Formation of 4H-azepine by the electrophilic reaction of a 2-methoxyazepinium ion and analysis of the sigmatropic isomerization

Article abstract of DOI:10.1002/ejoc.200600257

2-Aryl-2H-, 3-aryl-3H-, and 4-aryl-4H-azepine were formed by the novel, electrophilic, π_LUMO-controlled reaction of the 2-methoxyazepinium ion, generated in situ by the reaction of TiCl₄ with 2,7-dialkoxy-2H-azepine and an aryl compo

Full text of DOI:10.1002/ejoc.200600257

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200600257
Formation of 4H-Azepine by the Electrophilic Reaction of a
2-Methoxyazepinium Ion and Analysis of the Sigmatropic Isomerization
Christopher E. J. Cordonier,[a] Kyosuke Satake,*[a] Hideki Okamoto,[a] and
Masaru Kimura[a]
Keywords: Azepinium / 4H-Azepine / Density functional calculations / Heterocycles / 1,5-Hydrogen shift
2-Aryl-2H-, 3-aryl-3H-, and 4-aryl-4H-azepine were formed
by the novel, electrophilic, πLUMO-controlled reaction of the
2-methoxyazepinium ion, generated in situ by the reaction
hydrogen rearrangement of the 4H-azepine was measured.
The substitution and hydrogen shift of the azepinium ion
were analyzed with DFT studies.
of TiCl4 with 2,7-dialkoxy-2H-azepine and an aryl com- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
pound, for which the kinetic parameters of the sigmatropic
Germany, 2006)
further evidence for the character of this new family of aro-
matic cations. The reaction with hard nucleophiles has been
confirmed,[7] and the reaction with soft nucleophiles, as in
the Friedel–Crafts reaction, could also generate valuable in-
formation on the electrophilic character of these delocalized
cations. We recently reported the reaction via 2b with vari-
ous aromatic compounds, including benzene, where 2-
phenyl-2H-azepine 1e and 3-phenyl-3H-azepine 3a were
formed as shown in Scheme 1.[8] In order to attain a deeper
and more fundamental view of the azepinium ion character,
analysis of a less substituted cation free of bulky ring sub-
stituents must be conducted.
Introduction
In relation to the hydrocarbon analogue cycloheptatriene
(CHT),[1,2] reaction pathways that involve azepine com-
pounds have been a long-studied field of fundamental
chemistry, although vastly limited to theoretical studies as
experimental results have been rare. Among these pathways,
of particular interest are those that involve the aromatic
azepinium cation[3] and sigmatropic isomerization of the
azepine ring.[4,5] Recently, attention has been drawn to the
role of azepinium ions as intermediates in the reactions of
nitrenium ions with aryl compounds.[6] For the first time,
azepinium ions 2a,b have been generated in solution by the
dealkoxylation of 2H-azepines 1a,b with titanium tetrachlo-
ride (TiCl4).[7] In the same account, they were reported to
react electrophilically with water and an amine in situ to
give the corresponding 2-substituted 2H-azepines 1c,d
(Scheme 1).
Results and Discussion
Reaction of 2-isopropoxy-7-methoxy-2H-azepine (1f)
with TiCl4 in the presence of benzene gave a mixture of
7-methoxy-2-phenyl-2H-azepine (1g), 2-methoxy-3-phenyl-
3H-azepine (3b), and 7-methoxy-4-phenyl-4H-azepine (4)
(Scheme 2). This reaction presumably occurs with the initial
formation of the 2-methoxyazepinium ion (2c) followed by
electrophilic addition to benzene and then deprotonation.
The azepinium ring structure was confirmed by comparison
1
of the H- and 13C NMR signals of 2c and the previously
reported tert-butyl-substituted analogues 2a and 2b,[7] listed
in Table 1, in which chemical shifts coincided.
Scheme 1.
The minimal substitution and lack of steric restriction of
the azepinium ring in 2c offers a more fundamental exam-
ple of its reactivity and electrophilic character. Lack of a
substituent at the 5-position of the 2-methoxyazepinium
ring not only allowed for the clear experimental observation
of the πLUMO character, but also the formation of rare 4H-
azepines.[12] 2-Isopropoxy-2H-azepine 1f was prepared in
high yield from nitrobenzene by using the established meth-
ods,[13,14] and the products 1g, 3b, and 4 were easily sepa-
Investigation of the reactive properties of azepinium ions
could lead to a deeper understanding of them and provide
[a] Department of Chemistry, Faculty of Science, Okayama Uni-
versity,
3-1-1, Tsushima-Naka, Okayama, 700-8530 Japan
Fax: +81-86-251-7841
E-mail: satake@cc.okayama-u.ac.jp
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2006, 3803–3807
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3803

C. E. J. Cordonier, K. Satake, H. Okamoto, M. Kimura
SHORT COMMUNICATION
examples are rare. While accounts of the shift in 2H-azep-
ines have been reported, along with their kinetic analyses,[15]
there have been no accounts of the kinetic analysis on the
isomerization of 4H-azepines.[12a]
Isomerization of 4 apparently occurs as a two-step reac-
tion from 4 first to 7-methoxy-4-phenyl-3H-azepine (5a)
and then to 2-methoxy-5-phenyl-3H-azepine (3c). If the
isomerization occurs by a 1,5-H shift, then the step from 4
to 5a is a two-step reaction followed by one final reaction
from 5a to the most thermally stable 3c, as illustrated in
Scheme 3, where the second reaction is very fast. The time-
dependent concentration of 4, 5a, and 3c closely fit those
expected for a two-step first-order reaction system. The first
rate constant, k1, was estimated from the first-order reac-
tion equation and k3 was estimated from the best fit line of
the equation for a two-step first-order reaction system, with
an overall rate constant given by k1, to the experimental
data. The experimental data for the isomerization of 4 at
160 °C is compared to the simulation for k1 = 1.73·10–5 s–1
and k3 = 1.00·10–4 s–1 in Figure 1, for which a good corre-
lation is observed. Experimentally determined values of k1
and k3 are listed in Table 2. The kinetic parameters listed
in Table 3 for both observed steps (4 to 5a and 5a to 3c)
were estimated from the Arrhenius plots of k1 and k3.
Scheme 2.
Table 1. 13C- and 1H NMR chemical shifts observed for the ring
atoms of azepinium ions in CDCl3.[a]
Compound
13C Chemical shift [ppm]
C2
C3
C4
C5
C6
C7
2a
2b
2c
187.5
176.0
176.9
133.1
131.0
132.7
150.0
150.8
151.3
180.0
181.2
152.7
133.7
134.5
139.5
158.6
170.2
170.9
1H Chemical shift [ppm]
C3–H C4–H C5–H C6–H C7–H
2a
2b
2c
8.11
7.81
8.04
8.78
8.61
8.76
–
–
8.76
8.10
8.19
8.48
9.24
9.36
9.66
[a] Peak assignments were made on the basis of HMQC and
HMBC NMR correlations.
rated from the reaction mixture by medium-pressure liquid
chromatography on silica gel at 0 °C. The structures of all
the new compounds were confirmed by the analysis of the
3
characteristic JH,H coupling constants, from which the
connectivity of the azepine ring could be unambiguously
determined.
Scheme 3.
Compounds 1g and 4 were formed and isolated in good
The product distribution of the substitution reaction can
yields, which enabled the easy preparation of the 2H- and be best explained by an electrophilic substitution reaction
4H-azepines. Because 2H- and 4H-azepines isomerize by a controlled by the azepinium πLUMO as opposed to a ther-
[1,5] sigmatropic H shift, a characteristic feature of the modynamically controlled reaction. For analysis of this re-
azepine ring,[4] to give their more stable 3H- form, synthetic action, the azepinium πLUMO (Figure 2) and the potential
Figure 1. Isomerization of 4 at 162 °C. Points denote A = [4], B = [5a] and C = [3c] observed by 1H NMR spectroscopy. Solid lines
represent simulation of two consecutive reactions for k1 = 1.73·10–5 and k3 = 1.00·10–4 s–1.
3804
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3803–3807

Formation of 4H-Azepine and Analysis of the Sigmatropic Isomerization
Table 2. Rate constants at temperature T of the isomerization of 4.
SHORT COMMUNICATION
[a]
[b]
T [°C]
k1 [s–1]
k3 [s–1]
132
140
150
162
170
1.84·10–6
4.60·10–6
1.00·10–5
1.73·10–5
4.59·10–5
1.40·10–5
3.00·10–5
6.15·10–5
1.00·10–4
3.00·10–4
[a] k1 was determined from the linear correlation of time versus
ln[4] at T. [b] k3 was determined by fitting the two consecutive reac-
tion linear equation, given k1, to the time dependent [5a] and [3c]
1
determined by H NMR spectroscopy.
Figure 2. The B3LYP/6-31G(d,p) calculated πLUMO and eigenvec-
tors of the 2pz orbital for the ring atoms of the aromatic methoxy-
azepinium cation 2c.
Table 3. Kinetic parameters of [1,5] sigmatropic hydrogen shifts.[a]
Ea
∆H϶
∆S϶
∆G϶
Reaction
lnA
[kJ/mol]
[kJ/mol]
[J/mol]
[kJ/mol]
4 Ǟ 5a
5a Ǟ 3c
117.0
110.4
21.6
21.7
114.5
108.0
–73.3
–73.2
136.3
129.8
[a] Calculated from the Arrhenius equation, lnk = Ea/RT – lnA, for
the experimentally observed correlation of lnk and T–1
.
energies of the possible substitution products (Figure 3)
were calculated by DFT studies at the B3LYP/6-
31G(d,p)[9] level with Gaussian 03.[10] The relative stabilities
of the products decrease in the order: 3b Ͼ 1g Ͼ 4 Ͼ 5b Ͼ
6a, as was predicted, and thus the experimental results are
in contradiction with those expected for a thermodynami-
cally controlled substitution. On the other hand, the πLUMO
lobe coefficients of 2c are position-dependent and decrease
in size in the order: 7-position Ն 5-position Ͼ 3-position,
where 7-position ϾϾ 6-position and 5-position ϾϾ 4-posi-
tion, which correlates to the experimental product distribu-
tion. The πLUMO coefficient ratios and molar product ratios
were found to be close to each other, which suggests a kin-
etic control of the reaction. In a resonance theory, consider-
ation of azepinium ion 2c for the resonance structures
where the imino–ether moiety exists, a positive charge is
located at the 3-, 5-, and 7-positions (Scheme 4). Thus, a
stabilizing effect of the imino–ether form may explain the
product distribution. Similarly, the higher stability of the
Figure 3. Zero-point-corrected B3LYP/6-31G(d,p) energies (in a.u.)
and relative energies for the possible substitution products of ben-
zene by 2c.
Scheme 4.
Thermodynamic isomerization of 4 to the more stable
imino–ether-bearing products, 1g, 3b, and 4 in Figure 3 intermediate 8, as opposed to the least stable intermediate
compared with those without the imino–ether group, 5b 7, could be expected as depicted in Figure 4, but isomeriza-
and 6a, may also result from stabilization by the imino– tion via 7 is experimentally observed. The largest Ea in the
ether form. Substitution of benzene by the azepinium ion system for the transition of 8 to TS6 (170.1 kJ/mol) could
illustrates the relative increase in electrophilicity of the ion present a larger barrier to this pathway, although from 8,
compared to the hydrocarbon analogue, tropylium bromide, barrierless enamine–imine isomerization directly to the 3H-
which did not react.[11]
azepine 3c may be anticipated. Preference for the reaction
Again to confirm the experimental findings, the potential via 7 can be considered a result of the difference in Ea in the
energy surface of the [1,5] sigmatropic H-shift from 4 was kinetically controlled reaction from 4 to TS1 as opposed to
calculated at the B3LYP/6-31G(d,p)[9] level, shown in Fig- TS7 (164.4 kJ/mol). This difference can be elegantly ex-
ure 4, for which a correlation with the experimental data plained in terms of the frontier molecular orbital theory,[16]
was found. Transition states were all found to be the ex- by the MO interaction between the σLUMO of the C–H
pected 6-membered 4π+2σ electron cyclic states. The acti- bond and the πHOMO of the reacting diene moiety. The
vation energy, Ea, of 4 to TS1 was calculated to be 148.8 kJ/ πHOMO energies of 2-methoxy-1-azadiene and 1-methoxy-2-
mol, of 7 to TS2 97.7 kJ/mol, and of 5a to TS3 139.7 kJ/ azadiene are 10.163 eV and –9.365 eV, respectively, as calcu-
mol; therefore, as observed experimentally, k2 is expected to lated byAM1.[17] The reacting diene moiety in the transition
be very large in comparison to k1 and k3, and k3 is expected from 4 to TS1 is thus the more reactive 1-methoxy-2-azadi-
to be larger than, but comparable to, k1.
ene.
Eur. J. Org. Chem. 2006, 3803–3807
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3805

C. E. J. Cordonier, K. Satake, H. Okamoto, M. Kimura
SHORT COMMUNICATION
Figure 4. Isomerization pathway of 7-methoxy-4-phenyl-4H-azepine (4), calculated at the B3LYP/6-31G(d,p) level of theory.
K2CO3 (8.3 g, 60 mmol). Salts formed were removed by vacuum
filtration. The aqueous phase was then separated from the organic
phase and washed three times with CH2Cl2. The organic liquors
were dried with MgSO4, then the volatile components were re-
moved. The phenyl-substituted compounds were separated from
the remaining nonvolatile component by MPLC on ICN 32–63 sil-
ica gel at 0 °C eluted by 1:9 v/v AcOEt/hexane. The following were
isolated: 970 mg (49%) of 7-methoxy-2-phenyl-2H-azepine (1g) as
a colorless oil [1H NMR (600 MHz, CDCl3): δ = 3.82 (s, 3 H), 4.29
(d, J = 5.4 Hz, 1 H), 5.95 (dd, J = 9.8, 5.4 Hz, 1 H), 6.24 (dd, J =
9.8, 5.4 Hz, 1 H), 6.66 (d, J = 11.5 Hz, 1 H), 6.86 (dd, J = 11.5,
5.4 Hz, 1 H), 7.33 (d, J = 7.6 Hz, 1 H), 7.44 (dd, J = 7.6, 7.1 Hz,
2 H), 7.67 (d, J = 7.1 Hz, 2 H) ppm. 13C NMR (150 MHz, CDCl3):
δ = 53.7 (q), 58.1 (d), 125.9 (d), 126.2 (d), 126.68 (d), 126.70 (d),
128.3 (d), 137.6 (d), 137.9 (d), 144.3 (s), 161.7 (s) ppm], 56 mg (3%)
of 2-methoxy-3-phenyl-3H-azepine (3b) as a colorless oil [1H NMR
(600 MHz, CDCl3): δ = 3.31 (d, J = 6.4 Hz, 1 H), 3.68 (s, 3 H),
5.57 (dd, J = 9.0, 6.4 Hz, 1 H), 6.06 (dd, J = 8.1, 5.9 Hz, 1 H), 6.36
(dd, J = 9.0, 5.9 Hz, 1 H), 7.03 (d, J = 8.1 Hz, 1 H), 7.28–7.32 (m,
3 H), 7.36 (dd, J = 7.8, 7.6 Hz, 2 H) ppm. 13C NMR (150 MHz,
CDCl3): δ = 50.2 (d), 55.3 (q), 114.6 (d), 121.4 (d), 126.6 (d), 127.0
(d), 128.2 (d), 129.0 (d), 137.0 (d), 137.2 (s), 150.1 (s) ppm], and
854 mg (43%) of 7-methoxy-4-phenyl-4H-azepine (4) as a colorless
oil [1H NMR (600 MHz, CDCl3): δ = 3.49–3.52 (m, 1 H), 3.81 (s,
3 H), 5.10 (ddd, J = 7.1, 5.1, 1.0 Hz, 1 H), 5.90 (dd, J = 10.0,
1.7 Hz, 1 H), 6.13 (ddd, J = 10.0, 5.9, 1.0 Hz, 1 H), 6.66 (dd, J =
7.1, 1.7 Hz, 1 H), 7.29 (t, J = 7.3 Hz, 1 H), 7.32 (dd, J = 7.8,
7.3 Hz, 2 H), 7.37 (d, J = 7.8 Hz, 2 H) ppm. 13C NMR (150 MHz,
CDCl3): δ = 42.8 (d), 52.4 (q), 116.9 (d), 117.3 (d), 126.8 (d), 127.4
(d), 128.7 (d), 134.2 (d), 142.5 (d), 142.5 (s), 163.7 (s) ppm].
As displayed in Figure 4, the most stable transition state
is that in which the imino–ether moiety is intact, TS4. The
higher Ea requirement in the isomerizations of 4 to 5a and
5a to 3c in Table 3 and Figure 4, as opposed to that of 1h
to 3c in Figure 4, may also be a consequence of the disrup-
tion of the imino–ether conjugation. Azepinium substitu-
tion and 1,5-H isomerization of the azepine ring present
elegant examples of the correlation between experimental
data and DFT predictions.
Conclusions
The simplest form of azepinium so far, 2c, was synthe-
sized and the analysis of the electrophilic reactivity gave an
elegant example of the character of the πLUMO, for which
experimental data and DFT predictions coincide. The novel
4H-azepine 4 was one of the major products of the reaction
of 2c with benzene, providing a new synthetic route to this
rare class of compounds.
The kinetic analysis of the 1,5-hydrogen shift of 4, which
was the first reported for a 4H-azepine, was conducted and
correlated well with the model estimated by the B3LYP
theory with the 6-31G(d,p) basis set.
Experimental Section
Reaction of 1f with TiCl4 in the Presence of Benzene: TiCl4 (5.69 g,
30.0 mmol) was added to a solution of 1f (1.81 g, 10.0 mmol) in
benzene (40 mL) cooled in an ice bath. The reaction mixture was
then stirred at room temp. for 15 h before quenching with aq.
Supporting Information (see footnote on the first page of this arti-
cle): Experimental procedures for, characterization and spectral as-
signments of compounds 1f, 1g, 2c, 3b, 3c, 4, and 5a. Kinetic analy-
sis of the 1,5-hydrogen shift of 4, including calculations used.
3806
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 3803–3807

Formation of 4H-Azepine and Analysis of the Sigmatropic Isomerization
SHORT COMMUNICATION
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratch-
ian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Sal-
vador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Ba-
boul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Lia-
shenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W.
Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Pittsburgh PA,
2003.
D. Bryce-Smith, N. A. Perkins, J. Chem. Soc. 1962, 5295–5297.
a) U. Goeckel, U. Hartmannsgruber, A. Steigel, J. Sauer, Tetra-
hedron Lett. 1980, 21, 595–598; b) W. Steglich, H. Bauer, M.
Grosse-Bley, R. Jeschke, J. Josten, J. Klein, J. Heterocycl. Chem.
1990, 27, 107–110.
C. E. J. Cordonier, K. Satake, M. Atarashi, Y. Kawamoto, H.
Okamoto, M. Kimura, J. Org. Chem. 2005, 70, 3425–3436.
M. Masaki, K. Fukui, J. Kita, Bull. Chem. Soc. Jpn. 1977, 50,
2013–2016.
Acknowledgments
The authors thank the SC-NMR Laboratory of Okayama Univer-
sity for the service of NMR spectroscopy. This work is in part
supported by JSPS, Grant-in-aid for Scientific Research (C)
16550039.
[1] W. Von E. Doering, L. H. Knox, J. Am. Chem. Soc. 1954, 76,
3203–3206.
[2] T. Okajima, K. Imafuku, J. Org. Chem. 2002, 67, 625–632.
[3] D. A. Smith, J. Bitar, J. Org. Chem. 1993, 58, 6–8.
[4] R. K. Smalley, Comprehensive Heterocyclic Chemistry, Perga-
mon Press, Oxford, 1984, vol. 7, p. 481.
[11]
[12]
[5] a) K. Satake, R. Okuda, M. Hashimoto, Y. Fujiwara, H. Oka-
moto, M. Kimura, S. Morosawa, J. Chem. Soc., Perkin Trans.
1 1994, 1753–1757; b) K. Satake, R. Okuda, M. Hashimoto, Y.
Fujiwara, I. Watadani, H. Okamoto, M. Kimura, S. Moros-
awa, J. Chem. Soc., Chem. Commun. 1991, 1154–1156.
[13]
[6] M. Di Stefano, M. Rosi, A. Sgamellotti, F. Negri, Chem. Phys.
2004, 302, 295–308.
[7] K. Satake, Y. Kubota, C. E. J. Cordonier, H. Okamoto, M. Ki-
mura, Angew. Chem. Int. Ed. 2004, 43, 736–738.
[14]
[8] Y. Kubota, K. Satake, H. Okamoto, M. Kimura, Org. Lett.
2005, 7, 5215–5218.
[9] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) M. J.
Frisch, J. A. Pople, J. S. Binkley, J. Chem. Phys. 1984, 80, 3265–
3269.
[10] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tom-
asi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
[15] a) D. Hamprecht, J. Josten, W. Steglich, Tetrahedron 1996, 52,
10883–10902; b) D. W. Jones, M. Thornton-Pett, J. Chem. Soc.,
Perkin Trans. 1 1995, 809–815.
[16] a) K. Fukui, Science 1982, 218, 747–754; b) K. Fukui, Angew.
Chem. 1982, 94, 852–861; c) K. Fukui, Acc. Chem. Res. 1971,
4, 57–64.
[17] M. J. S. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stewart, J.
Am. Chem. Soc. 1985, 107, 3902–3909.
Received: March 23, 2006
Published Online: July 25, 2006
Eur. J. Org. Chem. 2006, 3803–3807
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3807