The intramolecular aromatic electrophilic substitution of aminocyclopropanes prepared by the Kulinkovich-de Meijere reaction

Article abstract of DOI:10.1002/ejoc.200500428

This article describes new examples of intramolecular Kulinkovich-de Meijere reactions applied to carboxylic amides bearing an olefin moiety and an aromatic ring at a suitable position. Upon heating, the aminocyclopropanes thus obtained undergo intramolec

Full text of DOI:10.1002/ejoc.200500428

FULL PAPER
The Intramolecular Aromatic Electrophilic Substitution of Aminocyclopropanes
Prepared by the Kulinkovich–de Meijere Reaction
Laurent Larquetoux,[a] Nouara Ouhamou,[a] Angèle Chiaroni,[a] and Yvan Six*[a]
Dedicated to Professor Samir Z. Zard on the occasion of his 50th birthday
Keywords: Titanium / Aromatic substitution / Cyclisation / Amides / Cyclopropanes
This article describes new examples of intramolecular Kulin-
kovich–de Meijere reactions applied to carboxylic amides
bearing an olefin moiety and an aromatic ring at a suitable
position. Upon heating, the aminocyclopropanes thus ob-
tained undergo intramolecular aromatic electrophilic substi-
tution to afford polycyclic systems. Among the various start-
ing materials prepared, best results are obtained from indole
and phenol derivatives. In each case, a benzylic quaternary
centre is introduced at the newly-formed ring junction. On
one example, the efficiency of the cyclisation has been dra-
matically improved using a catalytic amount of para-tolu-
enesulfonic acid.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
Introduction
The intramolecular version of the de Meijere variation[1]
of the Kulinkovich reaction[2,3] provides a straightforward
access to bicyclic aminocyclopropanes such as 1.[4,5] We
have recently described the transformation of these interest-
ing compounds into vinylogous amides 5 and 6 under heat-
ing in the presence of acetic anhydride (Scheme 1).[6] The
mechanism putatively involves the enamine intermediates 2
and 4, in equilibrium with the corresponding iminium ion
3.
We envisioned that the intramolecular trapping of cat-
ionic species like 3 with nucleophilic moieties would consti-
tute an attractive extension of our method. We became par-
ticularly interested in the possibility of carrying out intra-
molecular aromatic electrophilic substitution reactions.
This could provide an access to 1,2,3,4-tetrahydroisoquino-
line and 1,2,3,4-tetrahydro-2-carboline skeletons, as an
hopefully complementary alternative to the Pictet–Speng-
ler,[7] Bischler–Napieralski[7d,8] and Polonovski–Potier[7b,9]
reactions.
Scheme 1.
Results and Discussion
Synthesis of the Aminocyclopropane Cyclisation Precursors
[a] Institut de Chimie des Substances Naturelles, U.P.R. 2301 du
C.N.R.S.,
Several aminocyclopropanes bearing electron-rich aro-
matic groups were prepared (Figure 1). We have already re-
ported the synthesis of the indole derivatives 7 and 8 in
three steps from tryptamine.[6] As cyclopropane formation
by intramolecular Kulinkovich–de Meijere reaction per-
Avenue de la Terrasse, 91198 Gif-sur-Yvette, France
Fax: +33-1-69077247
E-mail: Yvan.Six@icsn.cnrs-gif.fr
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
4654
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200500428
Eur. J. Org. Chem. 2005, 4654–4662

Intramolecular Aromatic Electrophilic Substitution of Aminocyclopropanes
FULL PAPER
formed very well in these cases, the same method was ap-
plied to make 9, 10, and 11 (Scheme 2). Good results were
obtained, except in the case of 9, where the six-membered
ring formation proceeded in 37% yield only.[10] Cleavage of
the oxygen–methyl bond of 11 by tribromoborane granted
access to the free phenol derivative 12.
Scheme 3.
Figure 1. Aminocyclopropanes 7–14.
Scheme 4.
Cyclisation Studies
With the cyclisation precursors 7–14 in hand, the feasi-
bility of the intramolecular aromatic substitution could be
investigated. Indeed, upon heating in chlorobenzene at re-
flux, 7, 8, 9, and 12 cyclised in low to moderate yield to
the corresponding 1,2,3,4-tetrahydro-2-carboline or 1,2,3,4-
tetrahydroisoquinoline structures 17–20 (Scheme 5). In con-
trast to 12, the methoxy-substituted derivatives 10 and 11
Scheme 2.
Preparing the pyrrole compound 13 in an efficient fash-
ion proved more difficult. This compound was isolated in
only 10% yield when our standard conditions were used
from the corresponding acetamide. In order to improve this
result, we tried to pre-form the intermediate (η2-cyclopen-
tene)diisopropyloxytitanium complex[11] before adding the
acetamide. The reaction was complex. The isolated by-pro-
ducts 15 and 16 indicate that competitive processes become
important at low temperature (Scheme 3). Other examples
involving these types of reaction pathways can be found in
the literature.[12]
Protection of the nitrogen of the pyrrole ring with a tri-
isopropylsilyl group proved beneficial, since the corre-
sponding aminocyclopropane 14 was isolated in 44% yield.
On the basis of this result, we next developed a one-pot
experimental procedure to synthesise 13 efficiently, with the
pyrrole ring being transiently protected by a trimethylsilyl
group (Scheme 4).
Scheme 5.
Eur. J. Org. Chem. 2005, 4654–4662
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4655

L. Larquetoux, N. Ouhamou, A. Chiaroni, Y. Six
FULL PAPER
did not yield any of the expected adducts. This is not sur- NMR analysis of the crude product showed an increase in
prising, since phenols are known to be better partners than the amount of diketone 21, whereas enamine 23 was no
their O-methyl analogues in aromatic electrophilic substitu- longer observed. Moreover, when the reaction was repeated
tion processes such as the Pictet–Spengler reaction.[7a] The for 24 h with the reflux condenser deliberately open to the
regioselectivity of the cyclisation of 12 was good, 87:13 in air, the amounts of isolated ketone 21 and aminal 22 in-
favour of the regioisomer 20. The pyrrole derivatives 13 and creased to 38% and 13%, respectively, with again no en-
14 decomposed under the same conditions, leading to com- amine 23 being detected by NMR spectroscopy of the crude
plex mixtures of products.
product.
These facts suggest that the enamine–iminium equilib-
rium lies more towards the enamine than in the case of the
five-membered analogue 7.[18] In order to favour the imin-
ium reactive intermediate, 0.1 equivalent of para-toluenesul-
fonic acid was added, which indeed increased the yield of
the desired tetrahydrocarboline 19 to 68% (Scheme 6).
Further Studies on the Cyclisation of the
Aminocyclopropane 9
As this was a typical case where the cyclisation per-
formed poorly, a more detailed study of the reaction of 9
was undertaken. Beside the expected adduct 19, two side-
products were isolated: Ketone 21 (5%) and the pentacyclic
compound 22 (2%), whose relative stereochemistry was de-
termined by X-ray diffraction through a single crystal (Fig-
ure 2 and Figure 3). Only 6% of the starting material was
recovered. NMR analysis of the crude product revealed that
the major compound produced during the reaction had ac-
tually not been isolated after purification by chromatog-
raphy over silica gel, and was tentatively assigned as the
enamine structure 23.[13]
Scheme 6.
Stereochemical and Mechanistic Issues
In order to assign the relative configurations of both dia-
stereoisomers of tetrahydrocarboline 17 (Scheme 5), one of
them was mixed with 2,4,6-trinitrophenol to form the corre-
sponding salt, which was then re-crystallised in acetonitrile.
X-ray diffraction through a single crystal revealed that the
compound is the cis diastereoisomer, “cis” referring to the
relative configurations of the methyl groups (Figure 4). The
proposed structures of both diastereoisomers are also in
agreement with NOESY 2D NMR experiments and mol-
ecular calculations using the AM1 method,[19] predicting a
cis ring junction in both cases (Figure 5).
Figure 2. ORTEP drawing of 22. Displacement ellipsoids are shown
at the 30% probability level.
Figure 3. Ketone 21 and putative intermediates 23 and 24.
The formation of 21 and 22 could be explained by an
oxidation of the enamine 23 by traces of oxygen contained
in the argon we used,[14] possibly as a dioxacyclobutane 24
(Figure 3),[15] followed by subsequent transformations.[16]
There is literature precedent for such processes.[17] This is
supported by the fact that when the reaction was main-
Figure 4. ORTEP drawing of cis-17•picric acid. Displacement ellip-
soids are shown at the 30% probability level.
In contrast to 17, the homologous compound 18 was iso-
tained at reflux under argon for seven days instead of 24 h, lated as a single diastereoisomer. The relative configuration
4656
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4654–4662

Intramolecular Aromatic Electrophilic Substitution of Aminocyclopropanes
FULL PAPER
Figure 7. Observed NOE effects on cis- and trans-19.
Figure 5. Observed NOE effects on cis- and trans-17.
is trans, as shown again by X-ray diffraction through a sin-
gle crystal of the corresponding 2,4,6-trinitrophenol salt
(Figure 6).
Figure 8. Observed NOE effects on cis-20.
trophilic substitution onto an iminium ion.[21] The elemen-
tary steps leading to this iminium ion from the aminocyclo-
propane still need to be established. This reactive species
could arise from the direct protonation of the cyclopropane
or from the prior isomerisation of the aminocyclopropane
to an enamine, followed by protonation.[22] In any case, an
equilibrium between the iminium and enamine species is
expected by loss of proton/reprotonation if the cyclisation
Figure 6. ORTEP drawing of 18•picric acid. Displacement ellip-
soids are shown at the 30% probability level.
In the case of 19, molecular calculations using the AM1 step is slow enough. It has been shown in the case of the
method[19] predicted a more stable conformation featuring six-membered aminocyclopropane 9 that the addition of a
a cis ring junction for the cis diastereoisomer and a trans small amount of acid drives this equilibrium towards the
ring junction for the trans diastereoisomer. In the most po- iminium ion and speeds up the reaction. In the case of the
lar diastereoisomer, the high coupling constant (3JH,H
=
phenol derivative 12, the phenol function might play the
9 Hz) measured at the signal of the proton α to one of the same role, which would explain why this compound reacts
methyl groups is typical for an axial conformation for this significantly faster than the corresponding indole derivative
proton, and consequently an equatorial conformation for 7.[23]
the methyl group, consistent with the proposed structure of
As far as diastereoselectivity is concerned, the results
cis-19. NOESY two-dimensional NMR experiments run on show some discrepancy, the reaction appearing sometimes
both diastereoisomers also fully support this assignment highly, and sometimes rather poorly diastereoselective.
and the calculated conformations (Figure 7).
Moreover, different diastereoselectivities have been ob-
The tetrahydroisoquinoline 20 was isolated as a single served from the same substrate depending on the reaction
diastereoisomer. Although we failed to obtain crystals that conditions (see for instance the case of product 19 in
would be suitable for X-ray diffraction analysis, the cis rela- Scheme 6). In all cases, the cis diastereoisomers are ex-
tive configuration is tentatively assigned. The 1H NMR sig- pected to be the kinetic products. The varying diastereo-
nals of the protons borne by the saturated five-membered selectivities could be explained if the cyclisation process was
ring are very similar to those of compound cis-17,[20] and reversible and if the diastereoisomers formed could thus be
the proposed structure is in agreement with the observed inter-converted. Indeed, related reversible cyclisations are
NOE effects shown in Figure 8.
documented.[24] Further studies are nonetheless needed to
From a mechanistic point of view, all the results are con- get a better understanding and control over these diastereo-
sistent with a cyclisation by intramolecular aromatic elec- selectivities.
Eur. J. Org. Chem. 2005, 4654–4662
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4657

L. Larquetoux, N. Ouhamou, A. Chiaroni, Y. Six
FULL PAPER
100%, then methanol/ethyl acetate, gradient from 1% to 4%)
yielded pure 9 (0.29 g, 1.1 mmol, 37%), starting amide (0.12 g,
0.43 mmol, 14%), and a reductive dimerisation product[31] (0.17 g,
0.32 mmol, 21%). Analytically pure aminocyclopropane 9 (0.24 g,
0.95 mmol, 32%) was obtained by recrystallisation (ethyl acetate
Concluding Remarks
Various aminocyclopropanes were used as iminium pre-
cursors and cyclised upon heating onto electron-rich aro-
matic groups by intramolecular Friedel–Crafts alkylation.
The new polycyclic structures 17–20 thus obtained feature a
quaternary carbon at the ring-junction next to the aromatic
moiety. This is an interesting feature because creating a
quaternary benzylic centre is not easy to achieve using stan-
dard methods. Indeed, indirect processes were developed,
such as adding an excess of methyllithium to the cyclic im-
inium ion obtained at the end of the Bischler–Napieralski
process.[25] In the Pictet–Spengler reaction, some ketones
may actually be used instead of aldehydes,[26] but this is
essentially limited to a few types of compounds like pyruvic
acid derivatives.[27]
and
a few drops of dichloromethane). Colourless crystals.
C17H22N2 (254.4): calcd. C 80.27, H 8.72; found C 80.27, H 8.72.
M.p. 128.1–128.9 °C. MS (ES+): m/z = 293 [MK+], 294. HRMS
(ES+): calcd. for C H N [MH+] 255.1861; found 255.1850. IR ν
˜
17 23
2
= 3415, 2926, 2856, 1454, 1353, 1010 cm–1. 1H NMR δ = 0.35 (dd,
J = 9, 5 Hz, 1 H), 0.50 (t, J = 5 Hz, 1 H), 0.90 (m, 1 H), 1.25 (s, 3
H), 1.29–1.40 (m, 2 H), 1.57 (dddd, J = 13, 7, 6, 2 Hz, 1 H), 1.96
(dq, J = 13, 7 Hz, 1 H), 2.33 (m, 1 H), 2.66 (m, 1 H), 2.83 (m, 1
H), 2.91–3.10 (m, 3 H), 6.99 (d, J = 2 Hz, 1 H), 7.08–7.21 (m, 2
H), 7.31 (d, J = 8 Hz, 1 H), 7.63 (d, J = 7 Hz, 1 H), 8.18 (br. s, 1
H) ppm. 13C NMR δ = 14.0, 19.5, 19.7, 23.1, 24.6, 26.4, 38.3, 47.9,
54.7, 111.1, 114.9, 118.9, 119.1, 121.5, 121.8, 127.6, 136.2 ppm.
Using our method, it should be noted that the structural
diversity of the products synthesised so far is quite restric-
ted. All of them except one feature a ring-junction methyl
group and another methyl group in the β position. We are
currently devoting our efforts towards the extension to a
larger variety of functional groups. The results will be re-
ported in due course.
Aminocyclopropane 10:[29] Titanium isopropoxide (1.5 equiv.,
1.2 mmol, 0.36 mL) was added to a solution of N-but-3-enyl-N-[2-
(3,4,5-trimethoxyphenyl)ethyl]acetamide[30] (1.0 equiv., 0.81 mmol,
0.25 g) in THF (15 mL). Cyclopentylmagnesium chloride (1.9  in
Et2O, 4.5 equiv., 3.7 mmol, 1.9 mL) was then added dropwise over
10 minutes. After 10 minutes of stirring at 20 °C, the mixture was
diluted in diethyl ether (0.10 L) and water (0.10 L). The organic
layer was separated, and the aqueous extracted with diethyl ether
(2×0.10 L). The combined organic phases were dried with sodium
sulfate, filtered, and concentrated to afford a black oil (0.26 g). Pu-
rification by flash column chromatography (ethyl acetate/heptane,
gradient from 50% to 100%) yielded pure 10 (0.17 g, 0.58 mmol,
Experimental Section
General Remarks: NMR spectra were recorded with AC 250 (1H
at 250 MHz, 13C at 62.9 MHz), AM 300, AVANCE 300 (1H at
300 MHz, 13C at 75.5 MHz) and AMX 400 (1H at 400 MHz)
Bruker spectrometers. Chemical shifts are given in ppm, referenced
to the peak of tetramethylsilane, defined at δ = 0.00 ppm (1H
NMR), or the solvent peak of CDCl3, defined at δ = 77.1 ppm
(13C NMR). Infrared spectra were recorded with a Perkin–Elmer
BX FT-IR spectrometer. Mass spectra were obtained using HP MS
5972 (CI), Thermofinigan Automass (EI), LC/MS Thermoquest
Navigator (ES+) and LCT Micromass (low- and high-resolution
ES+) spectrometers. Melting points were determined using a Büchi
BS540 apparatus and were not corrected. Unless otherwise stated,
flash column chromatography was performed on SDS Chromagel
silica gel 60 (35–70 µm). All reactions were carried out under argon
unless otherwise stated. The temperatures mentioned are the tem-
peratures of the cold baths or the oil baths used. Analytical grade
dichloromethane and diethyl ether were purchased from SDS and
used as such. Chlorobenzene was distilled before use. THF was
distilled from sodium/benzophenone under argon. Cyclopen-
tylmagnesium chloride solution in diethyl ether was purchased
from Sigma–Aldrich or Fluka and titrated according to a pre-
viously reported method.[11] n-Butyllithium solution in hexane was
also titrated prior to use.[28]
72%). Yellow oil. IR: ν = 2937, 1589, 1508, 1458, 1420, 1335, 1239,
˜
1
1129, 1010 cm–1. H NMR: δ = 0.11 (dd, J = 8, 5 Hz, 1 H), 0.77
(t, J = 5 Hz, 1 H), 1.16 (dt, J = 8, 5 Hz, 1 H), 1.37 (s, 3 H), 1.79
(m, 1 H), 1.89–2.10 (m, 2 H), 2.29 (dt, J = 12, 8 Hz, 1 H), 2.74 (t,
J = 8 Hz, 2 H), 3.03–3.21 (m, 2 H), 3.82 (s, 3 H), 3.86 (s, 6 H),
6.44 (s, 2 H) ppm. 13C NMR: δ = 7.8, 19.6, 22.4, 26.0, 35.8, 46.2,
49.8, 53.8, 56.0, 60.7, 105.4, 136.1, 136.4, 153.0 ppm.
Aminocyclopropane 11: A similar procedure as for the preparation
of 10, starting from N-(but-3-enyl)-N-[2-(3-methoxyphenyl)ethyl]-
acetamide[30] (2.1 mmol, 0.53 g), yielded pure 11 (0.43 g, 1.9 mmol,
87%). Pale orange oil. MS (CI, NH3): m/z = 232 [MH+], 233, 234.
IR: ν = 2946, 2865, 2832, 2805, 1610, 1602, 1594, 1584, 1488, 1453,
˜
1
1437, 1259, 1165, 1152, 1054 cm–1. H NMR: δ = 0.09 (dd, J = 8,
5 Hz, 1 H), 0.75 (t, J = 5 Hz, 1 H), 1.14 (dt, J = 8, 4 Hz, 1 H),
1.35 (s, 3 H), 1.77 (m, 1 H), 1.88–2.07 (m, 2 H), 2.29 (ddd, J = 12,
9, 8 Hz, 1 H), 2.70 (t, J = 8 Hz, 2 H), 3.09 (ddd, J = 12, 9, 8 Hz,
1 H), 3.16 (t, J = 8 Hz, 1 H), 3.80 (s, 3 H), 6.71–6.81 (m, 3 H),
7.20 (t, J = 8 Hz, 1 H) ppm. 13C NMR: δ = 7.9, 19.8, 22.5, 26.2,
35.7, 46.3, 49.9, 53.9, 55.2, 111.2, 114.5, 121.1, 129.3, 142.4,
159.7 ppm.
Aminocyclopropane 12: Tribromoborane (1.0  in dichloromethane,
2.0 equiv., 1.4 mmol, 1.4 mL) was added dropwise at 0 °C to a solu-
Aminocyclopropane 9:[29] Titanium isopropoxide (1.5 equiv.,
4.5 mmol, 1.3 mL) was added to a solution of N-[2-(1H-indol-2-yl)- tion of the aminocyclopropane 11 (1.0 equiv., 0.69 mmol, 0.16 g) in
ethyl]-N-(pent-4-enyl)acetamide[30] (1.0 equiv., 3.0 mmol, 0.81 g) in
THF (30 mL). Cyclopentylmagnesium chloride (1.9  in Et2O,
4.0 equiv., 12 mmol, 6.3 mL) was then added dropwise over 30 min-
utes. After 20 minutes of stirring at 20 °C, the mixture was diluted
in dichloromethane (0.10 L) and water (0.10 L). The organic layer
was separated, and the aqueous extracted with dichloromethane
(2×0.10 L). The combined organic phases were dried with sodium
sulfate, filtered and concentrated to afford a brown viscous oil
(0.94 g). Purification by flash column chromatography on neutral
alumina, activity 2 (ethyl acetate/heptane, gradient from 50% to
dichloromethane (10 mL). The mixture was stirred at room tem-
perature for 4 h. Saturated NaHCO3 aqueous solution (10 mL) was
then added. The organic layer was separated, and the aqueous ex-
tracted with dichloromethane (2×10 mL). The combined organic
phases were dried with sodium sulfate, filtered, and concentrated
to afford an orange solid (0.15 g). Purification by flash column
chromatography (ethyl acetate/dichloromethane, gradient from 0%
to 100%) yielded pure 12 (0.11 g, 0.51 mmol, 73%). Brown crystals.
C14H19NO (217.3): calcd. C 77.38, H 8.81; found C 77.14, H 9.03.
M.p. 112.9–113.2 °C (cyclohexane). MS (CI, NH3): m/z = 218
4658
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4654–4662

Intramolecular Aromatic Electrophilic Substitution of Aminocyclopropanes
FULL PAPER
[MH+], 219. IR ν = 3036, 2934, 2865, 2710, 2602, 1592, 1483, 1456, NMR spectroscopy) of 15 and starting material (27 mg, 0.14 mmol,
˜
1
1388, 1253, 1157 cm–1. H NMR: δ = 0.15 (dd, J = 8, 6 Hz, 1 H),
20% and 7% respectively), and a 53:47 mixture (as determined by
0.80 (dd, J = 6, 5 Hz, 1 H), 1.27 (dt, J = 8, 5 Hz, 1 H), 1.33 (s, 3 1H NMR spectroscopy) of the aminocyclopropane 13 and the di-
H), 1.80 (m, 1 H), 1.87–2.09 (m, 2 H), 2.29 (ddd, J = 12, 9, 7 Hz,
1 H), 2.76 (t, J = 8 Hz, 2 H), 3.03 –3.20 (m, 2 H), 6.65–6.86 (m, 3
H), 7.14 (t, J = 8 Hz, 1 H), 7.77 (br. s, 1 H, OH) ppm. 13C NMR:
δ = 8.3, 18.8, 22.7, 25.9, 35.0, 46.7, 49.4, 54.0, 114.2, 115.5, 119.7,
129.9, 141.4, 157.4 ppm.
mer 16 (50 mg, 0.18 mmol, 19% and 33%, respectively). These two
compounds could be obtained in a pure form by performing a sec-
ond flash chromatography. The isolated yields for 13 and 16 were
14 mg (79 µmol, 15%) and 23 mg (59 µmol, 23%), respectively. 15:
1
Yellow oil. MS (EI): m/z = 194 [M+·]. H NMR: δ = 0.93 (t, J =
7 Hz, 3 H), 1.30 (sext, J = 7 Hz, 2 H), 1.56 (m, 2 H), 2.09 (s, 3 H),
3.23 (dd, J = 8, 6 Hz, 2 H), 4.35 (s, 2 H), 6.00–6.09 (m, 2 H) 6.69
(m, 1 H), 9.22 (br. s, 1 H, NH) ppm. 13C NMR: δ = 13.8, 20.1,
21.4, 30.7, 43.0, 49.0, 107.1, 107.2, 118.2, 129.0, 171.7 ppm. 16:
Colourless oil. MS (EI): m/z = 80, 95, 205, 264, 306, 343, 386 [M+·].
1H NMR: δ = 0.82 (d, J = 6 Hz, 6 H), 1.34 (m, 4 H), 1.56 (m, 2
H), 2.09 (s, 6 H), 3.11–3.33 (m, 4 H), 4.35 (AB system, ∆υ = 8, JAB
= 15 Hz, 4 H), 6.01–6.08 (m, 4 H), 6.70 (br. s, 2 H), 9.23 (br. s, 2
H, NH) ppm. 13C NMR: δ = 14.2, 21.4, 33.6, 34.9, 43.3, 48.1,
107.2, 107.3, 118.3, 129.0, 171.6 ppm.
Tetrahydrocarboline 17: A solution of the aminocyclopropane 7[6]
(1.0 equiv., 4.9 mmol, 1.2 g) in chlorobenzene (100 mL) was heated
at reflux for 11 h. After cooling, the solvent was removed under
reduced pressure to afford a brown viscous oil (1.2 g). Analysis of
this crude product by 1H and 13C NMR spectroscopy revealed a
41:59 ratio for the cis and trans products, defined according to the
relative configurations of the methyl groups. Purification by flash
column chromatography (concentrated ammonium hydroxide
aqueous solution/ethyl acetate, gradient from 0% to 1%) yielded
pure cis-17 (0.31 g, 1.3 mmol, 26%), a 40:60 mixture (as determined
by 1H NMR spectroscopy) of cis- and trans-17 (0.11 g, 0.44 mmol,
4% and 5% respectively), and pure trans-17 (0.38 g, 1.6 mmol,
32%). The total yield was thus 0.80 g (3.3 mmol, 67%). Less Polar
cis Isomer: Oil. MS (CI, NH3): m/z = 241 [MH+], 242, 257
[MH+·NH3]. HRMS (ES+): calcd. for C16H21N2 [MH+] 241.1705;
Aminocyclopropane 13: n-Butyllithium (1.2  in hexane, 1.1 equiv.,
0.57 mmol, 0.47 mL) was added dropwise at –70 °C to a solu-
tion
of
N-(but-3-enyl)-N-(1H-pyrrol-2-ylmethyl)acetamide[30]
(1.0 equiv., 0.52 mmol, 0.10 g) in THF (4.0 mL). After 30 minutes
of stirring at –70 °C, chlorotrimethylsilane (1.1 equiv., 0.57 mmol,
73 µL) was added to the yellow mixture, which was then warmed
to 20 °C for 50 minutes and became brown. Titanium isopropoxide
(1.5 equiv., 0.78 mmol, 0.23 mL) was then added, followed by cy-
clopentylmagnesium chloride (2.1  in Et2O, 4.5 equiv., 2.3 mmol,
1.1 mL) dropwise over 5 minutes. After 10 minutes of stirring at
20 °C, the mixture was diluted in diethyl ether (50 mL) and water
(50 mL). The organic layer was separated, and the aqueous ex-
tracted with diethyl ether (2×50 mL). The combined organic
phases were dried with sodium sulfate, filtered, and concentrated
to afford a brown oil (0.14 g). Purification by flash column
chromatography (methanol/ethyl acetate, 2%) yielded pure 13
(62 mg, 0.35 mmol, 68%). White crystals. C11H16N2 (176.3): calcd.
C 74.96, H 9.15; found: C 74.83, H 9.32. M.p. 93.0–94.0 °C (ethyl
acetate). HRMS (ES+): calcd. for C11H17N2 [MH+] 177.1392;
found 177.1389. IR: ν = 3181, 3094, 2928, 2863, 1447, 1355, 1253,
˜
1
1241, 1125, 1097, 1025 cm–1. H NMR: δ = 0.14 (dd, J = 8, 6 Hz,
1 H), 0.80 (dd, J = 6, 4 Hz, 1 H), 1.16 (dt, J = 8, 4 Hz, 1 H), 1.34
(s, 3 H), 1.68 (m, 1 H), 1.84–2.00 (m, 2 H), 2.78 (m, 1 H), 3.60 (AB
system, δA = 3.22, δB = 3.98, JAB = 13 Hz, 2 H), 6.00 (m, 1 H),
6.10 (q, J = 3 Hz, 1 H), 6.70 (q, J = 3 Hz, 1 H), 8.89 (br. s, 1 H,
NH) ppm. 13C NMR: δ = 8.0, 19.7, 23.1, 25.8, 46.1, 48.9, 50.3,
106.8, 107.8, 117.3, 129.9 ppm.
found 241.1677. IR: ν = 3408, 3271, 2924, 2850, 1619, 1451, 1346,
˜
1294, 1233, 1138 cm–1. 1H NMR: δ = 1.20 (d, J = 7 Hz, 3 H), 1.36
(s, 3 H), 1.52 (dtd, J = 12, 9, 7 Hz, 1 H), 1.89 (dtd, J = 12, 7, 3 Hz,
1 H), 2.24 (sext, J = 7 Hz, 1 H), 2.52 (ddd, J = 16, 4, 2 Hz, 1 H),
2.83–3.04 (m, 3 H), 3.13–3.34 (m, 2 H), 7.05–7.18 (m, 2 H), 7.31
(m, 1 H), 7.48 (m, 1 H), 7.62 (br. s, 1 H, NH) ppm. 13C NMR: δ
= 16.6, 16.9, 22.7, 32.4, 42.2, 43.4, 48.4, 60.8, 106.6, 110.7, 118.3,
119.4, 121.5, 127.2, 135.7, 141.1 ppm. More Polar trans Isomer: Oil.
MS (CI, NH3): m/z = 241 [MH+], 242. HRMS (ES+): calcd. for
Aminocyclopropane 14: A similar procedure as for the preparation
of 10, starting from N-(but-3-enyl)-N-[1-(triisopropylsilanyl)pyrrol-
2-ylmethyl]acetamide[30] (0.23 mmol, 79 mg), yielded pure 14
(34 mg, 0.10 mmol, 44%). Colourless oil. MS (EI) m/z = 115, 194,
235, 236, 237, 332 [M+·]. HRMS (ES+) calcd. for C20H37N2Si
[MH+] 333.2726, found 333.2739. IR ν = 2947, 2867, 1467, 1141,
˜
1
1064, 882 cm–1. H NMR: δ = 0.08 (dd, J = 8, 5 Hz, 1 H), 0.83 (t,
C16H21N2 [MH+] 241.1705; found 241.1711. IR: ν = 3412, 3283,
˜
J = 5 Hz, 1 H), 0.98–1.17 (m, 19 H), 1.33 (s, 3 H), 1.50 (m, 1 H),
1.59–1.72 (m, 3 H), 1.75–1.90 (m, 2 H), 2.77 (m, 1 H), 3.12 (d, J
= 13 Hz, 1 H), 3.98 (d, J = 13 Hz, 1 H), 6.16 (t, J = 3 Hz, 1 H),
6.22 (m, 1 H), 6.78 (dd, J = 3, 2 Hz, 1 H) ppm. 13C NMR δ = 7.7,
13.2, 18.4, 18.6, 19.7, 23.1, 25.8, 46.4, 50.2, 50.5, 108.9, 112.5,
125.8, 136.3 ppm.
2960, 2926, 1597, 1465, 1378, 1347, 1323, 1289, 1267, 1234, 1126,
1
1099 cm–1. H NMR: δ = 1.23 (d, J = 7 Hz, 3 H), 1.28 (m, 1 H),
1.46 (s, 3 H), 1.92 (dtd, J = 11, 8, 4 Hz, 1 H), 2.22 (dquint, J = 11,
7 Hz, 1 H), 2.50 (dd, J = 15, 4 Hz, 1 H), 2.96–3.25 (m, 5 H), 7.06–
7.18 (m, 2 H), 7.32 (d, J = 8 Hz, 1 H), 7.50 (d, J = 8 Hz, 1 H),
7.65 (br. s, 1 H, NH) ppm. 13C NMR: δ = 16.1, 16.1, 26.8, 31.1,
42.8, 44.6, 47.1, 63.2, 110.0, 110.7, 118.2, 119.3, 121.7, 127.1, 135.7,
136.4 ppm.
Amide 15 and Dimer 16: Cyclopentylmagnesium chloride (2.0  in
Et2O, 4.0 equiv., 2.1 mmol, 1.0 mL) was added dropwise at –70 °C
to a solution of titanium isopropoxide (2.0 equiv., 1.0 mmol,
0.31 mL) in diethyl ether (10 mL). The mixture was warmed to
–30 °C in 5 minutes and became black. A solution of N-(but-3-
Tetrahydrocarboline 18: A similar procedure as for the preparation
of 17, starting from aminocyclopropane 8[6] (1.0 equiv., 0.50 mmol,
0.13 g), gave pure trans-18 (73 mg, 0.27 mmol, 54%) after purifica-
enyl)-N-(1H-pyrrol-2-ylmethyl)acetamide[30] (1.0 equiv., 0.52 mmol, tion by flash column chromatography (ethyl acetate/dichlorometh-
0.10 g) in diethyl ether (1.0 mL) was then added. The brown-red
mixture was warmed to 20 °C for 1 h20, then diluted in diethyl
ether (0.10 L) and water (0.10 L). The organic layer was separated,
and the aqueous extracted with diethyl ether (2×0.10 L). The com-
bined organic phases were dried with sodium sulfate, filtered, and
concentrated to afford a black oil (92 mg). Purification by flash
column chromatography (ethyl acetate/dichloromethane, gradient
ane, gradient from 0% to 100%). A pyrrole by-product was also
isolated in 7% yield.[31] Orange gum. MS (ES+): m/z = 269 [MH+],
270. HRMS (ES+): calcd. for C18H25N2 [MH+] 269.2018; found
269.1995. IR: ν = 3413, 3292, 2956, 2930, 2870, 1624, 1460, 1347,
˜
1
1322, 1295, 1123 cm–1. H NMR: δ = 0.86 (t, J = 7 Hz, 3 H), 1.00
(d, J = 7 Hz, 3 H), 1.10–1.57 (m, 3 H), 1.76 (dd, J = 9, 7 Hz, 2 H),
1.94 (dtd, J = 12, 7, 6 Hz, 1 H), 2.28 (sext, J = 7 Hz, 1 H), 2.52
(ddd, J = 15, 4, 1 Hz, 1 H), 2.85–3.12 (m, 4 H), 3.19 (ddd, J = 4,
1
from 0% to 100%) yielded a 74:26 mixture (as determined by H
Eur. J. Org. Chem. 2005, 4654–4662
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4659

L. Larquetoux, N. Ouhamou, A. Chiaroni, Y. Six
FULL PAPER
5, 1 Hz, 1 H), 7.06–7.18 (m, 2 H), 7.32 (d, J = 8 Hz, 1 H), 7.51 (d, = 8 Hz, 1 H), 7.56 (br. s, 1 H, NH) ppm. 13C NMR δ = 15.4, 17.2,
J = 7 Hz, 1 H), 7.67 (br. s, 1 H, NH) ppm. 13C NMR: δ = 14.9,
16.7, 16.8, 18.1, 31.7, 42.9, 43.7, 45.6, 49.1, 66.2, 110.7, 111.1,
118.1, 119.2, 121.5, 127.1, 135.7, 136.4 ppm.
21.3, 21.8, 26.9, 36.6, 47.8, 48.9, 57.7, 108.3, 110.6, 118.0, 119.1,
121.1, 127.4, 135.9, 140.4 ppm. More Polar cis Isomer: Yellowish
oil. MS (ES+): m/z = 255 [MH+], 256, 271. HRMS (ES+): calcd.
for C H N [MH+] 255.1861; found 255.1842. IR: ν = 3421, 2923,
˜
17 23
2
Tetrahydrocarboline 19: para-Toluenesulfonic acid (0.10 equiv., 95
µmol, 16 mg) was added to a solution of aminocyclopropane 9
(1.0 equiv., 0.95 mmol, 0.24 g) in chlorobenzene (20 mL). The mix-
ture was heated at reflux for 24 h under a static pressure of argon
(balloon filled with argon connected to the top of the reflux con-
denser). After cooling, the solvent was removed under reduced
pressure to afford a brown viscous oil (0.29 g). Analysis of this
1
2852, 1460, 1297, 1271, 1184, 1101, 1008 cm–1. H NMR: δ = 1.16
(d, J = 7 Hz, 3 H), 1.42 (s, 3 H), 1.45–1.62 (m, 3 H), 1.71 (m, 1
H), 2.15 (dqd, J = 9, 7, 4 Hz, 1 H), 2.74 (dt, J = 15, 6 Hz, 1 H),
2.81 (dtd, J = 13, 4, 1 Hz, 1 H), 2.89 (dt, J = 15, 6 Hz, 1 H), 3.06
(ddd, J = 13, 10, 3 Hz, 1 H), 3.10 (dt, J = 12, 6 Hz, 1 H), 3.24 (dt,
J = 12, 6 Hz, 1 H), 7.08 (td, J = 8, 1 Hz, 1 H), 7.14 (td, J = 8,
1 Hz, 1 H), 7.31 (d, J = 8 Hz, 1 H), 7.48 (d, J = 8 Hz, 1 H), 7.73
(br. s, 1 H, NH) ppm. 13C NMR: δ = 17.7, 19.6, 20.7, 21.8, 30.0,
35.1, 46.4, 48.7, 57.8, 107.7, 110.6, 118.1, 119.2, 121.3, 127.3, 135.6,
140.0 ppm.
1
crude product by H and 13C NMR spectroscopy revealed a 60:40
ratio for the cis and trans products, defined according to the relative
configurations of the methyl groups. Purification by flash column
chromatography (ethyl acetate/heptane, gradient from 40% to
100%) yielded pure trans-19 (66 mg, 0.26 mmol, 27%) and pure
cis-19 (0.10 g, 0.39 mmol, 41%). Less Polar trans Isomer: Yellowish
oil. MS (ES+): m/z = 144, 255 [MH+], 256, 323, 325. HRMS (ES+):
Tetrahydroisoquinoline 20: A similar procedure as for the prepara-
tion of 17, with 4 h of heating aminocyclopropane 12 (1.0 equiv.,
0.50 mmol, 0.11 g) in chlorobenzene at reflux, afforded the crude
calcd. for C17H23N2 [MH+] 255.1861; found 255.1849. IR: ν = product (0.10 g). 1H and 13C NMR spectroscopy revealed the ratio
˜
3412, 2925, 2855, 2808, 1460, 1448, 1344, 1296, 1284, 1166, 1128,
of 20 and its regioisomer (with the OH ortho to the newly-formed
bond) to be about 87:13. Only one diastereoisomer of 20 could
1113 cm–1. 1H NMR: δ = 0.91 (d, J = 7 Hz, 3 H), 1.38 (s, 3 H),
1.44–1.52 (m, 2 H), 1.86–2.06 (m, 3 H), 2.60 (dd, J = 14, 4 Hz, 1 be detected. Purification by flash column chromatography (ethyl
H), 2.66 (m, 1 H), 2.75 (dd, J = 11, 6 Hz, 1 H), 2.85 (td, J = 12,
3 Hz, 1 H), 2.91 (ddd, J = 14, 11, 6 Hz, 1 H), 3.00 (td, J = 11,
4 Hz, 1 H), 7.06–7.14 (m, 2 H), 7.30 (d, J = 8 Hz, 1 H), 7.46 (d, J
acetate/dichloromethane, gradient from 0% to 100%) gave some-
what impure regioisomer of 20 (a few milligrams) and the tetra-
hydroisoquinoline 20 (52 mg, 0.24 mmol, 49%). 20: Brown crystals.
Table 1. Crystallographic data and parameters for the compounds cis-17·picric acid, 18·picric acid, and 22.
Compound
cis-17·Picric acid
18·Picric acid
22
Empirical formula
Molecular mass
Temperature [K]
Wavelength [Å]
Crystal system
Space group
Unit cell dimensions
a [Å]
b [Å]
c [Å]
α [°]
β [°]
(C16H21N2)+(C6H2N3O7)–·0.5H2O
(C18H25N2)+(C6H2N3O7)–
C17H22N2O
270.37
293(2)
0.71073
monoclinic
P21/n
478.46
497.51
293(2)
0.71073
triclinic
P–1
293(2)
0.71073
monoclinic
P2/a
13.604(11)
11.742(11)
14.467(11)
90
102.87(3)
90
7.962(4)
11.542(5)
14.131(7)
67.20(3)
85.42(3)
87.81(3)
1193.3(10)
2
10.191(4)
12.069(4)
11.744(5)
90
95.78(3)
90
γ [°]
Volume [Å3]
Z
2253(3)
4
1437.1(10)
4
Density, calculated
1.411
1.385
1.250
[Mg/m3]
Absorption coefficient [mm–1]
F(000)
0.108
1004
0.104
524
0.078
584
Crystal habit
Crystal size [mm]
θ range [°]
dark orange, prismatic
0.35×0.30×0.20
1.73 to 23.82
–15 Յ h Յ 15,
–13 Յ k Յ 13,
–16 Յ l Յ 16
15640/6675
3448 [0.0335]
2517
pale orange
0.62×0.50×0.30
1.95 to 30.05
–11 Յ h Յ 11,
–14 Յ k Յ 14,
–19 Յ l Յ 19
13761/11325
5748 [0.0226]
4190
pale yellow, prismatic
0.50×0.25×0.12
2.43 to 27.86
–13 Յ h Յ 13,
–15 Յ k Յ 12,
–15 Յ l Յ 15
17743/5838
3418 [0.0308]
2640
Index ranges
Reflections collected
Independent reflections [Rint
Reflections observed
[I Ͼ 2σ (I)]
]
Absorption correction
Refinement method
none
none
none
full-matrix least squares on F2
5747/0/328
Data/restraints/parameters
Goodness-of-fit on F2
R1, wR2 indices [I Ͼ 2σ (I)]
R1, wR2 indices (all data)
Extinction coefficient
3444/2/319
1.081
0.0724, 0.1707
0.1012, 0.1927
0.009(2)
3417/0/187
1.044
0.0420, 0.1030
0.0587, 0.1136
0.049(5)
1.030
0.0609, 0.1589
0.0832, 0.1780
0.010(6)
Largest diff. peak and hole [e·Å–3]
0.316 and –0.331
0.281 and –0.278
0.270 and –0.148
4660
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4654–4662

Intramolecular Aromatic Electrophilic Substitution of Aminocyclopropanes
FULL PAPER
C14H19NO (217.3): calcd. C 77.38, H 8.81; found C 77.15, H 8.91.
gram SHELXS86[33] and were refined with standard methods using
SHELXL93[34] with anisotropic parameters for the non-hydrogen
atoms. Crystallographic data and parameters of the refinements are
M.p. 150.1–152.6 °C (ethyl acetate). MS (EI): m/z = 147, 175, 186,
202, 203, 217 [M+·]. IR: ν = 2963, 2687, 2596, 1609, 1582, 1497,
˜
1
1452, 1379, 1355, 1296, 1247, 1157, 1127 cm–1. H NMR: δ = 1.20 listed in Table 1.
(d, J = 7 Hz, 3 H), 1.32 (s, 3 H), 1.50 (dq, J = 12, 9 Hz, 1 H), 1.91
(dtd, J = 12, 7, 3 Hz, 1 H), 2.24 (dquint, J = 9, 7 Hz, 1 H), 2.47
Acknowledgments
(dt, J = 16, 4 Hz, 1 H), 2.82 (dt, J = 9, 7 Hz, 1 H), 2.92 (ddd, J =
16, 11, 5 Hz, 1 H), 3.01 (ddd, J = 9, 7, 3 Hz, 1 H), 3.04–3.20 (m,
2 H), 6.47 (d, J = 3 Hz, 1 H), 6.64 (dd, J = 8, 3 Hz, 1 H), 6.75 (br.
s, 1 H, OH), 7.04 (d, J = 8 Hz, 1 H) ppm. 13C NMR: δ = 16.6,
24.2, 24.5, 32.0, 44.1, 44.2, 49.2, 64.0, 114.4, 115.1, 127.3, 134.7,
We wish to thank Professor Pierre Potier and Professor Jean-Yves
Lallemand for their kind interest in our work, for their generous
support and their constant encouragements.
1
136.3, 154.3 ppm. Regioisomer: H NMR: δ = 0.80 (d, J = 7 Hz, 3
H), 1.26 (m, 1 H), 1.28 (s, 3 H), 1.54 (dddd, J = 13, 8, 5, 1 Hz, 1
H), 2.19 (dq, J = 13, 8 Hz, 1 H), 2.65 (quint, J = 7 Hz, 1 H), 2.75
(m, 1 H), 2.86–3.04 (m, 4 H), 6.53 (d, J = 8 Hz, 1 H), 6.70 (d, J =
8 Hz, 1 H), 6.99 (t, J = 8 Hz, 1 H) ppm. The signal corresponding
to the OH group could not be identified clearly on the spectrum
we collected.
[1] V. Chaplinski, A. de Meijere, Angew. Chem. Int. Ed. Engl. 1996,
35, 413–414.
[2] a) O. G. Kulinkovich, S. V. Sviridov, D. A. Vasilevskii, T. S. Pri-
tytskaya, J. Org. Chem. USSR (Engl. Trans.) 1989, 25, 2027–
2028; b) O. G. Kulinkovich, S. V. Sviridov, D. A. Vasilevski,
Synthesis 1991, 234.
[3] Reviews: a) O. G. Kulinkovich, A. de Meijere, Chem. Rev. 2000,
100, 2789–2834; b) O. G. Kulinkovich, Chem. Rev. 2003, 103,
2597–2632; c) A. de Meijere, S. I. Kozhushkov, A. I. Savch-
enko, in: Titanium and Zirconium in Organic Synthesis (Ed.: I.
Marek), Wiley-VCH, Weinheim, 2002, pp. 390–434.
[4] J. Lee, J. K. Cha, J. Org. Chem. 1997, 62, 1584–1585.
[5] For further examples of intramolecular Kulinkovich-de Meijere
reactions, see: a) V. Chaplinski, H. Winsel, M. Kordes, A.
de Meijere, Synlett 1997, 111–114; b) B. Cao, D. Xiao, M. M.
Joullié, Org. Lett. 1999, 1, 1799–1801; c) H. B. Lee, M. J. Sung,
S. C. Blackstock, J. K. Cha, J. Am. Chem. Soc. 2001, 123,
11322–11324; d) M. Gensini, S. I. Kozhushkov, D. S. Yufit,
J. A. K. Howard, M. Es-Sayed, A. de Meijere, Eur. J. Org.
Chem. 2002, 2499–2507; e) G.-D. Tebben, K. Rauch, C. Strat-
mann, C. M. Williams, A. de Meijere, Org. Lett. 2003, 5, 483–
485; f) N. Ouhamou, Y. Six, Org. Biomol. Chem. 2003, 1, 3007–
3009; g) M. Gensini, A. de Meijere, Chem. Eur. J. 2004, 10,
785–790.
Ketone 21 and Pentacyclic Aminal 22: A solution of the aminocyclo-
propane 9 (1.0 equiv., 0.70 mmol, 0.18 g) in chlorobenzene (14 mL)
was heated at reflux for 24 h, with the top of the reflux condenser
open to the air. After cooling, the solvent was removed under re-
duced pressure to afford a brown viscous oil (0.20 g). Purification
by flash column chromatography (ethyl acetate/heptane, gradient
from 40% to 100%) yielded relatively pure trans-19 (14 mg, 55
µmol, 8%), pure 22 (24 mg, 88 µmol, 13%), and pure 21 (77 mg,
0.27 mmol, 38%). 21: Colourless crystals. M.p. 124.1–124.8 °C
(ethyl acetate and a few drops of dichloromethane). MS (ES+): m/
z = 287 [MH+], 309 [MNa+], 310, 449, 450, 595. HRMS (ES+):
calcd. for C17H22N2NaO2 [MNa+] 309.1579; found 309.1533. IR
(62:38 mixture of two rotamers): ν = 3251, 2925, 1710, 1613, 1483,
˜
1455, 1421, 1359, 1229, 1170, 1009 cm–1. 1H NMR (major rotamer
in the 62:38 mixture): δ = 1.85 (quint, J = 7 Hz, 2 H), 1.93 (s, 3
H), 2.14 (s, 3 H), 2.47 (t, J = 7 Hz, 2 H), 3.02 (t, J = 7.5 Hz, 2 H),
3.39 (t, J = 7 Hz, 2 H), 3.56 (t, J = 7.5 Hz, 2 H), 7.00 (d, J = 2 Hz,
1 H), 7.10–7.23 (m, 2 H), 7.38 (d, J = 8.5 Hz, 1 H), 7.58 (d, J =
[6] L. Larquetoux, J. A. Kowalska, Y. Six, Eur. J. Org. Chem. 2004,
3517–3525.
[7] Reviews: a) W. M. Whaley, T. R. Govindachari, in: Organic Re-
actions (Ed.-in-chief: R. Adams), John Wiley & Sons, New
York, 1951, 6, 151–206; b) L. E. Overman, D. J. Ricca, in:
Comprehensive Organic Synthesis (Eds: B. M. Trost, I. Fleming,
C. H. Heathcock), Pergamon, New York, 1991, vol. 2, pp.
1016–1022; c) G. Bringmann, C. L. J. Ewers, R. Walter, in:
Comprehensive Organic Synthesis (Eds: B. M. Trost, I. Fleming,
E. Winterfeldt), Pergamon, New York, 1991, vol. 6, pp. 736–
740; d) M. D. Rozwadowska, Heterocycles 1994, 39, 903–931;
e) E. D. Cox, J. M. Cook, Chem. Rev. 1995, 95, 1797–1842.
[8] Review: W. M. Whaley, T. R. Govindachari, in: Organic Reac-
tions (Ed.-in-chief: R. Adams), John Wiley & Sons, New York,
1951, 6, 74–150.
1
8 Hz, 1 H), 8.07 (br. s, 1 H, NH) ppm. H NMR (minor rotamer
in the 62:38 mixture): δ = 1.78 (quint, J = 7 Hz, 2 H), 2.12 (s, 3
H), 2.13 (s, 3 H), 2.39 (t, J = 7 Hz, 2 H), 3.03 (t, J = 7.5 Hz, 2 H),
3.16 (t, J = 7 Hz, 2 H), 3.62 (t, J = 7.5 Hz, 2 H), 7.04 (d, J = 2 Hz,
1 H), 7.10–7.23 (m, 2 H), 7.36 (d, J = 8.5 Hz, 1 H), 7.69 (d, J =
8 Hz, 1 H), 8.00 (br. s, 1 H, NH) ppm. 13C NMR (major rotamer
in the 62:38 mixture): δ = 21.3, 21.8, 24.7, 30.0, 40.8, 44.6, 49.1,
111.4, 112.1, 118.2, 119.6, 121.9, 122.2, 127.1, 136.3, 170.8,
208.4 ppm. 13C NMR (minor rotamer in the 62:38 mixture): δ =
21.6, 22.6, 23.6, 30.0, 39.9, 46.9, 48.4, 111.1, 113.3, 118.8, 119.3,
121.9, 122.2, 127.5, 136.3, 170.4, 207.4 ppm. 22: Pale yellow crys-
tals. M.p. 176.4–176.9 °C (ethyl acetate and a few drops of dichlo-
[9] Reviews: a) M. Lounasmaa, A. Koskinen, Heterocycles 1984,
22, 1591–1612; b) D. Grierson, in: Organic Reactions (Ed.-in-
chief: L. A. Paquette), John Wiley & Sons, New York, 1990,
39, 85–295.
romethane). MS (ES+): m/z = 255, 271 [MH+], 272. IR: ν = 2925,
˜
2851, 2359, 1610, 1487, 1469, 1222, 1152, 1041, 953 cm–1. 1H
NMR: δ = 0.95 (s, 3 H), 1.07 (m, 1 H), 1.10 (s, 3 H), 1.42 (td, J =
14, 4 Hz, 1 H), 1.83–1.97 (m, 3 H), 2.21 (td, J = 12, 6 Hz, 1 H),
2.85–3.05 (m, 3 H), 3.16 (ddd, J = 12, 8, 4 Hz, 1 H), 4.66 (br. s, 1
H, NH), 5.46 (d, J = 3 Hz, 1 H), 6.56 (d, J = 8 Hz, 1 H), 6.73 (t,
J = 7 Hz, 1 H), 7.05–7.13 (m, 2 H) ppm. 13C NMR: δ = 15.8, 19.7,
27.9, 35.1, 36.4, 45.9, 49.3, 68.0, 70.6, 83.7, 100.3, 108.5, 119.0,
125.0, 128.4, 130.2, 149.3 ppm.
[10] A similar observation has been reported before from simpler
substrates: see ref.[5e]
.
[11] Y. Six, Eur. J. Org. Chem. 2003, 1157–1171.
[12] See for instance: a) C. Averbuj, J. Kaftanov, I. Marek, Synlett
1999, 1939–1941; b) J. C. Lee, M. J. Sung, J. K. Cha, Tetrahe-
dron Lett. 2001, 42, 2059–2061; c) A. de Meijere, C. M. Wil-
liams, A. Kourdioukov, S. V. Sviridov, V. Chaplinski, M.
Kordes, A. I. Savchenko, C. Stratmann, M. Noltemeyer, Chem.
Eur. J. 2002, 8, 3789–3801.
X-ray Crystallographic Study:[32] Crystals suitable for X-ray diffrac-
tion studies were grown from acetonitrile for the salts of cis-17 and
18 with picric acid, and ethyl acetate/dichloromethane for 22. Data
were collected at room temperature on a Nonius Kappa CCD aera-
detector diffractometer (φ and ω scan mode), using graphite-mono-
chromated Mo-Kα radiation. The structures were solved by the pro-
[13] The 1H NMR spectrum of the crude product features two sing-
lets at δ = 1.65 and 1.79 ppm that could be attributed to the
methyl groups of enamine 23. In the 13C NMR spectrum, a
peak at δ = 133.9 ppm could be assigned as the α carbon of
the enamine.
Eur. J. Org. Chem. 2005, 4654–4662
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4661

L. Larquetoux, N. Ouhamou, A. Chiaroni, Y. Six
FULL PAPER
[14] The types of argon available in our institute are “Arcal 1” and
“Nertal” (Air Liquide), normally used for arc welding and
plasma cutting applications.
[23] The pKa of the phenol can be estimated at about 10: a) D. A.
Whiting, in: Comprehensive Organic Chemistry; The Synthesis
and Reactions of Organic Compounds, (Eds.: Sir D. Barton,
W. D. Ollis), Pergamon, Oxford, 1979, vol. 1, pp. 709–710. The
pKa of the cyclic iminium ion should lie in the range 10–12; b)
R. L. Hinman, Tetrahedron 1968, 24, 185–190.
[15] G. Pitacco, E. Valentin, in: The Chemistry of Functional Groups,
Supplement F: The Chemistry of Amino, Nitroso and Nitro
Compounds and their Derivatives, Part 1 (Ed.: S. Patai), John
Wiley & Sons, Chichester, 1982, pp. 639–640.
[24] L.-F. Tietze, K. Brüggemann, Angew. Chem. Int. Ed. Engl.
[16] Compounds 21 and 22 were present in the crude product as
1982, 21, 539–540.
[25] Z. Czarnocki, D. Suh, D. B. MacLean, P. G. Hultin, W. A. Sza-
rek, Can. J. Chem. 1992, 70, 1555–1561.
[26] N. Carrasco, A. Urzúa, B. K. Cassels, J. Org. Chem. 1973, 38,
4342–4343.
1
shown by H and 13C NMR analysis. Therefore, their forma-
tion cannot be explained by oxidation during column
chromatography.
[17] G. Massiot, F. Sousa Oliveira, J. Lévy, Tetrahedron Lett. 1982,
23, 177–180.
[27] See for instance: a) G. Hahn, K. Stiehl, Ber. Dtsch. Chem. Ges.
1936, 69, 2627–2654; b) K. Narayanan, L. Schindler, J. M.
Cook, J. Org. Chem. 1991, 56, 359–365.
[18] It has been established that five-membered ring cyclic enamines
are generally more basic and more reactive towards electro-
philes than their six-membered homologues: a) A. G. Cook,
M. L. Absi, V. K. Bowden, J. Org. Chem. 1995, 60, 3169–3171;
b) B. Kempf, N. Hampel, A. R. Ofial, H. Mayr, Chem. Eur. J.
2003, 9, 2209–2218. It should be noted, however, that unlike
our intermediates, the cyclic enamines studied in these refer-
ences do not include the nitrogen atom in the same ring as the
carbon–carbon double bond.
[28] W. G. Kofron, L. M. Baclawski, J. Org. Chem. 1976, 41, 1879–
1880.
[29] This is an adaptation of a literature procedure described in
ref.[4]
.
[30] Experimental details about the preparation of this compound
can be found in the supporting information.
[31] The structure and characterisation of this compound can be
found in the supporting information.
[32] CCDC-273312, -273313, and -273314 contain the supplemen-
tary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[33] G. M. Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467–473.
[34] G. M. Sheldrick, SHELXL93. Program for the refinement of
crystal structures, 1993, University of Göttingen, Germany.
Received: June 13, 2005
[19] AM1 calculations (M. J. S. Dewar, E. G. Zoebisch, E. F. Healy,
J. J. P. Stewart, J. Am. Chem. Soc. 1985, 107, 3902–3909) were
performed with the MOPAC 97 program (Fujitsu Limited:
1993–1997) in the CS Chem. 3D® Pro system (CambridgeSoft
Corporation, Cambridge, U. S. A.).
1
[20] The comparison of the H NMR spectra is shown in the sup-
porting information; for supporting information see also the
footnote on the first page of this article
[21] See for instance: F. Ungemach, J. M. Cook, Heterocycles 1978,
9, 1089–1119.
Published Online: September 16, 2005
[22] Only a trace amount of acid would be sufficient since a proton
is released after the cyclisation during the re-aromatisation
step.
4662
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 4654–4662