Acid-catalyzed rearrangement of 1-benzyl-2-methyl-3-piperidone to 1-benzyl-2-acetylpyrrolidine

Article abstract of DOI:10.1016/j.tet.2006.04.038

We report that 1-benzyl-2-methyl-3-piperidone, conveniently prepared from 3-hydroxy-2-methylpyridine, undergoes rearrangement to 1-benzyl-2-acetylpyrrolidine in aqueous 6 N HCl at reflux. Studies showing that the 2,2-dimethyl analog is inert under the same conditions support a mechanism of reversible tautomeric equilibria via ring-opened intermediates, one of which was independently synthesized and shown to be a kinetically competent intermediate to product.

Full text of DOI:10.1016/j.tet.2006.04.038

Tetrahedron 62 (2006) 6361–6369
Acid-catalyzed rearrangement of 1-benzyl-2-methyl-3-piperidone
to 1-benzyl-2-acetylpyrrolidine
*
Shengyin Zhao, Heung-Bae Jeon, Durgesh V. Nadkarni and Lawrence M. Sayre
Department of Chemistry, Case Western Reserve University, Cleveland, OH 44106, USA
Received 15 January 2006; revised 8 April 2006; accepted 11 April 2006
Available online 19 May 2006
Abstract—We report that 1-benzyl-2-methyl-3-piperidone, conveniently prepared from 3-hydroxy-2-methylpyridine, undergoes rearrange-
ment to 1-benzyl-2-acetylpyrrolidine in aqueous 6 N HCl at reflux. Studies showing that the 2,2-dimethyl analog is inert under the same con-
ditions support a mechanism of reversible tautomeric equilibria via ring-opened intermediates, one of which was independently synthesized
and shown to be a kinetically competent intermediate to product.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
our knowledge, been reported before. A possibly related
rearrangement was observed more than 50 years ago by
Leonard and co-workers in Clemmensen reduction of cyclic
amine 3-ones, exemplified in Eq. 2. The rearrangement pro-
ceeded smoothly for 2-unsubstituted, mono 2-alkyl, and
2,2-dialkyl reactants 3a–d.7 The mechanism of Clemmensen
reduction had only been postulated at that time, and Leonard
rationalized the rearrangement he observed in terms of car-
bonyl O-protonation followed by 1,2-shift of nitrogen to the
adjacent carbonyl carbon, with concomitant8 or subsequent9
delivery of a zinc-derived hydride equivalent to the migration
origin (Eq. 3). Such mechanism would not accommodate the
nonreductive rearrangement we are now reporting.
Piperidine and pyrrolidine rings are ubiquitous structural
features of many alkaloid natural products1 and drug candi-
dates,2 and their derivatives are important intermediates in
organic synthesis. We have been interested in the chemistry
of reactive intermediates generated in the oxidative metabo-
lism of the N-alkyl tertiary amine ring systems. In particular,
we have been interested in the oxygenation of endocyclic
enamines in equilibrium with the initial endocyclic iminium
forms that result from cytochrome P450-mediated metabo-
lism (e.g., Eq. 1). During an investigation into the chemistry
of N-benzyl-3-piperidones,3,4 resulting from autoxidation of
the corresponding N-benzyl-D2-tetrahydropyridines, we
found that the 2-methyl compound 1 rearranges in hot
aqueous HCl to 2-acetylpyrrolidine 2.
Although there may be no relationship between the rear-
rangements observed by Leonard and by us, two conceivable
mechanisms that could be considered for both are shown in
Eqs. 4 and 5, assuming that under Clemmensen conditions,
these 3-piperidones underwent rearrangement prior to
zinc-mediated reduction. The mechanism in Eq. 4 would ex-
plain our observed conversion of 1 to 2, as well as Leonard’s
observations for 3a and 3b, whereas for the 2,2-dialkyl reac-
tants 3c and 3d, reduction would have to occur at the level of
intermediate 6. Another possible mechanism (Eq.5) would
involve two consecutive 1,2-alkyl shifts, based on the aza
equivalent of the a-ketol rearrangement,10 namely the inter-
conversion between a-hydroxy imines and a-amino
ketones,11 which has been observed (a-amino aldehyde) to
be catalyzed by Lewis acids.12 Again, the mechanism in
Eq. 5 would explain our observed conversion of 1 to 2 and
Leonard’s observations for 3a and 3b, whereas in the 2,2-
dialkyl reactants 3c and 3d an additional rapid 1,2-alkyl shift
would be needed, with reduction occurring only after gener-
ation of an intermediate 6.
O
O2
aq. HCl
∆
P450
(1)
N
N
N
+
N
N
O
Ph
1
2
Ph
Ph
Ph
Ph
A number of nonoxidative5 and oxidative6 examples of
rearrangements of a-amino ketones have been observed, but
the exact type of rearrangement we observed has not, to
Keywords: Rearrangement; Piperidone; Pyrrolidone; Clemmensen reduc-
tion.
* Corresponding author. Tel.: +1 216 368 3704; fax: +1 216 368 3006;
e-mail: lms3@case.edu
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.04.038

6362
S. Zhao et al. / Tetrahedron 62 (2006) 6361–6369
Synthesis of the 2,2-dimethyl-3-piperidone 10 was achieved
O
according to the route shown in Scheme 2.16 Heating benzyl-
amine with ethyl 2-bromoisobutyrate in ethanol in the pres-
ence of a catalytic amount of KI gave 11 in 52% yield.
Conversion of secondary amine 11 to tertiary amine 12
with ethyl 4-bromobutyrate under the same conditions was
more sluggish, even in refluxing n-butanol, but could be
achieved by heating in solvent-free conditions. Dieckmann
condensation of diester 12 with sodium ethoxide in THF
gave keto ester 13 (as the enol) in a low yield, but a more
satisfactory result (95% yield) was obtained using 60%
sodium hydride. Hydrolytic decarboxylation (6 N HCl)
afforded 10 in 89% yield.
a: R' = R'' = H
aq. HCl
Zn(Hg)
R'
R''
b: R' = H, R'' = CH2CH3
(2)
(3)
R'
R''
N
CH3
c: R' = R' = CH3
d: R' = CH3, R'' = CH2CH3
N
CH3
3
4
+
OH
OH
R'
[H]Zn
R'
R''
H
N
R
N
R
N
R
R"
R''
R'
+
OH
OH
O
Ph
NH2
OH
H+
H2O
K2CO3, KI
Br(CH2)3CO2Et
KI, 48 h
+
R'
R''
R'
R''
R'
R''
OH
Ph
N
H
CO2Et
11
N
R
N
R
N
R
N
R
+
C2H5OH, 60 h
R''
CO2Et
R'
Br
5
EtO
O
CO2Et
CO2Et
OH
H+, - H2O
H
O
OH
R'
H+
O
CO2Et
R'' = H
O
+
(4)
(5)
R''
N
O
N
R
N
R
6 N HCl
reflux, 4 h
R'
60% NaH, THF
reflux, 4 h
R'
R
N
N
6
N
N
Ph
10
Ph
12
13
Ph
Ph
+
H
R'
O
O
O
OH
Scheme 2.
H+
- H+
R'
R''
R
R'
R''
N
R
N
R
N
R"
+N
R
R''
2.2. Rearrangement of 1-benzyl-2-methyl-3-piperidone(1)
to 1-benzyl-2-acetylpyrrolidine (2)
R
Heating 1 in 6 N aqueous HCl at reflux for 24 h under argon
resulted in its rearrangement in 90% yield (remaining is
recovered 1) to 2-acetylpyrrolidine 2 (Scheme 3), accom-
panied by a trace amount of 1-benzyl-2-pyrrolidine 14,
representing oxidation of rearrangement product 2.17
Compound 2 was also obtained directly from 9 by heating
the latter in 6 N HCl at reflux for 30 h. The identity of 2
was verified by comparison with an authentic sample
prepared from 14 by reduction of the latter with Red-Al
followed by treatment with aqueous HCN solution, and
reaction of the resulting nitrile 1518 with methyllithium in
benzene, followed by hydrolytic workup.
The purpose of this study was to clarify the mechanism of
the nonreductive rearrangement responsible for conversion
of 1 to 2. Based on these results, we further propose a mech-
anism for the rearrangement observed under Clemmensen
conditions that is consistent with recent views on the chem-
istry of this reduction process.
2. Results and discussions
2.1. Preparation of 1-benzyl-3-piperidones 1 and 10
The required 2-methyl-3-piperidone 1 was prepared from
commercially available 3-hydroxy-2-methylpyridine in
four steps (Scheme 1). The key N,O-dibenzyl-protected
intermediate 813 was most conveniently obtained by the two-
step benzylation sequence as shown. Yield of 8 obtained
without isolation of 7 was inferior. Reduction (NaBH4) of
8 and O-deprotection of the resulting 914 by treatment with
6 N HCl at room temperature gave 115 in 83% yield. This
scheme might be adapted to offer a useful synthesis of other
3-piperidone analogs.
O
6 N HCl
reflux, 24 h
N
N
CH3
1
Ph
O
O
N
+
N
Ph
O
Ph
Ph
6 N HCl
reflux, 30 h
2
14
CH3
9
Ph
1) CH3Li, ether
2) 1 N HCl
CN
1) Red-Al, ether
O
N
N
N
1) Bu4N+HO-, CH3CN
2) HCN, CH3CO2H
OH
O
Ph
O
3) K2CO3, pH 7
PhCH2Br, CH3CN
reflux, 24 h, 76%
r.t., 30 min
2
14
Ph
15 Ph
Ph
2) PhCH2Br, CH3CN
reflux, 1 h, 92%
N
CH3
N
O
CH3
Ph
Scheme 3.
7
O
Ph
O
Various reaction conditions were tested in order to evaluate
the ease of rearrangement of 1. Compound 1 was generated
from 9 by exposure to 6 N HCl at room temperature for 24 h,
and we verified that 1 is stable under these conditions.
Heating 1 at reflux in 6 N HCl for periods of time shorter
NaBH4, CH3OH
3 h, 76%
1) 6 N HCl, r.t., 24 h
2) K2CO3, 83%
N
CH3
N
CH3
N
CH3
Br-
Ph
Ph
Ph
1
8
9
Scheme 1.

S. Zhao et al. / Tetrahedron 62 (2006) 6361–6369
6363
than 24 h resulted in a lower yield of conversion to 2 (e.g.,
55% in 6 h). The importance of acid concentration in this re-
arrangement was easily recognized by the finding that heat-
ing 1 at reflux in 0.5 N HCl for 6 h gave 2 in only 38% yield.
enol/iminium stability of the intermediates, or the instability
of the aldehyde equivalents of D and 2.
Another possible pathway, based on the aza equivalent of the
a-ketol rearrangement,11,12 is that a-hydroxy iminium inter-
mediate B could undergo two subsequent 1,2-alkyl shifts,
first a ring contraction to give J and then a methyl shift to
give H. Although this mechanism would involve a lesser
number of steps, we believe that the sequence of tautomeri-
zations through D and E provides a more likely overall low
energy pathway.
2.3. Attempted rearrangement of 1-benzyl-2,2-
dimethyl-3-piperidone (10)
The mechanism considered in Eq. 5 would predict that 2,2-
dimethyl substitution would still allow rearrangement to a
2-acylpyrrolidine. According to Eq. 4, whereas 2,2-dimethyl
substitution would prevent rearrangement to a 2-acylpyrroli-
dine, the reaction could still proceed to the stage of inter-
mediates 5/6. Thus, according to these mechanisms, one
might predict that 1-benzyl-2,2-dimethyl-3-piperidone (10)
would undergo, in refluxing aqueous 6 N HCl, conversion
to either 2-acetyl-2-methylpyrrolidine or to a product derived
from 5/6, possibly the ring-opened amino ketone. However,
heating 10 in 6 N HCl at reflux for 24 h resulted in total
recovery of starting material. The inertness of 10 is inconsis-
tent with Eq. 4 or 5 as the mechanism of the nonreductive
rearrangement, and suggests consideration of a mechanism
relying on an initial enolization step. At the same time, we
found that the unsubstituted parent compound 1-benzyl-3-
piperidone also does not undergo the rearrangement.
To determine whether the sequence of tautomerizations
through D and E provides at least a kinetically competent
mechanism for the rearrangement of 1 to 2, it would be
important to show that one of the two proposed ring-opened
intermediates could convert to 2 at least as rapidly as 1 con-
verts to 2 under the same reaction conditions. We chose
to prepare intermediate D, which could be obtained by
deprotection of the dimethyl ketal 21, independently synthe-
sized as shown in Scheme 5. On the basis of a previous
report,22 g-chloroketone 19 was selected as the key synthetic
intermediate, and was obtained by alkylation with 1-bromo-
2-chloroethane of the lithium derivative formed from
deprotonation of the N-isopropylimine of butane-2,3-dione
monoketal 16. Without isolation of the intermediate 18 fol-
lowing workup, selective hydrolysis of the imino function in
the presence of the ketal function was conveniently achieved
using aqueous oxalic acid. Alkylation of benzylamine with
19, and NaBH4 reduction of the resulting amino ketone 20
without its isolation, afforded 21. An attempt to prepare 21
by first reducing 19 and then reacting with benzylamine
was abandoned due to sluggish and low yield reactions.
Heating 21 in 6 N HCl at reflux was indeed found to give
2, presumably through initial hydrolysis to intermediate D
and rearrangement of the latter.
2.4. Mechanism of rearrangement
Two plausible mechanisms that would explain the rearrange-
ment of 1 to 2 but the inertness of 10 are outlined in Scheme
4. In acid, the 3-piperidone would exist in equilibrium with
enol–enamine A and iminium form B. The latter would fur-
ther be in equilibrium with the hydrate C and ring-opened
ketone D, with the overall conversion of 1 to D correspond-
ing to the reverse of the well-known Amadori rearrangement
that occurs following condensation of reducing sugars with
amines.19 If ketol D could be interchanged in acid with ketol
E,20 then the latter would be in equilibrium with 2 through
the reverse sequence of analogous intermediates containing
a five- rather than a six-membered ring. This would suggest
that the conversion of 1 to 2 merely represents an equilib-
rium process and the thermodynamic predominance of 2.
Precedent for these types of proposed keto–enol tautomeri-
zations has been reported.21 Why the unsubstituted parent
1-benzyl-3-piperidone would not undergo rearrangement
according to this mechanism may reflect the insufficient
On the other hand, deprotection of 21 by treatment with 6 N
HCl for 24 h at room temperature followed by neutralization
gave 1 (Scheme 6). Assuming Scheme 4 is operative, this
result suggests that the interconversion between inter-
mediate D (six-membered ring manifold) and intermediate
E (five-membered ring manifold) is mainly rate-limiting
in the rearrangement of 1 that requires higher temperature.
At the same time, this result also suggests the possibility
that in acid, 21 first undergoes conversion to 1, which at
O
O
H
OH
O
OH
O
OH
H+
H+, - H+
H
N
N
N
N
∆
+
NH
Ph
Ph
Ph
C
Ph
A
B
1
D
- H+
Ph
O
O
J
OH
N
NH
E
Ph
H+
Ph
OH2
OH
OH
OH
H+
N
N
N
N
N
OH
Ph
OH
O
G
I
H
F
Ph
Ph
Ph
Ph
2
Scheme 4.

6364
S. Zhao et al. / Tetrahedron 62 (2006) 6361–6369
O
N
N
TiCl4, ether
LDA, THF
NH2
+
Cl
ClCH2CH2Br
CH3O OCH3
CH3O OCH3
CH3O OCH3
16
17
18
O
O
H
N
benzylamine
NEt3, CH3CN
oxalic acid
Cl
CH2Cl2/H2O
CH3O OCH3
CH3O OCH3
20
19
O
OH
H
1) 6 N HCl, reflux, 24 h
2) K2CO3, pH=8
N
NaBH4
CH3OH
N
CH3O OCH3
21
2
Scheme 5.
O
OH
OH
H
N
6 N HCl
r.t., 24 h
K2CO3
N
NH2
Cl-
pH = 8
CH3O OCH3
O
21
22
1
Scheme 6.
elevated temperature then proceeds to 2 along a route that
is distinct from that shown in Scheme 4. However, evidence
that D is a kinetically competent intermediate on the lowest
energy pathway from 1 to 2, was provided by following
the time course of conversion of both 1 and 21 to 2 under the
conditions of refluxing 6 N HCl. As shown in Figure 1, the
fraction of 2 (remainder is 1), as determined by neutraliza-
tion and extraction, was significantly greater at early time
(1–3 h) when starting with 21 than with 1. After 24 h, how-
ever, the amount of 2 (90%) was the same, and this appears
to be close to the equilibrium distribution between 1 and 2
under the conditions of refluxing 6 N HCl.
Additional information on the mechanism was provided by
following the fate of 21 in 6 N deuterium chloride–deute-
rium oxide in an NMR tube. At room temperature, deprotec-
tion occurred immediately to reveal signals expected for the
protonated form of 22. Signals at d 2.1 (CH3) and d 4.4 (CH)
gradually disappeared over the course of 24 h due to deute-
rium exchange, but otherwise no other changes were ob-
served. Although rapid neutralization and extraction at this
point affords 1, this result suggests that in the strongly acidic
medium at room temperature, ring-opened amino ketone D
does not interconvert with E or any of the ring-closed amino
ketone forms.
100
Compound 1 in 6 N HCl
Compound 21 in 6 N HCl
80
In the same manner, the solution forms of both the six- and
five-membered amino ketones were investigated, in these
cases most readily by 13C NMR spectroscopy in DCl–
D2O. 1-Benzyl-3-piperidone$HCl is known to exist as its
hydrate in aqueous solution,23 as verified in this study, where
the free base form exhibited a carbonyl signal at 207.1 ppm,
whereas for the DCl salt in D2O, this signal was replaced by
one at 91.6 ppm assigned to the carbon bearing two OH
groups. The 13C NMR spectrum of the 2-methyl analog 1
shows that it also exists mainly as the hydrate in 6 N DCl–
D2O, as apparent from the lack of a carbonyl signal and pres-
ence of a geminal dihydroxy 13C signal at 93.8 ppm. The
2,2-dimethyl analog 10 in 6 N DCl–D2O exists as a mixture
of ketone and hydrate forms (13C NMR signals are seen at
both 207.6 and 95.3 ppm). On the other hand, 1-benzyl-2-
acetylpyrrolidine 2 exists as such in 6 N DCl–D2O (13C
NMR carbonyl signal at 205.0 ppm). The tendency of the
3-piperidones but not 2 to form the hydrate likely reflects
60
40
20
0
0
5
10
15
20
25
Time (hr)
Figure 1. The relative amount of rearrangement product 2 formed from
3-piperidone 1 or compound 21 in 6 N HCl with heating at reflux for dif-
ferent reaction times, followed by neutralization and product analysis by
1H NMR spectroscopy.

S. Zhao et al. / Tetrahedron 62 (2006) 6361–6369
6365
the well-known tendency of six-membered rings to become
4. Experimental
entirely sp3 tetrahedral to minimize torsional strain,24,25 with
the lack of hydration of 2 indicating that there is no intrinsic
electronic preference for protonated a-amino ketones to
exist as their hydrates.
4.1. General
1
All melting points are uncorrected. H NMR spectra were
obtained with Varian Gemini 200 MHz (13C NMR at
50 MHz) or 300 MHz (13C NMR at 75 MHz) instruments.
In all cases, tetramethylsilane (TMS) or the solvent peak
served as internal standard for reporting chemical shifts,
which are expressed as parts per million downfield from
TMS (d scale). In the 13C NMR line listings, attached proton
test designations are given as (+) or (ꢁ) following the chem-
ical shifts. High-resolution mass spectra (HRMS) using
electron impact (EI) or fast atom bombardment (FAB),
were obtained at 20–40 eVon a Kratos MS-25A instrument.
TLC was performed on glass plates precoated with silica
gel 60GF254. Compounds on the developed plate were
visualized by short-wavelength UV light (l¼254 nm) or
by spraying with either ninhydrin or 2,4-dinitrophenyl-
hydrazine solutions. All evaporations were conducted at
reduced pressure using a rotary evaporator. All new com-
As expected from these results, when the NMR tube contain-
ing 21 in 6 N DCl–D2O was heated at 95 ꢀC, the 13C NMR
spectrum after 5 h revealed a mixture of 1 (as its hydrate) and
2, whereas continued heating resulted in mainly 2, as is seen
when starting with 1. The main importance of these experi-
ments is the demonstration that equilibration between the
ring-opened and ring-closed forms of both the six- and
five-membered ring manifolds shown in Scheme 4 occurs
only at elevated temperature. Thus, although the D#E in-
terconversion has been ascribed to be the main rate-limiting
step in the rearrangement, other steps in the overall mecha-
nism may additionally be partially rate-limiting.
Over the years, although the mechanism of Clemmensen re-
duction has still not been totally elucidated, the most recent
studies on this subject suggest a stepwise electron/proton
transfer mechanism via radical and probably alkyl chloride
intermediates,26 with the additional possible intermediacy
of an organozinc species. It is known that the corresponding
carbinols are not intermediates, though dimeric pinacol
side-products are sometimes observed, suggestive of cou-
pling of a a-oxycarbon radical intermediate. A possible
mechanism that follows this line of reasoning and that
would explain the ring-contractions for 3a–d and the race-
mization of 3d observed by Leonard and co-workers9 is
given in Eq. 6. Since the first of the four electron reduction
steps would, according to this mechanism, occur prior to any
rearrangement step, this would then clearly be a distinct
process from the nonreductive rearrangement we proposed
in Scheme 4.
1
pounds were determined to be >95% pure by H NMR
spectroscopy.
4.2. 3-Benzyloxy-2-methylpyridine (7)
A mixture of 2-methyl-3-pyridinol (2.0 g, 0.018 mol) and
tetra-n-butylammonium hydroxide (40% solution in water,
11.68 g, 0.018 mol) in 50 mL of CH3CN was stirred at room
temperature for 30 min and then evaporated to dryness, and
the resulting ammonium salt was dried in vacuo. Benzyl bro-
mide (2.14 mL, 0.018 mol) was added to a solution of the
ammonium salt in 100 mL of CH3CN, and the mixture
was heated at reflux for 1 h. The reaction mixture was cooled
and filtered, and the filtrate was evaporated to afford a yel-
lowish oil. The crude product was purified by silica gel flash
chromatography using EtOAc–CH2Cl2 (1:4) to give 7 as
a pale yellowish oil (3.36 g, 92%): 1H NMR (CDCl3)
d 2.54 (s, 3H), 5.06 (s, 2H), 7.08 (m, 2H), 7.32–7.43 (5H),
8.09 (d, 1H, J¼4.6 Hz); 13C NMR (CDCl3) d 19.6 (ꢁ),
64.9 (+), 117.8 (ꢁ), 121.7 (ꢁ), 127.2 (ꢁ), 128.1 (ꢁ),
128.7 (ꢁ), 136.6 (+), 140.7 (ꢁ), 149.3 (+), 153.0 (+);
EI HRMS m/z calcd for C13H13NO (M+H)+ 199.0998, found
199.0997.
O
OZnX
Zn, ET
H+
OZnX
HCl
Cl
R'
R''
R'
R''
Zn
ET
R'
R''
R'
R''
R'
R''
N
R
N
R
N
R
N
R
R'
- Cl-
Zn
ET
Zn, ET
H+
R'
R''
(6)
N
R
N
R
N
R
+
R''
4.3. 1-Benzyl-3-benzyloxy-2-methylpyridinium
bromide (8)
3. Conclusions
Benzyl bromide (2.4 mL, 0.02 mol) was added to a solution
of 7 (3.36 g, 0.017 mol) in 100 mL of CH3CN. The reaction
mixture was heated at reflux for 24 h under argon. The sol-
vent was then evaporated, and the residue was washed
with diethyl ether (3ꢂ10 mL) and then dried in vacuo to
give pyridinium salt 8 (4.8 g, 76%): mp 186–188 ꢀC;
1H NMR (CDCl3) d 2.72 (s, 3H), 5.28 (s, 2H), 6.20 (s,
2H), 7.24–7.35 (10H), 7.88 (dd, 1H, J¼6.2 and 8.6 Hz),
8.22 (d, 1H, J¼8.6 Hz), 9.16 (d, 1H, J¼6.2 Hz); 13C NMR
(CDCl3, two carbons overlapped) d 14.3 (ꢁ), 62.4 (+),
72.6 (+), 126.2 (ꢁ), 127.1 (ꢁ), 127.9 (ꢁ, 2C), 128.9 (ꢁ,
2C), 129.3 (ꢁ), 129.5 (ꢁ), 132.2 (+), 134.1 (+), 138.1 (ꢁ),
147.2 (ꢁ), 156.0 (+); FAB HRMS m/z calcd for C20H20NO
(M+) 290.1546, found 290.1548.
In summary, we have developed an efficient route to 1-alkyl-
2-methyl-3-piperidones and demonstrated that the 1-benzyl
compound undergoes rearrangement to 1-benzyl-2-acetyl-
pyrrolidine in refluxing 6 N HCl. The reaction reaches an
equilibrium state atw90% conversion, with the driving force
apparently being the thermodynamic preference of a five-
membered ring over a sp2 carbon-containing six-membered
ring because of torsional strain in the latter.24,25 A mecha-
nism has been proposed for the rearrangement involving a
series of acid-catalyzed tautomerization and hydration/
dehydration steps that is distinct from a seemingly similar
rearrangement known to accompany Clemmensen reduction
of the same starting materials.

6366
S. Zhao et al. / Tetrahedron 62 (2006) 6361–6369
4.4. 1-Benzyl-3-benzyloxy-2-methyl-1,2,5,6-tetrahydro-
pyridine (9)
4.7. 1-Benzyl-2-cyanopyrrolidine (15)18
To a solution of 1-benzyl-2-pyrrolidone 14 (ICN Pharma-
ceuticals, 15.0 g, 0.0856 mL) in 300 mL of dry ether
was added 15 mL of Red-Al (3.4 M solution in toluene,
0.051 mol) at 0 ꢀC. The reaction mixture was stirred at
0 ꢀC under argon for 3 h. A cold solution of NaCN
(12.0 g) in 8 mL of H2O was slowly added to the reaction
mixture, and the pH of the reaction mixture was adjusted
to 5 with glacial acetic acid. The solution was stirred for
1 h at room temperature and then extracted with diethyl ether
(3ꢂ50 mL). The combined ether extract was dried (Na2SO4)
and evaporated to yield an oily crude product that was puri-
fied by silica gel flash chromatography using CH2Cl2–
To a solution of 8 (3.0 g, 8.11 mmol) in 50 mL of methanol
at 0 ꢀC was added NaBH4 (0.613 g, 16.2 mmol) over 15 min.
The reaction mixture was stirred at room temperature for 3 h
and then methanol was evaporated. To the solid residue were
added 50 mL of ether and 25 mL of saturated aqueous
K2CO3 solution. The reaction mixture was stirred vigorously
at room temperature for 1 h and separated, and the aqueous
layer was extracted with diethyl ether (3ꢂ20 mL). The com-
bined ether extract was dried (Na2SO4) and evaporated
to furnish 9 as a brownish oil (1.8 g, 76%) that was of satis-
factory purity (NMR) to be used directly in the next step: 1H
NMR (CDCl3) d 1.36 (d, 3H, J¼6.6 Hz), 2.03–2.09 (m, 1H),
2.25–2.35 (m, 1H), 2.19 (m, 1H), 3.24 (q, 1H, J¼6.6 Hz),
3.67 (d, 1H, J¼13.5 Hz), 3.87 (d, 1H, J¼13.5 Hz), 4.73–
4.83 (m, 3H), 7.14–7.46 (10H); 13C NMR (CDCl3)
d 16.2 (ꢁ), 22.4 (+), 44.2 (+), 55.4 (ꢁ), 58.0 (+), 68.6 (+),
92.3 (ꢁ), 126.9 (ꢁ), 127.4 (ꢁ), 127.7 (ꢁ), 128.2 (ꢁ),
128.5 (ꢁ), 128.9 (ꢁ), 137.7 (+), 139.5 (+), 156.5 (+); EI
HRMS m/z calcd for C20H30NO (M+H)+ 293.1781, found
293.1781.
1
EtOAc (4:1) to give 15 as an oil (10.5 g, 66%): H NMR
(CDCl3) d 1.88–2.20 (4H), 2.59 (dd, 1H, J¼8.2 and
10.0 Hz), 2.95 (m, 2H), 3.70 (d, 1H, J¼13.1 Hz), 3.93 (d,
1H, J¼13.1 Hz), 7.35 (5H); 13C NMR (CDCl3) d 22.0 (+),
29.6 (+), 51.3 (+), 53.3 (ꢁ), 56.6 (+), 118.0 (+), 127.6 (ꢁ),
128.56 (ꢁ), 128.57 (ꢁ), 128.60 (ꢁ), 128.9 (ꢁ), 137.7 (+);
EI HRMS m/z calcd for C12H14N2 (M+) 186.1158, found
186.1157.
4.8. 1-Benzyl-2-acetylpyrrolidine (2, alternate method)
4.5. 1-Benzyl-2-methyl-3-piperidone (1)
To a solution of 1-benzyl-2-cyanopyrrolidine 15 (0.62 g,
3.33 mmol) in 30 mL of dry benzene at 0 ꢀC was added
2.84 mL of methyllithium (1.4 M solution in ether,
3.98 mmol) under argon. The reaction mixture was allowed
to warm to room temperature and was then heated at reflux
for 2 h. The reaction solution was hydrolyzed with 1 N
HCl, and then extracted with diethyl ether. The aqueous
phase was separated and neutralized with aqueous K2CO3
solution to pH 7. The solution was then extracted with
CH2Cl2. The combined organic extract was dried (Na2SO4)
and evaporated, and the crude oily product was purified by
silica gel column chromatography using CH2Cl2–EtOAc
(4:1) to give 2 as an oil (0.473 g, 70%). The NMR data
were identical with the data obtained for the rearrangement
product.
A solution of 9 (1.0 g, 3.7 mmol) in 25 mL of 6 N HCl was
stirred at room temperature for 24 h. The reaction mixture
was extracted with ether. The aqueous phase was separated
and neutralized with K2CO3 to pH 8. The solution was
then extracted with diethyl ether (3ꢂ20 mL). The combined
ether layer was dried (Na2SO4) and evaporated to give 1 as
an oil (0.58 g, 83%): 1H NMR (CDCl3) d 1.29 (d, 3H,
J¼6.8 Hz), 1.87–1.96 (2H), 2.30–2.38 (m, 1H), 2.40–2.57
(2H), 2.88–2.96 (m, 1H), 3.18 (q, 1H, J¼6.8 Hz), 3.47 (d,
1H, J¼13.5 Hz), 3.87 (d, 1H, J¼13.5 Hz), 7.23–7.39 (5H);
13C NMR (CDCl3) d 12.2 (ꢁ), 24.2 (+), 37.8 (+), 47.6 (+),
57.8 (+), 66.2 (ꢁ), 127.2 (ꢁ), 128.3 (ꢁ), 128.8 (ꢁ), 138.6
(+), 210.2 (+); EI HRMS m/z calcd for C13H17NO (M+H)+
203.1311, found 203.1310.
4.6. 1-Benzyl-2-acetylpyrrolidine (2)
4.9. Ethyl 2-benzylamino-2-methylpropionate (11)27
A solution of 1 (0.5 g, 2.5 mmol) in 25 mL of 6 N HCl was
heated at reflux for 24 h under argon. The reaction mixture
was extracted with diethyl ether. The aqueous phase was sep-
arated, neutralized with K2CO3 to pH 8, and extracted with
diethyl ether (3ꢂ15 mL). The combined ether layer was
dried (Na2SO4) and evaporated, and the residue was frac-
tionated by flash silica gel chromatography with CH2Cl2–
EtOAc (4:1) as eluent to yield 2 as an oil (0.39 g, 78%):
1H NMR (CDCl3) d 1.76–1.88 (3H), 2.03–2.11 (m, 1H),
2.14 (s, 3H), 2.29 (dd, 1H, J¼8.3 and 12.0 Hz), 3.04–3.13
(2H), 3.44 (d, 1H, J¼13.1 Hz), 3.81 (d, 1H, J¼13.1 Hz),
7.23–7.32 (5H); 13C NMR (CDCl3) d 23.6 (+), 25.3 (ꢁ),
28.8 (+), 53.9 (+), 59.4 (+), 73.6 (ꢁ), 127.2 (ꢁ), 128.4 (ꢁ),
129.0 (ꢁ), 138.6 (+), 212.1 (+); EI HRMS m/z calcd for
C13H17NO (M)+ 203.1311, found 203.1309. A trace of the
known 1-benzyl-2-pyrrolidone (14) was also obtained in
A solution of benzylamine (10.7 g, 0.1 mol), ethyl 2-bromo-
isobutyrate (12.5 mL, 0.085 mol), potassium carbonate
(13.8 g, 0.1 mol), and potassium iodide (166 mg, 1 mmol)
in 120 mL of ethanol was stirred at reflux for 60 h. After
filtration of ethanol to remove suspended salts, the filtrate
was evaporated under reduced pressure, and the residue
was subjected to silica gel column chromatography using
hexane–ethyl acetate (6:1) as eluent to give the known amino
1
ester 11 (9.5 g, 43%) as an oil: H NMR (CDCl3) d 1.31
(t, 3H), 1.38 (s, 6H), 3.63 (s, 2H), 4.18 (q, 2H), 7.33 (5H).
4.10. Ethyl 4-[N-benzyl-N-(1-ethoxycarbonyl-1-methyl-
ethyl)]aminobutyrate (12)
A mixture of compound 11 (4.4 g, 20 mmol), ethyl 4-bromo-
butyrate (3.3 mL, 22 mmol), and potassium iodide (166 mg,
1 mmol) was stirred at 120–130 ꢀC for 48 h. The mixture
was then subjected to silica gel column chromatography
using hexane–ethyl acetate (8:1) as eluent to give the product
12 (3.41 g, 51%): 1H NMR (CDCl3) d 1.20 (t, 3H,
1
this experiment: H NMR (CDCl3) d 1.96 (m, 2H), 2.42
(t, 2H), 3.24 (t, 2H), 4.43 (s, 2H), 7.27 (5H); 13C NMR
(CDCl3) d 17.8 (+), 31.0 (+), 46.6 (+), 46.7 (+), 127.6 (ꢁ),
128.2 (ꢁ), 128.7 (ꢁ), 136.6 (+), 175.0 (+).

S. Zhao et al. / Tetrahedron 62 (2006) 6361–6369
6367
J¼7.0 Hz), 1.29 (t, 3H, J¼7.0 Hz), 1.35 (s, 6H), 1.53 (m,
2H), 2.17 (t, 2H, J¼7.2 Hz), 2.62 (t, 2H, J¼7.5 Hz), 3.82
(s, 2H), 4.03 (q, 2H, J¼7.0 Hz), 4.14 (q, 2H, J¼7.0 Hz),
7.35 (5H); 13C NMR (CDCl3) d 14.3, 14.4, 25.2, 25.4,
31.8, 51.5, 55.8, 60.2, 60.5, 63.8, 126.4, 127.7, 128.1,
142.8, 173.7, 176.2; FAB HRMS m/z calcd for C19H30NO4
(M+H)+ 336.2175, found 336.2182.
20 h at ambient temperature. The reaction mixture was
poured into 100 mL of 0.05 N sodium hydroxide and
extracted three times with ether. The combined organic
extracts were dried with K2CO3, filtered, and evaporated to
afford N-(6-chloro-2,2-dimethoxy-3-hexylidene)isopropyl-
amine (18, 2.1 g, 89%) as an oil. The compound was used
without further purification in the next step.
4.11. 1-Benzyl-2,2-dimethyl-4-carbethoxy-3-piperidone
(13)
To a solution of imine 18 (2.1 g, 9 mmol) in 120 mL of
CH2Cl2 was added oxalic acid dihydrate (1.7 g, 13 mmol)
in 100 mL of water. The mixture was heated at reflux with
vigorous stirring for 1 h and extracted three times with
CH2Cl2. The combined organic layer was dried (Na2SO4)
and filtered, and the resulting oil obtained upon evaporation
of solvent was subjected to silica gel flash chromatography
using hexane–ethyl acetate (15:1) as eluent to give the
keto acetal 19 (0.81 g, 42% overall for two steps) as an
oil: 1H NMR (CDCl3) d 1.37 (s, 3H), 2.05 (quintuplet,
2H), 2.80 (t, 2H, J¼7.0 Hz), 3.24 (s, 6H), 3.59 (t, 2H,
J¼6.4 Hz); 13C NMR (CDCl3) d 19.9 (ꢁ), 26.0 (+), 35.0
(+), 44.4 (+), 49.7 (ꢁ), 102.6 (+), 206.1 (+); FAB HRMS
Sodium hydride (240 mg, 6.6 mmol, 60%) was suspended in
a solution of 12 (2.0 g, 6.0 mmol) in 50 mL of dry THF. The
mixture was heated at reflux for 4 h. After evaporation of the
solvent, the viscous product was treated with H2O and the re-
sulting mixture was acidified to pH 5, maintaining the tem-
perature under 10 ꢀC with an ice bath. The aqueous solution
was neutralized with solid K2CO3 and extracted with Et2O.
Evaporation of the dried (Na2SO4) organic extract yielded
13 (1.27 g, 74%) as an oil that NMR showed to exist as
the enol tautomer: 1H NMR (CDCl3) d 1.30 (t, 3H,
J¼7.0 Hz), 1.41 (s, 6H), 2.18 (t, 2H, J¼5.8 Hz), 2.52 (t,
2H, J¼5.8 Hz), 3.62 (s, 2H), 4.21 (q, 2H, J¼7.0 Hz), 7.35
(5H); 13C NMR (CDCl3) d 14.4 (ꢁ), 21.4 (ꢁ), 23.2 (+),
42.6 (+), 53.2 (+), 58.4 (+), 60.4 (+), 95.6 (+), 126.8 (ꢁ),
128.3 (ꢁ), 140.7 (+), 172.9 (+), 175.5 (+). FAB HRMS
m/z calcd for C17H24NO3 (MH)+ 290.1756, found 290.1756.
calcd for C7H ClO2 (M+ꢁOCH3) m/z 163.0526, found
35
12
163.0518.
4.15. 6-Benzylamino-2,2-dimethoxy-3-hexanol (21)
A solutionof 6-chloro-2,2-dimethoxy-3-hexanone 19(0.35 g,
1.8 mmol), potassium iodide (300 mg, 1.8 mmol), benzyl-
amine (0.29 g, 2.7 mmol), and triethylamine (0.4 mL,
2.9 mmol) in 50 mL of acetonitrile was heated at reflux for
20 h under argon. The reaction mixturewas filtered and evap-
orated to dryness. The residue was extracted with CH2Cl2,
and the organic extract was washed with water, dried
(Na2SO4), and concentrated. The residue was purified by
chromatography using CH2Cl2–methanol (30:1) as eluent to
give6-benzylamino-2,2-dimethoxy-3-hexanone(20, 185 mg,
38%) as an oil, which has a low stability and was used quickly
for the next step: 1H NMR (CDCl3) d 1.34 (s, 3H), 1.78 (quin-
tuplet, 2H), 2.61 (m, 4H), 3.19 (s, 6H), 3.77 (s, 2H), 7.29
(5H); 13C NMR (CDCl3) d 19.9, 23.2, 35.6, 48.4, 49.6,
53.5, 102.7, 127.1, 128.3, 128.4, 139.5, 209.1.
4.12. 1-Benzyl-2,2-dimethyl-3-piperidone (10)
A solution of 13 (0.58 g, 2.0 mmol) in 10 mL of 6 N HCl
was heated at reflux for 3 h. The reaction mixture was cooled
to room temperature, neutralized with solid K2CO3 to pH
8.0, and extracted with ether. The organic layer was washed,
filtered, and evaporated to give pure piperidone 10 (0.39 g,
1
90%) as an oil: H NMR (CDCl3) d 1.31 (s, 6H), 1.82 (m,
2H), 2.50 (t, 2H, J¼7.1 Hz), 2.65 (t, 2H, J¼6.1 Hz), 3.63
(s, 2H), 7.31 (5H); 13C NMR (CDCl3) d 20.0 (ꢁ), 23.9
(+), 36.2 (+), 44.9 (+), 53.7 (+), 66.6 (+), 126.9 (ꢁ), 128.2
(ꢁ), 128.3 (ꢁ), 140.5 (+), 212.2 (+); FAB HRMS m/z calcd
for C14H20NO (M+H)+ 218.1545, found 218.1545.
4.13. N-(3,3-Dimethoxy-2-butylidene)isopropylamine
(17)22
To a solution of 20 (270 mg, 1 mmol) in 10 mL of methanol,
sodium borohydride (57 mg, 1.5 mmol) was added portion-
wise at 0 ꢀC. The reaction mixture was stirred at room tem-
perature for 3 h. After removal of methanol, the viscous
product was treated with H2O and then extracted with ether.
The organic layer was dried (Na2SO4) and evaporated to
dryness. The residue was purified by chromatography with
CH2Cl2–methanol (15:1) as eluent to afford the product 21
The preparation of 17 was carried out exactly according to
the literature report.22 The final residual oil was distilled in
vacuum to afford the product 17 (15.3 g, 89%) as a colorless
liquid: bp 70–76 ꢀC/25 mmHg (lit.22 60–63 ꢀC/11 mmHg);
1H NMR (CDCl3) d 1.09 (d, 3H, J¼6.1 Hz), 1.12 (d, 3H,
J¼6.2 Hz), 1.33 (s, 3H), 1.81 (s, 3H), 3.17 (s, 3H), 3.19 (s,
3H), 3.65 (m, 1H); 13C NMR (CDCl3) d 13.0 (+), 20.9 (+),
23.1 (+), 49.2 (+), 50.9 (+), 102.3 (ꢁ), 165.4 (ꢁ).
1
(201 mg, 76%) as an oil: H NMR (CDCl3) d 1.29 (s, 3H),
1.25 (m, 2H), 1.78 (m, 2H), 2.61 (m, 1H), 2.86 (m, 1H),
3.18 (s, 3H), 3.23 (s, 3H), 3.61 (d, 1H, J¼9.8 Hz), 3.87 (s,
1H), 7.34 (5H); 13C NMR (CDCl3) d 16.3 (ꢁ), 26.7 (+),
30.3 (+), 48.2 (ꢁ), 48.5 (ꢁ), 48.6 (+), 53.0 (+), 72.3 (ꢁ),
102.7 (+), 127.8 (ꢁ), 128.7 (ꢁ), 128.9 (ꢁ), 137.0 (+);
FAB HRMS m/z calcd for C15H26NO3 (M+H)+ 268.1912,
found 268.1915.
4.14. 6-Chloro-2,2-dimethoxy-3-hexanone (19)
A solution of lithium diisopropylamide (6 mL of 2.0 M,
0.012 mol) in 30 mL of dry tetrahydrofuran at 0 ꢀC was
treated dropwise with 1.73 g (0.01 mol) of N-(3,3-di-
methoxy-2-butylidene)isopropylamine, dissolved in 3 mL
of dry tetrahydrofuran. The mixture was stirred for 2 h at
0 ꢀC after which 1.72 g (0.012 mol) of 1-bromo-2-chloro-
ethane was added dropwise. The solution was stirred for
4.16. The reactions of compound 21 in 6 N HCl
A solution of compound 21 (50 mg, 0.19 mmol) in 5 mL of
6 N HCl solution was heated at reflux for 24 h under argon.

6368
S. Zhao et al. / Tetrahedron 62 (2006) 6361–6369
The reaction mixture was extracted with ether. The aqueous
phase was separated and neutralized with K2CO3 to pH 8.
The solution was then extracted with ether, which was dried
(Na2SO4) and evaporated to give 2 (33 mg, 87%) as an oil. A
solution of compound 21 (50 mg, 0.18 mmol) in 5 mL of 6 N
HCl solution was stirred at room temperature for 24 h under
argon. The reaction mixture was neutralized with K2CO3 to
pH 8. The solution was extracted with ether. The combined
ether layer was dried (Na2SO4) and evaporated to give 1
(30 mg, 79%) as an oil.
76.44 (ꢁ), 129.64 (ꢁ), 130.10 (ꢁ), 130.35 (ꢁ), 130.84 (+),
215.16 (+).
Acknowledgements
We are grateful for support of the early stages of this work
by the National Institutes of Health grants NS 22688 and
GM 48812. We thank Drs. Ke-Qing Ling and Yuming
Zhang for helpful discussion.
4.17. Time profile for conversion of compounds 1 or 21 to
compound 2 in 6 N HCl solution
References and notes
A solution of compound 1 or 21 in 25 mL of 6 N HCl solu-
tion was heated at reflux under argon. Aliquots were taken at
1, 3, 6, 12, and 24 h. The aliquots were neutralized with
K2CO3 to pH 8 and extracted with ether. The ether layers
were dried (Na2SO4) and evaporated to give mixtures of 1
1. Weintraub, P. M.; Sabol, J. S.; Kane, J. M.; Borcherding, D. R.
Tetrahedron 2003, 59, 2953; O’Hagan, D. Nat. Prod. Rep.
2000, 17, 435; Plunkett, A. O. Nat. Prod. Rep. 1994, 11, 581.
2. Hu, L. Y.; Ryder, T. R.; Rafferty, M. F.; Feng, M. R.; Lotarski,
S. M.; Rock, D. M.; Sinz, M.; Stoehr, Sally J.; Taylor, Charles
P.; Weber, M. L.; Bowersox, S. S.; Miljanich, G. P.; Millerman,
E.; Wang, Y. X.; Szoke, B. G. J. Med. Chem. 1999, 42, 4239;
LaRoche, S. M.; Helmers, S. L. JAMA 2004, 291, 605;
Tepper, S. J.; Rapoport, A. M.; Sheftell, F. D. Arch. Neurol.
2002, 59, 1084.
1
and 2 (oils) that were analyzed by H NMR spectroscopy.
The fraction of rearrangement was calculated from the inte-
gral ratio of the ketone methyl group in 2 (d 2.1 ppm) and the
piperidine ring 2-methyl group in 1 (d 1.3 ppm).
4.18. NMR spectral data of compounds in DCl–D2O
3. Sayre, L. M.; Engelhart, D. A.; Nadkarni, D. V.; Babu,
M. K. M.; Klein, M. E.; McCoy, G. Xenobiotica 1995, 25, 769.
4. Sayre, L. M.; Engelhart, D. A.; Venkataraman, B.; Babu,
M. K. M.; McCoy, G. D. Biochem. Biophys. Res. Commun.
1991, 179, 1368.
5. Stevens, C. L.; Pillai, P. M.; Munk, M. E.; Taylor, K. G.
Mechanisms of Molecular Migrations; Thyagarajan, B. S.,
Ed.; Interscience: New York, NY, 1968; p 271.
4.18.1. 1-Benzyl-3-piperidone(freebase). 1H NMR (CDCl3)
d 1.94 (m, 2H), 2.36 (t, 2H), 2.65 (t, 2H), 3.01 (s, 2H), 3.58
(s, 2H), 7.29 (m, 5H); 13C NMR (CDCl3) d 24.04 (+), 38.81
(+), 51.60 (+), 62.60 (+), 64.63 (+), 127.41 (ꢁ), 128.42 (ꢁ),
129.08 (ꢁ), 137.26 (+), 207.12 (+).
4.18.1.1. 1-Benzyl-3-piperidone$HCl. 13C NMR (6 N
DCl–D2O) d 20.5 (+), 34.2 (+), 53.1 (+), 58.9 (+), 61.1
(+), 91.6 (+), 128.8 (+), 130.1 (ꢁ), 131.1 (ꢁ), 132.1 (ꢁ).
4.18.2. 1-Benzyl-2-methyl-3-piperidone$HCl. 13C NMR
(6 N DCl–D2O, mixture of cis and trans isomers) d 7.1 (ꢁ),
10.1 (ꢁ), 19.6 (+), 23.5 (+), 29.2 (+), 35.2 (+), 47.3 (+),
51.5 (+), 61.3 (ꢁ), 65.7 (ꢁ), 92.6 (+), 93.7 (+), 128.6 (+),
129.2 (+), 129.7 (ꢁ), 130.0 (ꢁ), 130.5 (ꢁ), 130.7 (ꢁ),
132.3 (ꢁ).
4.18.3. 1-Benzyl-2,2-dimethyl-3-piperidone$HCl. 13C
NMR (6 N DCl–D2O) d 16.07 (ꢁ), 19.18 (ꢁ), 19.52 (ꢁ),
20.13 (+), 20.59 (ꢁ), 20.85 (+), 31.33 (+), 35.31 (+), 46.93
(+), 47.67 (+), 55.89 (+), 55.93 (+), 71.15 (+), 73.01 (+),
95.29 (+), 130.08 (+), 130.21 (ꢁ), 130.29 (+), 130.48 (ꢁ),
131.15 (ꢁ), 131.33 (ꢁ), 132.21 (ꢁ), 132.81 (ꢁ), 207.57 (+).
6. Duhamel, P.; Kotera, M.; Monteil, T.; Marabout, B. J. Org.
Chem. 1989, 54, 4419.
7. Leonard, N. J.; Ruyle, W. V. J. Am. Chem. Soc. 1949, 71, 3094;
Leonard, N. J.; Berthel, E., Jr. J. Am. Chem. Soc. 1950, 72,
3632.
8. Leonard, N. J.; Wildman, W. C. J. Am. Chem. Soc. 1949, 71,
3089.
9. Rearrangement of optically active 1,2-dimethyl-2-ethyl-3-
piperidione gave racemic 1-methyl-2-sec-butylpyrrolidine,
indicating that there must be complete cleavage of the C–N
bond at some point in the rearrangement: Leonard, N. J.;
Sentz, R. C.; Middleton, W. J. J. Am. Chem. Soc. 1953, 75, 1674.
10. Brunner, H.; Kagan, H. B.; Kreutzer, G. Tetrahedron:
Asymmetry 2003, 14, 2177.
´
11. Compain, P.; Gore, J.; Vatele, J.-M. Tetrahedron 1996, 52,
`
6647.
12. Ooi, T.; Saito, A.; Maruoka, K. J. Am. Chem. Soc. 2003, 125,
3220.
13. Hagishita, S.; Yamada, M.; Shirahase, K.; Okada, T.;
Murakami, Y. J. Med. Chem. 1996, 39, 3636.
14. Lounasmaa, M.; Karvinen, E.; Koskinen, A.; Jokela, R.
Tetrahedron 1987, 43, 2135.
15. Goes, A. D. S.; Ferroud, C.; Santamaria, J. Tetrahedron Lett.
1995, 36, 2235.
16. Macchia, B.; Macchia, M.; Martinelli, A.; Martinotti, E.;
Orlandini, E.; Romagnoli, F.; Scatizzi, R. Eur. J. Med. Chem.
1997, 32, 231.
1
4.18.4. 1-Benzyl-2-acetylpyrrolidine$HCl. H NMR (6 N
DCl–D2O) d 1.94 (m, 2H), 2.16 (s, 3H), 2.22 (m, 1H),
2.67 (m, 1H), 3.38 (m, 1H), 3.68 (m, 1H), 4.37 (d, 2H),
4.75 (t, 1H), 7.49 (m, 5H); 13C NMR (6 N DCl–D2O)
d 23.34 (+), 27.20 (ꢁ), 28.30 (+), 56.12 (+), 59.37 (+),
72.96 (ꢁ), 129.96 (ꢁ), 130.22 (+), 130.87 (ꢁ), 131.34 (ꢁ),
205.03 (+).
4.18.5. 6-Benzylamino-2,2-dimethoxy-3-hexanol$HCl.
1H NMR (6 N DCl–D2O) d 1.8 (m, 4H), 2.15 (s, 3H,
exchangeable), 3.10 (m, 2H), 4.21 (s, 2H), 4.40 (m, 1H,
exchangeable), 7.45 (m, 5H); 13C NMR (6 N DCl–D2O)
d 22.05 (+), 26.15 (ꢁ), 29.57 (+), 47.21 (+), 51.55 (+),
17. We have shown independently that autoxidation of 2, acceler-
ated by base, results in oxygenation and breakdown to 14
and acetic acid: Nadkarni, D. V.; Babu, M. K. M.; Jeon, H.-B.;
Sayre, L. M. Unpublished.

S. Zhao et al. / Tetrahedron 62 (2006) 6361–6369
6369
18. Sanders, E. B.; Secor, H. V.; Seeman, J. I. J. Org. Chem. 1978,
43, 324.
19. Li, J. J. Name Reactions: A Collection of Detailed Reaction
Mechanisms, 2nd ed.; Springer: Berlin, 2003.
20. Fleming, S. A.; Carroll, S. M.; Hirschi, J.; Liu, R.; Pace, J. L.;
Redd, J. T. Tetrahedron Lett. 2004, 45, 3341.
21. Hofmann, T.; Schieberle, P. J. Agric. FoodChem. 1998, 46, 2270.
22. De Kimpe, N.; Stevens, C. Tetrahedron 1995, 51, 2387.
23. Ozdowska, Z. Pol. J. Chem. 1978, 52, 385; Emmett, G. C.;
Cain, G. A.; Estrella, M. J.; Holler, E. R.; Piecara, J. S.;
Blum, A. M.; Mical, A. J.; Teleha, C. A.; Wacker, D. A.
Synthesis 2005, 1, 92.
24. The energetic preference against an sp2 carbon in a six-
membered ring and the energetic preference for an sp2 carbon
in a five-membered ring arises because conversion of sp2 to sp3
in the former allows for complete staggering, whereas conver-
sion of sp3 to sp2 in the latter averts partial eclipsing inter-
actions: Carey, F. A.; Sundberg, R. J. Advanced Organic
Chemistry, Part A: Structure and Mechanism, 4th ed.; Kluwer
Academic: New York, NY, 2000; p 172.
25. Schneider, H.-J.; Thomas, F. J. Am. Chem. Soc. 1980, 102,
1424.
26. Di Vona, M. L.; Rosnati, V. J. Org. Chem. 1991, 56, 4269.
27. Julia, T. Bull. Soc. Chim. Fr. 1958, 661.