Catalytic asymmetric synthesis of 2,2-disubstituted oxetanes from ketones by using a one-pot sequential addition of sulfur ylide

Article abstract of DOI:10.1002/anie.200805473

(Chemical Equation Presented) Better the second time around: The title compounds were synthesized by using a one-pot double methylene transfer catalyzed by a heterobimetallic La/Li complex. Chiral amplification in the second step was the key to obtaining

Full text of DOI:10.1002/anie.200805473

Angewandte
Chemie
DOI: 10.1002/anie.200805473
Asymmetric Synthesis
Catalytic Asymmetric Synthesis of 2,2-Disubstituted Oxetanes from
Ketones by Using a One-Pot Sequential Addition of Sulfur Ylide**
Toshihiko Sone, Gang Lu, Shigeki Matsunaga,* and Masakatsu Shibasaki*
The catalytic enantioselective syntheses of epoxides
has been studied intensively over the last three
decades.[1] In striking contrast, there are few reports
on the catalytic enantioselective synthesis of higher
analogues—oxetanes. Oxetanes have recently gained
attention as attractive, stable, small, and less lipophilic
molecular modules for drug discovery.[2] Considering
the wide applicability of oxetane units in medicinal
chemistry, organic synthesis, and material science,[2–5]
the development of synthetic methods for making a
diverse array of oxetanes, especially optically active
oxetanes, is highly desirable. Highly enantioselective
Figure 1. Structures of (S)-RE-M3-tris(binaphthoxide) complexes (REMB,
RE=rare-earth metal).
syntheses of 2-monosubstituted oxetanes, involving
the asymmetric reduction of 3-haloketones using
chiral reagents or catalysts and subsequent cycliza-
tion, have been reported.[4] However such strategies are not
applicable to the synthesis of 2,2-disubstituted oxetanes.
Previously reported methods for chiral 2,2-disubstituted
oxetanes required a lengthy multistep synthetic route.[5]
Herein, we describe a straightforward one-pot catalytic
asymmetric synthesis of 2,2-disubstituted oxetanes from
ketones. A LaLi3tris(binaphthoxide) complex (LLB ((S)-
reaction of epoxide 2a with ylide 3 to afford (R)-oxetane 4a
in 16% yield and 24% ee (Table 1, entry 1). No reaction
proceeded in the absence of (S)-LLB under otherwise
identical conditions at 458C (Table 1, entry 2).[11] Both (S)-
LSB ((S)-1b) and (S)-LPB ((S)-1c; Figure 1) afforded only
trace amounts, if any, of oxetane 4a (Table 1, entries 3 and 4).
We then examined the effects of achiral additives, which are
useful for modifying the chiral environment of REMB
complexes.[12] NaI, which is effective in the Corey–Chaykov-
sky cyclopropanation of enones and a,b-unsaturated N-
acylpyrroles,[13] was not at all effective (Table 1, entry 5) for
generating oxetanes. In contrast, phosphine oxides, which are
effective in the Corey–Chaykovsky epoxidation of ketones,[14]
increased both the reactivity and selectivity of the reaction.[15]
Among the phosphine oxides screened (Table 1, entries 6–9),
=
1a), Figure 1) with an Ar3P O additive (Ar= 2,4,6-trimeth-
oxyphenyl) promoted the successive addition of dimethyloxo-
sulfonium methylide to ketones and intermediate epoxides.
Chiral amplification was observed in the second step to afford
oxetanes with greater than 99.5–99% ee, an enantiomeric
excess higher than that of the intermediate epoxides.
Initially, we screened chiral catalysts suitable for the
reaction of readily available racemic 2,2’-disubstituted epox-
ide 2a with dimethyloxosulfonium methylide (3).[6] Among
the catalysts screened, rare earth/alkali metal/binol hetero-
bimetallic complexes (REMB; binol = 2,2’-dihydroxy-1,1’-
binaphthyl)[7] gave the most promising results for a kinetic
resolution.[8–10] The results of the optimization studies are
summarized in Table 1. (S)-LLB ((S)-1a) promoted the
=
Ar3P O 5d (Ar= 2,4,6-trimethoxyphenyl) afforded the best
results, giving oxetane 4a in 30% yield and 59% ee (Table 1,
entry 9, krel = 5.0).[16] Selectivity was additionally improved by
changing the reaction solvent from THF to either THF/
toluene (1:1, Table 1, entry 10, krel = 7.5) or THF/n-hexane
(1:1, Table 1, entry 11, krel = 7.7). By increasing the catalyst
loading to 10 mol%, oxetane 4a was obtained in 41% yield
and 73% ee after 48 hours (Table 1, entry 12, krel = 10.4). In
entry 12 of Table 1, (R)-2a was recovered in 42% ee. (R)-2b,
having an alkyl substituent, was also applicable under the
same reaction conditions, and oxetane 4b was obtained in
43% yield and 78% ee (Table 1, entry 13, krel = 14.6).
Although the krel values observed in the kinetic resolution,
by using racemic epoxides 2a and 2b, were not high (Table 1),
we assumed that the selectivity was sufficient to obtain
oxetanes 4 in high enantiomeric excess by a double methylene
transfer starting from ketones. In our previous report, (S)-
LLB ((S)-1a) preferentially afforded (S)-2a and (S)-2b
through the addition of ylide 3 to ketones.[14] The results
shown in Table 1 suggest that (S)-1a preferentially promoted
[*] T. Sone, G. Lu, Dr. S. Matsunaga, Prof. Dr. M. Shibasaki
Graduate School of Pharmaceutical Sciences
The University of Tokyo
Hongo, Bunkyo-ku, Tokyo 113-0033 (Japan)
Fax: (+81)3-5684-5206
E-mail: smatsuna@mol.f.u-tokyo.ac.jp
mshibasa@mol.f.u-tokyo.ac.jp
Homepage: http://www.f.u-tokyo.ac.jp/~kanai/e_index.html
[**] This work was supported by Grant-in-Aid for Scientific Research (S)
(for M.S.) and Grant-in-Aid for Scientific Research on Priority Areas
(No. 20037010, Chemistry of Concerto Catalysis) (for S.M.) from
JSPS and MEXT.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200805473.
Angew. Chem. Int. Ed. 2009, 48, 1677 –1680
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1677

Communications
Table 1: Optimization studies on the kinetic resolution of racemic 2,2’-disubstituted epoxides 2 catalyzed by heterobimetallic REMB complexes.
[c]
Entry
Epoxide
REMB (mol%)
Additive 5 (mol%)
Solvent
ee of
recovered
2 [%]
Yield of 4
ee of 4
krel
[%][a]
[%][b]
1
2
3
4
5
6
7
8
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2b
(S)-LLB (5)
none
none
none
none
none
NaI (5)
THF
THF
THF
THF
THF
THF
THF
THF
1
0
0
0
3
1
2
6
19
16
19
42
65
16
0
trace
trace
26
15
19
19
30
22
27
41
43
24
–
–
1.7
–
–
(S)-LSB (5)
(S)-LPB (5)
(S)-LLB (5)
(S)-LLB (5)
(S)-LLB (5)
(S)-LLB (5)
(S)-LLB (5)
(S)-LLB (5)
(S)-LLB (5)
(S)-LLB (10)
(S)-LLB (10)
–
–
19
20
23
44
59
72
71
73
78
1.6
1.5
1.7
2.9
5.0
7.5
7.7
10.4
14.6
=
5a: Ph3P O (5)
5b: (2,4,6-Me3-C6H2)3P O (5)
5c: (2,6-(MeO)2-C6H3)3P O (5)
=
=
=
9
5d: (2,4,6-(MeO)3-C6H2)3P O (5)
THF
10
11
12
13
5d (5)
5d (5)
5d (5)
5d (5)
THF/toluene (1:1)
THF/n-hexane (1:1)
THF/n-hexane (1:1)
THF/n-hexane (1:1)
[a] Yield of oxetane 4 was calculated based on the amount of starting racemic epoxide (Æ)-2. Theoretical maximum yield of chiral 4 is 50%.
[b] Determined by chiral GC analysis. [c] See reference [16].
the reaction of (S)-2a and (S)-2b with ylide 3 to give oxetanes
4a and 4b. We hypothesized that chiral amplification in the
second reaction between the epoxide and the ylide would
occur to afford oxetanes with a higher enantioselectivity than
the intermediate epoxides. Because the optimized reaction
conditions for the kinetic resolution of epoxides 2 (Table 1)
and the best conditions for the Corey–Chaykovsky epoxida-
tion of ketones[14] were the same, except for the reaction
temperature and solvent, we investigated a one-pot sequen-
tial methylene-transfer process starting from ketone 6b
(Scheme 1). Catalytic asymmetric epoxidation using
5 mol% of (S)-1a in THF at room temperature produced
epoxide 2b in greater than 99% yield and 93% ee after
12 hours. To compensate for the low reaction rate in the
second reaction, an additional 15 mol% of (S)-1a and
additive 5d were added to the reaction mixture together
with an additional 1.0 equivalents of ylide 3 and n-hexane.
The reaction mixture, including a total of 20 mol% of (S)-1a/
5d, was then stirred at 458C for 72 hours to give oxetane 4b in
88% yield and 99% ee upon isolation. As expected, chiral
amplification was observed in the second step (93% ee of 2b
versus 99% ee of 4b).
The substrate scope of the one-pot sequential methylene-
transfer process is summarized in Table 2. With aryl methyl
Table 2: Substrate scope of oxetane 4 synthesis by a one-pot sequential
addition of ylide 3 to various methyl ketones 6.[a]
Entry
6
R
ee of 2
[%]
4
Yield of 4
ee of 4
[%][b]
[%][c]
1
2
3
6a
6c
6d
96
96
94
4a
4c
4d
74
62
86
99
99
99
4
6e
97
4e
85
99
5
6
6b
6 f
93
93
4b
4 f
88
68
99
99
7
8
6g
6h
96
97
4g
4h
58
62
>99.5
>99.5
[a] Reaction was performed with 1.2 equiv of ylide 3 in THF at room
temperature using 5 mol% of (S)-1a/5d (1:1) for the first step; for the
second step, an additional 15 mol% of (S)-1a/5d (1:1), 1.0 equiv of ylide
3, and n-hexane were added to the reaction mixture. [b] Yield of isolated
oxetanes 4 (from ketones 6) after purification by column chromatog-
raphy. [c] Determined by GC or HPLC analysis.
Scheme 1. One-pot sequential addition of ylide 3 to ketone 6b with
chiral amplification.
1678
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1677 –1680

Angewandte
Chemie
[7] A review on heterobimetallic asymmetric catalysis: a) M.
ketones 6a and 6c–6e, the enantioselectivity in the first
epoxidation step was 94–97% ee. Oxetanes 4a and 4c–4e
were obtained in 99% ee and 86–62% yield after the second
step, although the krel value for aryl methyl ketones was
modest, as shown in Table 1.[17] The enantioselectivity in the
first epoxidation step was slightly lower for linear alkyl
methyl ketones 6b and 6 f than those for aryl methyl ketones
(Table 2, entries 5 and 6, 93% ee). Fortunately, alkyl methyl-
substituted oxetanes 4b and 4 f–4h were also obtained in
greater than 99.5–99% ee and 88–58% yield through chiral
amplification in the second step (Table 2, entries 5–8).
In summary, we developed a straightforward synthetic
method for chiral 2,2-disubstituted oxetanes. A one-pot
sequential addition of a sulfur ylide to ketones and inter-
mediate epoxides was promoted by the heterobimetallic
catalyst 1a. Chiral amplification was observed in the second
reaction of the epoxides with the sulfur ylide and was the key
to affording 2,2-disubstituted oxetanes with excellent enan-
tioselectivity (> 99.5–99% ee). Additional trials to improve
the reactivity, catalyst loading, and substrate scope of both the
epoxides and ylides in the second oxetane-forming step are
ongoing.[18]
Shibasaki, N. Yoshikawa, Chem. Rev. 2002, 102, 2187; For
selected recent examples, see: b) N. Yamagiwa, S. Matsunaga, M.
Shibasaki, J. Am. Chem. Soc. 2003, 125, 16178; c) N. Yamagiwa,
S. Matsunaga, M. Shibasaki, Angew. Chem. 2004, 116, 4593;
Angew. Chem. Int. Ed. 2004, 43, 4493; d) N. Yamagiwa, H. Qin,
S. Matsunaga, M. Shibasaki, J. Am. Chem. Soc. 2005, 127, 13419.
[8] General review for nonenzymatic kinetic resolution: E. Vedejs,
M. Jure, Angew. Chem. 2005, 117, 4040; Angew. Chem. Int. Ed.
2005, 44, 3974.
[9] Kinetic resolution of 2,2-disubstituted terminal epoxides using a
chiral metal catalyst is rare. For resolution with an azide
nucleophile catalyzed by
a (salen)Cr complex, see: a) H.
Lebel, E. N. Jacobsen, Tetrahedron Lett. 1999, 40, 7303; for
reviews on biocatalytic kinetic resolutions, see: b) A. Stein-
reiber, K. Faber, Curr. Opin. Biotechnol. 2001, 12, 552; c) A.
Archelas, R. Furstoss, Curr. Opin. Chem. Biol. 2001, 5, 112.
[10] Examples of kinetic resolution using heterobimetallic com-
plexes: a) S.-y. Tosaki, K. Hara, V. Gnanadesikan, H. Morimoto,
S. Harada, M. Sugita, N. Yamagiwa, S. Matsunaga, M. Shibasaki,
J. Am. Chem. Soc. 2006, 128, 11776; b) H. Mihara, Y. Sohtome, S.
Matsunaga, M. Shibasaki, Chem. Asian J. 2008, 3, 359.
[11] Racemic synthesis of oxetane from ketones through the
sequential addition of in situ generated sulfur ylides was
reported. Oxirane ring-expansion reactions proceeded in
tBuOH at 508C without any promoter, but no reaction
proceeded in THF neither at 458C nor at 508C in the absence
of (S)-1a. K. Okuma, Y. Tanaka, S. Kaji, H. Ohta, J. Org. Chem.
1983, 48, 5133. See also reference [4c] for oxirane ring-expansion
with an ylide using a chiral 2-monosubstituted epoxide.
Received: November 10, 2008
Published online: January 22, 2009
Keywords: asymmetric synthesis · homogeneous catalysis ·
.
[12] Reviews: a) E. M. Vogl, H. Grꢂger, M. Shibasaki, Angew. Chem.
1999, 111, 1672; Angew. Chem. Int. Ed. 1999, 38, 1570; b) M.
Shibasaki, S. Matsunaga, N. Kumagai, Synlett 2008, 1583.
[13] H. Kakei, T. Sone, Y. Sohtome, S. Matsunaga, M. Shibasaki, J.
Am. Chem. Soc. 2007, 129, 13410.
oxetanes · rare-earth metals · ylides
[14] T. Sone, A. Yamaguchi, S. Matsunaga, M. Shibasaki, J. Am.
Chem. Soc. 2008, 130, 10078.
[1] Reviews and accounts on catalytic asymmetric epoxidations:
a) E. N. Jacobsen, M. H. Wu in Comprehensive Asymmetric
Catalysis (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto),
Springer, Berlin, 1999, p. 649; b) T. Katsuki in Catalytic Asym-
metric Synthesis, 2nd ed. (Ed.: I. Ojima), Wiley-VCH, New York,
2000, p. 287; c) C. Bonini, G. Righi, Tetrahedron 2002, 58, 4981;
d) V. K. Aggarwal, C. L. Winn, Acc. Chem. Res. 2004, 37, 611;
e) Q.-H. Xia, H.-Q. Ge, C.-P. Ye, Z.-M. Liu, K.-X. Su, Chem.
Rev. 2005, 105, 1603; f) O. A. Wong, Y. Shi, Chem. Rev. 2008, 108,
3958.
[2] Applications in medicinal chemistry: a) G. Wuitschik, M.
Rogers-Evans, K. Mꢀller, H. Fischer, B. Wagner, F. Schuler, L.
Polonchuk, E. M. Carreira, Angew. Chem. 2006, 118, 7900;
Angew. Chem. Int. Ed. 2006, 45, 7736; b) G. Wuitschik, M.
Rogers-Evans, A. Buckl, M. Bernasconi, M. Mꢁrki, T. Godel, H.
Fischer, B. Wagner, I. Parrilla, F. Schuler, J. Schneider, A. Alker,
W. B. Schweizer, K. Mꢀller, E. M. Carreira, Angew. Chem. 2008,
120, 4588; Angew. Chem. Int. Ed. 2008, 47, 4512.
[3] Selected recent examples in polymer and material science: a) Y.
Permana, K. Nakano, M. Yamashita, D. Watanabe, K. Nozaki,
Chem. Asian J. 2008, 3, 710; b) H. Bouchekif, M. I. Philbin, E.
Colclough, A. J. Amass, Macromolecules 2008, 41, 1989; c) J. V.
Crivello, J. Polym. Sci Part A 2007, 45, 4331.
[4] a) M. M.-C. Lo, G. C. Fu, Tetrahedron 2001, 57, 2621, and
references therein; b) K. Soai, S. Niwa, T. Yamanoi, H. Hikima,
M. Ishizaki, J. Chem. Soc. Chem. Commun. 1986, 1018; c) F.
Bertolini, S. Crotti, V. Di Bussolo, F. Macchia, M. Pineschi, J.
Org. Chem. 2008, ASAP; See also, d) K. Hintzer, B. Koppen-
hoefer, V. Schurig, J. Org. Chem. 1982, 47, 3850.
[15] Steric and electronic modification of achiral phosphine oxides
have beneficial effects in other rare-earth-metal-catalyzed
asymmetric reactions: a) K. Daikai, M. Kamaura, J. Inanaga,
Tetrahedron Lett. 1998, 39, 7321; b) R. Kino, K. Daikai, T.
Kawanami, H. Furuno, J. Inanaga, Org. Biomol. Chem. 2004, 2,
1822; c) J. Tian, N. Yamagiwa, S. Matsunaga, M. Shibasaki,
Angew. Chem. 2002, 114, 3788; Angew. Chem. Int. Ed. 2002, 41,
3636; d) N. Yamagiwa, J. Tian, S. Matsunaga, M. Shibasaki, J.
Am. Chem. Soc. 2005, 127, 3413; e) H. Kakei, R. Tsuji, T.
Ohshima, H. Morimoto, S. Matsunaga, M. Shibasaki, Chem.
Asian J. 2007, 2, 257; f) K. Hara, S.-Y. Park, N. Yamagiwa, S.
Matsunaga, M. Shibasaki, Chem. Asian J. 2008, 3, 1500; g) H.
Morimoto, T. Yoshino, T. Yukawa, G. Lu, S. Matsunaga, M.
Shibasaki, Angew. Chem. 2008, 120, 9265; Angew. Chem. Int. Ed.
2008, 47, 9125.
[16] The krel values in this manuscript were calculated based on the
conversion and ee values of oxetane 4 assuming first-order
kinetic dependence on 2. Kinetic studies are required to
determine accurate krel values. For discussion on the validity of
the calculated values, see: J. M. Keith, J. F. Larrow, E. N.
Jacobsen, Adv. Synth. Catal. 2001, 343, 5. See also Ref. [8]. In
Table 1, the observed ee value of recovered epoxide 2a was
lower than expected on the basis of the ee value of oxetane 4a,
possibly because of side reaction from (R)-2a.
[17] Alternatively, both the epoxidation and the ring-expansion
reaction can be performed simultaneously at 458C. The reaction
of acetophenone (6a) at 458C using 2.2 equivalents of ylide 3
and 20 mol% (S)-LLB from the beginning afforded oxetane 4a
in slightly lower yield (70% yield) than that of entry 1 in Table 2,
but in excellent enantioselectivity (99% ee) after 72 hours. The
[5] P. H. Dussault, T. K. Trullinger, F. Noor-e-Ain, Org. Lett. 2002, 4,
4591.
[6] E. J. Corey, M. Chaykovsky, J. Am. Chem. Soc. 1965, 87, 1353.
Angew. Chem. Int. Ed. 2009, 48, 1677 –1680
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1679

Communications
slightly lower yield and high ee value implied that chiral
gave the intermediate epoxide in 88% ee. The subsequent ring-
expansion reaction, however, gave the oxetane in only 26%
yield and 91% ee.
amplification worked nicely even though the enantiomeric
excess of the intermediate epoxide was a little bit lower than
the sequential process in Table 2.
[18] The use of ethyl ketones is difficult at the moment because of
their low reactivity in the second ring-expansion reaction and
insufficient chiral amplification. For example, the reaction of
propiophenone using 20 mol% of (S)-LLB at room temperature
1680 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1677 –1680