Asymmetrie substitutions of 2-lithiated N-boc-piperidine and N-boc-azepine by dynamic resolution

Article abstract of DOI:10.1002/chem.200903059

Proton abstraction of N-tertbutoxycarbonyl-piperidine (N-Boc-piperidine) with sBuLi and TMEDA provides a racemic organolithium that can be resolved using a chiral ligand. The enantiomeric organolithiums can interconvert so that a dynamic resolution occurs. Two mechanisms for promoting enantioselectivity in the products are possible. Slow addition of an electrophile such as trimethylsilyl chloride allows dynamic resolution under kinetic control (DKR). This process occurs with high enantioselectivity and is successful by catalysis with substoichiometric chiral ligand (catalytic dynamic kinetic resolution). Alternatively, the two enantiomers of this organolithium can be resolved under thermodynamic control with good enantioselectivity (dynamic thermodynamic resolution, DTR). The best ligands found are based on chiral diamino-alkoxides. Using DTR, a variety of electrophiles can be used to provide an asymmetric synthesis of enantiomerically enriched 2-substituted piperidines, including (after Boc deprotection) the alkaloid (+)-ss-conhydrine. The chemistry was extended, albeit with lower yields, to the corresponding 2-substituted sevenmembered azepine ring derivatives.

Full text of DOI:10.1002/chem.200903059

DOI: 10.1002/chem.200903059
Asymmetric Substitutions of 2-Lithiated N-Boc-piperidine and
N-Boc-azepine by Dynamic Resolution
Iain Coldham,*[a] Sophie Raimbault,[a] David T. E. Whittaker,[b] Praful T. Chovatia,[a]
Daniele Leonori,[a] Jignesh J. Patel,[a] and Nadeem S. Sheikh[a]
Abstract: Proton abstraction of N-tert-
butoxycarbonyl-piperidine (N-Boc-pi-
peridine) with sBuLi and TMEDA pro-
vides a racemic organolithium that can
be resolved using a chiral ligand. The
enantiomeric organolithiums can inter-
convert so that a dynamic resolution
occurs. Two mechanisms for promoting
enantioselectivity in the products are
possible. Slow addition of an electro-
phile such as trimethylsilyl chloride
allows dynamic resolution under kinet-
ic control (DKR). This process occurs
with high enantioselectivity and is suc-
cessful by catalysis with substoichio-
metric chiral ligand (catalytic dynamic
kinetic resolution). Alternatively, the
two enantiomers of this organolithium
can be resolved under thermodynamic
control with good enantioselectivity
(dynamic thermodynamic resolution,
DTR). The best ligands found are
based on chiral diamino-alkoxides.
Using DTR, a variety of electrophiles
can be used to provide an asymmetric
synthesis of enantiomerically enriched
2-substituted piperidines, including
(after Boc deprotection) the alkaloid
(+)-b-conhydrine. The chemistry was
extended, albeit with lower yields, to
the corresponding 2-substituted seven-
membered azepine ring derivatives.
Keywords: amines
·
asymmetric
synthesis · carbanions · lithium ·
organometallic compounds
Introduction
ditions. Low temperatures are normally required so as to
avoid enantiomerization.
Enantiomerically enriched products can be generated from
chiral organolithiums by stereoselective electrophilic
quench.[1] Typically the chiral organolithium is generated by
asymmetric deprotonation or stereospecific tin–lithium ex-
change. Asymmetric deprotonation requires that the sub-
strate is amenable to deprotonation at the required position
and that a suitable chiral base can be found to effect this
process. Tin–lithium exchange requires a method to prepare
the organostannane in enantioenriched form and relies on
the successful transmetallation to the organolithium species.
The organolithium formed by either of these methods must
maintain its configurational stability under the reaction con-
An alternative to these methods is the use of an asymmet-
ric substitution of the organolithium in the presence of a
chiral ligand (L*). A considerable advantage of this ap-
proach is that a racemic organolithium can be used which
ideally is configurationally unstable, so that dynamic equili-
bration of the enantiomeric organolithiums occurs
(Scheme 1; k1, k2 ¼ 0). Therefore low-temperature condi-
tions are not a necessity and indeed are not preferable
(except to impart greater chemical stability). Asymmetric in-
duction after quench with an electrophile (E+) arises from a
thermodynamic preference for one of the diastereomeric or-
ganolithium·chiral ligand complexes (known as dynamic
thermodynamic resolution, DTR),[2] and/or from a kinetic
preference (k3 ¼ k4 or k3ꢀ ¼ k4’) for reaction of one of the
diastereomeric organolithium·chiral ligand complexes with
the electrophile (known as dynamic kinetic resolution,
DKR). In all cases, the electrophilic quench needs to be ste-
reoselective (occurring with retention or inversion of config-
uration at the carbanion center).
[a] Prof. I. Coldham, Dr. S. Raimbault, Dr. P. T. Chovatia, D. Leonori,
Dr. J. J. Patel, Dr. N. S. Sheikh
Department of Chemistry, University of Sheffield
Brook Hill, Sheffield S3 7HF (UK)
Fax : (+44)114-222-9346
E-mail: i.coldham@sheffield.ac.uk
[b] Dr. D. T. E. Whittaker
AstraZeneca, Alderley Park
Macclesfield SK10 4TG (UK)
In dynamic resolution under thermodynamic control, elec-
trophilic quench is faster than interconversion of the diaste-
reomeric complexes (k3, k4 > k1, k2). This does not preclude
a kinetic preference for reaction of either of the complexes.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200903059.
4082
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4082 – 4090

FULL PAPER
Results and Discussion
The starting material, N-Boc-piperidine (16) is available
commercially or can be prepared by simple protection of pi-
peridine with Boc2O. Following the method described by
Beak and co-workers,[11] N-Boc-2-lithiopiperidine was pre-
pared by proton abstraction of N-Boc-piperidine with
sBuLi, TMEDA in Et2O at ꢀ788C. This organolithium
could be trapped with electrophiles to give a variety of race-
mic 2-substituted N-Boc-piperidines.
Scheme 1. Pathways for asymmetric induction with a chiral organolithi-
um.
In contrast to the five-membered ring analogue,[12] asym-
metric deprotonation of N-Boc-piperidine is typically low
yielding or occurs with low enantioselectivity.[13] Therefore,
asymmetric substitution could provide a useful strategy for
the formation of enantiomerically enriched 2-substituted pi-
peridines.
In dynamic kinetic resolution, electrophilic quench is slower
than interconversion of the diastereomeric complexes (k3, k4
< k1, k2). In reactions of chiral organolithiums it is impor-
tant to be aware that there can be a combination of both
thermodynamic and kinetic factors that affect the enantio-
mer ratio (er) of the product.
A selection of chiral ligands (Figure 1) was prepared (see
Supporting Information) to test the asymmetric substitution
reaction. Scheme 2 shows the general Scheme for the experi-
ments conducted under conditions for DTR and Table 1 out-
lines the results. When the chiral ligand has an OH group
then it was first treated with an equivalent of BuLi prior to
addition to N-Boc-2-lithiopiperidine. The mixture was
warmed to ꢀ408C for 1.5 h or ꢀ308C for 1 h to promote for-
mation of the thermodynamic ratio of the organolithium·L*
complexes 17 prior to cooling and electrophilic quench with
TMSCl. The organolithium complexes do not crystallize but
remain in solution. The chiral ligand could be recovered by
column chromatography and distillation. Experiments at
higher temperatures or for longer time periods did not im-
prove the results, suggesting that the thermodynamic ratio
had been reached. This was in agreement with the kinetic
parameters for dynamic resolution of this organolithium in
the presence of TMEDA and the ligand 3, in which the rate
constant k=5.5ꢂ10ꢀ4 sꢀ1 was determined, equating to a
half-life, t1/2 ꢁ 21 min at ꢀ308C.[8d]
The area of asymmetric organic synthesis with chiral orga-
nolithiums made significant steps forward with the discovery
of highly selective asymmetric deprotonations by using
sBuLi and the chiral ligand (ꢀ)-sparteine to give dipole-sta-
bilized a-alkoxy- and a-amino-organolithium species.[3]
These organolithium species are typically configurationally
stable at low temperatures such as ꢀ788C and react stereo-
selectively with various electrophiles.[4,5] The first examples
of highly selective asymmetric substitutions by dynamic res-
olution of chiral organolithiums were reported with benzylic
organolithiums.[2,6] Benzylic organolithium species are often
prone to racemize even at low temperature, thereby favour-
ing resolution through a dynamic process. Good results in
the dynamic resolution of allylic, a-thio- and a-seleno-orga-
nolithium species at low temperature also rely on the config-
urational lability of these species.[7] To effect dynamic reso-
lution of a-amino-organolithium species temperatures
higher than ꢀ788C are typically required. Recent kinetic
studies have determined the barriers to enantiomerization
of several chiral a-amino-organolithium species.[8] These
show that dipole-stabilized species such as N-tert-butoxycar-
bonyl-2-lithiopyrrolidine (N-Boc-2-lithiopyrrolidine) have
slightly lower barriers to inversion than so-called “non-stabi-
lized” species such as N-alkyl-2-lithiopyrrolidines. Pertinent
to this paper, the organolithium N-Boc-2-lithiopiperidine (in
Et2O with one equivalent of TMEDA) has been found to
have a barrier to enantiomerization, DG° ꢁ17 kcalmolꢀ1 at
ꢀ308C.[8d]
Previously, we have reported that various N-substituted 2-
lithiopyrrolidines can be resolved in the presence of a chiral
ligand.[9] High levels of enantioselectivity are possible by a
dynamic kinetic or a dynamic thermodynamic resolution,
depending on the N-substituent and the conditions. This
paper describes in full our results on the dynamic resolution
of the corresponding six- and seven-membered ring ana-
logues, namely N-Boc-2-lithiopiperidine and N-Boc-2-lith-
ioazepine. This builds on some preliminary studies commu-
nicated with N-Boc-2-lithiopiperidine.[10]
Figure 1. Chiral ligands L*.
Chem. Eur. J. 2010, 16, 4082 – 4090
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4083

I. Coldham et al.
Scheme 3. Preparation of ligand 24. i) MeNHOMe·HCl (1.5 equiv),
EDCI, DMAP, CH2Cl2, RT, 4 h, 78%; ii) PhMgCl (6 equiv), THF, 08C to
RT, 2 d, 93%; iii) LiAlH4, THF, ꢀ108C, 1 h then heat, 19 h, 84%, dr
11:1; iv) MsCl, Et3N, THF, 08C, 2 h, then evaporate and add CH2Cl2,
Et3N and proline methyl ester hydrochloride, heat, 25 h; v) LiAlH4, THF,
08C then heat, 15 h, 86% over two steps.
Scheme 2. Procedure for DTR of 17. i) sBuLi (1.2 equiv), Et2O, TMEDA
(1.2 equiv), ꢀ788C, 3 h, then L* (1–15) (1.2 equiv) (L* for 3–15 pre-treat-
ed with sBuLi in Et2O), then ꢀ408C, 90 min (or ꢀ308C, 1 h), then
ꢀ788C, 3 equiv Me3SiCl.
From Table 1 it can be seen that the best ligands were dia-
mino-alcohols 6 and 11, leading to opposite major enantio-
mers of the product 18a. The absolute configurations were
verified by preparation of an authentic sample using the
low-yielding asymmetric deprotonation of N-Boc-piperidine
with sBuLi and the chiral ligand (ꢀ)-sparteine.[13]
Ligand 24 was tested in the DTR reaction of the organo-
lithium 17. On the first attempt, the product 18a was
formed with high selectivity (er 92:8, S/R), which seemed to
verify our hypothesis. However we have been unable to re-
produce this enantioselectivity and this ligand has more con-
sistently provided the product 18a with an er 85:15.
Table 1. Enantiomer ratio (er) for 18a from DTR of 17 with ligands 1–
15.
Due to the lengthy synthesis of ligand 24, we selected
ligand 11 as the optimum ligand from those that we had
screened. Further optimization studies were conducted to
improve the conversion to the product 18a as, using condi-
tions in Scheme 2, this product was typically formed in 40–
50% yield (together with about 30% yield of recovered N-
Boc-piperidine 16). Addition of LiCl (10 mol%) or water
(5 mol%) gave slightly reduced yields but unchanged er.
Addition of 5–10 mol% LiOiPr did give a slight improve-
ment in the yield. Carrying out the resolution in the solvent
hexane (instead of Et2O) gave racemic product and using 2-
methyltetrahydrofuran (2-Me-THF) or tert-butyl methyl
ether (TBME) gave similar yields to the use of Et2O but
with slightly lower er (~70:30). However, we found im-
proved yields with good enantiomer ratio (er 85:15) by
adding, after addition of the chiral diaminoalkoxide ligand
(dissolved in Et2O), hexane [to provide a final solution of
Et2O/hexane 1:1 with concentration of 0.09m]. This gave the
silane 18a in 65% yield (the piperidine 16 was the only
other identifiable product recovered). We then screened a
selection of electrophiles using these optimized conditions
(Scheme 4) and the structures of the products with yields
and enantiomer ratios are given in Figure 2.
Ligand L*
er (S/R)[a]
Ligand L*
er (S/R)[a]
1
2
3
4
5
6
7
8
45:55
50:50
77:23
42:58
80:20
15:85
41:59
51:49
9
51:49
72:28
85:15
28:72[b]
52:48
51:49
81:19
10
11
12
13
14
15
[a] Determined by chiral GC. [b] Corrected from ref. [10].
A comparison of ligands 14 and 15 reveals that the extra
methyl group in 15 provides an enhancement to the enantio-
selectivity. In contrast to ligand 15, ligand 8 (which lacks the
phenyl group) was poor. Replacing the methyl group (ligand
8) with an isopropyl or isobutyl group (ligands 5 and 11)
gave significantly improved selectivities. Therefore, we won-
dered whether replacing the methyl group of ligand 15 with
an isopropyl or isobutyl group would enhance the selectivity
further.
To test this hypothesis, we selected ligand 24 as shown in
Scheme 3. The known amino-acid derivative 19 was pre-
pared in two steps from valine according to the literature.[14]
This was converted to the Weinreb amide 20 which was
treated with phenyl magnesium chloride to give the ketone
21. Reduction of the ketone with simultaneous reduction of
the N-Boc group gave the amino-alcohol 22. This was
formed with high diastereoselectivity (dr 11:1) in favour of
the desired isomer 22 (as determined by comparison with re-
lated amino alcohols).[15] After separation of the minor
isomer, the amino-alcohol 22 was converted to the ester 23
via the intermediate aziridinium ion.[16] Finally, reduction of
the crude ester gave the ligand 24. The structure of the
ligand 24 was confirmed by single crystal X-ray analysis (see
Supporting Information).
The organolithium 17 resolves to a ratio of about 85:15
(S/R) in the presence of the chiral ligand 11. By cooling this
Scheme 4. Optimized procedure for DTR of organolithium 17 (formed
by deprotonation of 16). i) sBuLi (1.2 equiv), Et2O, TMEDA (1.2 equiv),
ꢀ788C, 3 h then 11 (1.2 equiv) (pre-treated with sBuLi in Et2O), then
hexane, then ꢀ308C, 1 h then ꢀ788C, 3 equiv electrophile E+; E+
=
Me3SiCl, Bu3SnCl, CO2, PhSSPh, acetone (for 18e add MeOH prior to
warming slowly to RT).
4084
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4082 – 4090

Asymmetric Substitutions of Lithiated Compounds
FULL PAPER
complex to ꢀ788C equilibration is prevented and the enan-
tiomer ratio of the products reflects the thermodynamic
ratio of the organolithium·11 complexes. Good selectivities
were obtained using the electrophiles TMSCl, Bu3SnCl and
CO2 to give 18a–c (Figure 2). There was sometimes some
erosion in the er, as observed for the electrophile PhSSPh
(or with p-TolSSTol) to give 18d, or with acetone, which
gave the alcohol product 18e if methanol was added at
ꢀ788C. However, if the mixture was allowed to warm to
room temperature before addition of methanol, then the
product was the oxazolidinone 18 f. With MeI (or Me2SO4)
as the electrophile, the yield was very poor (<10%) but the
selectivity was high (er 84:16 S/R).
Scheme 5. Synthesis of (+)-b-conhydrine. i) sBuLi, Et2O, TMEDA,
ꢀ788C, 3 h then 11 (pre-treated with sBuLi in Et2O), then hexane, then
ꢀ308C, 1 h then ꢀ788C, propionaldehyde, then after 1 h MeOH and
warm to RT gave 18g, 77%, dr 62:38, er major 84:16, er minor 82:18,
yield of major diastereomer 48%; ii) using major diastereomer of 18g,
TFA, CH2Cl2, RT gave b-conhydrine, 100%, er ~84:16.
good enantioselectivity (er 87:13 for major isomer, as deter-
mined by 1H NMR spectroscopy with the Pirkle solvating
agent). A similar (but racemic) procedure has been conduct-
ed on a 4-substituted N-Boc-piperidine as part of a synthesis
of an NK1 antagonist.[18]
Scheme 6. Synthesis of products 18h and 25. i) sBuLi, Et2O, TMEDA,
ꢀ788C, 3 h then 11 (pre-treated with sBuLi in Et2O), then hexane, then
ꢀ308C, 1 h then ꢀ788C, benzaldehyde, then after 3.5 h MeOH and warm
to RT to give 18h, 43%, dr 58:42; or ii) as above but with warming to
RT prior to addition of MeOH to give 25, 48%, dr 80:20, er major 87:13,
er minor 75:25.
Figure 2. Products from DTR of 17 with ligand 11 and different electro-
philes.
In 2008, we reported that the organolithium derived from
deprotonation of N-Boc-piperidine with sBuLi and TMEDA
could be converted to its organozinc derivative and used in
Negishi-type coupling reactions with aryl bromides to give
2-arylpiperidines.[19] To our disappointment, this procedure
was unsuccessful using the DTR method. It appears that
transmetallation of the resolved organolithium 17·L* to the
organozinc species was taking place (since subsequent addi-
tion of CuCN·2LiCl then allylation was successful).[10b] How-
ever, in the presence of the chiral diamino-alkoxide ligand,
no transmetallation with the aryl palladium bromide takes
place, possibly due to the presence of the bulky ligand that
can coordinate to the zinc metal. Attempts to swamp this
species with excess TMEDA to allow ligand exchange fol-
lowed by Negishi reaction were unsuccessful.
Carrying out the DTR of the organolithium 17 with the
chiral ligand 11 followed by electrophilic quench with pro-
pionaldehyde gave the product 18g as a mixture of diaste-
reomers (dr 62:38) (Scheme 5). Chiral GC analysis gave
four separate peaks (one for each enantiomer of the diaste-
reomers) and revealed good enantioselectivity (major dia-
stereomer er 84:16, minor diastereomer er 82:18). The dia-
stereomers were separated by column chromoatography and
hydrolysis of the major diastereomer gave the alkaloid (+)-
b-conhydrine (er 84:16; NMR spectroscopic data matched
the literature for this diastereomer).17 The specific rotation,
[a]2D2 = +6.0 (1.0, EtOH) was in line with that expected
based on the literature value for enantiopure b-conhydrine,
[a]D20 = +8.3 (0.9, EtOH).[17c] This two-step procedure (over-
all yield from 16, 48%) represents a remarkably rapid
access to this alkaloid. As far as we are aware, there are
only four previous syntheses of enantioenriched (+)-b-con-
hydrine and these have required 4–11 steps.[17]
Therefore, to prepare enantiomerically enriched 2-arylpi-
peridines, we took the organostannane 18b (er 84:16) and
prepared the organolithium in the absence of the chiral
ligand by transmetallation with nBuLi and TMEDA
(Scheme 7). Subsequent addition of ZnCl2 then a mixture of
the bromobenzene, PdACHTUNGTRNENUG(OAc)2 and (tBu)3P·HBF4 gave the
Addition of benzaldehyde as the electrophile to the re-
solved organolithium 17·11 gave the desired alcohols 18h
(er not determined) (Scheme 6). By allowing the product
mixture to warm to room temperature before addition of
methanol, the oxazolidinones 25 were obtained directly
from N-Boc-piperidine 16. The oxazolidinones 25 were
formed with reasonable diastereoselectivity (dr 80:20) and
desired 2-phenylpiperidine (26a).[19] The same procedure
with 4-bromoanisole gave the piperidine 26b. In each case
the 2-arylpiperidines were formed with er 82:18 (determined
by chiral HPLC), showing that there was little or no racemi-
zation during the whole transmetallation and arylation pro-
cess. The absolute configuration of the products 26 was as-
sumed to be R on the basis of expected retention of configu-
Chem. Eur. J. 2010, 16, 4082 – 4090
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4085

I. Coldham et al.
ration[20] and by analogy with other electrophiles (Figure 2).
This was verified from the specific rotation of compound
26a which was measured as [a]2D2 = +63.0 (0.88, CHCl3),
and which compares with the reported value for (R)-26a,
[a]2D5 = +83.7 (0.98, CHCl3).[21a]
Notably, the major enantiomer was opposite to that ob-
tained in the DTR chemistry [for example, DTR with ligand
4 gave silane 18a, er 42:58 (S/R)]. This shows that the minor
diastereomeric complex is more reactive.
Using the optimized DTR procedure with the ligand 6 re-
sulted in the formation of the opposite major enantiomer of
the products 18a, 18b and 18 f (Scheme 8).
Table 2. Results for DKR of 17.[a]
Ligand L* Yield [%] 18a, er (S/R) Ligand L* Yield [%] 18a, er (S/R)
1
2
3
55, 47:53
86, 49:51
45, 28:72
4
6
7
62, 93:7
71, 82:18
60, 89:11
[a] Starting with racemic stannane 18b and using nBuLi, Et2O, TMEDA
then L*, ꢀ788C then warm to ꢀ408C or above then slow addition of
TMSCl.
Scheme 7. Synthesis of 2-arylpiperidines. i) nBuLi, Et2O, TMEDA,
The DKR protocol was then carried out using proton ab-
straction of N-Boc-piperidine (16) with sBuLi, Et2O and
TMEDA followed by addition of the ligand 4 (pre-treated
with an equivalent of BuLi). Slow addition of TMSCl at
ꢀ208C gave the desired product 18a (66%, er 92:8, S/R).
We found some improvement in the er by conducting the
DKR in the solvent THF (Scheme 9). Under these condi-
tions using chiral ligand 4, the product (S)-18a was formed
with er 95:5. Normally THF is a
ꢀ788C, 1 h then ZnCl2, warm to RT, then ArBr, Pd
ACHTUNGRTEN(NUNG OAc)2 and
(tBu)3P·HBF4, RT, 16 h, Ar=Ph 71%, er 82:18, Ar=p-MeOC6H4 56%,
er 82:18.
Scheme 8. Formation of the enantiomeric products using DTR with
ligand 6. i) sBuLi (1.2 equiv), Et2O, TMEDA (1.2 equiv), ꢀ788C, 3 h then
6 (1.2 equiv) (pre-treated with sBuLi in Et2O), then hexane, then ꢀ308C,
1 h then ꢀ788C, 3 equiv electrophile E+; E+ =Me3SiCl, Bu3SnCl or ace-
tone to give, respectively, 18a 59%, er 80:20, 18b 60%, er 80:20 or 18 f
33%, er 75:25.
poor solvent for asymmetric or-
ganolithium chemistry as it
competes with the chiral ligand
for complexation to the lithium
atom. Clearly in this case it is
slightly preferable for kinetic
resolution.
Scheme 9. Optimized
dure for DKR of organolithi-
um 17 (formed by deprotona-
proce-
Hence, by using DTR, the organolithium 17 can be re-
solved with good enantioselectivity and can be converted to
a selection of enantioenriched 2-substituted piperidines. At-
tempts were made to conduct the DTR of 17 with substoi-
chiometric amounts of chiral ligand[8c] (3, 5 or 11) using vari-
ous temperatures and equilibration times. However, after
electrophilic quench at low temperature with TMSCl, the
silane 18a was isolated with very little or no enantioselectiv-
ity.
The good results for the
DKR indicate that the organo-
lithium 17 complexed with the
ligand 4 is considerably more
reactive to TMSCl than any or-
ganolithium complexed with
THF, Et2O or TMEDA. To test
this, we wondered if the DKR
reaction could be conducted
tion
of
16).
i)
sBuLi
(1.2 equiv), THF, TMEDA
(1.2 equiv), ꢀ788C, 3 h then 4
(1.5 equiv) (pre-treated with
nBuLi in THF), then ꢀ208C,
then slow addition of TMSCl
(4 equiv) over 1 h, 18a 60%,
er 95:5 (S/R).
For comparison with the DTR chemistry, we wanted to
determine whether the organolithium 17, when complexed
to a chiral ligand, could undergo dynamic kinetic resolution
(DKR). It is quite possible that one of the diastereomeric
complexes 17·L* is more reactive than the other. To test for
DKR, we need the rate of electrophilic quench to be slower
than the rate of interconversion of the diastereomeric orga-
nolithium complexes. This can be achieved by adding the
electrophile slowly at a temperature in which equilibration
occurs. We screened several of the chiral ligands shown in
Figure 1 for DKR with the electrophile TMSCl, which was
added slowly (over ~1 h) at ꢀ10 or ꢀ408C (Table 2). The
ligand (ꢀ)-sparteine (1) and the diamine 2 gave poor results.
However the ligands 3, 4, 6 and 7 gave significant asymmet-
ric induction. The ligand 3 has previously been found to
give good selectivity for the DKR of N-Boc-2-lithiopyrroli-
dine.[9c] The best er value was obtained with the ligand 4.
with substoichiometric amounts of the chiral ligand 4. This
would rely on ligand exchange, a process that is known with
related organolithiums and used for catalytic asymmetric de-
protonation.[22] We were pleased to find that catalytic DKR
was possible using as little as 10 mol% 4 in Et2O or THF
using a slightly slower rate of addition of TMSCl (2 equiv
over 1.5 h) (Scheme 10). This gave the silane 18a in reasona-
ble yield (54%) and very good enantioselectivity (er 96:4 S/
R). Higher yields (77%, er 90:10) could be obtained by
using the racemic stannane 18b as the starting material
(with tin–lithium exchange using nBuLi to form the organo-
lithium 17). This result is an interesting example of a catalyt-
ic dynamic kinetic resolution. A reason for its success could
be due to the higher reactivity of the chiral ligand complex
over other complexes in solution and that ligand exchange is
faster than electrophilic quench. A possible catalytic cycle is
represented in Scheme 10 (the organolithiums are drawn as
4086
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4082 – 4090

Asymmetric Substitutions of Lithiated Compounds
FULL PAPER
monomers although the aggregation state is unknown). The
chiral ligand must be able to displace the solvent or
TMEDA (represented by Ln in Scheme 10) from the orga-
nolithium 17·Ln (since good er values are obtained by
DTR). With 10 mol% of the chiral ligand 4, a small amount
of the complexes 17·4 will be formed and interconversion
between the diastereomeric complexes at ꢀ208C leads to a
(slight) preference for (R)-17·4 (see Table 1, er 42:58 S/R).
Addition of TMSCl then leads to reaction preferentially
with (S)-17·4. This releases the lithium alkoxide of the
ligand 4 to allow re-complexation with 17·Ln and continua-
tion of the catalytic cycle.
An alternative explanation for the high selectivities with
silyl halides could be that the electrophile reacts first with
the chiral ligand and is then transferred to the organolithi-
um. This might be expected to be favoured for the silyl hal-
ides that could have a fast reaction with a lithium alkoxide.
We therefore wanted to test the ability of the trimethylsilyl
ether of the chiral ligand 4 to promote DKR. Previous work
using chiral ligands that lack the OH group (e.g. the methyl
ether of diamino-alcohol 3) has led to poor enantioselectivi-
ties in related dynamic resolutions.[9d]
The silyl ether 27 was prepared according to a related
method,[23] and was purified by two successive distillations
under reduced pressure. The organolithium 17 was prepared
by deprotonation of N-Boc-piperidine (16) using sBuLi,
THF and TMEDA at ꢀ788C, then a solution of the silyl
ether 27 (1 equiv) in THF was added (Scheme 11). In one
experiment the temperature was maintained at ꢀ788C for
2 h then was quenched by addition of MeOH. This gave the
silane product 18a in low yield (14%, with 75% recovered
16) but as essentially a single enantiomer (R enantiomer
could not be detected by chiral GC analysis). In another ex-
periment, the mixture was warmed to ꢀ208C. After 2 h,
MeOH was added and purification by column chromatogra-
phy gave the silane 18a in reasonable yield (48%, with 40%
recovered 16) and excellent enantiopurity (er 99:1 S/R).
Scheme 10. Catalytic DKR of 17. i, sBuLi (1.2 equiv), Et2O, TMEDA
(1.2 equiv), ꢀ788C, 3 h then 4 (0.1 equiv) (pre-treated with nBuLi in
Et2O), then ꢀ208C, then slow addition of TMSCl (2 equiv) over 1.5 h,
18a 54%, er 96:4 (S/R).
For a successful catalytic cycle, the electrophilic quench
must occur only with 17·4 and must be much slower with
17·Ln. This was possible with TMSCl, but low er values were
obtained with other electrophiles (Bu3SnCl, DMF, allyl bro-
mide, Me2SO4), even using stoichiometric amounts of the
chiral ligand 4. Good selectivity was obtained using slow ad-
dition of PhMe2SiCl to give N-Boc-2-(dimethylphenylsilyl)-
piperidine (58% yield, er 88:12). The poor selectivity with
electrophiles other than silyl chlorides shows that the enan-
tioselectivity is dependent on the electrophile, and this
would be expected for a kinetic resolution. Our understand-
ing is that more reactive electrophiles are less able to dis-
criminate between the diastereomeric complexes (which are
present in almost equal amounts for ligand 4). By using elec-
trophiles that react faster than TMSCl, or are less selective
for one of the complexes, there will be less effective kinetic
resolution.
There are, however, several other possible explanations
for the high selectivities obtained only with silyl halides.
Firstly, we wondered whether the presence of LiCl could be
a factor (which would be formed after reaction with TMSCl
as the electrophile) and could potentially provide a different
organolithium aggregate structure. Therefore, we added
LiCl (1.1 equiv in THF) to 17·4 followed by warming to
ꢀ208C and slow addition of Me2SO4. However this resulted
in racemic N-Boc-2-methylpiperidine.
Scheme 11. Kinetic resolution of organolithium 17 (formed by deprotona-
tion of 16) with silyl ether 27. i) sBuLi (1.2 equiv), THF, TMEDA
(1.2 equiv), ꢀ788C, 2 h then 27 (1 equiv), and either 2 h then MeOH to
give 18a 14% er >99.5:<0.5 (S/R), or warm to ꢀ208C, then 2 h then
MeOH to give 18a 48%, er 99:1 (S/R).
These results show that silyl ether 27 provides excellent
control of the enantioselectivity in the kinetic resolution of
the organolithium 17. At low temperature there must be a
50:50 mixture of the enantiomeric organolithiums to which
the silyl ether 27 was added. This silyl ether is not a good
electrophile and either selects to react (on the basis of rela-
tive kinetics) with one of the enantiomeric organolithiums
(coordinated to TMEDA), or this diamino-silyl ether can
coordinate to (S)-17, or more likely to both (S)- and (R)-17,
and then delivers the silyl group with a strong preference to
(S)-17. These results tend to favour an explanation for the
high enantioselectivity in the DKR experiments with silyl
electrophiles as involving an unusual intermolecular retro-
Brook type rearrangement. However, it is possible that,
after partial reaction, the released lithium alkoxide of ligand
4 could then be coordinating to the organolithium 17 and
(S)-17·4 could then react with the silyl ether 27 by an inter-
molecular process (possibly with double asymmetric induc-
tion). Unfortunately, low yields and no enantioselectivity
Chem. Eur. J. 2010, 16, 4082 – 4090
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4087

I. Coldham et al.
were obtained in the corresponding reaction of the tributyl-
stannyl ether of ligand 4.
Deprotonation a- to the nitrogen atom in N-acylpiperi-
dines is known to provide the 2-lithiated intermediate in
[13a,24]
ꢀ
which the carbon lithium bond is equatorial.
This
Scheme 13. Procedure for DTR of 28. i) sBuLi (1.2 equiv), Et2O,
TMEDA (1.2 equiv), ꢀ788C, 3 h then 11 (1.2 equiv) (pre-treated with
sBuLi in Et2O), then hexane, then ꢀ308C, 1 h then ꢀ788C, 3 equiv
Me3SiCl, 29a 33% er 92:8 (35% recovered 28); Bu3SnCl, 29b 29% er
90:10; CO2, 29c 18% er ~87:13.
allows coordination of the lithium atom to the carbonyl
oxygen atom. Subsequent reaction with an electrophile
occurs with retention of configuration to give equatorially
substituted products. Therefore, proton abstraction of N-
Boc-piperidine with sBuLi in Et2O/TMEDA provides a mix-
ture of the epimeric organolithiums (S)- and (R)-17
(Scheme 12, where L=TMEDA or an added chiral ligand).
These are drawn as monomers (the aggregation state is un-
known), as calculations on the related N-Boc-2-lithiopyrroli-
dine suggest that inversion occurs through the monomer.[8a]
A possible mechanism for interconversion between these
with similar yield and er. The acid 29c was formed in low
yield but again with similar er. In this case the er was esti-
mated from a comparison of the specific rotation with the
literature: found [a]2D2 = +46.3 (1.0, MeOH), lit. for S enan-
tiomer, er 95:5, [a]D20 = ꢀ59.1 (1.0, MeOH).[27] This verified
that the major enantiomer was (R)-29c, as shown in
Scheme 13, and as expected based on electrophilic quench
with retention of configuration from (S)-N-Boc-2-lithioaze-
pine (the S- organolithium was expected for ligand 11, as
found for the 2-lithiopiperidine 17).
The azepine substrate 28 appears to be a poorer substrate
for proton abstraction than the corresponding five- and six-
membered analogues, so yields tend to be low. However,
enantioselectivities are good in the DTR with diamino-alk-
oxide ligands, so there is scope for this chemistry in asym-
metric synthesis after finding an alternative method for the
formation of the organolithium.
ꢀ
species involves breakage of the carbon lithium bond to
give the carbanion and lithium cation which may be associ-
ated with the carbonyl oxygen atom (conducted tour type
mechanism[25]). Inversion of the carbanion is required
(drawn as sp3 hybridised but it could be sp2) together with a
ring-flip from one chair conformation to the other, so that
re-association of the lithium cation provides the equatorial
ꢀ
carbon lithium bond.
Conclusion
The organolithiums N-Boc-2-lithiopiperidine and N-Boc-2-
lithioazepine can be resolved in the presence of a chiral
ligand. The dynamic equilibration of the diastereomeric
complexes occurs within one hour at ꢀ308C. After electro-
philic quench, good enantiomer ratios of several 2-substitut-
ed N-Boc-piperidines and N-Boc-azepines can be obtained.
Using N-Boc-2-lithiopiperidine and TMSCl as the electro-
phile, a highly enantioselective catalytic dynamic kinetic res-
olution can be promoted.
Scheme 12. Possible interconversion mechanism for 17.
Finally, we studied the dynamic resolution of the corre-
sponding seven-membered ring. Racemic deprotonation of
N-Boc-azepine with sBuLi in Et2O/TMEDA is known and
electrophilic quench has been reported with TMSCl or
Bu3SnCl.[26] Hence, treatment of the organolithium derived
from deprotonation of N-Boc-azepine (28) with the chiral
ligand 11 (pre-treated with sBuLi) then (after addition of
hexane) warming to ꢀ308C for 1 h (to allow equilibration to
the thermodynamic ratio of complexes), followed by cooling
to ꢀ788C and electrophilic quench with TMSCl, Bu3SnCl or
CO2 gave the products 29a-c (Scheme 13).
We screened several of the chiral ligands for the forma-
tion of the silane 29a and found the best result using ligand
11, which gave 29a, er 92:8 (ligands 3, 5 and 24 gave 29a
with er 77:23, 83:17 and 89:11, respectively). This is an im-
proved enantioselectivity in comparison with the corre-
sponding DTR of the six-membered ring 16. The yield of
the silane 29a was low, although the DTR was not opti-
mized for temperature or time and some azepine 28 was re-
covered from the reactions. The stannane 29b was formed
Experimental Section
General procedure for the dynamic thermodynamic resolution of N-Boc-
2-lithiopiperidine: sBuLi (1.2 equiv, 1.3m in hexanes) was added to N-
Boc-piperidine 16 (1.0 equiv) and TMEDA (1.2 equiv) in Et2O (0.5m) at
ꢀ788C. After 3 h, the deprotonated ligand 11 [prepared by adding sBuLi
(1.25 equiv, 1.3m in hexanes) to a 0.5m solution of 11 (1.2 equiv) in Et2O
at 08C for 30 min] was added. The mixture was diluted by a factor of two
by the addition of hexane and was warmed to ꢀ308C. After 1 h, the mix-
ture was cooled to ꢀ788C and the electrophile (3.0 equiv) was added.
The mixture was allowed to warm slowly (over 18 h) to room tempera-
ture and MeOH was added. The solvent was evaporated and the residue
was purified by column chromatography on silica, eluting with hexane-
EtOAc, to give the product. The chiral ligand 11 was recovered by
column chromatography, eluting with CH2Cl2/MeOH 7:3, followed by
evaporation of the solvent, acid/base wash (acidify with 2m HCl, wash
4088
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4082 – 4090

Asymmetric Substitutions of Lithiated Compounds
FULL PAPER
with CH2Cl2, basify with NaOH pellets and extract with CH2Cl2) and dis-
tillation under reduced pressure.
hedron Lett. 2008, 49, 6327–6329; d) J. Clayden, L. Lemiꢃgre, M.
Pickworth, Tetrahedron: Asymmetry 2008, 19, 2218–2221; e) E. Ku-
liszewska, M. Hanbauer, F. Hammerschmidt, Chem. Eur. J. 2008, 14,
8603–8614; f) J. Clayden, U. Hennecke, Org. Lett. 2008, 10, 3567–
3570; g) P. OÇa-Burgos, I. Fernꢄndez, L. Roces, L. Torre-Fernꢄndez,
S. Garcꢅa-Granda, F. Lꢆpez-Ortiz, Org. Lett. 2008, 10, 3195–3198;
h) J. L. Bilke, P. OꢀBrien, J. Org. Chem. 2008, 73, 6452–6454; i) D.
Stead, P. OꢀBrien, A. Sanderson, Org. Lett. 2008, 10, 1409–1412;
j) I. Coldham, J. J. Patel, S. Raimbault, D. T. E. Whittaker, H.
Adams, G. Y. Fang, V. K. Aggarwal, Org. Lett. 2008, 10, 141–143.
[6] For some examples of benzylic organolithium species, see a) D. J.
Gallagher, H. Du, S. A. Long, P. Beak, J. Am. Chem. Soc. 1996, 118,
11391–11398; b) G. P. Lutz, H. Du, D. J. Gallagher, P. Beak, J. Org.
Chem. 1996, 61, 4542–4554; c) S. Thayumanavan, A. Basu, P. Beak,
J. Am. Chem. Soc. 1997, 119, 8209–8216; d) Y. Kim, E. Shin, P.
Beak, Y. S. Park, Synthesis 2006, 3805–3808; e) S.-C. Kuo, F. Chen,
D. Hou, A. Kim-Meade, C. Bernard, J. Liu, S. Levy, G. G. Wu, J.
Org. Chem. 2003, 68, 4984–4987; f) H. Lange, R. Huenerbein, R.
Frçhlich, S. Grimme, D. Hoppe, Chem. Asian J. 2008, 3, 78–87;
g) H. Lange, R. Huenerbein, B. Wibbeling, R. Frçhlich, S. Grimme,
D. Hoppe, Synthesis 2008, 2905–2918.
[7] a) I. Hoppe, M. Marsch, K. Harms, G. Boche, D. Hoppe, Angew.
Chem. 1995, 107, 2328–2330; Angew. Chem. Int. Ed. Engl. 1995, 34,
2158–2160; b) S. Nakamura, R. Nakagawa, Y. Watanabe, T. Toru, J.
Am. Chem. Soc. 2000, 122, 11340–11347; c) S. Nakamura, N. Hirata,
R. Yamada, T. Kita, N. Shibata, T. Toru, Chem. Eur. J. 2008, 14,
5519–5527; d) R. P. Sonawane, R. Frçhlich, D. Hoppe, Adv. Synth.
Catal. 2006, 348, 1847–1854; e) R. Otte, R. Frohlich, B. Wibbeling,
D. Hoppe, Angew. Chem. 2005, 117, 5629–5632; Angew. Chem. Int.
Ed. 2005, 44, 5492–5496; f) R. W. Hoffmann, W. Klute, Chem. Eur.
J. 1996, 2, 694–700; g) S. Nakamura, T. Aoki, T. Ogura, L. Wang, T.
Toru, J. Org. Chem. 2004, 69, 8916–8923; h) H. Lange, K. Bergand-
er, R. Frçhlich, S. Kehr, S. Nakamura, N. Shibata, T. Toru, D.
Hoppe, Chem. Asian J. 2008, 3, 88–101.
Procedure for the catalytic dynamic kinetic resolution of N-Boc-2-lithio-
piperidine: sBuLi (1.2 equiv, 1.3m in hexanes) was added to N-Boc-piper-
idine 16 (1.0 equiv) and TMEDA (1.2 equiv) in Et2O (0.5m) at ꢀ788C.
After 3 h, the deprotonated ligand
4 [prepared by adding sBuLi
(0.1 equiv, 1.3m in hexanes) to a 0.5m solution of 4 (0.1 equiv) in Et2O at
08C for 30 min] was added. The mixture was warmed to ꢀ208C and
TMSCl (2.0 equiv) was added slowly (over 1.5 h). Methanol was added,
the solvent was evaporated and the residue was purified by column chro-
matography on silica, eluting with hexane/EtOAc 98:2, to give the prod-
uct 18a.
General procedure for the dynamic thermodynamic resolution of N-Boc-
2-lithioazepine: sBuLi (1.2 equiv, 1.3m in hexanes) was added to N-Boc-
azepine 28 (1.0 equiv) and TMEDA (1.2 equiv) in Et2O (0.5m) at ꢀ788C.
After 3 h, the deprotonated ligand 11 [prepared by adding sBuLi
(1.2 equiv, 1.3m in hexanes) to a 0.5m solution of 11 (1.2 equiv) in Et2O
at 08C for 30 min] was added. The mixture was diluted by a factor of two
by the addition of hexane and was warmed to ꢀ308C. After 1 h, the mix-
ture was cooled to ꢀ788C and the electrophile (3.0 equiv) was added.
The mixture was allowed to warm slowly (over 18 h) to room tempera-
ture and MeOH was added. The solvent was evaporated and the residue
was purified by column chromatography on silica, eluting with hexane-
EtOAc, to give the product.
Procedures and data for the ligand syntheses 2–15, 24 and 27 are given in
the Supporting Information.
Data for the piperidines 18a–h, (+)-b-conhydrine, 25, 26a–b and aze-
pines 29a–c are given in the Supporting Information.
Acknowledgements
[8] a) N. J. Ashweek, P. Brandt, I. Coldham, S. Dufour, R. E. Gawley, F.
Haeffner, R. Klein, G. Sanchez-Jimenez, J. Am. Chem. Soc. 2005,
127, 449–457; b) T. I. Yousaf, R. L. Williams, I. Coldham, R. E.
Gawley, Chem. Commun. 2008, 97–98; c) T. K. Beng, T. I. Yousaf, I.
Coldham, R. E. Gawley, J. Am. Chem. Soc. 2009, 131, 6908–6909;
d) I. Coldham, D. Leonori, T. K. Beng, R. E. Gawley, Chem.
Commun. 2009, 5239–5241.
We thank the EPSRC (grants GR/S35202/01, EP/E012272/1 and EP/
F000340/1), the University of Sheffield and AstraZeneca for support of
this work. The Leverhulme Trust is acknowledged for a Study Abroad
Fellowship (RF/4/RFG/2009/0057). Steven Robinson is thanked for help
with the preparation of ligands 11 and 24 and Harry Adams is thanked
for the X-ray crystal structure of ligand 24.[28]
[9] a) I. Coldham, S. Dufour, T. F. N. Haxell, S. Howard, G. P. Vennall,
Angew. Chem. 2002, 114, 4043–4045; Angew. Chem. Int. Ed. 2002,
41, 3887–3889; b) I. Coldham, S. Dufour, T. F. N. Haxell, G. P. Ven-
nall, Tetrahedron 2005, 61, 3205–3220; c) I. Coldham, J. J. Patel, G.
Sanchez-Jimenez, Chem. Commun. 2005, 3083–3085; d) I. Coldham,
S. Dufour, T. F. N. Haxell, J. J. Patel, G. Sanchez-Jimenez, J. Am.
Chem. Soc. 2006, 128, 10943–10951.
[10] a) I. Coldham, J. J. Patel, S. Raimbault, D. T. E. Whittaker, Chem.
Commun. 2007, 4534–4536; b) I. Coldham, S. Raimbault, P. T. Cho-
vatia, J. J. Patel, D. Leonori, N. S. Sheikh, D. T. E. Whittaker, Chem.
Commun. 2008, 4174–4176.
[1] For some recent reviews, see a) R. E. Gawley, Top. Stereochem.
2010, 26, 93–133; b) I. Coldham, N. S. Sheikh, Top. Stereochem.
2010, 26, 253–293; c) W. K. Lee, Y. S. Park, P. Beak, Acc. Chem.
Res. 2009, 42, 224–234; d) D. Hoppe, G. Christoph in The Chemistry
of Organolithium Compounds (Eds.: Z. Rappoport, I. Marek),
Wiley, New York, 2004, pp. 1055–1164; e) R. E. Gawley, I. Coldham
in The Chemistry of Organolithium Compounds (Eds.: Z. Rappo-
port, I. Marek), Wiley, New York, 2004, pp. 997–1053.
[2] P. Beak, D. R. Anderson, M. D. Curtis, J. M. Laumer, D. J. Pippel,
G. A. Weisenburger, Acc. Chem. Res. 2000, 33, 715–727.
[11] P. Beak, W. K. Lee, J. Org. Chem. 1993, 58, 1109–1117.
[12] P. Beak, S. T. Kerrick, S. Wu, J. Chu, J. Am. Chem. Soc. 1994, 116,
3231–3239.
[13] a) W. F. Bailey, P. Beak, S. T. Kerrick, S. Ma, K. B. Wiberg, J. Am.
Chem. Soc. 2002, 124, 1889–1896; b) M. J. McGrath, J. L. Bilke, P.
OꢀBrien, Chem. Commun. 2006, 2607–2609; c) I. Coldham, P.
OꢀBrien, J. J. Patel, S. Raimbault, A. J. Sanderson, D. Stead, D. T. E.
Whittaker, Tetrahedron: Asymmetry 2007, 18, 2113–2119.
[14] A. V. Malkov, M. Figlus, S. Stoncius, P. Kocovsky, J. Org. Chem.
2007, 72, 1315–1325.
[15] D. S. Fraser, S. B. Park, J. M. Chong, Can. J. Chem. 2004, 82, 87–
101.
[3] a) D. Hoppe, F. Hintze, P. Tebben, Angew. Chem. 1990, 102, 1455–
1456; Angew. Chem. Int. Ed. Engl. 1990, 29, 1422–1424; b) S. T.
Kerrick, P. Beak, J. Am. Chem. Soc. 1991, 113, 9708–9710; c) D.
Hoppe, T. Hense, Angew. Chem. 1997, 109, 2376–2410; Angew.
Chem. Int. Ed. Engl. 1997, 36, 2282–2316; d) P. Beak, A. Basu, D. J.
Gallagher, Y. S. Park, S. Thayumanavan, Acc. Chem. Res. 1996, 29,
552–560.
[4] For some examples of a-alkoxy-organolithium species, see a) D. C.
Kapeller, F. Hammerschmidt, J. Org. Chem. 2009, 74, 2380–2388;
b) D. Hoppe, Synthesis 2009, 43–55; c) H. Lange, R. Huenerbein, B.
Wibbeling, R. Frçhlich, S. Grimme, D. Hoppe, Synthesis 2008, 2841–
2867.
[16] a) R. K. Dieter, N. Deo, B. Lagu, J. W. Dieter, J. Org. Chem. 1992,
57, 1663–1671; b) S. E. de Sousa, P. OꢀBrien, P. Poumellec, J. Chem.
Soc. Perkin Trans. 1 1998, 1483–1492.
[17] For previous syntheses of (+)-b-conhydrine, see a) V. Ratoveloma-
nana, J. Royer, H.-P. Husson, Tetrahedron Lett. 1985, 26, 3803–
[5] For some examples of a-amino-organolithium species, see a) J. L.
Bilke, S. P. Moore, P. OꢀBrien, J. Gilday, Org. Lett. 2009, 11, 1935–
1938; b) D. Passarella, S. Riva, G. Grieco, F. Cavallo, B. Checa, F.
Arioli, E. Riva, D. Comi, B. Danieli, Tetrahedron: Asymmetry 2009,
20, 192–197; c) K. Tomooka, T. Akiyama, P. Man, M. Suzuki, Tetra-
Chem. Eur. J. 2010, 16, 4082 – 4090
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4089

I. Coldham et al.
3806; b) D. L. Comins, A. L. Williams, Tetrahedron Lett. 2000, 41,
2839–2842; c) S. Lebrun, A. Couture, E. Deniau, P. Grandclaudon,
Tetrahedron: Asymmetry 2008, 19, 1245–1249; d) A. Kamal, S. R.
Vangala, N. V. S. Reddy, V. S. Reddy, Tetrahedron: Asymmetry 2009,
20, 2589–2593.
[24] a) D. Seebach, W. Wykypiel, W. Lubosch, H.-O. Kalinowski, Helv.
Chim. Acta 1978, 61, 3100–3102; b) N. G. Rondan, K. N. Houk, P.
Beak, W. J. Zajdel, J. Chandrasekhar, P. v. R. Schleyer, J. Org.
Chem. 1981, 46, 4108–4110; c) P. Beak, W. J. Zajdel, J. Am. Chem.
Soc. 1984, 106, 1010–1018; d) R. E. Gawley, G. C. Hart, L. J. Barto-
lotti, J. Org. Chem. 1989, 54, 175–181; e) P. Beak, W. K. Lee, J. Org.
Chem. 1990, 55, 2578–2580.
[18] M. Guillaume, J. Cuypers, J. Dingenen, Org. Process Res. Dev. 2007,
11, 1079–1086.
[25] a) D. J. Cram, L. Gosser, J. Am. Chem. Soc. 1964, 86, 5445–5457;
b) W. T. Ford, E. W. Graham, D. J. Cram, J. Am. Chem. Soc. 1967,
89, 690–692.
[26] a) P. Beak, W.-K. Lee, Tetrahedron Lett. 1989, 30, 1197–1200; b) G.
Pandey, J. K. Laha, G. Lakshmaiah, Tetrahedron 2002, 58, 3525–
3534.
[27] a) R. Pauly, N. A. Sasaki, P. Potier, Tetrahedron Lett. 1994, 35, 237–
240; b) S. H. Lim, S. Ma, P. Beak, J. Org. Chem. 2001, 66, 9056–
9062.
[28] CCDC 747987 (4) contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[19] I. Coldham, D. Leonori, Org. Lett. 2008, 10, 3923–3925.
[20] a) K. R. Campos, A. Klapars, J. H. Waldman, P. G. Dormer, C.
Chen, J. Am. Chem. Soc. 2006, 128, 3538–3539; b) A. Klapars, K. R.
Campos, J. H. Waldman, D. Zewge, P. G. Dormer, C. Chen, J. Org.
Chem. 2008, 73, 4986–4993.
[21] There are two reports in the literature for optically active 26a with
similar specific rotations but conflicting absolute configuration as-
signments: a) J. C. A. Hunt, P. Laurent, C. J. Moody, J. Chem. Soc.
Perkin Trans. 1 2002, 2378–2389; b) I. Abrunhosa-Thomas, O. Roy,
M. Barra, T. Besset, P. Chalard, Y. Troin, Synlett 2007, 1613–1615.
[22] a) M. J. McGrath, P. OꢀBrien, J. Am. Chem. Soc. 2005, 127, 16378–
16379; b) J. L. Bilke, P. OꢀBrien, J. Org. Chem. 2008, 73, 6452–6454.
[23] B. Porath, R. Mꢇnzenberg, P. Heymanns, P. Rademacher, R. Boese,
D. Blꢈser, R. Latz, Eur. J. Org. Chem. 1998, 1431–1440.
Received: November 6, 2009
Published online: February 19, 2010
4090
www.chemeurj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4082 – 4090