Selective transformations of β-keto esters, promoted by Dowex basic ion-exchange resins

Article abstract of DOI:10.1002/1099-0690(200210)2002:19<3359::AID-EJOC3359>3.0.CO;2-7

Depending on the nature of the Dowex basic ion-exchange resin, cyclic β-keto esters 1 react with α,β-unsaturated aldehydes 2 to give either the corresponding Michael adducts 3 or the highly functionalized bicyclo[3.3.1]nonanes 4 in a onepot Michael addition-intramolecular aldolization sequence. Some selective transformations of the highly functionalized systems 3 and 4 to provide amino azabicyclo[3.3.1]nonanones 9, polycyclic aminals 10, and azacyclooctene 12 are also presented. Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002).

Full text of DOI:10.1002/1099-0690(200210)2002:19<3359::AID-EJOC3359>3.0.CO;2-7

FULL PAPER
Selective Transformations of β-Keto Esters, Promoted by Dowex Basic
Ion-Exchange Resins
[a]
[b]
[b]
[a]
´
Celine Simon, Jean-Francois Peyronel, Francois Clerc, and Jean Rodriguez*
¸
¸
Keywords: Fused-ring systems / Ion-exchange resins / β-Keto esters / Michael additions
Depending on the nature of the Dowex basic ion-exchange
systems 3 and 4 to provide amino azabicyclo[3.3.1]nonano-
resin, cyclic β-keto esters 1 react with α,β-unsaturated alde- nes 9, polycyclic aminals 10, and azacyclooctene 12 are also
hydes 2 to give either the corresponding Michael adducts 3
or the highly functionalized bicyclo[3.3.1]nonanes 4 in a one-
pot Michael addition−intramolecular aldolization sequence.
Some selective transformations of the highly functionalized
presented.
( Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
2002)
Introduction
acyclic vinyl ketones promoted by Amberlyst A-21 or A-27
have also been proposed.[10]
Control of the production of toxic waste and by-products
in chemical transformations is a challenging problem from
industrial, academic, and social points of view,[1] and con-
stitutes an active field of investigation owing to the increas-
ing demands of environmental legislation. In recent years
the development of environmentally friendly heterogeneous
reagents, long known for their low cost and toxicity, has
received much attention.[2] Moreover, the products can eas-
ily be isolated in good chemical purity by simple filtration,
avoiding time-consuming and tedious extractive workup. In
this context, a large number of new inorganic,[3] organic,[4]
or hybrid[5] heterogeneous catalysts have been found to be
efficient in many important organic transformations, and
special attention has been given to the Michael
addition.[4bϪ4d,5,6] Quite surprisingly, the commercially
available and environmentally benign basic ion-exchange
resins have not been the center of much interest as Michael
addition promoters. The pioneering studies of Schmilde,[7]
Yamada,[8] and Bergmann[9] in the 1950s demonstrated
their efficiency in some cases, but also showed the limita-
tions of a number of basic ion-exchange resins in Michael
additions between simple donors Ϫ such as thiols, nitroal-
kanes, and malonic derivatives Ϫ and acrylic acceptors.
Since then, conjugate additions of thiols with the aid of
fluoride ion-modified Amberlyst or Dowex resins and
Michael additions of nitroalkanes to methyl acrylate and
Results and Discussion
Pursuing our interest in the Michael addition[11] we
started a collaborative program aimed at the selective and
efficient construction of functionalized systems of types 3
and 4 starting from simple cyclic β-keto esters 1 and α,β-
unsaturated aldehydes 2 (Scheme 1).
For this purpose, we decided to study the utilization and
the selectivity of two commercially available Dowex basic
ion-exchange resins (Table 1), which combined the advant-
ages of solid-phase synthesis and anionic activation and
should be easily amenable to large-scale preparations.
To test the feasibility of the Michael addition under mild
conditions we first selected Dowex 66, a macroporous,
weakly basic hydroxide resin, in the reaction between
piperidone 1a as its commercially available hydrochloride
and acrolein 2a in MeOH, which had proven to be the best
solvent.[12] Although no reaction took place with only 20%
(by weight) of resin (Table 1, run 1) a quantitative yield of
the expected Michael adduct 3a was obtained with 100% of
Dowex 66 (run 2) and the same result could be achieved on
scaling up the reaction to 15 grams of adduct. Interestingly,
use of a large excess of resin (400%) did not affect the trans-
formation at all and no further intramolecular aldolization
to hydroxy azabicyclo[3.3.1]nonane 4a was observed even
after two days at room temperature (run 3). β-Keto ester 1b
was found to be less reactive toward acrolein 2a, and 500%
of Dowex 66 was needed to achieve total conversion in this
case, after 24 h at room temperature, and to isolate a quant-
itative yield of the corresponding adduct 3b (run 4). Altern-
atively, on treatment of 1a with acrolein 2a, no reaction was
[a]
´
´
´
`
Laboratoire ReSo, Reactivite en Synthese organique, UMR Ϫ
´ ˆ
ˆ
CNRS 6516, Centre de St. Jerome, boıte D12,
13397 Marseille cedex 20, France
Fax: (33) 491 28 88 41;
E-Mail: jean.rodriguez@reso.u-3mrs.fr
[b]
´
´
Aventis Pharma, Departement de Chimie Medicinale, Center
de Recherches de Paris,
13 Quai Jules Guesde, BP-14, 94403 Vitry-sur-Seine, France
Eur. J. Org. Chem. 2002, 3359Ϫ3364  WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/1019Ϫ3359 $ 20.00ϩ.50/0
3359

C. Simon, J.-F. Peyronel, F. Clerc, J. Rodriguez
FULL PAPER
Scheme 1
Table 1. Treatment of 1 and 2 with Dowex resins
Run[a]
Resin (%)[b]
t (h)
Product
Yield (%)[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
Dowex 66 (20)
Dowex 66 (100)
Dowex 66 (400)
Dowex 66 (500)
Dowex 550A (20)
Dowex 550A (200)[e]
Dowex 550A (300)[f]
Dowex 550A (500)
Dowex 66 (100)
2
4
48
24
72
24
5
1a[d]
3a
3a
92
99
99
100
100
100
100
77
3b
1a[d]
3a
4a[g]
4b[g]
1c[d] ϩ 3c
3c
16[h]
24
24
3[h]
24
96
92[i]
80
95
90
92
Figure 1. 1H observations for 4a axial-OH and equatorial-OH
(E ϭ COOMe)
Dowex 66 (300)
Dowex 550A (300)
Dowex 550A (300)
Dowex 550A (300)
3c
4c[j]
4d[j]
Finally, the formation of the bridged bicyclic com-
pound[15] 4b, the product of the reaction between 1b and
2a, was achieved by use of an excess of Dowex 550A in
refluxing EtOH for 16 h (Table 1, run 8), once again show-
ing lower reactivity for 1b than for 1a. Not unexpectedly,
commercially available piperidone hydrochloride 1c dis-
played lower reactivity than 1a or 1b toward acrolein 2a.
Regardless of the nature and the quantity of Dowex resin,
the transformation always gave the Michael adduct 3c (runs
9, 10), which could not be transformed into the correspond-
[a]
Unless otherwise noted, all reactions were performed at room
[b]
temperature, in MeOH for 1a and in EtOH for 1b and 1c. Per-
[c]
centage by weight of 1. Isolated after filtration through a short
[d]
[e]
pad of Celite.
Isolated at its free HCl form.
Use of 100%
[f]
resulted in an incomplete transformation after 48 h. The same
result was obtained with 400% of resin after 4 h.
[g]
Two isomers
[h]
[i]
in 1.3:1 ratio for 4a and 1.2:1 ratio for 4b.
Reflux. Ratio 3c/
1c ϭ 2.3. [j] Four isomers in 8:6.3:1.3:1 ratio for 4c and 1.5:1.3:1.2:1
ratio for 4d.
observed with 20% of the more basic Dowex 550A resin ing bicyclic derivative even in the presence of an excess of
even after a prolonged reaction time (run 5), but 3a could Dowex 550A in refluxing EtOH (run 11). This result can
be obtained quantitatively with 200% of the same resin (run be explained by the presence of the basic nitrogen atom in
6). The expected azabicyclic[13] derivative 4a was isolated the position β to the carbonyl, which prevents the enoliz-
cleanly as a 1.3 to 1 mixture of axial and equatorial epimers ation needed for the intramolecular aldolization. Other al-
after 5 hours by simple filtration when 300% of Dowex dehydes such as crotonaldehyde 2b (run 12) and methacrol-
550A was used (run 7). The stereochemistry was corrobor- ein 2c (run 13) also gave very good results in the condensa-
ated by comparison of the NMR spectroscopic data with tion with 1a in the presence of Dowex 550A, allowing the
previous work from this laboratory (Figure 1).[14] More spe- one-pot syntheses of functionalized hydroxy azabicy-
cifically, the equatorial-OH-4a epimer shows a coupling clo[3.3.1]nonanes 4c and 4d in very good isolated yields and
3
constant JH2ϪH3 ϭ 10.6 Hz due to HϪC-2 at δ ϭ as mixtures of the four possible diastereomers in 8:6.3:1.3:1
4.05 ppm (dt, J ϭ 10.6, 5.3 Hz), which is in agreement with and 1.5:1.2:1.1:1 ratios, respectively. Interestingly enough,
a 1,2-trans-diaxial arrangement for the two protons. In the the closely related commercially available Amberlite resins
major axial-OH-4a isomer, on the other hand, HϪC-2 IRA 410 and the more basic IRA 400 were totally inactive
shows up at δ ϭ 4.34 ppm as a broad doublet with J ϭ for this transformation even after activation of the resins by
3.1 Hz, corroborating the proposed structures.
treatment with 1  NaOH solution.
3360
Eur. J. Org. Chem. 2002, 3359Ϫ3364

Selective Transformations of β-Keto Esters
FULL PAPER
As was to be expected, when methyl vinyl ketone 5 corresponding bridged compounds 9a and 9b in 99% and
(Scheme 2) was used in the presence of Dowex 66 (100%) 77% yields, respectively (Table 2, runs 1Ϫ3). Alternatively,
the corresponding Michael adduct 3d was formed after when ω-functionalized amines such as 2-aminoethanol (run
120 h in 92% yield, together with unchanged starting mat- 4), 1,3-diaminopropane (run 5), and 2-aminoethanethiol
erial 1a. On the other hand, use of Dowex 550A proved to (run 6) were used, acetalization of both carbonyl functions
be inefficient for the formation of bridged derivatives, the resulted in the formation of tricyclic aminals 10a؊c in syn-
only transformation being a Robinson annulation to pro- thetically useful yields after a simple filtration through a
vide a 1:1 mixture of fused hydroxy bicyclic derivative 6 and short pad of Celite, generally with very high chemical pur-
the corresponding olefin 7 in 80% overall yield (Scheme 2). ity. As recently shown in this laboratory, the mechanistic
pathway probably involves an intramolecular capture of an
iminium intermediate by the nucleophilic function of the
amine.[18] It is interesting to note that fused polycyclic struc-
tures including N/O- or N/N-aminals are found in biologic-
ally active natural and unnatural compounds[19] and also
constitute effective intermediates for the preparation of
chiral pyrrolidines and piperidines.[20] Unfortunately, β-al-
anine (run 7) was unreactive under the standard conditions
and only hydroxybicyclo[3.3.1]nonane 3a, arising from in-
tramolecular aldolization, was isolated.
Table 2. Treatment of 3 with amines (E ϭ CO2Et)[a]
Scheme 2. Reagents and conditions: i, Dowex 66 (100%), MeOH,
room temp., 120 h, 92%. ii, Dowex 550A (300%), EtOH, room
temp., 120 h, then reflux, 6 h, 80%
With an efficient and simple large-scale heterogeneous
preparation of both functionalized monocyclic and bicyclic
systems 3 and 4 now to hand, we turned our attention to
the reactivity of these new derivatives for the selective elab-
oration of other valuable heterocyclic compounds.
We first decided to take advantage of the 1,5-dicarbonyl
functionality of compounds 3 in the reaction with primary
amines for the selective construction of 6-amino-3-azabicy-
clo[3.3.1]nonanones 9 by intramolecular Mannich reaction
of the transient aldimines 8[16] (Scheme 3).
Scheme 3
[a]
Unless otherwise noted, all reactions were performed on a
˚
1 mmol scale in the presence of 4-A MS in refluxing toluene for
Although aniline in refluxing toluene in the presence of
24 h, with an adduct/amine ratio of 1:1.5. [b] Isolated after filtration
[17]
˚
4-A molecular sieves (MS) proved unreactive,
we were
[c]
[d]
through a short pad of Celite. Room temperature, 24 h.
Two
pleased to find that benzylamine and tert-butylamine did
indeed undergo the expected transformation, affording the 10c, and 1.5:1 for 3a. Only one isomer.
isomers in ratios of 1.5:1 for 9a, 1.2:1 for 9b, 1.2:1 for 10a, 2:1 for
[e]
Eur. J. Org. Chem. 2002, 3359Ϫ3364
3361

C. Simon, J.-F. Peyronel, F. Clerc, J. Rodriguez
FULL PAPER
a 250 µm coating. IR spectra were recorded neat or in CHCl3, and
NMR spectra were obtained at 200 MHz in CDCl3 with residual
CHCl3 as internal reference.
A previous report on the reactivity of bicyclo[3.3.1]non-
anes[21] prompted us to surmise that the selective frag-
mentation of compounds 4 might provide access to substi-
tuted azacyclooctanes, as found in some bioactive natural
products, such as β-carboline alkaloids.[22] In the car-
bocyclic series, it is known that hydroxybicyclo[3.3.1]non-
anones undergo clean fragmentation to eight-membered
rings under acetalization conditions with ethylene glycol.[23]
On the other hand, it has been reported that the corres-
ponding equatorial tosylates were easily cleaved[24] by al-
koxides through a Grob-type reaction[25] to give the corres-
ponding cyclooctenes (Scheme 4).
Material: Unless otherwise noted, all starting materials were ob-
tained from commercial suppliers and used without further puri-
fication. Both Dowex and Amberlite ion-exchange resins were pur-
chased from Aldrich.
General Procedure for Reactions in the Presence of Ion-exchange
Resins. Preparation of Michael Adduct
3 and Hydroxybicy-
clo[3.3.1]nonanes 4: The ion-exchange resin was introduced into a
solution of the β-dicarbonyl compound (1 mmol) in 25 mL of alco-
holic solvent. The Michael acceptor (1.5 mmol) was then added
and the reaction mixture was stirred either at room temperature or
at reflux (see Table 1). After completion, simple filtration through
a short pad of Celite and evaporation of the filtrate under reduced
pressure gave the product in very good chemical purity as estimated
by NMR (Ͼ95%), and an analytical sample was obtained by FC
on SiO2.[27]
Ethyl 1-Benzyl-4-oxo-3-(3-oxopropyl)piperidine-3-carboxylate (3a,
from 1b and 2a): Rf ϭ 0.54 (Et2O/pentane, 9:1). IR (neat): ν˜ ϭ
2813, 1725, 1463 cmϪ1. 1H NMR: δ ϭ 9.69 (s, 1 H), 7.35Ϫ7.24 (m,
5 H), 4.18 (q, J ϭ 7.2 Hz, 2 H), 3.56, (s, 2 H), 3.58 (dd, J ϭ 11.4,
2.5 Hz, 1 H), 3.08Ϫ2.59 (m, 3 H), 2.49Ϫ2.01 (m, 5 H), 1.89 (m, 1
H), 1.23 (t, J ϭ 7.1 Hz, 3 H) ppm. 13C NMR: δ ϭ 206.2, 200.9,
171.3, 137.7, 128.8, 128.2, 127.3, 61.7, 61.4, 61.3, 60.1, 53.5, 40.4,
39.4, 24.1, 14.0 ppm. Elemental analysis calcd. (%) for C18H23NO4
(317.4): C 68.12, H 7.30, N 4.41; found C 68.41, H 7.28, N 3.98.
Scheme 4
In this case, although the acetalization of 3a failed re-
gardless of the experimental conditions, equatorial-OTs-11
was, as expected, cleanly cleaved on treatment with diazab-
icycloundecene (DBU) in refluxing MeOH to give the de-
sired azacyclooctene 12 in 73% isolated yield (Scheme 5).[26]
Ethyl 1-Oxo-2-(3-oxopropyl)cyclohexane-2-carboxylate (3b)[28]
1
(from 1c and 2a): IR (neat): ν˜ ϭ 3461, 2938, 1714, 1447 cmϪ1. H
NMR: δ ϭ 9.70 (t, J ϭ 1.3 Hz, 1 H), 4.17 (dq, J ϭ 7.2, 2.0 Hz, 2
H), 2.62Ϫ2.28 (m, 6 H), 2.11 (dd, J ϭ 9.6, 5.5 Hz, 1 H), 1.97 (m,
1 H), 1.85 (dd, J ϭ 9.6, 5.5 Hz, 1 H), 1.60 (m, 2 H), 1.43 (m, 1 H),
1.24 (t, J ϭ 7.1 Hz, 3 H) ppm. 13C NMR: δ ϭ 207.7, 201.2, 171.8,
61.5, 59.8, 41.0, 39.3, 36.6, 27.5, 26.8, 22.5, 14.1 ppm.
Ethyl 1-Benzyl-3-oxo-4-(3-oxopropyl)piperidine-4-carboxylate (3c,
from 1d and 2a): IR (neat): ν˜ ϭ 3438, 2933, 2807, 1727, 1448, 1201,
1111 cmϪ1 1H NMR: δ ϭ 9.67 (s, 1 H), 7.24 (m, 5 H), 4.14 (q,
.
Scheme 5. Reagents and conditions: i, DBU (1.5 equiv.), MeOH,
reflux, 4 h
J ϭ 7.2 Hz, 2 H), 3.5 (m, 2 H), 3.15 (d, J ϭ 15.5 Hz, 1 H), 2.92
(d, J ϭ 15.5 Hz, 1 H), 2.80Ϫ1.50 (m, 8 H), 1.16 (t, J ϭ 7.2 Hz, 3
H) ppm. 13C NMR: δ ϭ 204.9, 201.2, 170.9, 137.1, 129.0, 128.5,
127.5, 62.5, 61.7, 61.5, 56.5, 48.7, 39.2, 31.6, 25.7, 14.1 ppm. Ele-
mental analysis calcd. (%) for C18H23O4N (317.4): C 68.12, H 7.30,
N 4.41; found C 67.58, H 7.40, N 4.29.
Conclusion
We have demonstrated the efficiency of two commercially
available Dowex basic hydroxide resins for Michael addition
Methyl 1-Benzyl-4-oxo-3-(3-oxobutyl)piperidine-3-carboxylate (3d,
from 1a and 5): Rf ϭ 0.40 (Et2O/pentane, 7:3). IR (neat): ν˜ ϭ 3440,
1
of cyclic β-keto esters to α,β-unsaturated carbonyl com- 2980, 2840, 1740, 1460, 1370, 1250 cmϪ1. H NMR: δ ϭ 7.30 (m,
5 H), 3.70 (s, 3 H), 3.6Ϫ2.08 (m, 12 H), 2.07 (s, 3 H) ppm. 13C
pounds, combining the advantages of solid-phase synthesis
NMR: δ ϭ 207.3, 206.0, 171.7, 137.5, 128.6, 128.2, 127.2, 61.4,
and anionic activation. This allows the facile and environ-
61.1, 60.7, 52.7, 57.2, 40.1, 38.5, 29.6, 25.4 ppm. Elemental analysis
mentally friendly preparation of highly functionalized sys-
calcd. (%) for C18H23NO4 (317.4): C 68.12, H 7.30, N 4.41; found
C 67.92, H 7.46, N 4.39.
tems, precursors of polycyclic heteroatomic derivatives of
potential synthetic and biological interest.
Methyl 3-Benzyl-6-hydroxy-9-oxo-3-azabicyclo[3.3.1]nonane-1-carb-
oxylate (equatorial-OH) (4a, from 1a and 2a): Rf ϭ 0.47 (Et2O/
Experimental Section
pentane, 9:1). IR (neat): ν˜ ϭ 3516, 2953, 1737, 1438 cmϪ1 1H
.
NMR: δ ϭ 7.25 (m, 5 H), 4.05 (dt, J ϭ 10.6, 5.3 Hz, 1 H), 3.63 (s,
3 H), 3.45 (m, 1 H), 3.51 (d, J ϭ 12.5 Hz, 1 H), 3.71 (d, J ϭ
12.5 Hz, 1 H), 3.08 (dd, J ϭ 12.5, 2.5 Hz, 1 H), 2.85 (m, 2 H), 1.25
(m, 1 H), 1.75 (m, 1 H) ppm. 13C NMR: δ ϭ 209.2, 172.2, 137.8,
General: Melting points were observed in open Pyrex capillary
tubes and are uncorrected. FC (flash chromatography) was per-
formed with Merck 60 silica gel (230Ϫ240 mesh).[27] TLC was per-
formed on Alugram SIL G/UV 254 silica gel analytical plates with 128.8, 128.6, 127.5, 72.4, 61.8, 61.3, 58.2, 54.2, 54.2, 52.4, 30.4, 30.2
3362 Eur. J. Org. Chem. 2002, 3359Ϫ3364

Selective Transformations of β-Keto Esters
ppm. Elemental analysis calcd. (%) for C17H21O4N (303.4): C 2 H) ppm. 13C NMR: δ ϭ 198.3, 173.1, 160.3, 138.4, 128.8, 128.2,
FULL PAPER
67.09, H 6.95, N 4.60; found C 66.87, H 7.02, N 4.48.
127.3, 124.7, 62.9, 62.1, 53.7, 52.5, 49.8, 34.6, 35.5, 30.9 ppm.
Methyl 3-Benzyl-6-hydroxy-9-oxo-3-azabicyclo[3.3.1]nonane-1-carb-
oxylate (axial-OH) (4a, from 1a and 2a): White powder; m.p.
107Ϫ109 °C; Rf ϭ0.31 (Et2O/pentane, 9:1). IR (neat): ν˜ ϭ 3516,
2953, 1737, 1438 cmϪ1. 1H NMR: δ ϭ 7.25 (m, 5 H), 4.34 (d, J ϭ
3.1 Hz, 1 H), 3.64 (s, 3 H), 3.40 (s, 2 H), 3.17Ϫ2.44 (m, 10 H) ppm.
13C NMR: δ ϭ 210.3, 171.1, 138.0, 128.8, 128.6, 127.4, 76.4, 61.9,
61.7, 58.2, 54.8, 56.6, 52.3, 31.7, 29.2 ppm. Elemental analysis
calcd. (%) for C17H21O4N (303.4): C 67.09, H 6.95, N 4.60; found
C 66.33, H 7.07, N 4.52.
Treatment of Michael Adducts with Amines. Preparation of Amino-
bicycles 9 and Cyclic Aminals 10: Molecular sieves (4 A, 6 g) were
˚
added to a solution of the Michael adduct 3 (0.70 mmol) in dry
toluene (25 mL). The amine was introduced by syringe
(1.05 mmol), and the reaction mixture was heated under reflux for
24 h. Simple filtration through a short pad of Celite and evapora-
tion of the filtrate under reduced pressure usually gave the product
with very good chemical purity as estimated by NMR (Ͼ 95%).
Analytical samples, although very sensitive toward chromatography
on silica gel, were obtained with Et3N-neutralized SiO2.
Ethyl 4-Hydroxy-9-oxobicyclo[3.3.1]nonane-1-carboxylate (4b, from
1c and 2a, two diastereomers): IR (neat): ν˜ ϭ 3437, 3056, 2933,
Ethyl 3-Benzyl-6-tert-butylamino-9-oxo-3-azabicyclo[3.3.1]nonane-
1-carboxylate (9a, two diastereomers): IR (neat): ν˜ ϭ 2955, 2789,
1
1725, 1451, 1262, 1073, 736 cmϪ1. H NMR: δ ϭ 4.26 (m, 1 H),
1
1720, 1445, 1357, 1253 cmϪ1. H NMR: δ ϭ 7.23 (m, 5 H), 4.07
4.15 (q, J ϭ 7.1 Hz, 4 H), 4.03 (m, 1 H), 2.80Ϫ2.71 (m, 1 H), 2.68
(m, 1 H), 2.55 (m, 3 H), 2.31 (m, 5 H), 1.99 (m, 7 H), 1.74 (m, 5
H), 1.55 (m, 3 H), 1.23 (t, J ϭ 7.2 Hz, 6 H) ppm. 13C NMR: δ ϭ
213.9, 212.9, 172.7, 76.0, 72.7, 61.33, 61.28, 58.1, 57.5, 54.4, 54.0,
36.5, 36.1, 31.2, 30.4, 30.4, 28.5, 28.0, 26.5, 20.5, 19.4, 14.2 ppm.
Elemental analysis calcd. (%) for C12H18O4 (226.3): C 63.70, H
8.02; found C 63.78, H 7.82.
(q, J ϭ 7.0 Hz, 2 H), 3.50 (m, 3 H), 3.15Ϫ1.45 (m, 5 H), 1.15 (t,
J ϭ 7.0 Hz, 3 H), 1.15Ϫ1.00 (m, 5 H), 0.92 (s, 9 H) ppm. 13C
NMR: δ ϭ 210.7, 170.9, 138.4, 128.8, 128.4, 127.3, 62.1, 62.0, 61.7,
61.6, 61.2, 61.1, 57.9, 55.5, 54.5, 509, 51.1, 32.4, 31.7, 30.0, 14.2
ppm. MS: m/z (%) ϭ 371 (8) [Mϩ], 112 (27), 91 (100), 57 (30),
29 (10).
Methyl 3-Benzyl-6-hydroxy-8-methyl-9-oxo-3-azabicyclo[3.3.1]non-
ane-1-carboxylate (4c, from 1a and 2b, four diastereomers): Rf ϭ
0.30 (Et2O/pentane, 7:3). IR (neat): ν˜ ϭ 3440, 2950, 2820, 1730,
Ethyl 3-Benzyl-6-benzylamino-9-oxo-3-azabicyclo[3.3.1]nonane-1-
carboxylate (9b, two diastereomers): IR (neat): ν˜ ϭ 2953, 2368,
1714, 1457 cmϪ1 1H NMR: δ ϭ 7.85 (m, 10 H), 4.11 (q, J ϭ
.
1
1500, 1450, 1365, 1270, 1160, 1080 cmϪ1. H NMR: δ ϭ 7.30 (m,
7.1 Hz, 2 H), 3.70Ϫ1.70 (m, 15 H), 1.20 (t, J ϭ 7.1 Hz, 3 H) ppm.
13C NMR: δ ϭ 210.5, 210.3, 170.7, 140.1, 138.3, 128.9, 128.6,
128.5, 128.2, 128.0, 61.2, 62.1, 61.8, 61.3, 61.2, C1: 58.6, 58.2, 53.6,
52.6, 49.2, 50.6, 32.5, 31.9, 29.0, 26.8, 14.3 ppm. Elemental analysis
calcd. (%) for C25H30N2O3 (406.5): C 73.86, H 7.44, N 6.89; found
C 73.54, H 7.43, N 7.03.
5 H), 4.33 (s broad, 1 H), 4.08 (m, 1 H), 3.70 (s, 3 H), 3.68Ϫ3.15
(m, 3 H), 3.05 (m, 2 H), 2.85Ϫ2.35 (m, 3 H), 3.45 (s, 2 H), 0.87 (s,
3 H) ppm. 13C NMR: δ ϭ 211.2, 210.1, 209.9, 207.9, 171.0, 170.
6, 170.5, 138.1, 138.0, 137.9, 137.6, 129.1, 129.0, 128.8, 128.6,
128.6, 128.5, 128.4, 127.5, 127.4, 70.5, 70.0, 63.0, 62.9, 62.3, 62.1,
62.0, 61.7, 61.2, 58.3, 56.8, 56.6, 55.9, 54.1, 54.3, 54.5, 51.7, 52.0,
52.2, 52.3, 38.8, 38.5, 38.1, 37.4, 37.1, 36.4, 35.3, 34.3, 16.3, 16.6,
18.8, 19.5 ppm. Elemental analysis calcd. (%) for C18H23NO4
(317.4): C 68.12, H 7.30, N 4.41; found C 67.68, H 7.52, N 4.19.
˜
Compound 10a (two diastereomers): IR (neat): ν ϭ 3503, 2955,
1
1731, 1652, 1454, 1362 cmϪ1. H NMR: δ ϭ 7.20 (m, 5 H), 4.60
(m, 1 H), 4.40 (s broad, 1 H), 4.15Ϫ3.80 (m, 2 H), 3.76Ϫ2.20 (m,
10 H), 2.20Ϫ1.80 (m, 2 H), 1.80Ϫ1.40 (m, 3 H), 1.28Ϫ0.80 (m, 4
H) ppm. 13C NMR: δ ϭ 174.1, 142.5, 138.8, 128.8, 128.3, 127.1,
99.8, 90.6, 67.1, 61.8, 60.9, 54.1, 53.7, 49.1, 46.3, 29.6, 25.6, 25.5,
14.3 ppm. MS: m/z (%)ϭ 283 (30), 265 (100), 191 (11), 91 (44).
Methyl 3-Benzyl-6-hydroxy-7-methyl-9-oxo-3-azabicyclo[3.3.1]non-
ane-1-carboxylate (4d, from 1a and 2c, four diastereomers): Rf ϭ
0.19 (Et2O/pentane, 7:3). IR (neat): ν˜ ϭ 3480, 3080, 2960, 2840,
1
1740, 1470, 1280, 1130 cmϪ1. H NMR: δ ϭ 7.27 (m, 5 H), 4.82
Compound 10b: IR (neat): ν˜ ϭ 2938, 2799, 1729, 1647, 1457, 1359,
1256, 1212 cmϪ1. 1H NMR: δ ϭ 7.18 (m, 5 H), 4.62 (m, 1 H), 4.05
(m, 2 H), 3.70Ϫ1.15 (m, 18 H), 0.98 (t, J ϭ 7.0 Hz, 3 H) ppm. 13C
NMR: δ ϭ 174.1, 143.5, 138.6, 128.6, 128.2, 126.8, 99.0, 74.1, 60.5,
60.4, 53.8, 48.9, 47.3, 44.7, 30.5, 30.3, 27.3, 14.1 ppm. MS: m/z
(%) ϭ 282 (20), 264 (100), 209 (12), 91 (61).
(s, 1 H), 4.56 (m, 1 H), 3.71 (s, 3 H), 3.45Ϫ1.40 (m, 8 H), 1.19 (d,
J ϭ 6.3 Hz, 3 H), 1.09 (d, J ϭ 6.3 Hz, 3 H) ppm. 13C NMR: δ ϭ
210.5, 209.2, 171.9, 171.0, 138.1, 138.0, 129.212, 129.0, 128.8,
128.6, 128.5, 128.4, 128.1, 127.6, 127.5, 127.4, 127.1, 80.0, 78.0,
61.8, 61.6, 61.1, 60.9, 59.6, 59.2, 58.5, 57.1, 54.6, 56.2, 56.1, 54.0,
53.5, 52.4, 41.1, 40.3, 39.3, 36.5, 36.0, 34.6, 19.5, 18.2 ppm. Ele-
mental analysis calcd. (%) for C18H23NO4 (317.4): C 68.12, H 7.30,
N 4.41; found C 68.47, H 7.43, N 4.29.
Compound 10c (two diastereomers): IR (neat): ν˜ ϭ 2931, 1725, 1653,
1555, 1448, 1209 cmϪ1 1H NMR: δ ϭ 7.21 (m, 5 H), 5.10Ϫ4.50
.
(m, 1 H), 4.09 (m, 2 H), 3.80Ϫ2.65 (m, 7 H), 2.50Ϫ1.18 (m, 5 H),
1.09 (t, J ϭ 7.1 Hz, 3 H) ppm. 13C NMR: δ ϭ 174.0, 141.2, 138.7,
128.8, 128.2, 127.1, 101.1, 65.4, 61.9, 60.9, 60.0, 53.3, 50.9, 48.2,
32.1, 29.4, 28.1, 14.3 ppm. MS: m/z (%) ϭ 285 (37), 267 (100),
91 (89).
Methyl 2-Benzyl-4a-hydroxy-6-oxooctahydroisoquinoline-8a-carb-
oxylate (6): Rf ϭ 0.30 (Et2O/pentane, 9:1). H NMR: δ ϭ 7.25 (m,
1
5 H), 4.36 (s broad, 1 H), 3.67 (s, 3 H), 3.59Ϫ3.42 (m, 1 H), 3.45
(ABq, J ϭ 12.5 Hz, 2 H), 2.92 (m, 2 H), 2.72Ϫ1.96 (m; 9 H) ppm.
13C NMR: δ ϭ 208.2, 176.9, 138.0, 128.8, 128.5, 127.5, 78.7, 62.6,
62.1, 56.1, 53.7, 52.3, 51.2, 37.9, 34.7, 31.8 ppm. It was not possible
to separate the compound from remaining starting material and
elemental analysis was not performed.
Formation of Tosylate 11:[30] The hydroxyazabicyclic compound
equatorial-OH-4a (1.64 mmol), pyridine (3.28 mmol), and tosyl
chloride (2.46 mmol) were dissolved in 5 mL of chloroform (previ-
ously filtered through a short pad of basic alumina). After the mix-
Methyl 2-Benzyl-6-oxo-3,4,6,7,8,8a-hexahydro-1H-isoquinoline-8a-
carboxylate (7):[29] Rf ϭ 0.44 (Et2O/pentane, 9:1). IR (neat): ν˜ ϭ ture had been stirred for
2
h
at room temperature, Et3N
2953, 2805, 1728, 1673, 1453, 1350, 1213 cm1. 1H NMR: δ ϭ (2.46 mmol) and tosyl chloride (2.46 mmol) were added and the
7.33Ϫ7.27 (m, 5 H), 5.89 (s, broad, 1 H), 3.68 (s, 3 H), 3.50 (ABq,
2 H, J ϭ 13.4 Hz), 3.38 (d, broad, J ϭ 10.9 Hz, 1 H), 2.95 (m, 1
reaction mixture was stirred at room temperature until completion
(TLC). The organic layer was washed successively with a saturated
H), 2.82 (m, 1 H), 2.34Ϫ2.09 (m, 5 H), 1.80 (d, broad, J ϭ 11.2 Hz, solution of NaHCO3 (10 mL) and a saturated solution of NH4Cl
Eur. J. Org. Chem. 2002, 3359Ϫ3364 3363

C. Simon, J.-F. Peyronel, F. Clerc, J. Rodriguez
FULL PAPER
43, 2335. M. Kawai, M. Onaka, Y. Izumi, Bull. Chem. Soc.
Jpn. 1988, 61, 2157. B. C. Ranu, S. Bhar, Tetrahedron 1992, 48,
1327. B. C. Ranu, M. Saha, S. Bhar, Tetrahedron Lett. 1993, 34,
1989. R. Sreekumar, P. Rugmini, R. Padmakumar, Tetrahedron
Lett. 1997, 38, 6557. S. Sebti, H. Boukhal, N. Hanafi, S. Bou-
laajaj, Tetrahedron Lett. 1999, 40, 6207.
C. J. Schmildle, R. C. Mansfield, Ind. Eng. Chem. 1952, 44,
1388.
S.-I. Yamada, I. Chibata, R. Tsurui, J. Pharm. Soc. Jpn. 1953,
73, 123.
(10 mL), dried with MgSO4, and filtered, and the solvent was evap-
orated under reduced pressure. The crude product was purified by
flash chromatography on SiO2.
Methyl
[3.3.1]nonane-1-carboxylate (11, equatorial-OTs): Rf ϭ 0.6 (Et2O/
pentane, 9:1). IR (neat): ν˜ ϭ 1735, 3747, 2361 cmϪ1 1H NMR:
3-Benzyl-9-oxo-6-(toluene-4-sulfonyloxy)-3-azabicyclo-
[7]
[8]
.
δ ϭ 7.27 (m, 9 H), 4.68 (m, 1 H), 3.71 (s, 3 H), 3.62 (d, J ϭ 12.0 Hz,
1 H), 3.43 (d, J ϭ 12.0, 1 H), 3.15 (dd, J ϭ 11.5, 2.3 Hz, 1 H), 2.90
(dd, J ϭ 11.6, 1.8 Hz, 1 H), 2.44 (s, 3 H), 2.12 (m, 3 H) ppm. 13C
NMR: δ ϭ 206.2, 170.3, 145.1, 137.5, 133.7, 130.0, 128.7, 128.5,
127.7, 127.5, 80.5, 61.6, 61.2, 58.1, 54.6, 52.5, 29.7, 27.8, 21.7 ppm.
Elemental analysis calcd. (%) for C24H27NO6S (457.5): C 62.86, H
5.93, N 3.05; found C 62.35, H 6.35, N 3.02.
[9]
E. D. Bergmann, R. Corett, J. Org. Chem. 1956, 21, 107; E. D.
Bergmann, R. Corett, J. Org. Chem. 1958, 23, 1507.
J. M. Miller, K.-H. So, J. H. Clark, J. Chem. Soc., Chem. Com-
mun. 1978, 466. R. Ballini, M. Petrini, G. Rosini, Synthesis
1987, 711. R. Ballini, P. Marziali, A. Mozzicafreddo, J. Org.
Chem. 1996, 61, 3209.
[10]
[11] [11a]
[11b]
Fragmentation of the Equatorial Tosylate 11. Preparation of Azacy-
clooctene (12): DBU (0.22 mmol) was added to a solution of 11
(0.22 mmol) in 15 mL of anhydrous methanol and the reaction mix-
ture was heated under reflux for 6 h. The solvent was removed and
distilled water (15 mL) was added before extraction of the mixture
with diethyl ether (3 ϫ 20 mL). The combined organic layers were
dried with MgSO4 and the solvents were evaporated under reduced
pressure. The crude product was purified by flash chromatography
on SiO2.
J. Rodriguez, Synlett 1999, 505.
D. Bensa, J.-M. Bru-
[11c]
nel, G. Buono, J. Rodriguez, Synlett 2001, 715.
C. Simon,
J.-F. Peyronel, J. Rodriguez, Org. Lett. 2001, 3, 2145.
A rapid screening showed the inefficency of CH2Cl2, THF, and
acetone, which gave no transformation at all even after 5 days
at room temperature, while only 10% conversion was observed
with iPrOH under the same conditions. EtOH was used in the
case of the (ethoxycarbonyl)piperidones 1b and 1c.
For a comprehensive review on azabicyclo[3.3.1]nonanes, see:
R. Jeyaraman, S. Avila, Chem. Rev. 1981, 81, 149.
M.-H. Filippini, R. Faure, J. Rodriguez, J. Org. Chem. 1995,
60, 6872.
[12]
[13]
[14]
[15]
[16]
Dimethyl 1-Benzyl-1,4,5,8-tetrahydro-2H-azocine-3,3-dicarboxylate
(12): Rf ϭ 0.63 (Et2O/pentane, 9:1). IR (neat): ν˜ ϭ 1724, 1456
For a comprehensive review on bicyclo[3.3.1]nonanes, see: J. A.
Peters, Synthesis 1979, 321.
1
cmϪ1. H NMR: δ ϭ 7.22 (m, 5 H), 5.67 (m, 1 H), 3.66 (s, 6 H),
3.61 (m, 2 H), 3.33 (m, 2 H), 2.62Ϫ2.14 (m, 7 H) ppm. 13C NMR:
δ ϭ 172.2, 140.1, 131.6, 128.4, 126.9, 126.7, 63.4, 57.3, 56.7, 54.67,
52.3, 32.9, 23.3 ppm. Elemental analysis calcd. (%) for: C18H23O4N
(317.4): C 68.12, H 7.30, N 4.41; found C 68.50, H 7.35, N 4.40.
For recent utilization of closely related aldimines derived from
3a, see: R. Grigg, M. Thornton-Pett, G. Yoganathan, Tetrahed-
ron 1999, 55, 1763 and 8129.
[17]
˚
For the formation of imines in the presence of 4 A molecular
sieves, see: K. Taguchi, F. H. Westheimer, J. Org. Chem. 1971,
36, 1570.
[18]
[19]
For a recent application from this laboratory of this reactivity
and a mechanistic discussion, see ref.[11c]
A. S. elAzab, T. Taniguchi, K. Ogasawara, Org. Lett. 2000, 2,
2757. X. Zhu, N. H. Greig, H. W. Holloway, N. F. Whittaker,
A. Brossi, Q.-S. Yu, Tetrahedron Lett. 2000, 41, 4861. A. R.
Katritzky, G. Qiu, H.-Y. He, B. Yang, J. Org. Chem. 2000, 65,
3683 and references cited therein.
Acknowledgments
We gratefully acknowledge Aventis-Pharma and the CNRS for pro-
viding generous support of this work and a research grant for C.
Simon’s Ph.D. thesis (BDI 740653/01).
[1]
[20]
´
´
See a special issue on environmental chemistry: Chem. Rev.
For a recent example, see: J. M. Andres, I. Herraiz-Sierra, R.
´
1995, 95, 1.
Pedrossa, A. Perez-Encabo, Eur. J. Org. Chem. 2000, 1719.
[2]
[21]
[22]
T. R. E. Kressman, The Industrial Chemist, 1960, p. 3. S. J.
N. A. Petasis, M. A. Patane, Tetrahedron 1992, 48, 5757.
For a review on medium-ring heterocycles, see: P. A. Evans, A.
B. Holmes, Tetrahedron 1991, 47, 9131. For a recent paper re-
lating to β-carboline alkaloids, see: K. A. El Sayed, M. Kelly,
U. A. K. Kara, K. K. H. Ang, I. Katsuyama, D. C. Dunbar, A.
A. Khan, M. T. Hamann, J. Am. Chem. Soc. 2001, 123, 1804.
A. Gambacorta, S. Turchetta, S. Stefanelli, Tetrahedron Lett.
1991, 32, 6805.
G. L. Buchanan, Topics in Carbocyclic Chemistry 1969, 199.
M. Hesse, Ring Enlargement in Organic Chemistry; VCH Ver-
lagsgesellschaft: Weinheim, 1991.
C. A. Grob, Angew. Chem. Int. Ed. Engl. 1969, 8, 535.
Under the same experimental conditions the axial-OTs-11 gave
a complex mixture of inseparable olefinic compounds, probably
through a combination of direct elimination and retro-Dieck-
mann fragmentation: see for example: G. L. Buchanan, G. W.
McLay, Tetrahedron 1966, 22, 1521.
Shuttleworth, S. M. Allin, P. K. Sharma, Synthesis 1997, 1217.
N. Mizuno, M. Misono, Chem. Rev. 1998, 98, 199. H. Hattori,
Chem. Rev. 1995, 95, 537. See also a special issue on heterogen-
eous catalysis: Chem. Rev. 1995, 95, 475.
[3]
G. H. Posner, Synthesis 1978, 487. P. Laszlo, Acc. Chem. Res.
[23]
[24]
1986, 19, 121. R. Kloetstra, H. V. Bekkum, J. Chem. Soc.,
Chem. Commun. 1995, 1005. G. W. Kabalka, R. M. Pagni, Tet-
rahedron 1997, 53, 7999.
[4]
[4a] P. Hodge, E. Khoshdel, J. Waterhouse, J. Chem. Soc., Chem.
[4b]
Commun. 1983, 2205.
B. P. Bandgar, M. B. Zirange, P. P.
[4c]
[25]
[26]
Wadgaonkar, Synlett 1996, 149.
D. J. Macquairrie, Tetra-
[4d]
hedron Lett. 1998, 39, 4125.
C. Fava, R. Galeazzi, E. M.
Gonzalez-Rosende, M. Orena, D. J. Macquairrie, Tetrahedron
Lett. 1998, 39, 4125.
[5]
A. McKillop, D. W. Young, Synthesis 1979, 401; A. McKillop,
D. W. Young, Synthesis 1979, 481. P. Laszlo, P. Pennetreau,
Tetrahedron Lett. 1985, 26, 2645. P. Laszlo, M.-T. Montaufier,
S. L. Randriamahefa, Tetrahedron Lett. 1990, 31, 4867. H. Kot-
suki, K. Arimura, Tetrahedron Lett. 1997, 38, 7583. J. E. G.
Mdoe, J. H. Clark, D. J. Macquarrie, Synlett 1998, 625. B. M.
Choudary, M. L. Kantan, B. Kativa, Ch. V. Reddy, F. Figueras,
Tetrahedron 2000, 56, 9357. J.-I. Tateiwa, A. Hosomi, Eur. J.
Org. Chem. 2001, 1445.
[27]
[28]
W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 2923.
C. Cope, M. E. Synerholm, J. Am. Chem. Soc. 1950, 72, 5228.
B. R. Baker, H. S. Shapiro J. Med. Chem. 1963, 6, 664.
D. I. MaGee, M. L. Lee, A. Decken, J. Org. Chem. 1999, 64,
2549.
[29]
[30]
G. L. Buchanan, G. W. McLay, Tetrahedron 1966, 22, 1521.
Received January 22, 2002
[6]
M. Iglesias, J. M. Marinas, J. V. Sinisterra, Tetrahedron 1987,
[O02028]
3364
Eur. J. Org. Chem. 2002, 3359Ϫ3364