Bis(oxazoline)-ligand-mediated asymmetric [2,3]-Wittig rearrangement of benzyl ethers: Reaction mechanism based on the hydrogen/deuterium exchange effect

Article abstract of DOI:10.1016/j.tet.2012.03.059

We have investigated the mechanism of chiral induction in the asymmetric [2,3]-Wittig rearrangement of allyl benzyl ether in the presence of a bis(oxazoline) chiral ligand [(S,S)-Box-tBu] by comparing the reaction of both enantiomers of monodeuterated ben

Full text of DOI:10.1016/j.tet.2012.03.059

Tetrahedron 68 (2012) 4280e4285
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Bis(oxazoline)eligand-mediated asymmetric [2,3]-Wittig rearrangement
of benzyl ethers: reaction mechanism based on the hydrogen/deuterium
exchange effect
*
Maria Kitamura, Yoshimi Hirokawa, Yuki Yoshioka, Naoyoshi Maezaki
Faculty of Pharmacy, Osaka Ohtani University, 3-11-1 Nishikiori-Kita, Tondabayashi, Osaka 584-8540, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We have investigated the mechanism of chiral induction in the asymmetric [2,3]-Wittig rearrangement
of allyl benzyl ether in the presence of a bis(oxazoline) chiral ligand [(S,S)-BoxetBu] by comparing the
reaction of both enantiomers of monodeuterated benzyl ether 1aed. As a result, we found that chirality
was induced via enantioselective deprotonation followed by efficient chirality transfer of the resulting
chiral benzyl carbanion with the inversion of stereochemistry. It was revealed that the chiral ligand
facilitates selective deprotonation as well as prevents the chiral carbanion from racemization. Moreover,
we examined the effect of the o-methoxy substituent on the benzene ring.
Received 11 February 2012
Received in revised form 15 March 2012
Accepted 17 March 2012
Available online 29 March 2012
Keywords:
Asymmetric synthesis
Carbanions
Ó 2012 Elsevier Ltd. All rights reserved.
Chirality
Reaction mechanisms
Rearrangement
1. Introduction
racemic a-monodeuterated benzyl ether compared with that of the
non-deuterated substrate.
The [2,3]-Wittig rearrangement is a useful method for stereo-
selectively constructing the carbonecarbon bond.1 This rear-
rangement proceeds via the formation of a carbanion followed by
[2,3]-sigmatropic rearrangement. Therefore, enantioselective for-
mation of the chiral carbanion was expected to enable asymmetric
[2,3]-Wittig rearrangement, which is a powerful tool to construct
asymmetric carbons from achiral substrates. Considerable efforts
have been devoted to develop the effective method.2 In addition,
the mechanistic research revealed that the chiral carbanion un-
dergoes the [2,3]-sigmatoropic rearrangement with inversion of
configuration.3 During the course of these studies, it has been ar-
guable whether the asymmetric induction occurs at the deproto-
nation step or the post-deprotonation step via the dynamic
thermodynamic resolution or dynamic kinetic resolution.4
Nakai2a,b and Kimachi2h,i have independently reported that the
[2,3]-Wittig rearrangement of benzyl ethers proceeds via enantio-
selective deprotonation by using bis(oxazoline) (Box) and
(ꢀ)-sparteine as an external chiral ligand, respectively. They as-
sumed that chiral induction is caused by the asymmetric depro-
tonation by observing a decrease in the enantioselectivity of
Their proof was based on the theory that enantiomeric excess
(ee) will decrease in the a-monodeuterated substrate if chiral in-
duction occurs at the deprotonation step. If chiral induction occurs
at the post-deprotonation step, enantioselectivity will not
change.2a However, the change in ee was small in their substrates
because the enantioselectivity of the non-deuterated substrates
was moderate.
In a preceding paper, we have reported the highly enantiose-
lective [2,3]-Wittig rearrangement of allyl benzyl ethers using the
chiral bis(oxazoline) ligand [(S,S)-Box-tBu: L1].5 Selectivity excee-
ded 85% ee except in the substrates with o-methoxybenzyl ether.
The representative examples are shown in Scheme 1.6
We expected that if the hydrogen/deuterium exchange effect
was considered in both the enantiomers of
a-monodeuterated
benzyl ether 1a-d, the mechanism of chiral induction will be elu-
cidated since the reaction of 1a proceeded with high enantiose-
lectivity (98% ee). Moreover, the reason for low selectivity in the
benzyl ether bearing an o-methoxy substituent would be revealed
using a-monodeuterated 1b.
Herein, we report the results of the hydrogen/deuterium ex-
change effect on the asymmetric [2,3]-Wittig rearrangement and
discuss how asymmetric induction occurs. Moreover, we discuss
the effect of the o-methoxy substituent on the chiral induction of
the carbanion.
* Corresponding author. Tel./fax: þ81 721 24 9541; e-mail address: maezan@
osaka-ohtani.ac.jp (N. Maezaki).
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2012.03.059

M. Kitamura et al. / Tetrahedron 68 (2012) 4280e4285
4281
reduction using Corey’s catalyst (2-Me-CBS).8 Thus,
a
-mono-
deuterated benzyl alcohols (R)- and (S)-5a-d were obtained in 66%
and 68% yield in two steps with 93% and 94% ee, respectively (Table
1, entries 1 and 2). Corresponding o-methoxybenzyl alcohols (R)-
and (S)-5b-d were synthesized by TPAP oxidation of dideuterated o-
methoxybenzyl alcohol (3b)7 followed by asymmetric hydrogena-
tion of purified a-deuterated o-methoxybenzaldehyde (4b) using
BINAP-Ru (II) complex, developed by Takaya and co-workers.9 By
the addition of MeOH as the cosolvent, the asymmetric hydroge-
nation afforded (R)- and (S)-5b-d in good yield with 90% ee
respevtively even under moderate H2 pressure (0.4 MPa) (entries 3
and 4).
The ee of these a-monodeuterated benzyl alcohols 5a-d and 5b-
d was determined by 1H NMR spectroscopy after conversion to
Mosher esters. The absolute configuration of 5a-d was determined
by comparison with the reported specific rotation10 and that of 5b-
d was assumed by Takeuchi’s protocol using Mosher ester instead of
CFTA ester.11,12 These alcohols, (R)- and (S)-5a-d, were allylated with
allylic bromide 6 using tBuOK as a base,5 affording (R)- and (S)-1a-
Scheme 1. Asymmetric [2,3]-Wittig rearrangement of 1a and 1b.
2. Results and discussion
d in 77% yield. Similarly, a-monodeuterated o-methoxybenzyl al-
cohols (R)- and (S)-5b-d were also converted to the corresponding
allylic ethers (R)- and (S)-1b-d in 69% and 67% yield, respectively.13
We planned to examine the [2,3]-Wittig rearrangement using
each of the enantiomeric isomers of -monodeuterated benzyl
a
On treatment of a-monodeuterated benzyl ethers (R)-1a-d and
(S)-1a-d with an achiral base [tBuLi (10 equiv), THF, ꢀ78 ꢁC],14 the
[2,3]-Wittig rearrangement proceeded to afford the rearranged
products (1R,2S)-2a-d (52% ee) and (1S,2R)-2a-d (53% ee), re-
spectively (Scheme 3). This suggests that CꢀC bond formation
proceeded with the inversion of the initially formed carbanion
because a chiral carbanion with high optical purity should be
formed via the selective abstraction of the benzylic proton rather
than deuterium, which was confirmed by the disappearance of the
benzylic proton in the 1H NMR spectral data of the rearranged
product 2a-d.15 Because partial racemization to about 50% ee was
observed in the products, the configuration of the carbanion was
labile during the time course of the rearrangement reaction.
In contrast, in the presence of the chiral bis(oxazoline) ligand
[L1 (1 equiv), tBuLi (10 equiv), hexane, ꢀ78 ꢁC], only (R)-1a-
d afforded the product with 97% ee16 along with the inversion of the
stereogenic center. Only trace amount of [2,3]-Wittig rearranged
product was obtained from (S)-1a-d, but the decomposition of the
substrate occurred.
ethers. Because the abstraction of deuterium is slower than that of
hydrogen, the enantiomerically pure carbanion would be initially
formed by selective deprotonation. If the rearrangement reaction
was sufficiently faster than the racemization of the carbanion, the
product would maintain high optical purity. By examining the re-
lationship between the configuration of the carbanion and that of
the product, it would be confirmed that the rearrangement pro-
ceeded in retention or inversion of the chiral carbanion. If race-
mization was faster than rearrangement, optical purity would
decrease depending on the rate of racemization. In the presence of
the chiral ligand, matching of the enantiomeric substrates and the
chiral base would be confirmed by comparing the reaction of the
enantiomers.
Both enantiomers of
1b-d were synthesized as shown in Scheme 2.
benzyl alcohol (3a)7 was oxidized by tetrapropylammonium per-
ruthenate (TPAP). The resulting -deuterated benzaldehyde (4a)
a
-monodeuterated benzyl ethers 1a-d and
a,a
-Dideuterated
a
was unstable and was immediately subjected to asymmetric
On the basis of the aforementioned results, we assumed the
reaction mechanism of previously reported [2,3]-Wittig rear-
rangement of non-deuterated substrate 1a as follows.5 The pro-S
proton of 1a was abstracted predominately by the chiral base that
was formed from tBuLi and (S,S)-Box-tBu L1, and the resulting
chiral carbanion underwent stereoinversive CeC bond formation.
Furthermore, because no racemization was observed in (R)-1a-d by
the use of the base/chiral ligand complex in contrast to the achiral
base’s case, the chiral ligand appears to prevent the racemization of
the carbanion (Scheme 4).
On the other hand, a-monodeuterated o-methoxybenzyl ether
1b-d exhibited results that were significantly different from benzyl
ether 1a-d. The results are shown in Scheme 5. In the absence of the
bis(oxazoline) L1, the selective abstraction of hydrogen instead of
deuterium and the stereoinversion of the initially formed carban-
ion were observed in 1b-d again, affording (1R,2S)-2b-d and
(1S,2R)-2b-d from (R)- and (S)-1b-d, respectively, in moderate
yields (67%e69%). However, the rate of racemization was higher
than that of 1a-d to give each the enantiomers of 2b-d with 33% ee.
In the presence of L1, the yield from (R)-1b-d (24%e28%) was
higher than that from (S)-1b-d (8%e11%) although the yield was
low. The base/chiral ligand complex differentiated both the enan-
tiomers (R)- and (S)-1b-d to some extent, but the differentiation
was obviously low because even the mismatched substrate (S)-1b-
d was deprotonated. Furthermore, the chiral ligand restricted the
Scheme 2. Preparation of (R)- and (S)-benzyl ethers 1a-d and 1b-d.

4282
M. Kitamura et al. / Tetrahedron 68 (2012) 4280e4285
Table 1
Asymmetric reduction of a-deuterated benzaldehydes 4a and 4b
Entry
1a
Aldehyde
Conditions
Yield [%]c
66
ee [%]d
93
Config.
(R)-(ꢀ)
4a
(S)-2-MeeCBS (0.3 equiv), catecholborane (2.0 equiv),
tolueneemethylcyclohexaneeCH2Cl2 (2:2:1), ꢀ78 ꢁC
(R)-2-MeeCBS (0.3 equiv), catecholborane (2.0 equiv),
tolueneemethylcyclohexaneeCH2Cl2 (2:2:1), ꢀ78 ꢁC
(S)-2-Ru(OAc)2(BINAP) (3 mol %), H2 (0.4 MPa),
THF/MeOH (4:1), 0.2 N HCl (15 mol %), 25 ꢁC
2a
3b
4b
4a
4b
4b
68
83
85
94
90
90
(S)-(þ)
(R)-(ꢀ)
(S)-(þ)
(R)-2-Ru(OAc)2(BINAP) (3 mol %), H2 (0.4 MPa),
THF/MeOH (4:1), 0.2 N HCl (15 mol %), 25 ꢁC
a
Crude aldehyde was used.
Purified aldehyde was used.
b
c
Yield from alcohol 3a or 3b in two steps.
d
Determined by 1H NMR spectroscopy after conversion to Mosher ester.
Scheme 3. Asymmetric [2,3]-Wittig rearrangement of (R)- and (S)-1a-d.
Scheme 5. Asymmetric [2,3]-Wittig rearrangement of (R)- and (S)-1b-d.
Scheme 4. Proposed mechanism of asymmetric [2,3]-Wittig rearrangement of 1a.
racemization to some extent in (R)-1b-d (61% ee) but promoted the
racemization (2%e17% ee) in its enantiomer (S)-1b-d.
We suggest that low selectivity at the deprotonation step as well
as the facile racemization at the post-lithiation step caused low
enantioselectivity of the non-deuterated substrates with an o-
methoxybenzyl group, such as 1b (Scheme 6).6 The racemization
mechanism is ambiguous. However, considering that the re-
placement of the o-methoxy group with an o-ethyl group exhibited
high enantioselectivity,5 the o-methoxy group may cause low se-
lective deprotonation by coordinating to the lithium cation of the
base. Moreover, the substituent would disturb the lithiated sub-
strate/chiral ligand complex by coordinating to the lithium cation.
Scheme 6. Proposed mechanism of asymmetric [2,3]-Wittig rearrangement of 1b.
3. Conclusion
We investigated the hydrogen/deuterium exchange effect to
elucidate the reaction mechanism of chiral induction on the

M. Kitamura et al. / Tetrahedron 68 (2012) 4280e4285
4283
carbanion. As a result, we concluded that the bis(oxazoline) chiral
Colorless oil (66% in two steps). ½a D24
ꢂ
þ1.37 (c 0.57, CHCl3) [lit.10
ligand plays an important role in the highly enantioselective
deprotonation as well as the prevention of racemization before the
[2,3]-Wittig rearrangement. On the other hand, low enantiose-
lectivity observed in the o-methoxybenzyl substrates was revealed
to be caused by poor differentiation of deprotonation as well as
destabilization of the chiral carbanion.
½ ꢂ þ1.4 (c 3.0, CHCl3)].
a 2D5
4.1.3. (R)-
a-Deuterio-2-methoxybenzyl alcohol [(R)-5b-d]. TPAP
(53 mg, 0.15 mmol) was added to a mixture of
a,a
-dideuterio-2-
methoxybenzyl alcohol 3b (2.09 g, 14.9 mmol), 4 A molecular
sieves (7.5 g), and NMO (2.62 g, 22.4 mmol) in CH2Cl2 (50 mL) with
stirring at rt under Ar and the whole was stirred at the temperature
for 2.5 h. NMO (175 mg, 1.5 mmol) was added to the solution and
the additional stirring was continued to 1.5 h. After diluted with
CH2Cl2, the mixture was filtered through a pad of Celite and the
filtrate was washed with 10% HCl, water, saturated NaHCO3, and
brine prior to drying (MgSO4) and solvent evaporation. The crude
was chromatographed on silica gel eluting with n-hexane/EtOAc
4. Experimental section
4.1. General information and materials
Optical rotations were measured by using a JASCO P-1020 digital
polarimeter. 1H, 2D and 13C NMR spectra were recorded in CDCl3 so-
lution at 400, 61 and 100 MHz, respectively, with a JEOL JNM-AL-400
spectrometer. Chemical shifts of 1H NMR are expressed in parts per
million downfield from tetramethylsilane as an internal standard
(5:1) to give 2-methoxybenzaldehyde-a-d 4b (1.79 g, 88%) as
a colorless oil. (S)-Ru(OAc)2(binap) (105 mg, 0.124 mmol) and
aqueous 0.2 N HCl solution (3.1 mL, ca. 5 equiv to Ru) were added to
a solution of 4b (568 mg, 4.14 mmol) in THF/MeOH (4:1) (20 mL).
The mixture was stirring at room temperature for 24 h under H2 at
0.4 MPa. Then, the solution was poured to water and organic layer
was separated. The aqueous layer was extracted with EtOAc and the
combined organic layer was washed with brine prior to drying and
solvent evaporation. The residue was chromatographed on silica gel
eluting with n-hexane/EtOAc (3:1/2:1) to give (R)-5b-d (558 mg,
(
d
¼0). Chemical shifts of 2D and 13C NMR are expressed as ppm in
CDCl3 as an internal standard (
d
¼7.26 and 77, respectively). The fol-
lowing abbreviations are used: broad¼br, singlet¼s, doublet¼d,
triplet¼t, quartet¼q, and multiplet¼m. IR absorption spectra (FT:
diffuse reflectance spectroscopy) were recorded with KBr powder
with a JASCO FT-6300 IR spectrophotometer, and only noteworthy
absorptions (cmꢀ1) are listed. Mass spectrawere obtained with a JEOL
GC-Mate II mass spectrometer. Synthesis of compounds 1a, 1b, 2a,
and 2b have been reported in the previous paper.5a Purification of the
crude products was carried out by flash column chromatography. Fuji
Silysia Silica Gel BW-300 was used as an adsorbent for column
chromatography. Forpreparative TLC (PTLC), Silica gel60F254 (Merck)
was used. All air- or moisture-sensitive reactions were carried out in
flame-dried glassware under an atmosphere of Ar or N2. All organic
extractsweredriedoveranhydrousMgSO4, filtered, andconcentrated
under reduced pressure with rotary evaporator.
97%) as a pale yellowish green oil. ½a D22
ꢂ
ꢀ1.6 (c 3.14, CHCl3); 1H NMR
d
: 2.30 (s,1H), 3.87 (s, 3H), 4.67 (s,1H), 6.89 (d, J¼7.8 Hz,1H), 6.95 (t,
J¼7.4 Hz, 1H), 7.26e7.31 (m, 2H); 13C NMR
d: 55.2, 61.7 (t, J(C,
¼22.3 Hz), 110.2, 120.6, 128.7, 128.9, 129.0, 157.4; IR (KBr) cmꢀ1
:
D)
3341, 2131, 1603, 1492; MS (EI) m/z: 139 [M]þ; HRMS (EI) m/z: calcd
for C8H9DO2: 139.0736, found: 139.0743 [M]þ.
4.1.4. (S)-
a-Deuterio-2-methoxybenzyl
alcohol
[(S)-5b-d]. The
compound was synthesized in a similar manner that as described in
(R)-5b-d. The spectral data was identified with those of the (R)-
enantiomer. Pale yellowish green oil (85% in two steps). ½a D22
ꢂ þ1.6 (c
4.1.1. (R)-
a
-Deuteriobenzyl alcohol [(R)-5a-d]. TPAP (56 mg,
-dideuteriobenzyl al-
3.14, CHCl3).
0.16 mmol) was added to a mixture of
a,a
cohol 3a (1.76 g, 16.0 mmol), 4 A molecular sieves (8.0 g), and
NMO (2.88 g, 24.0 mmol) in CH2Cl2 (56 mL) with stirring at rt
under Ar and the whole was stirred at the temperature for 16 h.
After diluted with CH2Cl2, the mixture was filtered through a pad
of Celite and the filtrate was washed with 10% HCl, water, satu-
rated NaHCO3, and brine prior to drying (Na2SO4) and solvent
evaporation. The crude was used without further purification
because of instability of the product for air oxidation to give
4.2. General procedure of synthesis of substrates for [2,3]-
Wittig rearrangement: (R,E)-1-[ -deuterio(benzyl)oxy]-2-
(triisopropylsilyloxymethyl)but-2-en [(R)-1a-d]
a
tBuOK (438 mg, 3.90 mmol) was added to a solution of benzyl
alcohol (R)-5a-d (963 mg, 3.00 mmol) and bromide 6 (357 mg,
3.30 mmol) in dry THF (20 mL) with stirring at rt under Ar. After the
stirring was continued for 1 d, the reaction was quenched with
saturated NH4Cl. The mixture was partitioned between EtOAc and
water. The organic layer was separated and the aqueous layer was
extracted with EtOAc. The combined organic layer was washed
with water and brine prior to drying and solvent evaporation. The
residue was chromatographed on silica gel with hexane/EtOAc
(50:1) to give ether (R)-1a-d (812 mg, 77%) as a colorless oil. 1H
benzaldehyde-
catecholborane (24.0 mL, 24.0 mmol) was added dropwise to
a mixture of (S)-2-methyl-CBS-oxazaborolidine (1 toluene so-
a-d 4a (1.70 g) as a gray oil. A 1 M THF solution of
M
lution) (4.96 mL, 4.96 mmol) and 4a (1.70 g) in a mixed solvent of
tolueneemethylcyclohexaneeCH2Cl2 (2 : 2: 1, v/v) (178 mL) with
stirring at ꢀ78 ꢁC. After the mixture was stirred for 3.5 h at this
temperature, 4 M HCl in AcOEt (3 mL) and MeOH (19 mL) were
added to the mixture. The mixture was washed with 1 M NaOH
aqueous solution (30 mL) prior to drying (MgSO4) and solvent
evaporation. The residue was chromatographed on silica gel
eluting with n-hexane/EtOAc (3:1) to give (R)-5a-d (1.15 g, 66% in
NMR
d
: 0.90e1.20 (m, 21H), 1.69 (d, J¼7.1 Hz, 3H), 4.09 (s, 2H), 4.25
(s, 2H), 4.45 (s, 1H), 5.80 (q, J¼7.1 Hz, 1H), 7.23e7.37 (m, 5H); 13C
NMR
d
: 12.0 (3C), 13.0, 10.0 (6C), 64.9, 65.5, 71.6 (t, J(C,D)¼21.5 Hz),
123.7, 127.5, 127.7 (2C), 128.3 (2C), 135.8, 138.5; IR (KBr) cmꢀ1: 2116,
1495; MS (FAB) m/z: 350 [MþH]þ; HRMS (FAB) m/z: calcd for
C21H36DO2Si: 350.2626, found: 350.2653 [MþH]þ.17
two steps) as a colorless oil. ½a D24
ꢂ
ꢀ1.35 (c 0.57, CHCl3) [lit.10
:1.70 (d, J¼4.9 Hz, 1H), 4.68
½
a 2D5
ꢂ
ꢀ1.4 (c 3.0, CHCl3)]; 1H NMR:
d
(br s, 1H), 7.26e7.42 (m, 5H); 13C NMR:
d
: 64.3 (t, J(C,D)¼21.5 Hz),
4.2.1. (S,E)-1-[a-Deuterio(benzyl)oxy]-2-(triisopropylsilyloxymethyl)
126.8 (2C), 127.3, 128.3 (2C), 140.7; IR (KBr) cmꢀ1: 3308, 2135,
1496; MS (EI) m/z: 109 [M]þ; HRMS (EI) m/z: calcd for C7H7DO:
109.0638, found: 109.0635[M]þ.
but-2-en [(S)-1a-d]. The procedure was same as that was described
in the procedure of (R)-1a-d. Colorless oil (77%). The 1H and 13C
NMR spectral data was identical to those of the (R)-enantiomer.17
4.1.2. (S)-
a
-Deuteriobenzyl alcohol [(S)-5a-d]. The compound was
4.2.2. (R,E)-1-[
loxymethyl)but-2-en [(R)-1b-d]. Colorless oil (69%). 1H NMR
1.06e1.19 (m, 21H, TIPS), 1.71 (br d, J¼6.9 Hz, 3H), 3.82 (s, 3H), 4.13
a-Deuterio(2-methoxybenzyl)oxy]-2-(triisopropylsily-
synthesized in a similar manner that as described in (R)-5a-d. The
spectral data was identified with those of the (R)-enantiomer.
d:

4284
M. Kitamura et al. / Tetrahedron 68 (2012) 4280e4285
Nakai, T.; Tomooka, K. Pure Appl. Chem. 1997, 69, 595e600; (e) Nakai, T.; Mi-
kami, K. Org. React. 1994, 46, 105e209; (f) Marshall, J. A. In; Trost, B. M., Fleming,
I., Eds. Comprehensive Organic Synthesis; Pergamon: New York, 1991; Vol. 3,
pp 975e1014.
(s, 2H), 4.26 (br s, 2H), 4.49 (s, 1H), 5.80 (q, J¼6.9 Hz, 1H), 6.86 (dd,
J¼8.1, 0.9 Hz, 1H), 6.94 (td, J¼7.4, 0.9 Hz, 1H), 7.25 (td, J¼8.1, 1.7 Hz,
1H), 7.37 (d, J¼7.4, 1.7 Hz, 1H); 13C NMR
d: 12.0 (3C), 13.0, 18.0 (6C),
2. For enantioselective [2,3]-Wittig rearrangement of allyl propargyl ether-type sub-
strates, see: (a) Tomooka, K.; Komine, N.; Nakai, T. Chirality 2000, 12, 505e509; (b)
Tomooka, K.; Komine, N.; Nakai, T. Tetrahedron Lett. 1998, 39, 5513e5516; (c)
Manabe, S. Chem. Pharm. Bull. 1998, 46, 335e336; (d) Manabe, S. Chem. Commun.
1997, 737e738; (e) Kang, J.; Cho, W. O.; Cho, H. G.; Oh, H. J. Bull. Korean Chem. Soc.
1994, 15, 732e739. For enantioselective [2,3]-Wittig rearrangement of allyl benzyl
ether-type substrates, see: (f) Barrett, I. M.; Breeden, S. W. Tetrahedron: Asymmetry
2004,15, 3015e3017;(g) Tsubuki, M.;Takahashi, K.; Honda, T. J. Org. Chem. 2003, 68,
10183e10186; (h) Kawasaki, T.; Kimachi, T. Tetrahedron 1999, 55, 6847e6862; (i)
Kawasaki, T.;Kimachi, T. Synlett1998,1429e1431andreferences2a, b, c, d. Forother
asymmetric [2,3]-Wittig rearrangement, see: (j) Li, Y.-J.; Ho, G.-M.; Chen, P.-Z. Tet-
rahedron: Asymmetry 2009, 20, 1854e1863; (k) Sasaki, M.; Higashi, M.; Hyuma, M.;
Yamaguchi, K.; Takeda, K. Org. Lett. 2005, 7, 5913e5915; (l) Gibson, S. E.; Ham, P.;
Jefferson, G. R. Chem. Commun.1998,123e124; (m) Marshall, J. A.; Wang, X.-J. J. Org.
Chem. 1992, 57, 2747e2750.
55.3, 65.3, 65.5, 66.3 (t, J(C,D)¼21.5 Hz), 110.1, 120.4, 123.4, 127.0,
128.5, 128.9, 136.1, 157.1; IR (KBr) cmꢀ1: 2119, 1603, 1492; MS (FAB)
m/z: 380 [MþH]þ; HRMS (FAB) m/z: calcd for C22H38DO3Si:
380.2731, found: 380.2750 [MþH]þ.17
4.2.3. (S,E)-1-[a-Deuterio(2-methoxybenzyl)oxy]-2-(triisopropylsi-
lyloxymethyl)but-2-en [(S)-1b-d]. Colorless oil (67%). The 1H and
13C NMR spectral data were identified to those of the (R)-
enantiomer.17
4.3. General procedure of [2,3]-Wittig rearrangement:
(1R,2S)-1-deuterio-2-methyl-1-phenyl-3-
3. (a) Gawley, R. E.; Zhang, Q.; Campagna, S. J. Am. Chem. Soc. 1995, 117,
11817e11818; (b) Verner, E. J.; Cohen, T. J. Am. Chem. Soc. 1992, 114, 375e377; (c)
Hoffmann, R.; Bruckner, R. Angew. Chem., Int. Ed. Engl. 1992, 31, 647e649; (d)
(triisopropylsilyloxymethyl)but-3-en-1-ol [(1R,2S)-2a-d]
€
Tomooka, K.; Igarashi, T.; Watanabe, M.; Nakai, T. Tetrahedron 1992, 33,
5795e5798.
tBuLi (1.58 M in pentane, 0.80 mL, 1.27 mmol) was added to
a solution of allyl benzyl ether (R)-1a-d (44.4 mg, 0.127 mmol) and
(S,S)-Box-tBu (37.4 mg, 0.127 mmol) in dry hexane (0.64 mL) with
stirring at ꢀ78 ꢁC under Ar. The stirring was continued for 2 h at
this temperature. The reaction mixture was quenched with satu-
rated aqueous NH4Cl and allowed to warm to room temperature.
The resulting mixture was extracted with EtOAc. The combined
extracts were washed with saturated aqueous NH4Cl, water, and
brine prior to drying and solvent evaporation. The crude was
purified by PTLC (SiO2) with hexane/EtOAc (7:1) to give (1R,2S)-
4. For examples of the configurationally stable carbanion, see: (a) Hammerschmidt,
€
F.; Hanninger, A.; Simov, B. P.; Vollenkle, H.; Werner, A. Eur. J. Org. Chem. 1999,
3511e3518; (b) Hammerschmidt, F.; Hanninger, A.; Vollenkle, H. Chem.dEur. J.
1997, 3, 1728e1732; (c) Coldham, I.; Hufton, R.; Snowden, D. J. J. Am. Chem. Soc.
1996, 118, 5322e5323; (d) Still, W. C.; Sreekumar, C. J. Am. Chem. Soc. 1980, 102,
€
1201e1202; (e) Hoppe, D.; Carstens, A.; Kramer, T. Angew. Chem., Int. Ed. Engl.
1990, 29, 1424e1425. For examples of configurationally labile carbanion, see: (f)
Lee, W. K.; Park, Y. S.; Beak, P. Acc. Chem. Res. 2009, 42, 224e234; (g) Lange, H.;
€
Bergander, K.; Frohlich, R.; Kehr, S.; Nakamura, S.; Shibata, N.; Toru, T.; Hoppe, D.
€
Chem.dAsian. J. 2008, 3, 88e101; (h) Lange, H.; Huenerbein, R.; Frohlich, R.;
Grimme, S.; Hoppe, D. Chem.dAsian. J. 2008, 3, 78e87 For review on configu-
€
rational stability of chiral carbanions, see: (i) Dorwald, F. Z. Side Reactions in
2a-d (28.0 mg, 63%) as a colorless oil (97% ee). ½a D27
ꢂ ꢀ11.7 (c 1.45,
Organic Synthesis; Wiley-VCH: Weinheim, 2005; 197e203.
CHCl3); 1H NMR
d
: 0.99 (d, J¼7.1 Hz, 3H), 1.05e1.16 (m, 21H), 2.60
5. (a) Kitamura, M.; Hirokawa, Y.; Maezaki, N. Chem.dEur. J. 2009, 15, 9911e9917; (b)
Hirokawa, Y.; Kitamura, M.; Maezaki, N. Tetrahedron: Asymmetry 2008, 19,
1167e1170.
6. The yield and selelctiviy of 2b reported in Ref. 5 were revised as shown in
Schemes 1 and 6.
7. The dideuterated benzylic alcohols 3a and 3b were prepared by reduction with
LiAlD4 from ethyl benzoate and methyl o-methoxybenzoates, respectively.
8. (a) O’Hagan, D.; Goss, R. J. M.; Meddour, A.; Courtieu, J. J. Am. Chem. Soc. 2003,
125, 379e387; (b) Sato, I.; Omiya, D.; Saito, T.; Soai, K. J. Am. Chem. Soc. 2000,
122, 11739e11740 and references cited therein.
9. Ohta, T.; Tsutsumi, T.; Takaya, H. J. Organomet. Chem. 1994, 484, 191e193.
10. Sato, I.; Omiya, D.; Saito, T.; Soai, K. J. Am. Chem. Soc. 2000, 122, 11739e11740.
11. Takeuchi, Y.; Fujisawa, H.; Noyori, R. Org. Lett. 2004, 6, 4607e4610.
12. The stereochemistry of monodeuterated o-methoxybenzyl alcohol 5b-d was
speculated by comparison of the chemical shifts of the benzylic proton and
deuterium in the 1H and 2D NMR spectra, respectively, after conversion to (S)-
and (R)-Mosher esters. The Dd value is the difference of the chemical shifts
(q, J¼7.1 Hz, 1H), 3.04 (s, 1H), 4.08 (d, J¼12.9 Hz, 1H), 4.22 (d,
J¼12.9 Hz, 1H), 4.98 (br s, 1H), 5.20 (br s, 1H), 7.21e7.37 (m, 5H);
13C NMR
d: 11.9 (3C), 12.5, 18.0 (6C), 44.8, 66.0, 75.5 (t,
J(C,D)¼21.5 Hz), 112.6, 126.2 (2C), 126.9, 127.9 (2C), 142.9, 150.5; IR
(KBr) cmꢀ1: 3411, 2129, 1651, 1603, 1494; MS (FAB) m/z: 350
[MþH]þ; HRMS (FAB) m/z: calcd for C21H36DO2Si: 350.2626,
found: 350.2612 [MþH]þ.
4.3.1. (1R,2S)-1-Deuterio-1-(2-methoxyphenyl)-2-methyl-3-(triiso-
propylsilyloxymethyl)-but-3-en-1-ol [(1R,2S)-2b-d]. Colorless oil
(61% ee). ½a 2D6
ꢂ
ꢀ6.2 (c 0.46, CHCl3); 1H NMR
: 1.01 (d, J¼7.1 Hz, 3H),
d
1.06e1.16 (m, 21H), 2.65 (q, J¼7.1 Hz, 1H), 3.00 (s, 1H), 3.83 (s, 3H),
4.10 (d, J¼13.6 Hz, 1H), 4.19 (d, J¼13.6 Hz, 1H), 5.02 (s, 1H), 5.20 (s,
1H), 6.85 (d, J¼7.9 Hz, 1H), 6.94 (t, J¼7.4 Hz, 1H), 7.21 (td, J¼7.9
between (S)- and (R)-MTPA esters (Dd
¼
dS dR). dH and dD are the chemical shift
ꢀ
difference (Dd) in the 1H and 2D NMR spectra. In the equilibrium between two
favorable conformations of the MTPA esters of (S)-5b-d, the benzylic proton of
(R)-MTPA ester and deuterium of (S)-MTPA ester would be shielded by the
phenyl group (See the scheme below). In the Mosher ester of (R)-5b-d, deu-
terium of (R)-MTPA ester and the benzylic proton of (S)-MTPA ester were
shielded. Therefore, we assumed that the enantiomer with positive DdH values
(þ0.07 ppm) and negative DdD values (ꢀ0.04 ppm) was assigned (S) config-
uration. The enantiomer exhibiting an opposite sign, that is, negative DdH
values (ꢀ0.07 ppm) and positive DdD values (þ0.06 ppm), was assigned (R)
configuration. A similar discussion was reported using CFTA ester by Takeuchi
and co-workers in Ref. 11. In addition, the comparison of the specific rotation
,1.5 Hz, 1H), 7.40 (dd, J¼7.4, 1.5 Hz, 1H); 13C NMR
d: 12.0 (3C), 12.7,
18.0 (6C), 42.1, 55.2, 65.6, 71.8 (t, J(C,D)¼22.3 Hz), 110.1, 111.0, 120.4,
127.8, 127.8, 131.0, 151.3, 156.1; IR (KBr) cmꢀ1: 3423, 2163, 1650,
1602, 1491; MS (FAB) m/z: 380 [MþH]þ; HRMS (FAB) m/z: calcd for
C22H38DO3Si: 380.2731, found: 380.2744 [MþH]þ.
Acknowledgements
with
a-monodeuterated benzyl alcohols [(R)- and (S)-5a-d] support the
assignment.
We acknowledge the financial support of a Grant-in-Aid for
Scientific Research (C) from the Japan Society for the Promotion of
Science (No. 21590032) and Osaka Ohtani University Research Fund
(Pharmaceutical Sciences).
Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2012.03.059.
References and notes
1. For reviews of [2,3]-Wittig rearrangement, see: (a) Tomooka, K. Chem. Orga-
nolithium Compd. 2004, 2, 749e828; (b) Hiersemann, M.; Abraham, L.; Pollex, A.
Synlett 2003, 1088e1095; (c) McGowan, G. Aust. J. Chem. 2002, 55, 799; (d)

M. Kitamura et al. / Tetrahedron 68 (2012) 4280e4285
4285
13. From the 1H NMR spectral data, the deuterium exchange ratio of both enan-
tiomers of 1a-d and 1b-d is high.
14. Ten equivalents of base were used to complete the [2,3]-Wittig rearrangement.
15. Hoppe reported the large H/D kinetic isotope effect on deprotonation; see:
Hoppe, D.; Paetow, M.; Hintze, F. Angew. Chem., Int. Ed. Engl. 1993, 32, 394e396.
16. Since ee of the substrate is estimated as 93% ee from that of (R)-5a-d, the ee
of the resulting (1R,2S)-2a-d would be increased to 98% ee by kinetic
resolution.
17. Specific rotation was considerably small to be measured with accuracy.