Stereoselective anti-SN2′ Mitsunobu reaction of α-hydroxy-α-alkenylsilanes

Article abstract of DOI:10.1016/j.tetlet.2010.11.080

A novel silyl group-directed anti-S_N2′ reaction of allylic alcohols under Mitsunobu reaction conditions is described. The Mitsunobu reaction of α-hydroxy-α-alkenylsilanes with a TBS or TIPS group gave the anti-S_N2′ product, in which regio- and stereochemical outcomes of the reaction depended on the steric bulkiness of the silyl group.

Full text of DOI:10.1016/j.tetlet.2010.11.080

Tetrahedron Letters 52 (2011) 422–425
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Stereoselective anti-SN20 Mitsunobu reaction of
a-hydroxy-
a-alkenylsilanes
⇑
⇑
Masato Higashino, Naoko Ikeda, Tetsuro Shinada, Kazuhiko Sakaguchi , Yasufumi Ohfune
Graduate School of Science, Department of Material Science, Osaka City University, Sugimoto, Sumiyoshi, Osaka 558-8585, Japan
a r t i c l e i n f o
a b s t r a c t
A novel silyl group-directed anti-SN20 reaction of allylic alcohols under Mitsunobu reaction conditions is
Article history:
Received 14 October 2010
Revised 11 November 2010
Accepted 15 November 2010
Available online 18 November 2010
described. The Mitsunobu reaction of a-hydroxy-a-alkenylsilanes with a TBS or TIPS group gave the anti-
SN20 product, in which regio- and stereochemical outcomes of the reaction depended on the steric bulk-
iness of the silyl group.
Ó 2010 Elsevier Ltd. All rights reserved.
Mitsunobu reaction of secondary alcohols:
The nucleophilic substitution reaction of alcohols by various
types of hetero-atoms (O, N, S) or carbon nucleophiles, which is
typically effected by a triaryl- or trialkylphosphine and an azodi-
carboxylate as the stoichiometric reagent, is well-known as the
Mitsunobu reaction.1 The Mitsunobu reaction of a chiral secondary
alcohol has become an important method to prepare its inversion
product in an SN2 fashion (Eq. (1)). Limited numbers of this method
were applied to secondary allylic alcohols in which the stereo-
OH
OPPh3
Nu
Nu
DEAD, PPh3
ð1Þ
ð2Þ
R1 R2
R1 R2
R1 R2
SN2
Mitsunobu reaction of secondary allylic alcohols:
OH
Nu
Nu
DEAD, PPh3
Nu
γ
+
R1
R2
R1
R2
R1
OH
γ
R2
α
α
OH
chemical outcome is dependent on the
a- and/or c-substituents.
γ
α
10% H2SO4 aq.
THF, rt
recovery of 1
86% (90% ee)
In some cases, the regio- and stereo-selectivities were not well-
documented since the Mitsunobu reaction of the secondary allylic
alcohols involves two major pathways to the regio- and stereo-
+
TBS
α
Me
γ
TBS
2 (10%, 29% ee)
Me
1 (90% ee)
ð3Þ
ð4Þ
selectivities, that is, an SN2 or SN1 reaction at the
attached to the hydroxy group, or a syn-SN20, anti-SN20 or SN10 reac-
tion at the
-allylic carbon (Eq. (2)).2,3 Therefore, the Mitsunobu
reaction of the secondary allylic alcohol has remained as an impor-
tant subject.
a-carbon
TBS
Me
TBS
Me
α-silyl cation
γ-silyl cation
c
This work
Mitsunobu
reaction
OH
Nu
γ
During the course of our recent studies regarding the cationic
R3Si
α
R'
R3Si
γ
R'
reactions of optically active a
-hydroxysilanes,4,5 we reported that
SN2' product
A
the treatment of
a-hydroxy-a-alkenylsilanes 1 (90% ee) with 10%
H2SO4 gave the
c
-hydroxyvinylsilane 2 (10% yield, 29% ee) to-
To examine the mode of the Mitsunobu reaction of
a-hydroxy-
gether with the recovery of the starting alcohol 1 (86%, 90% ee)
with no loss of its optical purity (Eq. (3)).5 Despite the cationic
reaction conditions, the chirality of the starting 1 was partially
transferred to the product 2 (syn-SN20 product). We have postu-
lated that the instability of the putative
ing force for the rearrangement to the more stable
(Eq. (3)). These results led us to explore whether the Mitsunobu
reaction of the -hydroxy-
-alkenylsilane A facilitates the SN20
reaction (Eq. (4)). Herein, we wish to report the first example of
the silyl group-directed anti-SN20 Mitsunobu reaction of
-hydro-
a
-alkenylsilanes, the chiral TBS-substituted (R)- -hydroxy-a-
a
alkenylsilane 3a (78% ee) was employed as the model substrate
(Table 1, entry 1). The reaction of 3a (78% ee) with 4-nitrobenzoic
acid under the standard Mitsunobu reaction conditions (DEAD,
a
-silyl cation6 is the driv-
-silyl cation
PPh3, benzene, rt)7 gave the
c-nitrobenzoyloxy-(E)-vinylsilane 5a
c
as the major product together with the -substituted isomer 4a
a
(91%, 4a/5a = 16:84).8,9 The optical purity and absolute configura-
tion of the products, 5a and 4a, were determined to be 73% ee with
R-configuration for 5a and 78% ee and S for 4a, respectively, by
converting them into the corresponding (R)- and (S)-MTPA esters
followed by the modified Mosher method10 (Supplementary data).
To elucidate whether the isolated (R)-5a and (S)-4a were the Mits-
unobu products, these compounds were subjected to the same
reaction conditions.11 Since both isomers were recovered without
the loss of their optical purity, the inter-conversions such as the
a
a
a
xy-a-alkenylsilanes.
⇑
Corresponding authors. Tel.: +81 6 6605 2571; fax: +81 6 6605 2522 (K.S.).
E-mail address: sakaguch@sci.osaka-cu.ac.jp (K. Sakaguchi).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.11.080

M. Higashino et al. / Tetrahedron Letters 52 (2011) 422–425
423
Table 1
Mitsunobu reaction of
a-hydroxysilanes
no inter-conversions
OH
OCOAr
Conditionsa
γ
OCOAr
Ph
Ph
+
Ph
α
Si
α
Si
γ
Si
Ar = 4-NO2-C6H4-
(R)-3
(S)-4
Yieldb
(R)-5
Entry
Si
Solvent
Time
4:5c
Es of 4d
Es of 5d
1
2
3
4
5
6
7
8
TBS (3a)e
TBS (3a)e
TBS (3a)e
TBS (3a)e
TBS (3a)e
TIPS (3b)h
DMPS (3c)j
TMS (3d)k
Benzene
THF
DME
CH2Cl2
CH3CN
Benzene
Benzene
Benzene
12 h
12 h
12 h
12 h
12 h
72 hi
2 h
91%
84%
65%f
26%f
14%f
57%
78%
95%
16:84
14:86
10:90
6:94
14:86
9: 91
69:31
80:20
100% (4a)
100% (4a)
94% (5a)
96% (5a)
86% (5a)
85% (5a)
76% (5a)
100% (5b)
65% (5c)
74% (5d)
g
—
g
—
—
g
91% (4b)
94% (4c)
95% (4d)
0.5 h
a
4-Nitrobenzoic acid (1.5 equiv), DEAD (1.5 equiv), PPh3 (1.5 equiv), rt.
Combined yields as an inseparable mixture of 4 and 5.
The ratio was determined by 1H NMR.
b
c
d
e
f
Es (enantio specificity) was defined by the following equation: es = ee (product)/ee (s.m.) ꢀ 100.
78% ee.
The starting material was recovered: 19% (DME), 57% (CH2Cl2), 49% (CH3CN).
Not determined.
74% ee.
DEAD (5.0 equiv), PPh3 (5.0 equiv) and ArCO2H (5.0 equiv) were used.
80% ee.
78% ee.
g
h
i
j
k
Table 2
Mitsunobu reaction of other
acyl rearrangement between 5a and 4a did not occur at all. These
experiments indicated that the
-substituted 5a was an anti-SN20
Mitsunobu product (94% es) and the -substituted 4a was an SN2
Mitsunobu product (100% es), with high degrees of chirality trans-
fer from the chiral 3a.
Next, we examined the effect of the solvent and silyl substituent
for the regio- and stereo-selectivities of the Mitsunobu products.
The use of THF gave almost the same result as that of benzene
a
-hydroxysilanes
c
OCOAr
OCOAr
OH
Conditionsa
a
γ
+
γ
α
TBS
R
α
TBS
R
TBS
R
7
Ar = 4-NO2-C6H4-
8
6
Entry
R
Time
Yieldb
7:8c
Es of 7
Es of 8
1d
2
i-Pr (6a)e
Ph (6b)f
H (6c)h
5 h
4 h
1 h
93%
96%
64%
15:85
10:90
84:16
—
—
(Table 1, entry 2). Although a higher
c-selectivity was observed
6% (7b)
96% (7c)
5% (8b)
—
3g
using DME, CH2Cl2, or CH3CN, the reactions were slow which re-
sulted in decreased yields (14–65%) and the es (76–86% es, entries
3–5). Thus, benzene or THF was used as the solvent for further
studies.12 The sterically more bulky TIPS-substituted 3b showed
a
b
c
d
e
f
4-Nitrobenzoic acid (1.5 equiv), DEAD (1.5 equiv), PPh3 (1.5 equiv), C6H6, rt.
Combined yields as an inseparable mixture of 7 and 8.
The ratio was determined by 1H NMR.
THF was used as the solvent.
Racemate.
an increased regioselectivity and the complete es of the c-substi-
tuted product (R)-5b (4b/5b = 9:91, 100% es), while the reaction re-
quired a larger amount of reagents and prolonged reaction time
(entry 6). On the other hand, the reaction of the less sterically
bulky DMPS-substituted 3c and TMS-substituted 3d13 gave a mix-
93% ee, R form.
g
4-Nitrobenzoic acid (1.2 equiv), DEAD (1.2 equiv) and PPh3 (1.2 equiv) were
used.
h
95% ee, S form.
ture of the substitution products in which the
4d were the major isomers, respectively, and the stereoselectivity
of the -substituted 5c and 5d decreased (entries 7 and 8).
A significant -selectivity was also observed using other alkyl or
phenyl-substituted olefins, 6a and 6b (Table 2). However, the ee of
the cinnamyl alcohol derivative 6b (93% ee) was poorly transferred
to the products (6% es for 7b and 5% es for 8b), suggesting that the
phenyl group participated in the stabilization of the allylic cation
that facilitated the SN1 and SN10-type reactions.14,15 The reaction
a-substituted 4c and
c
OCOAr
OCOAr
OH
Conditions
c
R
R
+
TBS
22 : 78
R
TBS
TBS
THF ,12 h
73%
(S)-10
(es not determined)
(S)-5a (8% es)
(R)-9 (>95% ee)
OH
OCOAr
OCOAr
R
Conditions
+
R
benzene, 20 h
87%
R
of the terminal olefin 6c showed a high
probably due to the low electrophilicity, that is, instability of the
primary cation at the -position.
a-selectivity (entry 3)
32 : 68
(R)-11 (96% ee)
(S)-12 (73% es)
(R)-13 (46% es)
Ar = 4-NO2-C6H4-
R = CH2CH2CH2Ph
c
Conditions: ArCO2H (1.5 equiv), DEAD (1.5 equiv),
PPh3 (1.5 equiv), rt.
To obtain some insights into the reaction mechanism of the no-
vel anti-SN20 reaction, several secondary alcohols were examined
for the Mitsunobu reaction. Under the standard reaction condi-
tions, the chiral (R,Z)-TBS-substituted 9 gave the SN20 product 5a
Scheme 1.
as the major product in which the
creased (10/5a = 22:78) in comparison to that of the (E)-isomer
= 16:84, Table 1). Significant loss of its es (8%) was observed
(Scheme 1). Although the -primary or secondary alkyl substituted
c-selectivity was slightly de-
allylic alcohols mainly afforded the normal SN2 Mitsunobu prod-
uct,2 the reaction of the carbon analog (R)-11 possessing a t-Bu
group, which is more sterically bulky in A-value than the TMS
(a/c
a

424
M. Higashino et al. / Tetrahedron Letters 52 (2011) 422–425
group,16 has not been documented. The reaction gave a mixture of
the SN2 and SN20 products, in which the
-selectivity was moderate
and the es of the products decreased [12 (73% es)/13 (46%
es) = 32:68]. On the other hand, the -alkynyl- -hydroxysilane
14 exclusively gave the SN2 product 15, and the -cyclopropyl-
-hydroxysilane 16 and saturated -hydroxysilane 17 were inert
the attack of the nucleophile at the
of the less bulky DMPS and TMS groups, the decreased steric repul-
sion in C would allow the attack of the nucleophile from the -face
in a competitive manner (anti and syn-SN20) resulting in a de-
creased stereoselectivity at the -position. The significant decrease
in the es observed in the carbon analogs (73% es for 12 and 46% es
for 13) would be due to the stability of the carbocation since an a-
a-position (SN2). In the case
c
a
a
a
a
c
a
a
under the reaction conditions and the starting materials were
recovered (Scheme 2).
alkyl-substituted carbocation is much more stable than that of an
a
The above experimental results indicated that the silyl group
was mostly attributed to the regio- and stereo-selectivities of the
Mitsunobu reaction of the (E)-allylic alcohols. The characteristic
points of the reactions are summarized as follows: (i) the degrees
of SN2 vs SN20 depended on the silyl substituent, that is, the steri-
cally bulky TBS and TIPS substrates underwent the anti-SN20 reac-
tion as the major pathway while the less sterically bulky TMS
-silyl-substituted carbocation.6,18
In summary, the Mitsunobu reaction of the optically active
a-
hydroxy-a-alkenylsilane possessing a TBS or TIPS group and (E)-
olefin underwent the anti-SN20 reaction to give the (S,E)-vinylsilane
in a highly regio- and stereo-selective manner. The silyl group
played an essential role in the regio- and stereo-selective attack
of the nucleophile where the competitive reactions (SN2 and
syn-SN20) were observed when the sterically less bulky silyl group
(DMPS or TMS) was employed. Further efforts to expand the scope
of the present SN20-type Mitsunobu reaction using other substrates
and nucleophiles, such as nitrogen, sulfur, and carbon are
underway.
and DMP substrates preferred the SN2 reaction. (ii) The a-substitu-
tions occurred in a complete inversion (SN2) and the high degrees
of chirality transfer (anti-SN20) were observed when the sterically
bulky TBS and TIPS groups were employed. (iii) The SN20 reaction
exclusively produced the E-vinylsilanes. (iv) The presence of the
alkenyl
p
-conjugated system is essential to promoting the SN20
Mitsunobu reaction since the propargyl, alkyl, and cyclopropyl
Acknowledgement
groups did not afford the SN20-products.
Based on these results, we propose the putative reaction path-
way for the present regio- and stereo-selective Mitsunobu reac-
tion. As shown in Figure 1, the breaking C–O bond is placed
perpendicular to the olefin plane and the silyl group is located out-
side of the olefin to minimize the allylic-1,3-strain in the transition
state of the SN20 reaction. The conformer B is more favorable than
the conformer C because of the large steric repulsion between the
bulky silyl (TBS and TIPS) and PPh3 groups. In the conformer B, the
This study was financially supported by Grants-in-Aid
(16201045 and 19201045) for Scientific Research from the Japan
Society of the Promotion of Science (JSPS).
Supplementary data
Supplementary data (Full experimental details and character-
ization data of synthetic compounds) associated with this article
can be found, in the online version, at doi:10.1016/j.tetlet.2010.
11.080.
a
face of the olefin is shielded by the PPh3 group, therefore, the at-
tack of the nucleophile would occur from the opposite side of the
C–O bond to give the (R,E)-vinylsilane as the anti-SN20 product.17
The sterically bulky silyl groups would also effectively prevent
References and notes
1. (a) Mitsunobu, O.; Yamada, M.; Mukaiyama, T. Bull. Chem. Soc. Jpn 1967, 40,
935–939; (b) Mitsunobu, O.; Yamada, M. Bull. Chem. Soc. Jpn 1967, 40, 2380–
2382; (c) Mitsunobu, O.; Eguchi, M. Bull. Chem. Soc. Jpn 1971, 44, 3427–3430;
(d) Hughes, D. L. Org. React. 1992, 42, 335–656; (e) Swamy, K. C. K.; Kumar, N.;
Balaraman, E. Chem. Rev. 2009, 109, 2551–2651.
2. The SN2 Mitsunobu reaction of secondary allylic alcohols: (a) Danishefsky, S.;
Berman, E. M.; Ciufolini, M.; Etheredge, S. J.; Segmuller, B. E. J. Am. Chem. Soc.
1985, 107, 3891–3898; (b) Ito, Y.; Sawamura, M.; Hayashi, T. Tetrahedron Lett.
1988, 29, 239–240.
OH
OCOAr
Conditions
THF, rt, 2.5 h
81%
TBS
TBS
(R)-14 (>95% ee)
(S)-15 (92% ee)
OH
Conditions
recovery of 16
recovery of 17
3. The SN20 Mitsunobu reaction in 2-methylidene systems, see: (a) Charette, A. B.;
Cote, B. Tetrahedron Lett. 1993, 34, 6833–6836; (b) Aditya, S.; Gary, A. S.
Tetrahedron Lett. 1994, 35, 3661–3664; (c) Charette, A.; Cote, B. J. Org. Chem.
1995, 60, 6888–6894; Cyclic systems, see: (d) Burke, S. D.; Pacofsky, G. J.
Tetrahedron Lett. 1986, 27, 445–448; (e) Shull, B. K.; Sakai, T.; Nichols, J. B.;
Koreeda, M. J. Org. Chem. 1997, 62, 8294–8303; (f) Kwon, Y. U.; Chung, S. K. Org.
Lett. 2001, 3, 3013–3016; (g) Kwon, Y. U.; Lee, C.; Chung, S. K. J. Org. Chem. 2002,
67, 3327–3338; Vinyl boronates, see: (h) Berrée, F.; Gernigon, N.; Hercouet, A.;
Lin, C. H.; Carboni, B. Eur. J. Org. Chem. 2009, 329–333.
4. (a) Sakaguchi, K.; Fujita, M.; Ohfune, Y. Tetrahedron Lett. 1998, 39, 4313–4316;
(b) Sakaguchi, K.; Suzuki, H.; Ohfune, Y. Chirality 2001, 13, 357–365; (c)
Sakaguchi, K.; Yamada, T.; Ohfune, Y. Tetrahedron Lett. 2005, 46, 5009–5012; (d)
Sakaguchi, K.; Okada, T.; Yamada, T.; Ohfune, Y. Tetrahedron Lett. 2007, 48,
3925–3928; (e) Sakaguchi, K.; Okada, T.; Shinada, T.; Ohfune, Y. Tetrahedron
Lett. 2008, 49, 25–28.
TBS
benzene, reflux, 3 h
16 (racemate)
OH
+
94% (78% ee)
OCOAr
Conditions
Ph
benzene, rt, 45 h
TBS
Ph
(R)-17 (78% ee)
TBS
18 trace
Ar = 4-NO2-C6H4-
Conditions: ArCO2H (1.5 equiv), DEAD (1.5 equiv),
PPh3 (1.5 equiv).
Scheme 2.
5. Sakaguchi, K.; Higashino, M.; Ohfune, Y. Tetrahedron 2003, 59, 6647–6658.
6. (a) Wierschke, S. G.; Chandrasekhar, J.; Jorgensen, W. L. J. Am. Chem. Soc. 1985,
107, 1496–1500; (b) Apeloig, Y.; Stanger, A. J. Am. Chem. Soc. 1985, 107,
2806–2807; (c) Apeloig, Y.; Biton, R.; Abu-Freih, A. J. Am. Chem. Soc. 1993, 115,
2252–2253.
Nu
(anti-SN2')
Nu
(anti-SN2')
(SN2)
Nu
7. General procedure of the reactions is described in Supplementary data.
8. The diene 19 (mixture of the E/Z isomers, ca. 1:1) was a by-product (<10%).
R'
R'
H
H
γ
γ
Si
Si
H
H
H
H
α
O
Ph
19
TBS
O
(syn-SN2')
PPh3
(syn-SN2')
Ph3P
Nu
Nu
B
C
9. The reaction at a lower temperature (ꢁ20 °C) was very slow (23 h, 52%, 4a/
5a = 19: 81), however, the complete chirality transfer was observed in 5a (99%
es).
Figure 1.

M. Higashino et al. / Tetrahedron Letters 52 (2011) 422–425
425
10. Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113,
4092–4093.
16. The A-value of t-Bu (>4.5 kcal/mol) is greater than that of TMS (2.5 kcal/mol):
Kitching, W.; Olszowy, H. A.; Drew, G. M.; Adcock, W. J. Org. Chem. 1982, 47,
5153–5156.
17. It is not clear at this stage why the reaction of (R,Z)-9 resulted in a significant
loss of its es (8% es). Possible explanations for the present low es are under
investigation.
11. The reaction without DEAD or PPh3 did not proceed at all.
12. The combination of the Mitsunobu reagents, DEAD and PPh3, was suitable for
the present reaction. Other combinations, for example, 1,10-(azodicarbonyl)-
dipiperidine (ADDP), N,N,N0,N0-tetramethylazodicarboxamide (TMAD), PBu3,
PMePh2, resulted in a recovery of the starting materials. The reaction using
18. The reaction of the TBS-containing-c-hydroxyvinylsilane 21 (69% ee) under the
benzoic acid and 4-methoxybenzoic acid as
a
nucleophile also gave the
standard conditions gave the SN2 product (S)-5a (100% es) as the sole isomer
(77%) together with the recovery of 21 (22%). The DMPS and TMS-substituted
corresponding Mitsunobu products (Supplementary data).
13. The relative steric bulk of trialkylsilyl groups; see: Brook, M. A. In Silicon in
Organic, Organometallic, and Polymer Chemistry; Wiley & Sons: New York, 2000.
Chap. 8, pp 199–201.
14. Using CH2Cl2 as the solvent, 7b (6%, 11% es) and 8b (59%, 8% es) were obtained.
15. The reaction of 20 (racemate) under the standard conditions gave a mixture of
rac-7b and rac-8b (92%, 7b/8b = 11:89), which is accordance in the yield and
regioselectivity of the reaction of 6b (Table 2, entry 2). Both 7b and 8b were
confirmed as the kinetic products, because no inter-conversions between them
occurred under the reaction conditions.
analogs of 21 also gave the
c-substituted products as the sole isomer in
excellent yields, respectively (90%-quant). However, the reaction of the t-Bu
substituted analog 22 (racemate) gave a mixture of rac-12 and rac-13 (91%, 12/
13 = 10:90).
ArCO2H
DEAD, PPh3
(S)-5a
recovery of 21
OH
(from 21)
+
(77%, 100% es)
(22%)
Ph
R
rac-12 + rac-13 (from 22)
(91%, 12/13 = 10 : 90)
THF, rt, 12 h
21: R = TBS (69% ee)
22: R = t-Bu (racemate)
Ar = 4-NO2-C6H4-
ArCO2H
DEAD, PPh3
OH
rac-7b
+
rac-8b
TBS
20 (racemate)
Ph
THF, rt, 12 h, 92%
Ar = 4-NO2-C6H4-
11 : 89