB) m/z found 195.1167, calcd for
C10H15N2O2 (M + H)+ 195.1134. Anal. Calcd for C10H14N2O2:
C, 61.84; H, 7.27; N, 14.42. Found: C, 61.88; H, 7.39; N, 14.36.
Reaction of 1b in the Presence of N-Methylacetamide.
According to procedure A, NBS (540 mg, 3.03 mmol) was
reacted with 1b (500 mg, 2.79 mmol) in the presence of
N-methylacetamide (410 mg, 5.61 mmol) at -98 °C, which
gave a mixture of 3b and 5b. Separation of the reaction
mixture by MPLC on alumina at 0 °C eluted by 1:1 v/v THF-
hexane gave 3b (160 mg, 21%) and 5b (540 mg, 77%).
Recrystallization of 5b from AcOEt and hexane gave colorless
prisms.
2,7-Dimethoxy-4-methyl-2H-azepine (2c): colorless prisms;
1
mp 38.5-40.5 °C; H NMR (300 MHz, CDCl3) δ 1.87 (3H, s),
3.47 (3H, s), 3.71 (3H, s), 4.30 (1H, br d, J ) 4.0 Hz), 5.71 (1H,
br d, J ) 4.0 Hz), 6.45 (1H, d, J ) 11.5 Hz), 6.63 (1H, d, J )
11.5 Hz); 13C NMR (75 MHz, CDCl3) δ 20.8 (q), 53.8 (q), 54.9
(q), 87.5 (d), 124.1 (d), 131.4 (d), 132.4 (s), 141.5 (d), 158.5 (s);
IR (KBr) νmax 1638, 1444, 1247, 1205, 1087 cm-1; UV-vis
(EtOH) λmax 280 nm (log ꢀ 3.08); MS (FAB) m/z 168 (M + H)+.
Anal. Calcd for C9H13NO2: C, 64.65; H, 7.84; N, 8.38. Found:
C, 63.87; H, 7.71; N, 8.68. Instability of 2c results in unsat-
isfactory elemental analysis data.
Reaction of 1b in the Presence of tert-Butyl Alcohol.
According to procedure A, NBS (540 mg, 3.03 mmol) was
reacted with 1b (500 mg, 2.79 mmol) in the presence of tert-
butyl alcohol (410 mg, 5.53 mmol) at -98 °C, which gave a
mixture of 3b and 6. Hexane extraction of the nonvolatile
component gave 6 (640 mg, 91%) as a slightly yellow solid.
The remaining insoluble 3b (60 mg, 8%) upon recrystallization
from AcOEt and hexane gave colorless prisms. Purification of
6 by MPLC on ICN 32-63 silica gel eluted by 1:9 v/v AcOEt-
hexane gave a white solid (470 mg, 67%). Recrystallization of
6 from hexane gave colorless plates.
2-tert-Butoxy-4-tert-butyl-7-methoxy-2H-azepine (6):
colorless prisms; mp 70.5-71.5 °C; 1H NMR (600 MHz, CDCl3)
δ 1.06 (9H, s), 1.24 (9H, s), 3.64 (3H, s), 4.47 (1H, d, J ) 4.2
Hz), 5.71 (1H, dd, J ) 4.2, 1.0 Hz), 6.44 (1H, d, J ) 12.0 Hz),
6.91 (1H, dd, J ) 12.0, 1.0 Hz); 13C NMR (150 MHz, CDCl3) δ
29.0 (q), 29.6 (q), 34.2 (s), 53.5 (q), 74.0 (s), 81.3 (d), 124.4 (d),
130.5 (d), 139.1 (d), 143.5 (s), 157.2 (s); IR (KBr) νmax 1632,
1468, 1446, 1247, 1205, 1096 cm-1; UV-vis (EtOH) λmax 278
nm (log ꢀ 3.22); MS (FAB) m/z 252 (M + 1)+. Anal. Calcd for
4-tert-Butyl-7-methoxy-2-(N-methylacetamido)-2H-
azepine (5b): colorless prisms; mp 113-116 °C. E-5b: 1H
NMR (600 MHz, CDCl3) δ 1.10 (9H, s), 2.01 (3H, s), 3.14 (3H,
s), 3.64 (3H, s), 4.74 (1H, d, J ) 4.6 Hz), 5.65 (1H, d, J ) 4.6
Hz), 6.52 (1H, d, J ) 12.0 Hz), 6.96 (1H, d, J ) 12.0 Hz); 13C
NMR (150 MHz, CDCl3) δ 21.8 (q), 28.8 (q), 29.5 (q), 34.7 (s),
53.6 (q), 70.4 (d), 125.0 (d), 125.4 (d), 138.9 (d), 147.2 (s), 159.7
(s), 170.1 (s). Z-5b: 1H NMR (600 MHz, CDCl3) δ 1.08 (9H, s),
2.13 (3H, s), 3.21 (3H, s), 3.64 (3H, s), 5.51 (1H, d, J ) 4.9
Hz), 5.70 (1H, d, J ) 4.9 Hz), 6.46 (1H, d, J ) 12.0 Hz), 6.91
(1H, d, J ) 12.0 Hz); 13C NMR (150 MHz, CDCl3) δ 22.5 (q),
29.5 (q), 31.3 (q), 34.6 (s), 53.4 (q), 65.1 (d), 124.9 (d), 125.6
(d), 138.7 (d), 147.2 (s), 160.0 (s), 170.8 (s); IR (KBr) νmax 1650,
1634, 1446, 1400, 1255, 1214, 1017, 996 cm-1; UV-vis (EtOH)
λmax 278 nm (log ꢀ ) 3.14); MS (FAB) m/z 251 (M + H)+. Anal.
Calcd for C14H22N2O2: C, 67.17; H, 8.86; N, 11.19. Found: C,
67.09; H, 9.00; N, 11.27.
Reaction of 1a in the Presence of Methanol. According
to procedure A, NBS (400 mg, 2.25 mmol) was reacted with
J. Org. Chem, Vol. 70, No. 9, 2005 3433

Cordonier et al.
C15H25NO2: C, 71.67; H, 10.02; N, 5.57. Found: C, 71.51; H,
10.03; N, 5.77.
7-Methoxy-2-propylthio-2H-azepine (9a): colorless oil;
1H NMR (600 MHz, CDCl3) δ 0.99 (3H, dd, J ) 7.6, 7.3 Hz),
1.65 (2H, m, J ) 8.1, 7.8, 7.6, 7.3, 6.8, 6.6 Hz), 2.64 (1H, ddd,
J ) 12.7, 8.1, 6.8 Hz), 2.72 (1H, ddd, J ) 12.7, 7.8, 6.6 Hz),
3.68 (3H, s), 4.54 (1H, d, J ) 5.9 Hz), 5.90 (1H, dd, J ) 9.8,
5.9 Hz), 6.13 (1H, dd, J ) 9.8, 5.4 Hz), 6.48 (1H, d, J ) 11.7
Hz), 6.70 (1H, dd, J ) 11.7, 5.4 Hz); 13C NMR (150 MHz,
CDCl3) δ 13.6 (q), 23.1 (t), 32.5 (t), 53.9 (q), 59.7 (d), 126.0 (d),
126.5 (d), 134.8 (d), 137.7 (d), 162.1 (s); IR (film) νmax 1624,
1444, 1245, 1197 cm-1; UV-vis (EtOH) λmax 279 nm (log ꢀ
3.15); MS (FAB) m/z 198 (M + H)+; HRMS (FAB) m/z found
198.0919, calcd for C10H16NOS (M + 1)+ 198.0953. Anal. Calcd
for C10H15NOS: C, 60.88; H, 7.66; N, 7.10. Found: C, 60.57;
H, 7.18; N, 7.32.
Reaction of 1a in the Presence of 2,4,6-Trichlorophe-
nol. According to procedure B, NBS (760 mg, 4.27 mmol) was
reacted with 1a (500 mg, 4.06 mmol) in the presence of 2,4,6-
trichlorophenol (1.50 g, 7.60 mmol) at -98 °C, which prior to
extraction gave 3.88 g of a brown viscous oil that was estimated
1
to contain 3a (430 mg, 48%) and 7a (610 mg, 47%) by the H
NMR spectrum. The remaining components of the oil were
triethylammonium bromide, succinimide, and TCP. The 920
mg nonvolatile component from the hexane extract was
1
estimated to contain 7a (480 mg, 37%) determined by the H
NMR spectrum.
7-Methoxy-2-(2′,4′,6′-trichlorophenoxy)-2H-azepine
(7a): 1H NMR (600 MHz, CDCl3) δ 3.53 (3H, s), 5.33 (1H, dd,
J ) 3.7, 2.0 Hz), 6.15 (1H, ddd, J ) 10.0, 5.6, 2.0 Hz), 6.30
(1H, dd, J ) 10.0, 3.7 Hz), 6.44 (1H, d, J ) 11.5 Hz), 6.74 (1H,
dd, J ) 11.5, 5.6 Hz), 7.30 (2H, s); 13C NMR (150 MHz, CDCl3)
δ 53.9 (q), 89.1 (d), 124.3 (d), 125.6 (d), 128.4 (d), 129.4 (s),
131.2 (s), 134.7 (d), 138.3 (d), 148.4 (s), 158.2 (s); MS (FAB)
m/z 317 (M)+; HRMS (FAB) m/z found 316.9774, calcd for
C13H10NO2Cl3 (M) 316.9777.
Reaction of 1b in the Presence of Propane Thiol.
According to procedure A, NBS (540 mg, 3.03 mmol) was
reacted with 1b (500 mg, 2.79 mmol) in the presence of
propane thiol (420 mg, 5.51 mmol) at -98 °C, which gave a
mixture of unreacted 1b and propyl disulfide.
Reaction of 1c in the Presence of Propane Thiol.
According to procedure A, NBS (710 mg, 3.99 mmol) was
reacted with 1c (500 mg, 3.64 mmol) in the presence of propane
thiol (550 mg, 7.22 mmol) at -98 °C, which gave a mixture of
unreacted 1c and propyl disulfide.
General Procedure for Preparation of 1,4-Adducts of
3H-Azepines. NBS (2 mmol) was dropped into a solution of 1
(2 mmol) and additive (4 mmol) in CD2Cl2 (5 mL) stirred under
a nitrogen atmosphere at the specified reaction temperature
then the solution was stirred at the specified reaction tem-
perature for 10 h. A portion of the solution was then trans-
ferred to an NMR sample tube for spectral analysis. Excess
additive and CD2Cl2 were then removed. The nonvolatile
component was then analyzed by FAB-MS and FAB-HRMS
spectroscopy.
Reaction of 1b in the Presence of 2,4,6-Trichlorophe-
nol. According to procedure B, NBS (540 mg, 3.03 mmol) was
reacted with 1b (500 mg, 2.79 mmol) in the presence of 2,4,6-
trichlorophenol (1.10 g, 5.57 mmol) at -98 °C, which gave 1.65
g of an amber oil that was estimated to contain 3b (50 mg,
1
6%) and 7b (670 mg, 64%) by the H NMR spectrum.
2-(2′,4′,6′-Trichlorophenoxy)-4-tert-butyl-7-methoxy-
1
2H-azepine (7b): H NMR (600 MHz, CDCl3) δ 1.13 (9H, s),
3.53 (3H, s), 5.20 (1H, d, J ) 4.0 Hz), 6.10 (1H, dd, J ) 4.0,
1.5 Hz), 6.41 (1H, d, J ) 12.0 Hz), 6.94 (1H, dd, J ) 12.0, 1.5
Hz), 7.29 (2H, s); 13C NMR (150 MHz, CDCl3) δ 29.4 (q), 34.7
(s), 53.7 (q), 89.7 (d), 124.5 (d), 127.1 (d), 128.4 (d), 129.3 (s),
131.2 (s), 139.5 (d), 145.4 (s), 148.5 (s), 158.1 (s); MS (FAB)
m/z 374 (M + H)+.
Preparation of 5-Bromo-5,6-dihydro-7-methoxy-2-suc-
cinimido-2H-azepine (14a). 14a was prepared from 1a at
1
-41 °C. H NMR (600 MHz, CD2Cl2) δ 2.70 (4H, s), 3.05 (1H,
ddd, J ) 13.7, 4.9, 2.0 Hz), 3.56 (1H, dd, J ) 13.7, 12.5 Hz),
3.57 (3H, s), 4.90 (1H, dddd, J ) 12.5, 4.9, 3.2, 2.7 Hz), 5.75
(1H, ddd, J ) 11.7, 2.7, 1.7 Hz), 5.90 (1H, dddd, J ) 11.7, 3.2,
2.7, 2.0 Hz), 6.45 (1H, dt, J ) 2.9, 2.7 Hz); 13C NMR (150 MHz,
CD2Cl2) δ 29.9 (t), 40.8 (t), 42.0 (d), 53.9 (q), 62.0 (d), 131.3
(d), 131.7 (d), 163.1 (s), 176.6 (s).
Reaction of 1a in the Presence of Acetic Acid. Accord-
ing to procedure B, NBS (800 mg, 4.49 mmol) was reacted with
1a (500 mg, 4.06 mmol) in the presence of acetic acid (490 mg,
8.16 mmol) at -41 °C, which gave a mixture of 3a and 8.
Hexane extraction of the nonvolatile component gave 8 (590
mg, 80%) as a brown oil. Further purification of the hexane
extract by MPLC on ICN 32-63 silica gel at 0 °C eluted by 1:4
v/v AcOEt-hexane gave 8 (380 mg, 52%) as a colorless oil.
The remaining insoluble material was redissolved in CH2Cl2,
washed once with water and twice with aq K2CO3, then dried
over MgSO4. Removal of the solvent gave 3a (160 mg, 18%) as
a brown solid that upon recrystallization from AcOEt and
hexane gave colorless prisms. Throughout the procedure, care
was taken not to expose 8 to temperatures above 20 °C.
2-Acetoxy-7-methoxy-2H-azepine (8): colorless oil; 1H
NMR (600 MHz, CDCl3) δ 2.12 (3H, s), 3.62 (3H, s), 5.66 (1H,
dd, J ) 4.2, 2.0 Hz), 5.89 (1H, dd, J ) 10.0, 4.2 Hz), 6.08 (1H,
ddd, J ) 10.0, 5.4, 2.0 Hz), 6.48 (1H, d, J ) 11.5 Hz), 6.72
(1H, dd, J ) 11.5, 5.4 Hz); 13C NMR (150 MHz, CDCl3) δ 21.3
(q), 53.9 (q), 81.5 (d), 124.6 (d), 126.1 (d), 133.6 (d), 138.0 (d),
159.2 (s), 170.2 (s); IR (film) νmax 1744, 1634, 1446, 1253, 1203,
1044 cm-1; UV-vis (EtOH) λmax 271 nm (log ꢀ 3.31); MS (FAB)
m/z 182 (M + H)+; HRMS (FAB) m/z found 182.0826, calcd
for C9H12NO3 (M + H)+ 182.0817.
Preparation of 5-Bromo-4-tert-butyl-5,6-dihydro-7-
methoxy-2-succinimido-2H-azepine (14b). 14b was pre-
pared from 1b at -98 °C. Extraction of impurities and
decomposition products from the residue with acetone gave
1
insoluble 14b as a colorless solid: mp 140-141 °C; H NMR
(500 MHz, CDCl3) δ 1.10 (9H, s), 2.76 (4H, s), 3.05 (1H, ddd,
J ) 18.6, 3.7, 2.8 Hz), 3.16 (1H, ddd, J ) 18.6, 4.0, 2.8 Hz),
3.52 (3H, s), 4.85 (1H, ddd, J ) 4.0, 3.7, 2.1 Hz), 5.84 (1H, dd,
J ) 3.4, 2.1 Hz), 7.00 (1H, dt, J ) 3.4, 2.8 Hz); 13C NMR (75
MHz, CDCl3) δ 28.0 (q), 28.3 (t), 37.1 (s), 39.7 (t), 41.9 (d), 52.7
(q), 60.5 (d), 129.2 (d), 149.1 (s), 158.5 (s), 176.2 (s); IR (KBr)
νmax 1711, 1692, 1400, 1373. 1207, 1187 cm-1; UV-vis (n-
hexane) λmax 222 nm (log ꢀ 3.41); MS (FAB) m/z 357 (M + H)+;
HRMS (FAB) m/z found 357.0853, calcd for C15H22BrN2O3 (M
+ H)+ 357.0814.
Preparation of 5-Bromo-5,6-dihydro-2,7-dimethoxy-
2H-azepine (15a). 15a was prepared from 1a in the presence
of methanol at -41 °C. 1H NMR (600 MHz, CD2Cl2) δ 3.08
(1H, dd, J ) 13.4, 3.4 Hz), 3.33 (1H, dd, J ) 13.4, 12.0 Hz),
3.36 (3H, s), 3.69 (3H, s), 4.81 (1H, ddddd, J ) 12.0, 4.9, 4.4,
3.4, 1.2 Hz), 5.44 (1H, dd, J ) 4.4, 2.2 Hz), 5.76 (1H, ddd, J )
11.8, 2.2, 1.2 Hz), 5.80 (1H, dd, J ) 11.8, 4.9 Hz); 13C NMR
(125 MHz, CD2Cl2) δ 40.8 (t), 42.6 (d), 53.4 (q), 54.5 (q), 86.4
(d), 131.1 (d), 134.6 (d), 161.2 (s); MS (FAB) m/z 234 (M + H)+;
Reaction of 1a in the Presence of Propane Thiol.
According to procedure A, NBS (800 mg, 4.49 mmol) was
reacted with 1a (500 mg, 4.06 mmol) in the presence of propane
thiol (620 mg, 8.14 mmol) at -41 °C, which gave a mixture of
unreacted 1a, 9a, and propyl disulfide. Separation of the
reaction mixture by MPLC on ICN 32-63 silica gel at 0 °C
eluted by 1:9 v/v AcOEt-hexane gave 1a (370 mg, 3.00 mmol),
9a (170 mg, 21%) as a colorless oil, and propyl disulfide (500
mg, 3.33 mmol).
HRMS (FAB) m/z found 234.0091, calcd for C
8H13NO2Br (M
+ H)+ 234.0130.
Preparation of 5-Bromo-4-tert-butyl-5,6-dihydro-2,7-
dimethoxy-2H-azepine (15b). 15b was prepared from 1b in
the presence of methanol at -98 °C. 1H NMR (500 MHz, CD2-
3434 J. Org. Chem., Vol. 70, No. 9, 2005

Synthesis of 2-Substituted 2H-Azepines
Cl2) δ 1.09 (9H, s), 3.03 (1H, ddd, J ) 18.9, 3.7, 2.7 Hz), 3.09
(1H, ddd, J ) 18.9, 4.0, 2.4 Hz), 3.48 (3H, s), 3.61 (3H, s), 4.86
(1H, ddd, J ) 4.0, 3.7, 2.1 Hz), 5.75 (1H, ddd, J ) 3.1, 2.7, 2.4
Hz), 5.86 (1H, dd, J ) 3.1, 2.1 Hz); 13C NMR (125 MHz, CD2-
Cl2) δ 28.2 (q), 37.0 (s), 40.3 (t), 43.1 (d), 52.5 (q), 54.6 (q), 84.8
(d), 133.4 (d), 147.9 (s), 162.1 (s).
Reaction of 1b with 0.5 equiv of NBS. NBS (100 mg,
0.56 mmol) was dropped into a solution of 2-methoxy-3H-
azepine 1b (200 mg, 1.12 mmol) in CH2Cl2 (10 mL) stirred
under a nitrogen atmosphere at room temperature then the
solution was stirred for 5 h at room temperature before
treatment with aq K2CO3. The organic phase was separated
and dried over MgSO4. Separation of the reaction mixture by
MPLC on Woelm 32-63 silica gel eluted with 1:4 v/v AcOEt-
hexane at 0 °C gave unreacted 1b (50 mg, 0.28 mmol) and 13
(120 mg, 63%, 84% yield based on converted 1b). Recrystalli-
zation of 13 from AcOEt and hexane gave colorless plates.
(d), 127.9 (d), 139.7 (d), 146.6 (s), 152.1 (s), 159.6 (s), 171.3 (s).
IR (KBr) νmax 1671, 1636, 1466, 1448, 1267, 1201 cm-1; UV-
vis (EtOH) λmax 265 nm (log ꢀ 3.10); MS (FAB) m/z 453 (M +
H)+; HRMS (FAB) m/z found 453.1757, calcd for C22H34BrN2O3
(M + 1)+ 453.1752.
Elimination of HBr from 16. A solution of 16 (270 mg,
0.60 mmol) in 5 mL of 1.0 M methanolic tetrabutylammonium
hydroxide was stirred at room temperature for 2.5 days
(reaction progress monitored by TLC). The mixture was then
concentrated under vacuum, diluted with water, extracted with
CH2Cl2, then dried over MgSO4. Purification by MPLC on
Woelm 32-63 silica gel eluted with 3:7 v/v AcOEt-hexane gave
a 20:47 molar ratio isomeric mixture of 17 (150 mg, 68%).
Recrystallization of the isomeric mixture of 17 from AcOEt
and hexane gave colorless plates with no change in molar ratio
of the isomers.
5-tert-Butyl-7-methoxy-3H-azepin-2-yl 4-tert-butyl-7-
methoxy-2H-azepin-2-yl ether (17): colorless plates; mp
140.5-141.0 °C. Major isomer: 1H NMR (600 MHz, CDCl3) δ
1.09 (9H, s), 1.11 (9H, s), 2.71 (1H, dd, J ) 12.1, 5.6 Hz), 3.14
(1H, dd, J ) 12.1, 7.6 Hz), 3.53 (3H, s), 3.72 (3H, s), 5.40 (1H,
dd, J ) 7.6, 5.6 Hz), 5.47 (1H, s), 5.59 (1H, d, J ) 4.4 Hz),
5.88 (1H, d, J ) 4.4 Hz), 6.35 (1H, d, J ) 12.0 Hz), 6.90 (1H,
d, J ) 12.0 Hz); 13C NMR (50 MHz, CDCl3) δ 29.6 (q), 29.6 (q),
34.4 (s), 34.7 (s), 37.4 (t), 53.3 (q), 55.8 (q), 67.8 (d), 94.0 (d),
112.9 (d), 123.8 (d), 131.2 (d), 139.1 (d), 144.7 (s), 147.4 (s),
154.2 (s), 159.4 (s), 169.1 (s). Minor isomer: 1H NMR (600
MHz, CDCl3) δ 1.06 (9H, s), 1.11 (9H, s), 2.79 (1H, dd, J )
11.7, 4.9 Hz), 3.19 (1H, dd, J ) 11.7, 7.8 Hz), 3.68 (3H, s),
3.77 (3H, s), 5.44 (1H, dd, J ) 7.8, 4.9 Hz), 5.58 (2H, s), 5.61
(1H, s), 6.47 (1H, d, J ) 12.0 Hz), 6.86 (1H, d, J ) 12.0 Hz);
13C NMR (50 MHz, CDCl3) δ 29.5 (q), 29.6 (q), 34.4 (s), 34.7
(s), 37.4 (t), 53.9 (q), 56.4 (q), 66.5 (d), 95.9 (d), 114.2 (d), 124.7
(d), 125.3 (d), 138.5 (d), 146.4 (s), 146.9 (s), 152.7 (s), 160.0 (s),
170.4 (s). IR (KBr) νmax 1673, 1642, 1462, 1439, 1412, 1367,
1255, 1220, 1151 cm-1; UV-vis (EtOH) λmax 247 nm (log ꢀ
4.00); MS (FAB) m/z 373 (M + H)+. Anal. Calcd for C22H32N2O3:
C, 70.94; H, 8.66; N, 7.52. Found: C, 71.08; H, 8.76; N, 7.70.
5-tert-Butyl-3H-azepin-2-yl 4-tert-butyl-7-methoxy-2H-
1
azepin-2-yl ether (13): colorless plates; mp 115-116 °C; H
NMR (600 MHz, CDCl3) δ 1.04 (9H, s), 1.13 (9H, s), 2.62 (1H,
dd, J ) 12.2, 6.1 Hz), 3.16 (1H, dd, J ) 12.2, 7.9 Hz), 3.67
(3H, s), 5.29 (1H, d, J ) 4.9 Hz), 5.53 (1H, dd, J ) 7.9, 6.1
Hz), 5.73 (1H, d, J ) 4.9 Hz), 6.23 (1H, d, J ) 9.3 Hz), 6.48
(1H, d, J ) 12.0 Hz), 6.93 (1H, d, J ) 12.0 Hz), 7.03 (1H, d, J
) 9.3 Hz); 13C NMR (150 MHz, CDCl3) δ 29.3 (q), 29.6 (q), 34.4
(s), 34.6 (s), 37.5 (t), 53.6 (q), 65.4 (d), 115.4 (d), 117.4 (d), 124.8
(d), 127.1 (d), 127.8 (d), 139.0 (d), 146.7 (s), 147.8 (s), 160.3
(s), 168.3 (s); IR (KBr) νmax 1678, 1630, 1446, 1394, 1367, 1267,
1241, 1216, 1181 cm-1; UV-vis (EtOH) λmax 252 nm (log ꢀ
3.90); MS (FAB) m/z 343 (M + H)+. Anal. Calcd for C21H30N2O2:
C, 73.65; H, 8.83; N, 8.18. Found: C, 73.82; H, 8.97; N, 8.47.
Reaction of 13 with NBS. NBS (110 mg, 0.62 mmol) was
dropped into a solution of 13 (200 mg, 0.58 mmol) in CH2Cl2
(5 mL) stirred under a nitrogen atmosphere at 0 °C then the
solution was stirred for 2 h. Separation of the reaction mixture
by MPLC on Woelm 32-63 silica gel at 0 °C eluted with 1:1
v/v AcOEt-hexane gave 10 (1.8 mg, 0.7%), cis-12 (203 mg,
67%), and trans-12 (63 mg, 21%).
Elimination of HBr from 12. A solution of isomeric 12
(280 mg, 0.538 mmol) and DBU (1.00 g, 6.57 mmol) in CH2Cl2
was stirred at room temperature for 2.5 days (reaction progress
monitored by TLC). Volatile components were removed under
vacuum, and isolation by column chromatography on silica gel
eluted with 1:4 v/v AcOEt-hexane gave 11 (220 mg, 93%).
Recrystallization from hexane gave 11 as colorless plates.
Reaction of 13 with NBS in the Presence of Methanol.
NBS (140 mg, 0.79 mmol) was dropped into a solution of 13
(250 mg, 0.73 mmol) and methanol (1 mL, 24.7 mmol) in CH2-
Cl2 (5 mL) stirred under a nitrogen atmosphere at 0 °C then
the solution was stirred for 2 h before treatment with aq K2-
CO3. The organic phase was separated and dried over MgSO4.
Removal of volatile components gave an isomeric mixture of
16 (330 mg, 100%) with a 4:1 molar ratio as a colorless solid.
5-Bromo-4-tert-butyl-5,6-dihydro-2-methoxy-2H-azepin-
7-yl 4-tert-butyl-7-methoxy-2H-azepin-2-yl ether (16):
colorless solid; mp 75-78 °C. Major isomer: 1H NMR (600
MHz, CDCl3) δ 1.09 (9H, s), 1.27 (9H, s), 2.85 (1H, dd, J )
13.9, 6.8 Hz), 3.50 (3H, s), 3.64 (3H, s), 3.95 (1H, dd, J ) 13.9,
1.5 Hz), 4.83 (1H, d, J ) 6.8, 1.5 Hz), 5.40 (1H, d, J ) 4.9 Hz),
5.65 (1H, d, J ) 7.1 Hz), 5.74 (1H, d, J ) 4.9 Hz), 6.10 (1H, d,
J ) 7.1 Hz), 6.47 (1H, d, J ) 12.0 Hz), 6.92 (1H, d, J ) 12.0
Hz); 13C NMR (150 MHz, CDCl3) δ 29.4 (q), 30.9 (q), 34.6 (s),
36.4 (s), 42.3 (d), 44.2 (t), 53.8 (q), 56.1 (q), 66.2 (d), 80.4 (d),
124.5 (d), 124.8 (d), 128.5 (d), 139.1 (d), 146.5 (s), 152.9 (s),
160.3 (s), 171.4 (s). Minor isomer: 1H NMR (600 MHz, CDCl3)
δ 1.04 (9H, s), 1.25 (9H, s), 2.89 (1H, dd, J ) 13.9, 6.8 Hz),
3.46 (3H, s), 3.61 (3H, s), 3.81 (1H, dd, J ) 13.9, 1.5 Hz), 4.79
(1H, d, J ) 6.8, 1.5 Hz), 5.38 (1H, d, J ) 4.9 Hz), 5.59 (1H, d,
J ) 7.1 Hz), 5.83 (1H, d, J ) 4.9 Hz), 6.06 (1H, d, J ) 7.1 Hz),
6.44 (1H, d, J ) 12.0 Hz), 6.92 (1H, d, J ) 12.0 Hz); 13C NMR
(150 MHz, CDCl3) δ 29.5 (q), 30.8 (q), 34.6 (s), 36.5 (s), 41.8
(d), 44.8 (t), 53.4 (q), 56.0 (q), 67.7 (d), 82.7 (d), 124.4 (d), 124.7
Qualitative Nucleophilic Substitution of 7a, 7b, and
8. To a CDCl3 solution of 7a, 7b, or 8 (10 mg) in an NMR
sample tube, a nucleophile was introduced in molar excess.
1
The mixture was periodically observed by H NMR spectros-
copy until no further change in the spectra could be detected.
The results are displayed in Table 6. The products were
purified by MPLC on silica gel at 0 °C.
4-tert-Butyl-7-methoxy-2-propylthio-2H-azepine (9b):
colorless liquid; 1H NMR (200 MHz, CDCl3) δ 1.00 (3H, t, J )
7.4 Hz), 1.09 (9H, s), 1.66 (2H, hext, J ) 7.4 Hz), 2.54-2.83
(2H, m), 3.70 (3H, s), 4.35 (1H, d, J ) 5.8 Hz), 5.66 (1H, d, J
) 5.8 Hz), 6.46 (1H, d, J ) 12.0 Hz), 6.93 (1H, d, J ) 12.0 Hz);
13C NMR (50 MHz, CDCl3) δ 13.6 (q), 23.0 (t), 29.4 (q), 32.3
(t), 34.6 (s), 53.6 (q), 60.2 (d), 124.6 (d), 127.0 (d), 138.9 (d),
147.5 (s), 161.4 (s); IR (film) νmax 1628, 1444, 1243, 1203 cm-1
;
UV-vis (EtOH) λmax 277 nm (log ꢀ 3.36); MS (FAB) m/z 254
(M + H)+; HRMS (FAB) m/z calcd for C14H23NOS (M)+
253.1500, found 253.1464. Anal. Calcd for C14H23NOS: C,
66.36; H, 9.15; N, 5.53. Found: C, 66.17; H, 9.24; N, 5.73.
4-tert-Butyl-7-methoxy-2-propylamino-2H-azepine
(18): colorless solid; mp 75.5-77.0 °C; 1H NMR (600 MHz,
CDCl3) δ 0.95 (3H, t, J ) 7.2 Hz), 1.07 (9H, s), 1.50-1.60 (2H,
m), 1.70 (1H, br s), 2.51-2.57 (1H, m), 2.93-2.99 (1H, m), 3.66
(3H, s), 3.75 (1H, d, J ) 4.8 Hz), 5.62 (1H, d, J ) 4.8 Hz), 6.46
(1H, d, J ) 12.0 Hz), 6.92 (1H, d, J ) 12.0 Hz); 13C NMR (150
MHz, CDCl3) δ 12.0 (q), 23.4 (t), 29.6 (q), 34.5 (s), 48.0 (t), 53.4
(q), 70.3 (d), 124.5 (d), 128.9 (d), 139.0 (d), 145.8 (s), 159.5 (s);
IR (KBr) νmax 3260, 1624, 1441, 1241 cm-1; UV-vis (EtOH)
λmax 276 nm (log ꢀ 3.07); MS (FAB) m/z 237 (M + H)+; HRMS
(FAB) m/z calcd for C14H25N2O (M + H)+ 237.1967, found
237.1989.
J. Org. Chem, Vol. 70, No. 9, 2005 3435

Cordonier et al.
Preparation of 9b. NBS (220 mg, 1.24 mmol) was dropped
into a solution of 1b (210 mg, 1.17 mmol) in the presence of
2,4,6-trichlorophenol (450 mg, 2.28 mmol) in CH2Cl2 (20 mL)
stirred under a nitrogen atmosphere at -98 °C then the
solution was stirred at -98 °C for 10 h before triethylamine
(360 mg, 3.56 mmol) was introduced and stirring was contin-
ued at room temperature for 5 h. Propane thiol (900 mg, 11.8
mmol) was then added to the reaction mixture and stirring
was continued for 6 h. Excess triethylamine, propane thiol and
CH2Cl2 were then removed under reduced pressure then the
nonvolatile component redissolved in CH2Cl2, washed once
with water then twice with aq K2CO3, then dried over MgSO4.
Volatile components were then removed under reduced pres-
sure and separation of the remaining material by column
chromatography on silica gel eluted by 1:9 v/v AcOEt-hexane
gave 9b (130 mg, 44%) as a colorless oil.
Acknowledgment. The authors thank the SC-NMR
Laboratory of Okayama University for the service of
NMR spectroscopy. This work is in part supported by
JSPS, Grant-in-aid for Scientific Research (C) 16550039.
JO0500232
3436 J. Org. Chem., Vol. 70, No. 9, 2005