Efficient deracemization of pipecolic acid amides through enantioselective protonation of their lithium enolates: Insights into the origin of the transferred proton

Article abstract of DOI:10.1002/ejoc.200900726

A detailed study of the deracemization of pipecolic acid amides is reported. Enantioselective protonation of the lithium enolates of these amides with use of commercially available ephedrines led to enantiomeric excesses (ee values) higher than 99%. The success of the reaction was strongly dependent on the following parameters: the base, the reaction temperature, the structure of the chiral source, and the achiral quenching reagent. sBuLi and the bimetallic base "potassium alkoxide/nBuLi" were the only bases to allow complete formation of the enolate in conjunction with high stereocontrol of the protonation. Experiments with (+)- or (-)-ephedrine derivatives as chiral sources and deuteriated reagents gave evidence that both the OH and NH protons of ephedrine were involved in the stereoinduction. External delivery of the proton was mainly operative with the aniline derivative (+)-5, as shown by deuterium labeling experiments, whereas internal quenching was the major pathway observed with ephedrine (6). Finally, the deracemization procedure was successfully applied to prepare both enantiomers of N-protected pipecolic acid from racemic pipecolic acid (51 % overall yield, 99% ee).

Full text of DOI:10.1002/ejoc.200900726

FULL PAPER
DOI: 10.1002/ejoc.200900726
Efficient Deracemization of Pipecolic Acid Amides through Enantioselective
Protonation of Their Lithium Enolates: Insights into the Origin of the
Transferred Proton
Juliette Martin,[a] Jean-Christophe Plaquevent,[b] Jacques Maddaluno,[c] Jacques Rouden,[a]
and Marie-Claire Lasne*[a]
Keywords: Asymmetric synthesis / Protonation / Deracemization / Amides / Lithium enolates / Enantioselectivity
A detailed study of the deracemization of pipecolic acid
amides is reported. Enantioselective protonation of the lith-
reagents gave evidence that both the OH and NH protons
of ephedrine were involved in the stereoinduction. External
ium enolates of these amides with use of commercially avail- delivery of the proton was mainly operative with the aniline
able ephedrines led to enantiomeric excesses (ee values)
higher than 99%. The success of the reaction was strongly
dependent on the following parameters: the base, the reac-
tion temperature, the structure of the chiral source, and the
achiral quenching reagent. sBuLi and the bimetallic base
“potassium alkoxide/nBuLi” were the only bases to allow
complete formation of the enolate in conjunction with high
stereocontrol of the protonation. Experiments with (+)- or
(–)-ephedrine derivatives as chiral sources and deuteriated
derivative (+)-5, as shown by deuterium labeling experi-
ments, whereas internal quenching was the major pathway
observed with ephedrine (6). Finally, the deracemization pro-
cedure was successfully applied to prepare both enantiomers
of N-protected pipecolic acid from racemic pipecolic acid
(51% overall yield, 99% ee).
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
nervous system, we prepared both enantiomers of AF-
DX 384 (1) via amides 2 derived from (S)-(–)- or (R)-(+)-
pipecolic acids (3, Figure 1).[9] However, the high cost of
these homochiral acids was an important drawback to their
use as starting materials on a large scale. Moreover, because
of the ubiquitous structural feature of the piperidine-2-
carboxyl unit in peptides,[10] immunosuppressor agents,[11]
and biologically active compounds,[12] there is still a de-
mand for alternatives to the numerous routes[13] to the en-
antio-enriched acids (+)- or (–)-3. A few years ago we suc-
cessfully tested the deracemization of pipecolamides[14]
through enantioselective protonation of their enolates
(Scheme 1). In this paper we summarize our efforts to ad-
dress the key issues of this reaction to define the roles of
the various reaction parameters and to gain insight into
Introduction
Deracemization of ketones, acids, amino acid derivatives,
esters, thioesters, or phosphane oxides through enantiose-
lective protonation of their corresponding enolates or enol-
ate equivalents is well documented,[1–3] and some catalytic
versions[4] have been developed. In the case of amides, how-
ever, this methodology remains limited; the only reported
examples involve γ,δ unsaturated substrates[5,6] or a particu-
lar lactam.[7] Vedejs et al.[8,5c,5d] demonstrated in an in-
depth study that the enantiomeric excesses (ee values) are
strongly dependent on the substrate, on the base used to
generate the enolate, and on the temperature of the reac-
tion.
As a part of our ongoing program on the synthesis of
ligands for the study of muscarinic receptors in the central
[a] Laboratoire de Chimie Moléculaire et Thio-organique, UMR
6507, Institut Normand de Chimie Moléculaire, Médicinale et
Macromoléculaire, FR 3038, ENSICAEN, Université de Caen
Basse-Normandie, CNRS,
6 Boulevard du Maréchal Juin, 14050 Caen, France
Fax: +33-2-31452877
E-mail: marie-claire.lasne@ensicaen.fr
[b] Université Paul Sabatier Toulouse 3, CNRS, Synthèse et phys-
ico-chimie de molécules d’intérêt biologique, bat. 2R1,
118 Route de Narbonne, 31062 Toulouse Cedex 4, France
[c] Université de Rouen, CNRS, Institut de Recherche en Chimie
Organique Fine,
Rue Tesniére, 76821 Mont St Aignan, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900726.
Figure 1. (S)-(+)-AF-DX 384 from (S)-(–)-pipecolic acid (3).
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Eur. J. Org. Chem. 2009, 5414–5422
5414

Deracemization of Pipecolic Acid Amides
Scheme 1. Deracemization of pipecolamides 2.
the identities of the proton donors. The efficiency of the reach T3 and the quenching agent was added. Comparison
methodology was demonstrated in the synthesis of N-pro- of Entries 1 and 4 suggests that a reorganization of the
tected (S)- or (R)-pipecolic acids.
“sBuLi/enolate/chiral source” complex was necessary for
high ee values to be achieved. The enantioselection de-
creased dramatically with a higher temperature T2 (En-
tries 3 and 4), particularly when T3 was also high (Entry 2).
These results could be explained by reversible deproton-
ation of the amide 2a at temperatures above 0 °C.[5b]
Results and Discussion
The amide 2a (Scheme 1) was used as a model substrate
for detailed deracemization optimization experiments. From
Vedejs’ studies[5,8] and our preliminary results,[14] we chose
the aniline derivative (+)-5 as the chiral “acid” for the enan-
tioselective protonation of the enolate 4a. For the sake of
reproducibility[8] we used 2 equiv. of chiral source. More-
over, in this study and for each experiment, the extent of
amide enolization was evaluated by 1H NMR measurement
of deuterium incorporation (“deuterium test”, Scheme 1,
route a; see Exp. Sect.).
Choice of Base
The ee values being strongly dependent on the quality of
the sBuLi,[15] we searched for an alternative base for the
preparation of intermediate 4a. No enantioselection was
observed with lithium diisopropylamide (LDA) or lithium
hexamethyldisilazide (LHMDS) whatever the number of
equivalents used. Because an exchange of proton between
the lithiated reagent and the amine is known to occur,[16,17]
we did not pursue our investigations with these bases. No
asymmetric induction was observed with any of nBuLi,
MeLi, or tBuLi, although complete deprotonation of amide
2a was observed (deuterium test). Addition of BF3·Et2O
(2 equiv.), known to increase electron demand in the amine/
anion complex and to force internal proton return,[18] gave
low ee values (Ͻ27%). Addition of lithium chloride, which
is able to form mixed aggregates and to modify the reaction
intermediate,[17] was no more successful (ee values Ͻ8%).
sBuLi in THF, however, yielded the amide (S)-(–)-2a in
87% ee (Table 2, Entry 1). No improvement was observed
in the presence of BF3·Et2O (Entry 2), but when LiBr[19–22]
(1.75 equiv.) was added the ee reached 95% (Entry 3). The
use of sBuLi (1.75 equiv.) was necessary to achieve com-
plete deprotonation. Indeed, under the same conditions as
in Entry 1 but with 1 equiv. of base, only a 54% yield of
the enolate (deuterium test) was formed and no enantiose-
lection was observed (Entry 4). Similar results were ob-
served when the reaction was performed with sBuLi
(1 equiv.) at –30 °C rather than –78 °C (data not shown).
Attempts to carry out the reaction in other solvents (diethyl
ether or toluene or a diethyl ether/THF mixture) failed (En-
try 5).
Influence of Temperature
The temperatures of the different reaction steps are cru-
cial parameters in enantioselective protonations.[2,5,8] In
Table 1 we report some data relating to the influence of the
temperature on the obtained ee values of amide 2a. The
enolate 4a was generated at temperature T1 (–78 °C) and
the chiral source (+)-5 was added. After 60 min at T1, the
solution was warmed to temperature T2 over a 30 min
period. After 5–10 min at T2, the mixture was allowed to
Table 1. Obtained ee values of 2a as a function of temperature.
Entry[a]
T1 [°C]
Enolate forma-
tion and (+)-5
addition
T2 [°C]
T3 [°C]
ee [yield]
Reaction mixture H2O addition
[%][b]
1
2
3
4
–78
–78
–78
–78
–78
+15
+15
–78
+15
–78
–78
1 [88]
5 [77]
54 [75]
96 [78]
–30 or –20
[a] sBuLi (1.75 equiv.), LiBr (1.75 equiv.), amide 2a, chiral source
(+)-5 (2 equiv.). [b] Isolated yields; ee values were determined by
chiral HPLC. Major isomer: (S)-(–)-2a.
Eur. J. Org. Chem. 2009, 5414–5422
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5415

J. Martin, J.-C. Plaquevent, J. Maddaluno, J. Rouden, M.-C. Lasne
Table 2. Deracemization of pipecolamide 2a. Influences of base and additives.
FULL PAPER
Entry Base [equiv.][a]
Added reagent
[equiv.][b]
A*H
[2 equiv.]
Solvent
2a
ee
Yield [%][c]
[%][d,e]
1
2
3
4
5
6
sBuLi [1.75]
sBuLi [1.75]
sBuLi [1.75]
sBuLi [1]
no
(+)-5
(+)-5
(+)-5
(+)-5
(+)-5
no
THF
THF
THF
THF
87
26
87
86
95
0
0
0
BF3·Et2O [2]
LiBr [1.75]
no
LiBr [1.75]
no
78
n.d.[g]
sBuLi [1.75]
Et2O, toluene, or Et2O/THF[f] 62–68
KH/nBuLi/(–)-ephedrine
[1.25:2.25:1.25]
KH/nBuLi/(+)-ephedrine
[1.25:2.25:1.25]
KH/nBuLi/(–)-ephedrine
[1.25:2.25:1.25]
THF
THF
THF
67
65
53
7
8
no
no
(+)-6
(–)-6
92
88
[a] Amide 2a and the additive in the solvent were cooled to –78 °C and stirred at this temperature for 40–60 min. The base was added
and after 15 min a deuterium incorporation test was carried out on an aliquot (see Exp. Sect.). The chiral source (2 equiv.) was then
added. After 30 min at –78 °C (T1) the mixture was warmed over a 30 min period to –20 or –30 °C and then, after 5–10 min, cooled
again to –78 °C before quenching with H2O. [b] Number of equivalents vs. amide 2a. [c] Isolated yields. [d] Determined by chiral HPLC.
[e] The enantiomer (S)-(–)-2a was formed from (+)-5 or (+)-6. [f] Ratio 1:1 v/v. [g] Because only 54% enolate generation had occurred,
the yield of the reaction was not determined.
It is noteworthy that no deracemization occurred when group (9-H) and the protons of the methyl (12-H) and
the enolate 4a was not completely formed. This observation methylene (10-H) units of the propyl chain on the hand and
was checked by addition of amide 2a (0.2 equiv.) to a pre- between 6-H and 9-H on the other hand. However, no
formed enolate before addition of the chiral source (+)-5. cross-coupling between protons 10-H and 3-H was ob-
Under standard conditions of temperature and time served. The observed correlations suggested that the major
(Table 2, Entry 3), amide 2a was isolated as a racemate. enolate had the E configuration (Figure 3).[28] A relative
This result suggests either an equilibrium between enolate 60:40 ratio for the rotamers (similar to that observed for
1
4a and amide 2a or the participation of amide 2a in the the amide 2a) was deduced from the H NMR spectrum.
complex formed in the transition state.
Finally, in order to test strong deprotonating agents, we
prepared a mixed lithium/potassium base by treatment of
(+)- or (–)-ephedrine [(+)- or (–)-6] with KH (1.2 equiv.,
room temp., 5 min) in THF followed by nBuLi (2.25 equiv.,
–40 °C, 30 min) under Schlosser conditions.[23] No chiral in-
duction was observed, as shown in Entry 6. Treatment,
however, of amide 2a (–78 °C, 30 min) with the “superbase”
followed by addition of ephedrine to enolate 4a prior to
Figure 3. Cross-couplings observed in NMR of enolates 4a (E).
quenching with H2O at –78 °C yielded (S)- or (R)-amide 2a
in 92 and 88% ees, respectively (Table 2, Entries 7, 8). The
ee values in these asymmetric protonations were similar to
those obtained with sBuLi (Entries 1–3). The alkoxide- Influence of Nitrogen Substituents
amide “superbase” thus appeared to be a good alternative
Amides 2b, 2c, and 2d were prepared in order to allow
to sBuLi for deracemization of pipecolamides.
comparison of the steric demands of the nitrogen systems.
Attempts to trap the enolate intermediates 4a (Figure 2)
The results of all the experiments performed under the best
by treatment variously with chlorotrimethylsilane, chlorodi-
conditions (Table 2, Entry 3) are presented in Table 3. The
methyl-tert-butylsilane, or acetic anhydride failed.[8] Few
data for lithium enolates of amides are available,[24–27] so
1
1
we undertook H/1H COSY and H/1H NOESY NMR ex-
periments. Carried out at –60 °C in [D8]THF, these showed
strong cross-couplings between the protons of the methoxy
Table 3. Influence of nitrogen substitutions on the obtained ee val-
ues of amides 2.
Amide
Entry[a]
R1
R2
A*H
Product Yield[b] [%] ee[c] [%]
1
2
3
4
5
6
2b –(CH2)5– Me
2b –(CH2)5– Me
(+)-5
(–)-6
(+)-5
(–)-6
(+)-5
(–)-6
(S)-2b
(R)-2b
(S)-2c
(R)-2c
(S)-2d
(R)-2d
65
56
74
65
85
89
Ͼ99
Ͼ99
96
2c Me
2c Me
2d nPr
2d nPr
Me
Me
tBu
tBu
95
86[d]
85[d]
[a] Standard conditions: Table 2, Entry 3. [b] Isolated yield. [c] De-
termined by chiral HPLC. [d] From optical rotations of the N-de-
protected compound.
Figure 2. Enolate intermediates 4a.
5416
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5414–5422

Deracemization of Pipecolic Acid Amides
piperidinyl amide 2b was obtained in ee values higher than prochiral enolates of 2-alkyltetralones. High ee values
99% whereas the dimethyl amide 2c gave the same selectiv- (91%) were obtained with a triamine/acetic acid pair,[19]
ity as its n-propyl analogue 2a. A small erosion of the ee whereas a 3-aminopyrrolidine derivative[31] or a C2-symmet-
values was observed when the methoxycarbonyl component rical cyclohexyldiamine[29] in combination with acetic acid
was substituted by the easily removable tert-butoxycarbonyl were less efficient (ee values around 40%). The use of a
group in compound 2d (Entries 5, 6).
catalytic amount of the BINAP·AgF complex with MeOH
The excellent ee values obtained with amide 2b are note- as the proton source induced enantioselective protonation
worthy. The moderate yields observed in Entries 1 and 2 of silyl enolates of 2-substituted cyclohexanones with ee val-
were significantly improved when working with amide 2b ues higher than 99%.[32]
on a gram scale (see below). However, one could not ex-
We have shown that both enantiomers of the commer-
clude contamination by a side-product resulting from a re- cially available and inexpensive ephedrine (6) have the same
action between sBuLi and the carbamate.[6]
efficiency as aniline (+)-5[33] in the asymmetric protonation
of pipecolamide enolates 4. In order to understand the
course of the proton transfer, we first examined the role of
each proton in the amine and alcohol functions of 6 (Fig-
ure 4). Typical results for the asymmetric protonation of
enolate 4a in the presence of the ephedrine derivatives 6–10
as sources of chirality are summarized in Table 5. A com-
parison of experiments performed with aniline (+)-5
(Table 2, Entries 1–2) and with 6 (Table 5, Entries 1–3)
show that the effect of BF3·Et2O on the yield was less pro-
nounced with (+)-ephedrine [(+)-6]; however, BF3·Et2O sig-
nificantly affected the ee (Entry 3 vs. 2). Norephedrine (7),
N-methylephedrine (8), and O-methylephedrine (9) gave no
asymmetric induction (Table 5, Entries 4, 5, 6 respectively),
and pseudoephedrine (10) was less efficient (ee = 26%) than
ephedrine itself (Entry 7). Thus, both the N–H and the O–
H units and the unlike configuration of the stereogenic cen-
ters of ephedrine were essential for high induction in the
asymmetric protonation of enolate 4a.
Influence of the Amount of Chiral Source
In a few experiments, the amount of chiral source was
decreased while the 1.75 equiv. of sBuLi was maintained.
Under the standard conditions (Table 2, Entry 3), the high-
est ee values were obtained with 2 or 3 equiv. of chiral
source (+)-5. These results (Table 4, Entries 1, 3, 4) sug-
gested that two molecules either of aniline (+)-5 or of eph-
edrine (6) could be involved in the mixed aggregates.[26,27]
Table 4. Effect of the amount of the chiral source.
Entry[a]
A*H
A*H [equiv.]
ee [%][a,b]
1
2
3
4
5
6
(+)-5
(+)-5
(+)-5
2
95
0
96
89; 93
87
61
1 or 1.75
3
2
1.25
1
(+)-6; (–)-6
(+)-6
(+)-6
[a] Standard conditions Table 2, Entry 3, isolated yields: 68–87%.
[b] Determined by chiral HPLC.
The two chiral sources appear to behave differently, how-
ever: with 1.75 or 1 equiv. of aniline (+)-5 (Table 4, Entry 2)
no induction was observed, whereas with 1.25 and 1 equiv.
of (+)-ephedrine [(+)-6, Entries 5, 6], 2a was isolated in 87
and 61% ee values, respectively. Deuterium labeling experi-
ments (see below) confirmed that the proton transfer
mechanisms with these two proton sources are quite dif-
ferent.
Figure 4. Ephedrine derivatives as chiral sources.
Origin of the Transferred Proton
Table 5. Obtained ee values of amide 2a as a function of ephedrine
The asymmetric induction relies on the use either of a
chiral protic compound (“internal quench”) or of an achiral
protic agent coupled with a chiral ligand (“external
quench”).[2d,2e,4f] Many effective chiral protonating agents
A*H (chiral acids, diols, functionalized alcohols or amines,
cinchona alkaloids, hydroxy sulfoxides or selenoxides,
imides, phenolic amides) have been developed[2,29,4c] in or-
der to achieve high enantioselectivities in deracemization
through asymmetric protonation. In some cases (see, for in-
stance,[8,30]), deuterium-labeled sources have been used to
demonstrate the origin of the proton delivery. The “external
quench” is far less common than the “internal quench”.
derivatives.
Entry A*H[a] Additive Yield 2a [%][b] ee [%][c]
Config.
1
2
3
4
5
6
7
(–)-6
73
74
68
68
71
78
73
93
89
44
0
0
5
R-(+)
S-(–)
S-(–)
(+)-6
(+)-6 BF3·Et2O
(+)-7
(+)-8
(+)-9
(+)-10
S-(–)
S-(–)
26
[a] Reaction conditions: amide 2a in THF was cooled to –78 °C
and sBuLi (1.73 equiv.) was added. After 15 min at –78 °C, A*H
(2 equiv.) in THF was added and the mixture was warmed to
– 20 °C over a 30 min period. After cooling again to –78 °C, H2O
Most of the reports on the “external quench” have used the was added. [b] Isolated yield. [c] Determined by chiral HPLC.
Eur. J. Org. Chem. 2009, 5414–5422
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5417

J. Martin, J.-C. Plaquevent, J. Maddaluno, J. Rouden, M.-C. Lasne
FULL PAPER
In our next investigation, asymmetric protonation of lith- 5, if the deuterolysis was carried out at –20 °C rather than
ium enolate 4a was performed under the standard condi- at –78 °C (Table 6, Entry 5), or if all the reaction was en-
tions, but D2O was used to quench the reaction mixture in tirely conducted at –78 °C (Entry 4), the deuterium levels
order to evaluate the amount of deuterium incorporation reached 90% and 70%, respectively, although no asymmet-
at the α-carbon.[34,35] In a preliminary experiment, we ric induction was observed.
checked that the reaction between D2O and 4a, at –78 °C,
Whereas norephedrine (7) and N-methylephedrine (8)
led to the deuteriated amide 2a-D in 100% yield (Table 6, were not efficient in enantioselective protonation of 4a, they
Entry 1, Scheme 1, route “a”). The results summarized in showed different behavior in deuteriation experiments. In-
Table 6 show that, under the same conditions of tempera- deed, deuterium was transferred with up to 90% effective-
ture and time, quenching of the reaction mixture with D2O ness from N-methylephedrine (Table 6, Entry 10) but with
led to deuterium levels of 75–80% with diamine (+)-5 (En- only 22% effectiveness from the more acidic norephedrine
tries 2, 3) and 25% with (–)-ephedrine [(–)-6, Entry 7]. (Entry 9). Substitution of the ephedrine hydroxy group for
These tests showed that in the case of aniline (+)-5 the α- a methoxy group in (+)-9 raised the level of deuterium in-
proton in (S)-2a was delivered mainly from the quenching corporation over that obtained with (–)-ephedrine [(–)-6, 50
reagent (“external quench”). This process was less impor- vs. 25%, Entries 11 and 7] and similar to that observed
tant with ephedrine (6, Entries 6, 7) or norephedrine (7, En- when 6-D was used (Entry 6). This result confirmed the role
try 9), probably because of the higher acidity of the alcohol of the hydroxy group in the delivery of the proton.
function relative to that of the secondary amine of aniline
Finally we studied the effects of the acidities of the
(+)-5.[36,37] The 50% deuterium level observed when eph- quenching reagents on the ee values obtained from the
edrine derivative (+)-6D was used confirmed the role of the asymmetric protonations. The roles of those reagents are
OD/OH and NH groups in the reaction pathway and that usually to regenerate the chiral sources. Experiments (En-
of H2O as source of protonating agent of the enolate. From tries 12–17; cf. Entries 2 or 3) carried out with aniline (+)-5
these observations we were able to suggest that the main and different protic reagents (alcohols, phenol, acetic acid)
process for asymmetric protonation with aniline (+)-5 in- showed similar deuterium levels with use of D2O, ROD, or
volved the formation of a chiral complex including the achi- PhOD, but the degree of enantioselectivity was significantly
ral quenching reagent (H2O or D2O; Scheme 1, route d and lower with the more acidic phenol (Entries 15 and 16). On
then e) whereas with ephedrine (6) an internal quench moving to acetic acid, no asymmetric induction was ob-
(route b) was mainly operating.
served (Entry 17). Moreover, in one experiment, AcOH was
added immediately after the addition of water (Table 6, En-
try 8). In that case a high ee was obtained, suggesting that
enantioselective protonation had been complete before the
addition of AcOH at –78 °C. All these experiments were in
good agreement with our previous hypothesis on the par-
Table 6. Obtained ee values of amide 2a as a function of the
quenching reagent.
Entry
A*H[a]
Quenching reagent
2a-D deuterium ee [%][d]
[T/°C][b]
content [%][c]
1
2
3
4
5
6
7
8
–
D2O [–78]
D2O [–78]
D2O [–78]
D2O [–78]
D2O [–20]
H2O [–78]
D2O [–78]
D2O then AcOH [–78]
D2O [–78]
100
80
75
70
95
50
25
50
22
90
50
70
70
60
–
–
97
80[e]
0
(+)-5
(+)-5
(+)-5[f]
(+)-5
(+)-6D
(–)-6
(+)-6
(+)-7
(+)-8
(+)-9
(+)-5
(+)-5
(+)-5
(+)-5
(+)-5
(+)-5
0
88[e]
92
85[e]
0
9
10
11
12
13
14
15
16
17
D2O [–78]
D2O [–78]
0
5
MeOD [–78]
tBuOD [–78]
tBuOD [–78]
PhOH [–78]
PhOD [–78]
AcOD [–78]
81[e]
87[e]
90
63
66
0[e]
70
50
[a] Reaction conditions: the chiral source was added at –78 °C to
enolate 4a prepared from amide 2a (0.3 mmol), LiBr, and sBuLi in
THF (see Exp. Sect.). [b] 36 equiv. and temperature of addition.
1
[c] Determined by H NMR, the chemical yields being higher than
70%. [d] Determined by chiral HPLC. The enantiomer (S)-(–)-2a
was formed from (+)-5 or (+)-6. [e] Experiment carried out without
LiBr. [f] The reaction was entirely conducted at –78 °C.
Scheme 2. Reagents and conditions. i) ClCOOMe, NaOH (2 ),
pH 8–9, room temp., 14 h, 97%; ii) ClCOOiBu, NEt3, CHCl3, 1 h,
room temp., then piperidine; 56%; iii) sBuLi (1.75 equiv.), LiBr
(1.75 equiv.), rac-2b (3.94 mmol) –78 °C, (–)-(1R,2S)-6 (2 equiv.).
–20 °C, then –78 °C, addition of H2O; 98%; ee: 99%. iv) HCl (6 ),
The deuterium labeling confirmed our previous observa-
tions (Table 1) on the effects of temperature on the obtained
ee values of amide 2a. With the chiral aniline derivative (+)- reflux, 3 d; 99%, ee: 99%.
5418
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5414–5422

Deracemization of Pipecolic Acid Amides
ticipation of the quenching reagent in the enantioselective
process. Proton/deuterium exchanges (equilibrium d in
Scheme 1) have previously been postulated to explain the
deuterium levels observed in asymmetric protonations of
lactones.[38]
Experimental Section
General Methods: Unless otherwise stated, all reactions were per-
formed under argon in glassware oven-dried overnight at 110 °C.
THF was distilled from benzophenone ketyl under argon prior to
use. Diisopropylamine and BF3·Et2O were distilled from calcium
hydride and diethyl ether from lithium aluminium hydride. Lithium
chloride and lithium bromide were dried in vacuo (0.1 Pa) at 50 °C
for 14 h. Zinc bromide was dried at 300 °C in vacuo (1 Pa) for 1 h
and was then sublimed (0.1 Pa). nBuLi and sBuli were titrated by
the described methods.[40] (1S,2R)-(+)-Ephedrine [(+)-6], (1R,2S)-
(–)-ephedrine [(–)-6], (1S,2R)-(+)-norephedrine (7), (1S,2R)-(+)-N-
methylephedrine (8), and (1S,2S)-(+)-pseudoephedrine (10) were
commercially available. (R)-(+)-[5-Chloro-2-(methylamino)phenyl]-
1,2,3,4-tetrahydroisoquinoline was obtained from its commercially
available tartrate.[5] (1S,2R)-O-Methylephedrine (9),[41] methyl 2-
(dipropylcarboxamido)piperidine-1-carboxylate (2a),[9a,14] and
methyl 2-(piperidinecarboxamido)piperidine-1-carboxylate (2b)[14]
have been described previously. All other reagents or catalysts were
used as obtained from commercial sources (purity Ͼ 98%). Thin
layer chromatography (TLC) was performed on silica gel plates
(60 F-254, 0.1 mm) with iodine and/or UV detection. Flash
chromatography was carried out on silica gel (SI 60, 0.040–
0.063 mm, Merck). Optical rotations were measured with a Perkin–
Elmer 241 polarimeter with a sodium lamp (589 nm) as the light
source. Infra-red (IR) spectra were recorded with a Perkin–El-
mer 684 FT-IR spectrometer. 1H NMR and 13C NMR spectra were
recorded at 250 MHz (1H) or 62 MHz (13C) with TMS and residual
protic solvent (CHCl3) as the reference and solvent respectively.
Chemical shifts (δ) are given in parts per million (ppm) and cou-
pling constants (J) are given in Hertz (Hz). The proton spectra
are reported as follows: δ (ppm), number of protons, multiplicity,
coupling constant J (Hz), assignment. 13C NMR spectra are re-
ported as follows: δ (ppm), assignment. DEPT and two-dimen-
sional NMR spectroscopy were used, where appropriate, to aid in
Enantio-Enriched Pipecolic Acids
As a practical application of this method, we prepared
the racemic amide 2b (Scheme 2) on a gram scale and then
subjected it to deracemization under the standard condi-
tions (Table 2, Entry 3) in the presence of (–)-ephedrine
[(–)-6]. The amide (+)-2b was obtained in an excellent yield
and ee (98 and 99%, respectively).
Of the different attempts to hydrolyze the amide 2b, the
acidic conditions (HCl, 6 , reflux, 3 d) gave the best results
in terms both of yields and ee values. The enantiomeric pu-
rity, checked by HPLC, was completely retained. The N-
protected pipecolic acid (+)-(R)-11 (Scheme 2) was ob-
tained from (Ϯ)-pipecolic acid (3) in 51% overall yield and
99% ee. The efficiency of this synthetic process makes it
attractive for the synthesis of enantio-enriched pipecolic
acid derivatives.
Conclusions
An efficient method for deracemization of pipecolamides
2 (ee Ͼ 95%) has been developed. If several results, such as
those relating to the effects of temperature and the choice
of the deprotonating base, are reminiscent of those found
by Vedejs et al., salient observations have been made. a) An
“external quench” by water, added to hydrolyze the reaction
mixture, appeared to be the major process with the aniline
derivative (+)-5, whereas both internal and external
quenches could be involved with the more acidic ephedrines
(6). It would be premature to propose a model for the mixed
aggregates assumed to undergo the enantioselective proton-
ation at this stage, knowledge about aggregates in solution
still being in its infancy. b) The enolate 4a was characterized
by 1H and 13C NMR spectroscopy. NOESY and COSY ex-
periments were in favor of the formation of the E stereoiso-
mer as two rotamers. The efficiency of the enantioselective
protonation did not reflect the ratio of the E/Z configura-
tions of the enolates. It seems that in this case the approach
of the proton is controlled by only one of the two trigonal
centers in the enolate (the carbon in the 2-position).[39]
c) Optimization of the formation of enolate 4a suggested
that a mixed potassium-lithium alkoxide could be an alter-
native to the use of sBuLi. d) The asymmetric protonation
of pipecolamides was not strongly dependent on the substi-
tution of the exo- or endocyclic nitrogens. Finally, we have
shown that deracemization of pipecolamide 2b with the aid
of commercially available (+)- or (–)-ephedrine [(+)- or (–)-
6] can offer an efficient route to the preparation of both
the assignments of signals in the H and 13C NMR spectra. Low-
resolution mass spectra were recorded on a NermagR10 instrument
(EI, 70 eV). Chiral HPLC analyses were performed on a Waters
instrument with chiral stationary columns from Chiralcel.
1
Representative Procedure for the Synthesis of Amides: Triethylamine
(0.77 mL, 5.35 mmol, 1 equiv.) and tert-butyl chloroformate
(0.7 mL, 5.4 mmol, 1.01 equiv.) were added dropwise to a solution
of
(1-methoxycarbonyl)piperidine-2-carboxylic
acid
(1 g,
5.35 mmol) in chloroform (15 mL), cooled to 0 °C. The mixture
was stirred for 1 h at 0 °C. The secondary amine (5.35 mmol,
1 equiv.) was added and the reaction mixture was stirred for 3 h at
0 °C. The reaction was quenched with water, followed by extraction
with dichloromethane. The combined organic layers were washed
successively with brine (20 mL), dilute hydrochloric acid (3 ,
2ϫ20 mL), aqueous sodium hydrogencarbonate (saturated,
2ϫ20 mL), and brine (20 mL) and dried with magnesium sulfate.
After concentration in vacuo, the residue was purified by flash col-
umn chromatography to afford the desired product (Figure 5).
enantiomers of N-protected pipecolic acid on a gram scale. Figure 5. Numbering of carbon atoms of pipecolamides.
Eur. J. Org. Chem. 2009, 5414–5422
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5419

J. Martin, J.-C. Plaquevent, J. Maddaluno, J. Rouden, M.-C. Lasne
FULL PAPER
Methyl 2-(Dimethylcarboxamido)piperidine-1-carboxylate (2c): Two
conformers (38%), colorless oil, b.p. 80 °C (0.1 Pa). [α]2D2 = –29.6
acetate and dilute hydrochloric acid (3 ). The organic layer was
successively washed with dilute hydrochloric acid (3 ), sodium hy-
(c = 3, CHCl3). 1H NMR (250 MHz, CDCl3, 20 °C): δ = 1.42–1.90 droxide (2 ), and brine and was then dried with magnesium sul-
(m, 6 H, 3-H, 4-H, 5-H), 2.94 (s, 3 H, NCH3), 3.06 (s, 3 H, NCH3), fate. After filtration, evaporation of the solvent, and purification by
3.47–3.68 (m, 1 H, 6-H), 3.70 (s, 3 H, OCH3), 3.74–3.93 (m, 1 H, flash chromatography (eluent: CH2Cl2/MeOH, 97:3), the enantio-
6-H), 5.06 (s, 1 H, 2-H) ppm. 13C NMR (62 MHz, CDCl3, 20 °C): enriched amide was isolated. The yields and ee values are shown
δ = 19.5 (4-C), 24.9 (5-C), 26.5 (3-C), 37.1 and 35.8 (10-C, 10Ј-C), in Tables 1, 2, 3, 4, 5, and 6. The ee values were determined by
42.0 (6-C), 51.0 (2-C), 52.7 (8-C), 156.8 (7-C), 175.4 (9-C) ppm. IR chiral HPLC as follows.
(film): ν
= 1698 (NCOO), 1652 (CON) cm–1. MS (EI): m/z (%)
˜
max
Compound 2a: Chiralcel OD, n-heptane/iPrOH 95:5; λ = 220 nm,
flow rate: 0.6 mLmin–1. tR (+)-(R): 11 min; tR (–)-(S): 18 min.
= 215 [M+1] (2), 214 (11), 142 (100) cm–1. C10H18N2O3 (214):
calcd. C 56.06, H 8.46, N 13.07, O 22.40; found C 55.89, H 8.35,
N 13.04, O 22.26.
Compound 2b: Chiralcel OD, n-heptane/iPrOH 95:5; λ = 215 nm,
flow rate: 0.6 mLmin–1. tR (+)-(R): 27 min; tR (–)-(S): 48 min.
tert-Butyl 2-(Dimethylcarboxamido)piperidine-1-carboxylate (2d):
Di-tert-butyl dicarbonate (270 mg, 1.39 mmol, 1.18 equiv.) in
dichloromethane (1 mL) was added to N,N-dipropylpiperidine-2-
carboxamide[9a] (222 mg, 1.18 mmol) in dichloromethane (6 mL).
The mixture was stirred at room temperature for 24 h and the vola-
tile compounds were evaporated. The residue was diluted with
dichloromethane and the solution was washed with diluted hydro-
chloric acid (3 , 3ϫ10 mL), satd. aqueous sodium hydrogen car-
bonate (10 mL), and then brine (10 mL). The organic layer was
dried with magnesium sulfate. After evaporation of the solvent,
amide 2d was isolated (120 mg, 33% yield) as a colorless liquid. 1H
NMR (250 MHz, CDCl3, 20 °C): δ = 4.90–5.10 (m, 1 H, 2-H),
4.10–3.80 (m, 1 H, 1 6-H), 3.29–3.18 (m, 5 H, 1 6-H, 4 10-H), 1.84–
1.47 (m, 10 H, 2 3-H, 2 4-H, 2 5-H, 4 11-H), 0.85 (t, J = 7 Hz, 3
H, 15-H), 0.81 (t, J = 7.3 Hz, 3 H, 15Ј-H) ppm. 13C NMR
(62 MHz, CDCl3, 20 °C): δ = 171.2 (9-C), 157.4 (7-C), 79.6 (8-C)
49.5 (2-C), 47.8 (6-C), 43.8 (2 10-C), 28.5 (3 13-C), 27.2 (3-C), 22.3
Compound 2c: Chiralcel OB, n-hexane/iPrOH 99:1; λ = 210 nm,
flow rate: 0.6 mLmin–1. tR (+)-(R): 35 min; tR (–)-(S): 53 min.
Compound 2d: The ee values were determined from the optical rota-
tion of N,N-dipropylpiperidine-2-carboxamide (12, cf. above).
Deuterium Test. Typical Procedure: In the deracemization pro-
cedure, an aliquot (0.3 mL) of the reaction mixture (base, amide 2,
and solvent) was removed after 10–15 min stirring at –78 °C and
rapidly transferred into a vial containing deuterium oxide (0.5 mL).
The mixture was vigorously stirred and then extracted with deuter-
iochloroform. The solvent was evaporated and the residue was ana-
lyzed by 1H NMR spectroscopy. When the proton in the 2-position
had completely disappeared in the spectrum, the deracemization
experiment was carried out.
Deracemization of 2a with a Bimetallic Base: (–)-Ephedrine [(–)-6,
47 mg, 0.28 mmol, 1.25 equiv.] in THF (0.9 mL) was added at room
temperature to KH (11 mg, 0.28 mmol, 1.25 equiv.). The mixture
was cooled to –40 °C and nBuLi was added (0.31 mL, 0.495 mmol,
2.25 equiv.). After stirring for 40 min at this temperature, the mix-
ture was cooled to –78 °C and amide 2a (60 mg, 0.22 mmol) in
THF (1 mL) was added dropwise. The mixture was stirred for
30 min and (–)-ephedrine [(–)-6, 98 mg, 0.44 mmol, 2 equiv.] in
THF (3 mL) was added. The procedure was then identical to that
described above for the deracemization.
(5-C), 21.0 (4-C), 19.7 (2 11-C), 11.4 (2 12-C) ppm. IR (film): ν
˜
max
= 1701 (NCOO), 1655 (CON) cm–1.
N,N-Dipropylpiperidine-2-carboxamide (12)[9] from Amide 2d: En-
antio-enriched amide (+)- or (–)-2d (40 mg, 0.13 mmol) in chloro-
form (2 mL) was cooled to 0 °C. Trifluoroacetic acid (5 mL) was
added and the solution was stirred at 0 °C for 30 min. The mixture
was allowed to warm to room temp. and the volatile compounds
were evaporated. Chloroform was added to the residue. The or-
ganic layer was washed with sodium hydroxide (2 , 2ϫ5 mL) and
then brine, dried with magnesium sulfate, filtered and concentrated.
From amide (R)-(+)-2d {[α]2D5 = +10 (c = 4, CHCl3) obtained with
(–)-ephedrine 6 for the deracemization}, (R)-(+)-N,N-dimethylpipe-
peridine-2-carboxamide [(+)-12] was obtained in 89% yield and
85% ee {[α]2D5 = +8.9 (c = 2.4, in CHCl3)}.
From amide (S)-(–)-2d {[α]2D5 = –12.5 (c = 4.35, in CHCl3) obtained
with aniline (+)-5 for the deracemization}, (S)-(–)-N,N-dipropyl-
piperidine-2-carboxamide [(–)-12] was obtained in 85% yield and
86% ee {[α]2D5 = –9.0 (c = 2.53, in CHCl3); ref.[9] [α]2D5 = –10.4 (c =
4, in CHCl3)}.
Li Enolate 4a: Methyllithium in diethyl ether (2.2 , 0.15 mL,
0.33 mmol, 3 equiv.) was added by syringe to an 1H NMR tube
previously dried, stoppered with a rubber septum, and flushed with
nitrogen. The diethyl ether was removed under vacuum (2 Pa,
40 min). A white solid was obtained and cooled to –40 °C. [D8]-
THF [0.25 mL, distilled from a few drops of nBuLi in hexanes
(1.6 ) before use] was added. The mixture was cooled to –70 °C
and amide 2a (30 mg, 0.11 mmol) in [D8]THF (0.25 mL) was
1
added. H NMR spectra of the mixture of rotamers (70:30) were
then recorded at –60 °C. The signals were assigned by comparison
1
with those observed in the H NMR of amide 2a recorded under
the same conditions. The enolate was stable for 18 h at –60 °C and
the 1H NMR signals were not modified upon warming until
–40 °C. Signals coalesced at –30 °C.
Deracemization of Amides 2 through Enantioselective Protonation.
Typical Procedure with LiBr as Additive: A solution of amide 2a
(0.15 , 80 mg, 0.3 mmol) and LiBr (145 mg, 1.75 equiv.) in THF
(2 mL) was cooled to –78 °C for 1 h. sBuLi (1.3  in cyclohexane,
1.75 equiv., 0.41 mL) was added over 10 min. After 15 min, a deute-
rium test (see below) was carried out on an aliquot. The yellow
mixture was stirred for 45 min at –78 °C and the chiral proton
source (2 equiv., 0.15 ) in THF (4 mL) was then added over 30–
40 min at this temperature. After 30 min at –78 °C, the yellow solu-
tion was warmed to –30 °C over a 30 min period. The orange-red
mixture was stirred for 5 min at –30 °C and then cooled again to
–78 °C for addition of water (0.2 mL, 37 equiv.[42]). The mixture
was allowed to warm to room temp. The volatile compounds were
removed in vacuo and the residue was partitioned between ethyl
Supporting Information (see also the footnote on the first page of
this article): Preparation and 1H NMR of lithium enolate 4a. Prep-
aration of amide 2b on a gram scale.
Acknowledgments
The authors thank P. and L. Duhamel for helpful discussions and
H. Oulyadi for NMR experiments at low temperature. The work
was supported by the Réseau Interrégional de Chimie Organique
Fine (Contrat de Plan Etat-Régions Haute-Normandie, Basse-Nor-
5420
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5414–5422

Deracemization of Pipecolic Acid Amides
[13]
For reviews, see: a) F. Couty, Amino Acids 1999, 16, 297–320;
b) C. Kadouri-Puchot, S. Comesse, Amino Acids 2005, 29, 101–
130.
mandie), the Centre National de la Recherche Scientifique (CNRS),
and the Ministry of Education and Research.
[14]
[15]
[16]
J. Martin, M.-C. Lasne, J.-C. Plaquevent, L. Duhamel, Tetrahe-
dron Lett. 1997, 38, 7181–7182.
For reproducible results it was necessary to use a freshly
opened bottle of sBuLi.
T. Laube, J. D. Dunitz, D. Seebach, Helv. Chim. Acta 1985, 68,
1373–1393.
[1] Pioneer work: L. Duhamel, C. R. Acad. Sci. 1976, 282C, 125–
127.
[2] For reviews, see: a) L. Duhamel, P. Duhamel, J.-C. Launay, J.-
C. Plaquevent, Bull. Soc. Chim. Fr. 1984, 2, 421–430; b) C.
Fehr, Angew. Chem. Int. Ed. Engl. 1996, 35, 2567–2587; c) A.
Yanagisawa, K. Ishihara, H. Yamamoto, Synlett 1997, 411–
420; d) J. Eames, N. Weerasooriya, Tetrahedron: Asymmetry
2001, 12, 1–24; e) L. Duhamel, P. Duhamel, J.-C. Plaquevent,
Tetrahedron: Asymmetry 2004, 15, 3653–3691.
[3] For recent deracemizations through enantioselective proton-
ations, see: a) D. E. Ward, O. T. Akinnusi, I. Q. Alarcon, V.
Jheengut, J. Shen, J. W. Quail, Tetrahedron: Asymmetry 2004,
15, 2425–2430; b) G. S. Coumbarides, J. Eames, J. E. W.
Scheuermann, K. F. Sibbons, M. J. Suggate, M. Watkinson,
Bull. Chem. Soc. Jpn. 2005, 78, 906–909; c) R. Melgar-Fernan-
dez, E. Juaristi, Tetrahedron Lett. 2005, 46, 1221–1222; d) A.
Yanagisawa, T. Touge, T. Arai, Pure Appl. Chem. 2006, 78, 519–
523; e) C. Fehr, H. Randall, Chem. Biodiversity 2008, 5, 942–
957.
[4] a) A. Yanagisawa, T. Kikuchi, T. Watanabe, T. Kuribayashi, H.
Yamamoto, Synlett 1995, 372–374; b) C. H. Cheon, H. Yama-
moto, J. Am. Chem. Soc. 2008, 130, 9246–9247; c) K. Mitsu-
hashi, R. Ito, T. Arai, A. Yanagisawa, Org. Lett. 2006, 8, 1721–
1724; d) J. Gil, M. Medio-Simon, G. Mancha, G. Asensio, Eur.
J. Org. Chem. 2005, 1561–1567; e) Y. Yamashita, Y. Emura, K.
Odashima, K. Koga, Tetrahedron Lett. 2000, 41, 209–213; f)
M. Medio-Simon, P. Aleman, A. Cuenca, J. Gil, N. Rodriguez,
G. Asensio, ARKIVOC 2005, 266–286; g) T. Poisson, V. Dalla,
F. Marsais, G. Dupas, S. Oudeyer, V. Levacher, Angew. Chem.
Int. Ed. 2007, 46, 7090–7093; h) T. Poisson, S. Oudeyer, V.
Dalla, F. Marsais, V. Levacher, Synlett 2008, 2447–2450; i) M.
Morita, L. Drouin, R. Motoki, Y. Kimura, I. Fujimori, M.
Kanai, M. Shibasaki, J. Am. Chem. Soc. 2009, 131, 3858–3859.
[5] a) E. Vedejs, N. Lee, J. Am. Chem. Soc. 1991, 113, 5483–5485;
b) E. Vedejs, N. Lee, S. T. Sakata, J. Am. Chem. Soc. 1994, 116,
2175–2176; c) E. Vedejs, N. Lee, J. Am. Chem. Soc. 1995, 117,
891–900; d) E. Vedejs, A. W. Kruger, E. Suna, J. Org. Chem.
1999, 64, 7863–7870; e) E. Vedejs, A. W. Kruger, J. Org. Chem.
1998, 63, 2792–2793.
[6] G. S. Coumbarides, J. Eames, S. Ghilagaber, M. J. Suggate, Tet-
rahedron Lett. 2004, 45, 9469–9474.
[7] R. Melgar-Fernandez, R. Gonzalez-Olvera, E. Juaristi, Tetra-
hedron 2005, 61, 4329–4333.
[8] E. Vedejs, A. W. Kruger, N. Lee, S. T. Sakata, M. Stec, E. Suna,
J. Am. Chem. Soc. 2000, 122, 4602–4607, and references cited
therein.
[9] a) J. Martin, A. Deagostino, C. Perrio, F. Dauphin, C. Du-
candas, C. Morin, P.-L. Desbene, M.-C. Lasne, Bioorg. Med.
Chem. 2000, 8, 591–600; b) M.-C. Lasne, L. Barré, C. Huard,
C. Ducandas, E. T. MacKenzie, J. Labelled Compd. Ra-
diopharm. 1994, 35, 425.
[10] e.g a) S. Gupta, S. B. Krasnoff, D. W. Roberts, J. A. A.
Renwick, L. S. Brinen, J. Clardy, J. Am. Chem. Soc. 1991, 113,
707–709; b) K. K. Lee, J. B. Gloer, J. A. Scott, D. Malloch, J.
Org. Chem. 1995, 60, 5384–5385; c) D. L. Boger, J.-H. Chen,
K. W. Saionz, J. Am. Chem. Soc. 1996, 118, 1629–1644; d) W.
Gu, M. Cueto, P. R. Jensen, W. Fenical, R. B. Silverman, Tetra-
hedron 2007, 63, 6535–6541.
[11] P. J. Belshaw, S. D. Meyer, D. D. Johnson, D. Romo, Y. Ikeda,
M. Andrus, D. G. Alberg, L. W. Schultz, J. Clardy, S. L. Schrei-
ber, Synlett 1994, 381–392.
[12] See, for example: a) anesthetic (S)-bupivacaine: B. Adger, U.
Dyer, G. Hutton, M. Woods, Tetrahedron Lett. 1996, 37, 6399–
6402; b) anticonvulsants: B. Ho, P. M. Venkatarangan, S. F.
Cruse, C. N. Hinko, P. H. Andersen, A. M. Crider, A. A. Ad-
loo, D. S. Roane, J. P. Stables, Eur. J. Med. Chem. 1998, 33, 23–
31.
[17]
[18]
D. Seebach, Angew. Chem. Int. Ed. Engl. 1988, 27, 1624–1654.
IPR: the amine proton in an amine–anion complex becomes
reattached to the original carbon more rapidly than the carban-
ion can interact with the external proton source; see ref.[5,16]
P. Riviere, K. Koga, Tetrahedron Lett. 1997, 38, 7589–7592.
T. Yasukata, K. Koga, Tetrahedron: Asymmetry 1993, 4, 35–
38.
G. Asensio, P. A. Aleman, L. R. Domingo, M. Medio-Simón,
Tetrahedron Lett. 1998, 39, 3277–3280, and references cited
therein.
A. Yanagisawa, T. Kikuchi, H. Yamamoto, Synlett 1998, 174–
176.
a) M. J. Schlosser, J. Organomet. Chem. 1967, 8, 9; b) M.
Schlosser, Organometallics in Synthesis Wiley, Chichester, 1994,
p. 72.
W. Bauer, T. Laube, D. Seebach, Chem. Ber. 1985, 118, 764–
773.
J.-Y. Valnot, J. Maddaluno, Patai Series: The Chemistry of
Organolithium Compounds, vol. 2, in: The Chemistry of Func-
tional Groups (Eds.: Z. Rappoport, I. Marek), Wiley, Chiches-
ter, 2006, p. 525.
For a recent characterization of a chiral enolate aggregate, see:
D. Li, C. Sun, P. G. Williard, J. Am. Chem. Soc. 2008, 130,
11726–11736, and references cited therein.
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
P. G. Williard, M. J. Hintze, J. Am. Chem. Soc. 1987, 109,
5539–5541.
From the 13C NMR spectrum, the (E)-4a/(Z)-4a ratio was
higher than 80:20.
For recent chiral proton sources in deracemization through
asymmetric protonation: a) β-hydroxy esters: B. M. Kim, H.
Kim, W. Kim, K. Y. Im, J. K. Park, J. Org. Chem. 2004, 69,
5104–5107; b) C2-symmetric sulfonamides: E. Boyd, G. S. Co-
umbarides, J. Eames, A. Hay, R. V. H. Jones, R. A. Stenson,
M. J. Suggate, Tetrahedron Lett. 2004, 45, 9465–9468.
A. Yanagisawa, T. Kikuchi, T. Kuribayashi, H. Yamamoto,
Tetrahedron 1998, 54, 10253–10264.
K. Flinois, Y. Yuan, C. Bastide, A. Harrison-Marchand, J.
Maddaluno, Tetrahedron 2002, 58, 4707–4716.
A. Yanagisawa, T. Touge, T. Arai, Angew. Chem. Int. Ed. 2005,
44, 1546–1548.
An alternative to the aniline derivative (+)-5 has been sug-
gested: E. Vedejs, P. Trapencieris, E. Suna, J. Org. Chem. 1999,
64, 6724–6729.
For C-deuteriation of ketone enolates, see: a) G. S. Coumbar-
ides, J. Eames, N. Weerasooriya, J. Labelled Compd. Ra-
diopharm. 2002, 45, 965–973; b) J. Eames, G. S. Coumbarides,
M. J. Suggate, N. Weerasooriya, Eur. J. Org. Chem. 2003, 634–
641; c) G. S. Coumbarides, J. Eames, M. J. Suggate, J. Labelled
Compd. Radiopharm. 2004, 47, 359–371; d) G. S. Coumbarides,
J. Eames, M. J. Suggate, N. Weerasooriya, J. Labelled Compd.
Radiopharm. 2006, 49, 641–652; e) M. Begum, S. Chavda, G. S.
Coumbarides, M. Dingan, J. Eames, M. J. Suggate, N. Weera-
sooriya, J. Labelled Compd. Radiopharm. 2006, 49, 707–732.
a) J. Eames, N. Weerasooriya, Tetrahedron Lett. 2000, 41, 521–
523; b) J. Eames, N. Weerasooriya, Chirality 1999, 11, 787–789.
For intramolecular vs. intermolecular proton delivery in keton-
ization of enolates, see: a) H. E. Zimmerman, J. Cheng, Org.
Lett. 2005, 7, 2595–2597; b) H. E. Zimmerman, J. Cheng, J.
Org. Chem. 2006, 71, 873–882 and references cited therein.
[30]
[31]
[32]
[33]
[34]
[35]
[36]
Eur. J. Org. Chem. 2009, 5414–5422
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5421

J. Martin, J.-C. Plaquevent, J. Maddaluno, J. Rouden, M.-C. Lasne
FULL PAPER
[37]
[40] a) L. Duhamel, J. C. Plaquevent, J. Organomet. Chem. 1993,
448, 1–3; b) J. Suffert, J. Org. Chem. 1989, 54, 509–510.
[41] L. Xiao, R. Kitzler, W. Weissensteiner, J. Org. Chem. 2001, 66,
8912–8919.
[42] This amount could be reduced to 5 equiv. The ee values de-
creased dramatically with lower amounts.
pKa of ephedrine: 9.56, pseudoephedrine: 9.74, norephedrine:
8.47. T. H. Nguyen, R. J. Ansell, Org. Biomol. Chem. 2009, 7,
1211–1220.
[38] U. Gerlach, T. Haubenreich, S. Hünig, Chem. Ber. 1994, 127,
1981–1988.
[39] P. Duhamel, L. Duhamel, C. R. Acad. Sci. Paris 1995, 320,
Received: June 30, 2009
Published Online: September 16, 2009
689–694.
5422
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5414–5422