Chirality transfer in Brook rearrangement-mediated SE2′ solvolytic protonation and its use in estimation of the propensity for racemization of the α-lithiocarbanions of the substituents

Article abstract of DOI:10.1016/j.tet.2013.05.043

Chirality transfer from an α-silylalcohol to α-carbmoyloxy- and α-siloxyallyl-carbanions was investigated using a Brook rearrangement-mediated S_E2′ protonation in γ- carbamoyloxy- and γ-siloxy-α-silylallyl alcohols. We proposed a hypothesis that the reaction proceeds along one of two pathways that involves (1) a concerted protonation of a silicate intermediate and (2) a concerted lithiation of the intermediate followed by protonation with retention or by protonation after racemization. Comparison of the extent of the chirality transfer provides a new method for semi-quantitative evaluation of the propensity for racemization of lithiocarbanions next to a conjugative electron-withdrawing group.

Full text of DOI:10.1016/j.tet.2013.05.043

Tetrahedron 69 (2013) 5823e5828
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Chirality transfer in Brook rearrangement-mediated SE20 solvolytic
protonation and its use in estimation of the propensity for
racemization of the a-lithiocarbanions of the substituents
Michiko Sasaki a, Misato Fujiwara a, Yuri Kotomori a, Masatoshi Kawahata b,
,
Kentaro Yamaguchi b, Kei Takeda a
*
a Graduate School of Medical Sciences, Hiroshima University, 1-2-3 Kasumi, Minami-Ku, Hiroshima 734-8553, Japan
b Tokushima Bunri University, 1314-1 Shido, Sanuki, Kagawa 769-2193, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 April 2013
Received in revised form 8 May 2013
Accepted 13 May 2013
Available online 18 May 2013
Chirality transfer from an
a
-silylalcohol to
a
-carbmoyloxy- and
a
-siloxyallyl-carbanions was investigated
using a Brook rearrangement-mediated SE20 protonation in
g
-carbamoyloxy- and -siloxy- -silylallyl
g
a
alcohols. We proposed a hypothesis that the reaction proceeds along one of two pathways that involves
(1) a concerted protonation of a silicate intermediate and (2) a concerted lithiation of the intermediate
followed by protonation with retention or by protonation after racemization. Comparison of the extent of
the chirality transfer provides a new method for semi-quantitative evaluation of the propensity for ra-
cemization of lithiocarbanions next to a conjugative electron-withdrawing group.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Carbanion
Chirality transfer
Rearrangement
Electrophilic substitution
1. Introduction
peting reaction pathways: anti-attack protonation to the CeSi
bond in silicate intermediate 2a and the intermediacy of a lith-
iocarbanion 3a, in which the latter is responsible for observed
partial racemization.
In a previous paper,1 we reported that a Brook rearrangement2-
mediated SE20 solvolytic protonation in
-silylallylalcohols 1a, in
a
which the resulting allyl carbanion is potentially planarized by
a neighboring CN group,3 proceeds in an anti-fashion to the CeSi
bond to afford enantioenriched nitrile derivative 4a in 75% ee
(Scheme 1).4 The findings that configurationally extremely labile
Our recent study, however, showed that a chiral carbanion of an
acyclic nitrile generated by deprotonation from an enantioenriched
O-carbamoyl cyanohydrin 5 is able to be trapped in situ by ethyl
cyanoformate with 80% ee (Scheme 2).7 The enantioselectivity
observed in 7 apparently originates from a combination of the
powerful chelating ability of a carbamoyl group8,9 and rapid trap-
ping of a carbanion by an electrophile. Consequently, we wondered
about the possibility of the chirality transfer observed in 4a being
due at least partially to the intermediacy of a lithiocarbanion (S)-3a,
which is generated via anti-mode SE20-type lithiation10,11 and has
a sufficient configurational stability to allow trapping by t-BuOH
without complete racemization followed by a retentive protonation
(Scheme 3, (S)-3/(S)-4).12 The key question is whether the strong
chelation effect of a carbamoyl group can effectively operate for
fixing the configuration of a chiral carbanion even in a protic sol-
a
-nitrile carbanions5,6 can be trapped without complete racemi-
zation led us to postulate a mechanism that involves two com-
Scheme 1. Brook rearrangement-mediated SE20 solvolytic protonation of 1a.
vent and for a more configurationally labile allylic a-nitrile carb-
anion than 6. To gain an insight into the effects of a carbamoyl
group on the configurational stability of a carbanion in the reaction
of 1a, we investigated the influence of a substituent next to a chiral
carbanion, an O-substituent, a counter cation, and a solvent on the
chirality transfer.
* Corresponding author. Tel./fax: þ81 82 257 5184; e-mail address: takedak@
hiroshima-u.ac.jp (K. Takeda).
0040-4020/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2013.05.043

5824
M. Sasaki et al. / Tetrahedron 69 (2013) 5823e5828
Table 1
Brook rearrangement-mediated SE20 solvolytic protonation or deuteration of 1a, 10a,
1b and 10b
Scheme 2. Enantioselective trapping of an
a-chiral carbanion of acyclic nitrile by
a carbon electrophile.
Entry
1
M
Base (equiv)
H/D
Yield (%)
ee (%)a
1
2
3
4
5
6
7
8
1a
1a
Li
Li
Li
Li
Li
Li
Na
Na
K
K
Li
Li
Na
Na
K
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
1.0
H
D
H
D
H
D
H
D
H
D
H
D
H
D
H
D
91
91
69
84
97
96
81
87
87
91
50
70
58
56
55
55
63
65
36
18
96
94
98
98
97
97
44
36
76
59
68
44
10a
10a
1b
1b
1b
1b
1b
1b
10b
10b
10b
10b
10b
10b
9
10
11
12
13
14
15
16
1.0
1.0
1.0
1.0
Scheme 3. Stereochemical pathways for SE20 protonation.
K
1.0
a
Corrected for the ee of the starting materials (77e100% ee).
2. Results and discussion
without loss of stereochemical information leading to (S)-4, or by
protonation after racemization via planarization and/or a kete-
nimiate intermediate rac-3019 leading to rac-4 ((S)-3/rac-4). The
possibility of non-stereoselective protonation of (S)-3 leading to
2.1. Mechanism of SE20 solvolytic protonation of 1
We first examined the reaction by using phosphono derivative
1b in addition to 1a and their O-tert-butyldimethylsilyl derivatives
rac-4 is unlikely, since we observed that protonation of an a-car-
1a and 10b. Our previous studies suggest that an
a
-phosphono
bamoyloxy-substituted lithiocarbanion generated between allyl
and phenyl groups with MeOH proceeds with retention.12a
carbanion has configurational stability similar to that of an -nitrile
a
carbanion.13 Since a silyl group has much less ability for fixing the
configuration of a carbanion in comparison with a carbamoyl
group,14 replacement of a carbamoyl group by a silyl group should
cause extensive lowering of the ee in the products if a lith-
iocarbanion intermediate is involved in the major reaction
pathway.
In the case of carbamoyloxy derivatives, the lithium ion in 2 can
be delivered to the opposite face of the CeSi bond with co-
ordination to the carbamoyl group to form (S)-3 (path B),7,20 which
is protonated with retention of stereochemistry to give (S)-4 or
undergoes racemization to give rac-4. The degree of racemization
in (S)-3, which can compete with retentive protonation leading to
(S)-4, is associated with the configurational stability of a lith-
iocarbanion (S)-3. In siloxy derivatives, since both migration of
a lithium atom conducted by a siloxy group leading to (S)-3 and
configurational stabilization of a carbanion in (S)-3 by fixing
a lithium cation with the aid of a siloxy group are less favorable,
path A should be the only pathway leading to non-racemic (S)-4.
Consequently, the ees (36% and 44%, Table 1, entries 3 and 11) ob-
tained from siloxy derivatives 10a and 10b may reflect the extent of
the reaction proceeding through path A. A smaller difference in ee
between silyl derivatives 40a and 40b than that between carbamoyl
derivatives 4a and 4b, which can be formed via path B involving
a lithiocarbanion intermediate (S)-3, also supports the notion that
in silyl derivatives, path A is the only pathway leading to (S)-4 and
path B leads to racemization via rac-3. In contrast, the greater dif-
ference (96% and 63%, Table 1, entries 5 and 1) observed on going
from phosphono derivative 4b to cyano derivative 4a in the car-
bamoyl derivatives led us to assume that in the case of carbamoyl
derivatives, path B is a major pathway and the observed ee reflects
the ratio of retentive protonation to racemization in (S)-3 that
potentially varies depending on the configurational stability of the
lithiocarbanion next to group X.
Treatment of carbamoyloxy-substituted nitrile derivative (R)-
1a1 or siloxy-substituted nitrile derivative (R)-10a (vide infra) with
a catalytic amount (0.05 equiv) of t-BuOLi in t-BuOH at 25 ꢀC
afforded 4a or 40a15 in 63% ee or 36% ee, respectively (Table 1, en-
tries 1 and 3), suggesting that configurational stabilization by
a carbamoyl group can be at least partly responsible for the ste-
reospecificity observed in the reaction of carbamoyloxy-substituted
nitrile derivative 1a. On the other hand, reactions of phosphono
derivatives were even more remarkable. Thus, whereas
carbamoyloxy-substituted phosphono derivative 1b (vide infra)
afforded 4b16,17 in 96% ee (Table 1, entry 5), indicating an
a-phos-
phonyl anion to be configurationally much more stable than an a-
nitrile carbanion, siloxy-substituted phosphono derivative 10b (vide
infra) resulted in significant loss of enantiopurity (44% ee,15 Table 1,
entry 11).
To explain these observations, we propose working hypotheses
as shown in Scheme 3. Thus, Brook rearrangement-mediated pro-
ton transfer reactions in 1a,b and 10a,b proceed via two competing
pathways, path A and path B, depending on the chelating ability of
the O-substituent and on the electronic character of a group X. Path
A involves a concerted anti-attack protonation by the solvent in
a silicate intermediate 2 leading to (S)-4. Although the possibility of
syn-attack protonation cannot be ruled out, it is unlikely on the
basis of results of our previous studies.18 Path B involves the initial
formation of a lithiocarbanion intermediate (S)-3 in an anti-attack
lithiation to the CeSi bond followed by retentive protonation11
On the basis of these analyses, we next conducted reactions
using t-BuOD in place of t-BuOH as a solvent,21 which can affect the
chirality transfer in several ways. Thus, a decreased electrophilic
quenching rate should provide more chances for racemization in
lithiocarbanion (S)-3 and should increase the ratio of reaction that

M. Sasaki et al. / Tetrahedron 69 (2013) 5823e5828
5825
Table 2
proceeds via path B, which is also more prone to racemization.
Effects of solvents on chirality transfer in reactions of 1a,b and 10a,b
Whereas a lowering of the ee was observed in silyl derivatives
(Table 1, entries 3, 4, 11, and 12), the ee remained almost unchanged
in the case of carbamoyl derivatives (Table 1, entries 1, 2, 5, and 6).
The former result is readily understandable in terms of an increased
ratio of path B. The latter is consistent with the above assumption
that the carbamoyl derivatives proceed almost through path B, if
lithiocarbanion (S)-3 has sufficient configurational stability to allow
offsetting of a decreased electrophilic quenching rate.
Entry
1
R
X
Base
Solvent
Yield [%] eea
(equiv)
(SM)
[%]
1
2
1a
1a
i-Pr2NCO CN
i-Pr2NCO CN
0.05
0.05
t-BuOH
THF/t-BuOH
(10:1)
91
96
63
35
Change in the counter cation from lithium to sodium or potas-
sium can also affect the chirality transfer by enhancing the re-
activity of a silicate intermediate and that of a metallocarbanion
3
4
5
1a
1a
1a
i-Pr2NCO CN
i-Pr2NCO CN
i-Pr2NCO CN
0.05
0.05
0.05
Et2O/t-BuOH
(10:1)
Toluene/t-BuOH 93
(10:1)
Hexane/t-BuOH
(10:1)
t-BuOH
THF/t-BuOH
(10:1)
92
47
54
56
and by increasing the configurational lability of
a metal-
locarbanion.22 Thus, whereas the enhanced reactivity of the anionic
species should accelerate the solvolytic protonation of the silicate
intermediate to increase the proportion of reaction following path
A, leading to a higher ee, an increase in reactivity, and configura-
tional lability of a metallocarbanion should operate for the chirality
transfer in the opposite direction in path B. The increase in ee (Table
1, entries 11, 13, and 15) in silyl derivative 10b when the counter ion
was changed from lithium to sodium or potassium suggests an
increased ratio of reaction following path A and/or acceleration of
retentive protonation in path B. On the other hand, the ees obtained
with carbamoyl derivative 1b were almost unchanged (Table 1,
entries 5, 7, and 9). The latter case indicates the possibility that the
sodio and potassio carbanions still have a sufficient lifetime to al-
low trapping without racemization or that path A became the
major reaction pathway.
91
6
7
10a TBS
10a TBS
CN
CN
0.05
0.05
69
96
36
5
8
10a TBS
10a TBS
10a TBS
CN
CN
CN
0.05
0.05
0.05
0.05
0.05
1.0
Et2O/t-BuOH
(10:1)
Toluene/t-BuOH e (91)
(10:1)
Hexane/t-BuOH
(10:1)
12 (79)
21
d
9
10
11
12
13
14
13 (80)
97
30
96
70
44
35
1b
1b
i-Pr2NCO P(O)
(OEt)2
i-Pr2NCO P(O)
(OEt)2
P(O)
t-BuOH
THF/t-BuOH
(10:1)
t-BuOH
92
10b TBS
10b TBS
50
(OEt)2
P(O)
1.0
THF/t-BuOH
83
We recently reported that the configurational stability of lith-
iocarbanions is greatly affected by solvents, particularly their
electron-donating ability.23 Consequently, examination of the ef-
fects of solvents on chirality transfer in the reaction of (R)-1 would
provide further information for the intermediacy of lith-
iocarbanions. The results obtained using a mixed-solvent system
containing 10% (v/v) t-BuOH as a proton source are shown in
Table 2. Particularly notable is that a greater extent of racemization
in 1a was observed in THF than in the other solvent systems (Table
2, entries 1e5). This result can be understood by considering ra-
cemization of lithiocarbanion (S)-3 via a solvent-separated ion pair
driven by THF, although it cannot exclude the possibility of change
in the preference for path A or B depending on the solvent. De-
creased chemical yields along with recovery of the starting material
in silyl derivative 10a may also support the idea that a carbamoyl
group can promote a Brook rearrangement in 2 by delivering
(OEt)2
(10:1)
a
Corrected for the ee of the starting materials (90e100% ee).
The propensity for racemization of the carbanions next to con-
jugative electron-withdrawing groups such as an ester, an amide, or
a nitrile group has received very little attention because it has been
considered impossible to generate a chiral carbanion in an a-position
of the groups whose configurational stability is high enough to allow
it to undergo an electrophilic substitution without racemization.24
One of the most effective methods for testing configurational
stability of carbanions is the Hoffmann test, which is based on ki-
netic resolution during the electrophilic substitution step using
racemic and enantioenriched aldehydes as electrophiles.25 Since
the test shows whether a given carbanion is configurationally sta-
ble on the timescale of the reaction with an aldehyde, it is not
applicable to configurationally unstable carbanions on timescales
such as those next to conjugative electron-withdrawing groups and
it is intrinsically impossible to quantitatively compare the sub-
stituent effects on the stability.
We previously reported that the extent of chirality transfer
(Scheme 4, 8/10) in [2,3]-Wittig rearrangement using an enan-
tioenriched b-substituted allyl benzyl carbanion 9 can be used as
a tool for estimation of the effects of substituents on the configu-
rational stability of a chiral lithiocarbanion through a double
bond.13 The method showed that the configurational stability of the
a lithium ion around the silicon atom to the
group using its chelating ability.
a-position of the nitrile
Although it is difficult to draw definitive conclusions concerning
the detailed reaction mechanism at the present stage, it seems
reasonable to assume that (1) carbamoyl derivatives proceed mainly
through path B and the extent of chirality transfer observed in 4
depends on the configurational stability of lithiocarbanion and (2) in
the case of silyl derivatives, path A is the only pathway capable of
transferring chirality and complete racemization occurs in path B.
2.2. Estimation of the propensity for racemization of the
lithiocarbanions of the substituents
a-
a
-lithiocarbanion in Et2O decreases in the order of Me>SiMe3>
Ph>P(O)(OEt)2, P(O)Ph2, SO2Ph, and CN. Although it should be
noted that even the -nitrile lithiocarbanion partially retained
The above analysis led us to the idea that the effects of sub-
stituents on the propensity for racemization of chiral lith-
iocarbanions can be evaluated semi-quantitatively on the basis of
the extent of chirality transfer in the reaction of carbamoyl de-
rivative 1, in which a substituent X is a conjugative electron-
a
chirality without complete racemization (9% ee), it was difficult to
reliably assign the ranking of their stabilities in the latter four cases
because of their low ee values and because of small differences in
their ee values (9e14%). Thus, the method cannot estimate the
withdrawing group, such as a ketone, ester or amide, or an
a
-an-
difference in the configurational stability between a-carbanions of
ion-stabilizing heteroatom substituent. Thus, the ee observed in 4
should reflect the effects of substituents on the configurational
stability of a lithiocarbanion (S)-3.
sulfone and nitrile groups, although the former is known to have
configurational stability on the timescale of the reactions with re-
active electrophiles.26

5826
M. Sasaki et al. / Tetrahedron 69 (2013) 5823e5828
equilibration between the products and a trace amount of basic
species depending on their acidity, the products 4a, 4c, and 4e were
treated under the same reaction conditions over a period of 24 h
and their ees remained unchanged. Therefore, the enantiomeric
excess should reflect the degree of racemization of lithiocarbanion
3 before protonation by t-BuOH. Consequently, the results shown in
Table 3 suggest that the propensity for racemization of a lith-
iocarbanion next to a conjugative electron-withdrawing group or
an anion-stabilizing heteroatom substituent increases in the fol-
lowing order: SO2p-TolzP(O)(OEt)2<CN<CONMe2<CO2Et<COPh.
It is noteworthy that (1) this method allows a comparison of the
propensity for racemization of lithiocarbanions next to sulfonyl and
phosphono groups and a cyano group, which showed almost the
same propensity for racemization in the method using a [2,3]-
Wittig rearrangement, and (2) this method is applicable to even
Scheme 4. Chirality transfer in the [2,3]-Wittig rearrangement of 8.
Since in the system using 1, the proton source for trapping
a lithiocarbanion is an alcohol that is present in large excess as
a solvent and the rate of trapping is apparently much faster than
[2,3]-Wittig rearrangement, it would be much less influenced by
the differences in chemical reactivity of a lithiocarbanion and in
steric bulk of the substituent X. Consequently, this system should
be applicable to a semi-quantitative estimation of configurational
stability of much more configurationally labile carbanions than an
a
-nitrile lithiocarbanion. Although the mechanism for racemization
a very configurationally labile lithiocarbanion such an
carbanion, enabling comparison of the propensity for racemization
of carbanions next to cyano, ester, and amide groups. Although
phosphinoyl27 and -sulfonyl3e,9e,28 lithiocarbanions have been
a-ester
and the detailed structure of a lithiocarbanion including a hetero-
atom-lithiation can change depending on the nature of the sub-
stituents,9e they would not be critical factors in the estimation
because the system can provide information not on the intrinsic
configurational stability of an a-carbanion of a given substituent
but on the propensity for racemization on the timescale of elec-
trophilic quenching.
We chose each of the following electron-withdrawing groups as
the substituent X in view of the expected ease in preparation: CO2Et
(1c), C(O)Ph (1d), CONMe2 (1e) and SO2p-Tol (1f) (Scheme 5). A
method previously reported1 for the corresponding cyano de-
rivative 1a that involves base-induced ring opening of epoxysilane
12 was used to prepare the ester derivative 1c. For the preparation
of substrates other than 1a,c, a method that involves a ring opening
of diol derivatives 14b,def was applied because they resisted all
attempts to open the epoxide ring (see: Supplementary data).
a
-
a
shown to be configurationally unstable and stable, respectively, by
the Hoffmann test, the present study, for the first time, provides
semi-quantitative information regarding the configurational sta-
bility of a lithiocarbanion next to a cyano group and two carbonyl
functional groups.
3. Conclusions
We have investigated chirality transfer to differently substituted
a-carbmoyloxy- and a-siloxyallyl-lithiocarbanions in a Brook
rearrangement-mediated SE20 solvolytic protonation in enan-
tioenriched allylsilicates and the configurational stability of the
carbanions. We proposed a hypothesis that the reaction proceeds
along one of two pathways that involves (1) a concerted pro-
tonation of a silicate intermediate (path A) and (2) a concerted
lithiation of the intermediate followed by protonation with re-
tention or by protonation after racemization (path B), depending on
the chelating ability of an O-substituent and on the electronic
character of a group X. The results of experiments using t-BuOD and
of changing the counter cation provided support for the hypothesis.
On the basis of the findings, we have developed a method that
makes it possible to roughly evaluate the propensity for racemi-
zation of a carbanion next to a conjugative electron-withdrawing
group such as ester and nitrile groups, which has been impossible
by other means.
Scheme 5. Preparation of starting materials 1bef.
With the substrates in hand, we undertook a proton transfer
experiment via a Brook rearrangement using 1aef. Treatment of
1aef with t-BuOLi (0.05 equiv) in t-BuOH at 25 ꢀC for 15 min
afforded enol silyl ether 4aef in the range of 0e97% ee (Table 3). To
rule out the possibility that the lowering of ees is partly due to an
4. Experimental section
4.1. General experimental details
All moisture-sensitive reactions were performed under a posi-
tive pressure of nitrogen. Anhydrous MgSO4 was used for drying all
organic solvent extracts in workup, and removal of the solvents was
performed with a rotary evaporator. Dry solvents and reagents
were obtained by using standard procedures. For routine chroma-
tography, the following adsorbents were used: silica gel 60N of
Table 3
Reactions of 1aef with a catalytic amount of t-BuOLi in t-BuOH
particle size 63e210 mm or 40e50 mm and precoated silica gel 60 F-
254 plates for analytical thin-layer chromatography. Liquid chro-
matography under medium pressures (MPLC) was carried out using
prepacked columns (22 mmꢁ100 mm (5
22 mmꢁ300 mm (10
silica gel)). 1H NMR spectra were taken in
CDCl3 with reference to CHCl3 ( 7.26) or in C6D6 with reference to
C6H6
7.20). 13C NMR spectra were measured in CDCl3 with ref-
erence to the CDCl3 triplet ( 77.2) or in C6D6 with reference to the
C6DH5 triplet (
128.0). The assignment of 1H and 13C NMR spectra
m silica gel) or
1
X
Yield [%]
eea [%]
m
f
SO2p-Tol
P(O)(OEt)2
CN
CONMe2
CO2Et
100
97
91
85
92
57
97
96
63
34
32
0
d
b
a
e
c
d
(d
d
d
C(O)Ph
was based on HeH decoupling and HMQC experiments. Mass
spectra (HRMS) were obtained on an Orbitrap mass spectrometer.
a
Corrected for the ee of the starting materials (81e100% ee).

M. Sasaki et al. / Tetrahedron 69 (2013) 5823e5828
5827
4.2. General procedure for SE20 protonation of
hydroxyallylsilane
a
-
IPA¼30:1, flow rate 0.50 mL/min, detection at 254 nm, tR¼21.4 min
14
(62.5%) and 24.1 min (31.4%); [
a
]
D
þ18.6 (c 1.11, CHCl3).
Reaction of 1a with t-BuOLi in t-BuOH: To a solution of 1a
(30 mg, 88.1 mol, 93% ee) in t-BuOH (990 L) was added t-BuOLi
(1.0 M in hexane, 4.4
L) at 25 ꢀC. After being stirred at the same
temperature for 15 min, CH3COOH (1.0 M in Et2O, 176 L) was
added. The mixture was stirred for 5 min and then saturated
aqueous NH4Cl (5 mL) was added. The aqueous phase was
extracted with Et2O (5 mLꢁ3). The combined organic phases were
washed with H2O (5 mL) and saturated brine (5 mL), dried, and
concentrated. The residual oil was subjected to column chroma-
tography (2.5 g of silica gel, elution with hexane/Et2O¼6:1) to
give 4a (27 mg, 91%); ChiralpakÒ IC (25 cm), hexane/IPA¼20:1,
flow rate 1.0 mL/min, detection at 254 nm, tR¼6.4 min (70.4%) and
10.1 min (18.2%)
4.2.5. (E)-3-((tert-Butyldimethylsiloxy)-1-tosylallyl)diisopro-
m
m
pylcarbamate (4f). A clear colorless oil: Rf¼0.25 (hexane/
m
AcOEt¼5:1); IR (neat) 2961, 2932, 1711 and 1656 cmꢂ1
;
1H NMR
m
(CDCl3)
d
0.15 (s, 6H), 0.90 (s, 9H), 1.00 (d, J¼6.0 Hz, 3H), 1.08e1.20
(m, 9H), 2.41 (s, 3H), 3.63e3.71 (br, 1H), 3.85e3.93 (br, 1H), 5.11 (dd
J¼11.9, 9.4 Hz, 1H), 6.15 (d, J¼9.4 Hz, 1H), 6.67 (d, J¼11.9 Hz, 1H),
7.30 (d, J¼8.2, 2H), 7.78 (d, J¼8.2, 2H); 13C NMR (C6D6)
ꢂ5.2, ꢂ5.1,
d
18.3, 20.2, 20.4, 21.2, 21.5, 21.8, 25.6, 46.4, 46.6, 86.2, 99.3, 129.6,
129.9, 133.8, 145.0, 150.1, 152.0; HRMS-ESI (m/z): [MþNa]þ calcd for
C23H39NO5SSi, 492.22104, found 492.22110; ChiralpakÒ AD-H
(25 cm), hexane/IPA¼20:1, flow rate 0.78 mL/min, detection at
19
254 nm, tR¼10.9 min (1.53%) and 14.7 min (87.8%); [
a
]
þ1.03 (c
D
1.75, CHCl3).
4.2.1. (R,E)-3-((tert-Butyldimethylsiloxy)-1-(diethoxyphosphoryl)al-
4.2.6. (S,E)-2,4-Bis((tert-butyldimethylsilyl)oxy)but-3-enenitrile
(40a)..29 ChiralcelÒ OD-H column (25þ25 cm), hexane, flow rate
0.55 mL/min, detection at 230 nm, tR¼16.1 min (30.9%, minor) and
lyl)diisopropylcarbamate (4b). A pale yellow oil: Rf¼0.38 (hexane/
AcOEt¼1:2); IR (neat) 2964, 2933 and 1690 cmꢂ1
;
1H NMR (C6D6)
0.02 (s, 6H), 0.90 (s, 9H), 1.11 (t, J¼6.9 Hz, 6H), 1.13e1.21 (br, 12H),
d
17.8 min (63.5%, minor); [
a
]
20 þ1.18 (c 0.78, CHCl3).
D
3.64e3.80 (br, 2H), 4.03e4.14 (m, 4H), 5.65 (1H, ddd, J¼18.1, 11.9,
10.0 Hz, 1H), 6.11 (dd, J¼12.0, 10.1 Hz, 1H), 6.91 (dd, J¼11.9, 3.2 Hz,
4.2.7. (R,E)-Diethyl 1,3-bis(tert-Butyldimethylsiloxy)prop-2-enylpho
1H); 13C NMR (C6D6)
d
ꢂ5.6, ꢂ5.6, 16.3, 18.1, 20.5, 21.4, 25.5, 46.5,
sphonate (40b). A clear colorless oil: Rf¼0.34 (hexane/AcOEt¼2:1);
46.7, 61.9, 62.0, 62.1, 62.2, 66.5, 68.0, 104.5, 148.6, 148.7, 154.1, 154.2;
HRMS-ESI (m/z): [MþNa]þ calcd for C20H42NO6PSi, 474.24112,
found 474.24106; ChiralpakÒ IA (25 cm), hexane/IPA¼35:1, flow
IR (neat) 2932, 1659, 1470 cmꢂ1
; d 0.08 (s, 6H), 0.10 (s, 6H), 0.15 (s,
6H), 0.90 (s, 9H), 0.91 (s, 9H), 1.31 (t, J¼7.1 Hz, 6H), 4.08e4.20 (m,
4H), 4.34 (dd, J¼12.4, 8.5 Hz, 1H), 5.11 (ddd, J¼12.1, 8.5, 6.0 Hz, 1H),
rate 0.85 mL/min, detection at 254 nm, tR¼8.7 min (2.7%) and
6.48 (dd J¼12.1, 4.6 Hz, 1H); 13C NMR (CDCl3)
d
ꢂ5.1, ꢂ5.0, ꢂ4.7,
18
10.5 min (97.3%); [
a]
ꢂ0.61 (c 2.09, CHCl3).
ꢂ4.3, 16.7 (d, JCeP¼4.8 Hz), 16.8 (d, JCeP¼5.7 Hz), 18.4, 18.4, 25.8,
25.9, 62.7 (d, JCeP¼6.7 Hz), 62.9 (d, JCeP¼7.7 Hz), 68.4 (d,
JCeP¼177.4 Hz), 107.5, 145.0 (d, JCeP¼16.2 Hz); HRMS-ESI (m/z):
[MþNa]þ calcd for C19H43O5NaPSi2, 461.22789, found 461.22794;
ChiralpakÒ IC (25), hexane/IPA¼40:1, flow rate 0.5 mL/min, de-
D
4.2.2. (E)-Ethyl-4-((tert-butyldimethylsiloxy)-2-((diisopropylcarba-
moyl)oxy))but-3-enoate (4c). A pale yellow oil: Rf¼0.53 (hexane/
AcOEt¼5:1); IR (neat) 2963, 2934, 1753, 1698 and 1664 cmꢂ1
;
1H
NMR (CDCl3)
d
0.16 (s, 6H), 0.91 (s, 9H), 1.15e1.35 (br, 12H), 1.25 (t,
tection at 230 nm, tR¼17.3 min (73.9%, major) and 19.2 min (19.2%,
25
J¼7.1 Hz, 3H), 3.72e3.85 (br, 1H), 4.00e4.10 (br, 1H), 4.13e4.24 (m,
minor); [
a]
þ12.4 (c 0.34, CHCl3).
D
2H), 5.12 (dd J¼11.9, 9.4 Hz, 1H), 5.32 (d, J¼9.4 Hz, 1H), 6.68 (d,
J¼11.9 Hz, 1H); 13C NMR (C6D6)
d
ꢂ5.4, 14.1, 18.3, 20.6, 21.3, 21.5,
Acknowledgements
25.6, 46.0, 46.5, 60.9, 72.0, 105.8, 147.7, 154.6, 170.3 HRMS-ESI (m/z):
[MþNa]þ calcd for C19H37NO5Si, 410.23332, found 410.23331;
This research was partially supported by a Grant-in-Aid for
Scientific Research (B) 22390001 (K.T.), a Grant-in-Aid for Chal-
lenging Exploratory Research 2365900700 (K.T.) and a Grant-in-Aid
for Young Scientists (B) 22790011 (M.S.) from the Ministry of Ed-
ucation, Culture, Sports, Science and Technology (MEXT), the Hoan
Sha Foundation (M.S.), the Hayashi Memorial Foundation for Fe-
male Natural Scientists (M.S.), and the Takeda Science Foundation
(M.S.). We are greatly indebted to Professor R.W. Hoffmann of
ChiralpakÒ AD-H (25 cm), hexane/IPA¼20:1, flow rate 0.80 mL/min,
25
detection at 254 nm, tR¼5.4 min (61.4%) and 6.9 min (35.8%); [
a]
D
þ16.3 (c 0.98, CHCl3).
4.2.3. (R,E)-4-((tert-Butyldimethylsiloxy)-1-oxo-1-phenylbut-3-en-
2-yl)diisopropylcarbamate (4d). A yellow oil: Rf¼0.57 (hexane/
AcOEt¼3:1); IR (neat) 2962, 2933, 1693 and 1658 cmꢂ1
;
1H NMR
(C6D6)
d
e0.04 (s, 3H) ꢂ0.03 (s, 3H), 0.83 (s, 9H), 1.11e1.26 (br, 12H),
€
Philipps-Universitat Marburg and Professor I. Coldham of Univer-
3.84e3.96 (br, 2H), 5.53 (dd, J¼11.9, 9.4 Hz, 1H), 6.61 (dd, J¼9.4,
0.5 Hz, 1H), 6.76 (dd, J¼11.9, 0.5 Hz, 1H), 7.00e7.10 (m, 3H), 8.01
sity of Sheffield for valuable advice and comments. We thank the
staff of Natural Science Center for Basic Research and Development
(N-BARD), Hiroshima University for the use of their facilities.
(ddd, J¼7.1, 3.2, 1.5 Hz, 1H); 13C NMR (C6D6)
d
ꢂ5.4, 18.3, 20.5, 21.3,
21.6, 25.6, 46.1, 46.4, 74.1, 106.2, 128.9, 132.6, 136.0, 148.6, 154.7,
195.3; HRMS-ESI (m/z): [MþNa]þ calcd for C23H37NO4Si 442.23841,
found 442.23868; ChiralpakÒ IC (25 cm), hexane/IPA¼50:1, flow
rate 1.0 mL/min, detection at 254 nm, tR¼8.5 min (49.5%) and
11.0 min (49.7%).
Supplementary data
Procedure for the preparation of compounds 1bef, 10a and 10b;
determination of E/Z geometries and absolute configurations of
substrates; copies of the 1H and 13C NMR spectra of all compounds.
Supplementary data associated with this article can be found in the
online version, at http://dx.doi.org/10.1016/j.tet.2013.05.043.
4.2.4. (E)-3-((tert-Butyldimethylsiloxy)-1-(dimethylcarbamoyl)-
prop-2-enyl)diisopropylcarbamate (4e). White powder (recrystal-
lized from hexane): mp 91e92 ꢀC; Rf¼0.50 (hexane/AcOEt¼1:1); IR
(neat) 2961, 2933, 1696 and 1647 cmꢂ1 1H NMR (C6D6)
; d 0.05 (s,
References and notes
6H), 0.91 (s, 9H), 1.14e1.35 (br, 12H), 2.55 (s, 3H), 2.64 (s, 3H),
3.81e3.92 (br, 1H), 3.94e4.05 (br, 1H), 5.60 (dd, J¼11.9, 9.2 Hz, 1H),
1. Sasaki, M.; Shirakawa, Y.; Kawahata, M.; Yamaguchi, K.; Takeda, K. Chem.dEur. J.
2009, 15, 3363e3366.
2. For reviews on Brook rearrangement, see: (a) Brook, M. A. Silicon in Organic,
Organometallic, and Polymer Chemistry; Wiley & Sons: New York, NY, 2000; (b)
Brook, A. G.; Bassindale, A. R. In Rearrangements in Ground and Excited States; de
Mayo, P., Ed.; Academic: New York, NY, 1980; pp 149e221; (c) Brook, A. G. Acc.
5.97 (d, J¼9.2 Hz,1H), 6.77 (d, J¼11.9 Hz,1H); 13C NMR (C6D6)
ꢂ5.1,
d
18.5, 20.8, 21.5, 25.6, 35.6, 36.1, 46.2, 46.4, 69.5, 107.1, 147.7, 155.2,
169.3; HRMS-ESI (m/z): [MþNa]þ calcd for C19H38N2O4Si,
409.24931, found 409.24988; ChiralcelÒ OD-H (25þ25 cm), hexane/

5828
M. Sasaki et al. / Tetrahedron 69 (2013) 5823e5828
Chem. Res. 1974, 7, 77e84; (d) Moser, W. H. Tetrahedron 2001, 57, 2065e2084;
14. Sasaki, M.; Kawanishi, E.; Shirakawa, Y.; Kawahata, M.; Yamaguchi, K.; Takeda,
K. Eur. J. Org. Chem. 2008, 3061e3064.
(e) Ricci, A.; Degl’Innocenti, A. Synthesis 1989, 647e660; (f) Patrocinio, A. F.;
Moran, P. J. S. J. Braz. Chem. Soc. 2001, 12, 7e31; (g) Schaumann, E.; Kirschning,
A. Synlett 2007, 177e190.
15. The absolute configurations of the major enantiomers of 40a and 40b were
determined by the modified Mosher method after derivatization. See
Supplementary data for details: (a) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kaki-
sawa, H. J. Am. Chem. Soc. 1991, 113, 4092e4096; (b) Kusumi, T. J. Synth. Org.
Chem. Jpn. 1993, 51, 462e470.
3. (a) Fleming, F. F.; Shook, B. C. Tetrahedron 2002, 58, 1e23; (b) Arseniyadis, S.;
Kyler, K. S.; Watt, D. S. Org. React. 1984, 31, 1e364; (c) Boche, G.; Harms, K.;
Marsch, M. J. Am. Chem. Soc. 1988, 110, 6925e6926; (d) Scott, R.; Granander, J.;
Hilmersson, G. J. Am. Chem. Soc. 2004, 126, 6798e6805; (e) Boche, G. Angew.
Chem., Int. Ed. Engl. 1989, 28, 277e297; (f) Carlier, P. R.; Lo, C. W.-S. J. Am. Chem.
Soc. 2000, 122, 12819e12823; (g) Carlier, P. R.; Lucht, B. L.; Collum, D. B. J. Am.
Chem. Soc. 1994, 116, 11602e11603; (h) Fleming, F. F.; Gudipati, S.; Zhang, Z.; Liu,
W. O.; Steward, W. J. Org. Chem. 2005, 70, 3845e3849; (i) Mycka, R. J.; Eck-
enhoff, W. T.; Steward, O. W.; Barefoot, N. Z.; Fleming, F. F. Tetrahedron 2013, 69,
366e376.
4. Regarding description of enantioselectivities, we use “ee” rather than “er”
throughout this article for the sake of convenience for comparison in Table 3,
although the former seems to be replacing by the latter ( Gawley, R. E. J. Org.
Chem. 2006, 71, 2411e2416 ).
5. Sasaki, M.; Takeda, K. Synlett 2012, 2153e2164.
6. The only example of successful generation and trapping of a discrete chiral a-
nitrile carbanion other than chiral cyclopropyl nitrile carbanions is ours (see:
Ref. 7). The latter carbanions have exceptional configurational stability that
arises from enhanced angle strain in the inversion transition state. For chiral
cyclopropyl nitrile carbanions, see: (a) Walborsky, H. M.; Youssef, A. A.; Motes,
J. M. J. Am. Chem. Soc. 1962, 84, 2465e2466; (b) Walborsky, H. M.; Motes, J. J. M.
J. Am. Chem. Soc. 1970, 92, 2445e2450; (c) Carlier, P. R. Chirality 2003, 15,
340e347; (d) Fleming, F. F.; Zhang, Z. Tetrahedron 2005, 61, 747e789; (e) Carlier,
P. R.; Zhang, Y. Org. Lett. 2007, 9, 1319e1322; (f) Patwardhan, N. N.; Gao, M.;
Carlier, P. R. Chem.dEur. J. 2011, 17, 12250e12253.
16. The absolute configuration of the major enantiomer of 4b was determined after
derivatization to a sample of known absolute configuration. See Supplementary
data for details.
17. For the stereochemistry of protonation of
a-phosphono carbanions, see: Ka-
peller, D.; Barth, R.; Mereiter, K.; Hammerschmidt, F. J. Am. Chem. Soc. 2007, 129,
914e923 and references cited therein.
18. We have shown that SE20-type protonation of alkynyl silicates proceeds in anti-
attack to the CeSi bond Sasaki, M.; Kondo, Y.; Kawahata, M.; Yamaguchi, K.;
Takeda, K. Angew. Chem., Int. Ed. 2011, 50, 6375e6378.
19. Although keteniminate can be chiral due to it having an axis of chirality, the
energy barriers for racemization seem to be too low to be trapped under
conditions without racemization. (a) Yavari, I.; Tahmassebi, D.; Zonouzia, A.;
Nori-Shargh, D. J. Chem. Res., Synop. 1997, 476e477; (b) Jochims, J. C.; Lam-
brecht, J.; Burkert, U.; Zsolnai, L.; Huttner, G. Tetrahedron 1984, 40, 893e903.
20. (a) Whisler, M. C.; MacNeil, S.; Snieckus, V.; Beak, P. Angew. Chem., Int. Ed. 2004,
43, 2206e2225; (b) Antoniotti, P.; Canepa, C.; Tonachini, G. J. Org.
Chem. 1994, 59, 3952e3959.
21. Schowen, R. L. Prog. Phys. Org. Chem. 1972, 9, 275e332.
22. To our knowledge, there are no reports on a direct comparison of the re-
activity and the configurational stability among metallocarbanions where the
metal cation is lithium, sodium, or potassium. For the reactivity of the
corresponding enolates, see: Zook, H. D.; Gumby, W. L. J. Am. Chem. Soc.
1960, 82, 1386e1389; For NMR studies of the corresponding 1- and 2-
naphthylmethylmetals, see: Kronzer, F. J.; Sandel, V. R. J. Am. Chem. Soc. 1972,
94, 5750e5759.
23. Ikemoto, H.; Sasaki, M.; Takeda, K. Eur. J. Org. Chem. 2010, 6643e6650.
24. For configurational stability of organolithiums, see: (a) Clayden, J. In Organo-
lithiums: Selectivity for Synthesis; Pergamon: Oxford, UK, 2002; pp 169e240; (b)
Gawley, R. E. In Stereochemical Aspects of Organolithium Compounds; Gawley, R.
E., Siegel, J., Eds.; Topics in Stereochemistry; Wiley: New York, 2010; Vol. 26,
pp 93e133; (c) Kizirian, J.-C. In Stereochemical Aspects of Organolithium Com-
pounds; Gawley, R. E., Siegel, J., Eds.; Topics in Stereochemistry; Wiley: New
York, NY, 2010; Vol. 26, pp 189e251; (d) Ref. 9e.
7. Sasaki, M.; Takegawa, T.; Ikemoto, H.; Kawahata, M.; Yamaguchi, K.; Takeda, K.
Chem. Commun. 2012, 2897e2899.
8. Hoppe, D. Angew. Chem., Int. Ed. Engl. 1984, 23, 932e948.
€
9. (a) Hoppe, D.; Marr, D. F.; Bruggemann, M. In Organolithiums in Enantioselective
Synthesis; Hodgson, D. M., Ed.; Springer: New York, NY, 2003; pp 61e137; (b)
€
Hoppe, D.; Kramer, T. Angew. Chem., Int. Ed. 1986, 25, 160e162; (c) Dreller, S.;
Drybusch, M.; Hoppe, D. Synlett 1991, 397e400; (d) Hoppe, D.; Carstens, A.;
€
Kramer, T. Angew. Chem., Int. Ed. 1990, 29, 1424e1425; (e) Basu, A.; Thayuma-
navan, S. Angew. Chem., Int. Ed. 2002, 41, 716e738.
10. We have discussed the stereochemical pathway of reactions involving allylsi-
licates and related systems: (a) Tanaka, K.; Masu, H.; Yamaguchi, K.; Takeda, K.
Tetrahedron Lett. 2005, 46, 6429e6432; (b) Nakai, Y.; Kawahata, M.; Yamaguchi,
K.; Takeda, K. J. Org. Chem. 2007, 72, 1379e1387.
25. (a) Hoffmann, R. W. In Stereochemical Aspects of Organolithium Compounds,
Topics in Stereochemistr; Gawley, R. E., Siegel, J., Eds.; Wiley: New York, NY,
2010; Vol. 26, pp 165e188; (b) Hoffmann, R. W.; Lanz, J.; Metternich, J. R.;
Tarara, G.; Hoppe, D. Angew. Chem., Int. Ed. Engl. 1987, 26, 1145e1146; (c) Hirsch,
11. For the corresponding SE20 reactions of allylsilanes, see: (a) Buckle, M. J. C.;
Fleming, I.; Gil, S.; Pang, K. L. C. Org. Biomol. Chem. 2004, 2, 749e769; (b) Masse,
C. E.; Panek, J. S. Chem. Rev. 1995, 95, 1293e1316; (c) Fleming, I.; Barbero, A.;
€
R.; Hoffmann, R. W. Chem. Ber. 1992, 125, 975e982; (d) Hoffmann, R. W.; Ruhl,
ꢀ
Walter, D. Chem. Rev. 1997, 97, 2063e2192; (d) Fleming, I.; Dunogues, J.;
T.; Harbach, J. Liebigs Ann. Chem. 1992, 725e730; (e) Hoffmann, R. W.; Julius, M.;
Chemla, F.; Ruhland, T.; Frenzen, G. Tetrahedron 1994, 50, 6049e6060; (f) Basu,
A.; Gallagher, D.; Beak, P. J. Org. Chem. 1996, 61, 5718e5719; (g) Beak, P.; Basu,
A.; Gallagher, D. J.; Park, Y. S.; Thayumanavan, S. Acc. Chem. Res. 1996, 29,
552e560.
Smithers, R. Org. React. 1989, 37, 57e575.
12. For the stereochemistry of electrophilic substitution of allylbenzyllithiums or
benzyllithiums, see: (a) Ikemoto, H.; Sasaki, M.; Kawahata, M.; Yamaguchi, K.;
Takeda, K. Eur. J. Org. Chem. 2011, 6553e6557; (b) Carstens, A.; Hoppe, D. Tet-
€
€
rahedron 1994, 50, 6097e6108; (c) Hoppe, D.; Carstens, A.; Kramer, T. Angew.
26. Scholz, R.; Hellmann, G.; Rohs, S.; Ozdemir, D.; Raabe, G.; Vermeeren, C.; Gais,
Chem., Int. Ed. 1990, 29, 1422e1424; (d) Derwing, C.; Frank, C. H.; Hoppe, D. Eur.
J. Org. Chem. 1999, 3519e3524; (e) Hoppe, D.; Kaiser, B.; Stratmann, O.; Frohlich,
R. Angew. Chem., Int. Ed. 1997, 36, 2784e2786; (f) Gawley, R. E.; Zhang, Q. J. J.
Org. Chem. 1995, 60, 5763e5769; (g) Faibish, N. C.; Peak, Y. S.; Lee, S.; Beak, P. J.
Am. Chem. Soc. 1997, 119, 11561e11570; (h) Thayumanavan, S.; Lee, S.; Liu, C.;
Beak, P. J. Am. Chem. Soc. 1994, 116, 9755e9756.
H.-J. Eur. J. Org. Chem. 2010, 2010, 4588e4616.
€
27. (a) O’Brien, P.; Powell, H. R.; Raithby, P. R.; Warren, S. J. Chem. Soc., Perkin Trans. 1
1997, 1031e1039; Also, see: (b) Denmark, S. E.; Dorow, R. L. J. Org. Chem. 1990,
55, 5926e5928.
28. (a) Gais, H.-J.; Vollhardt, J.; Lindner, H. J. Angew. Chem., Int. Ed. Engl. 1986, 25,
939e941; (b) Gais, H.-J.; Hellmann, G. J. Am. Chem. Soc. 1992, 114, 4439e4440.
29. Sasaki, M.; Kawanishi, E.; Nakai, Y.; Matsumoto, T.; Yamaguchi, K.; Takeda, K. J.
Org. Chem. 2003, 68, 9330e9339.
13. Sasaki, M.; Ikemoto, H.; Kawahata, M.; Yamaguchi, K.; Takeda, K. Chem.dEur. J.
2009, 15, 4663e4666.