Enantioselective O-nitroso aldol reaction of silyl enol ethers

Article abstract of DOI:10.1002/anie.200705679

(Chemical Equation Presented) New nucleophiles for the O-nitroso aldol reaction in the form of readily prepared disilanyl enol ethers make this transformation more practicaland more versatile. A silver catalyst with a chiral biaryl phosphite ligand promotes the title reaction with high enantio- and regioselectivity (see scheme). R¹,R² = H, Ar; TMS = trimethylsilyl.

Full text of DOI:10.1002/anie.200705679

Angewandte
Chemie
DOI: 10.1002/anie.200705679
Asymmetric Catalysis
Enantioselective O-Nitroso Aldol Reaction of Silyl Enol Ethers**
Masanori Kawasaki, Pingfan Li, and Hisashi Yamamoto*
Table 1: Effect of the silyl group on the nitroso aldol reaction.[a]
The regio- and stereoselective introduction of a hydroxy
group at the a position to a carbonyl group is an important
transformation in organic synthesis.[1] Recently, we reported
that the asymmetric nitroso aldol (NA) reaction is a powerful
method for this endeavor.[2,3] Although NA reactions of
nitrosobenzene with tin enolates in the presence of silver
catalysts proceeded with excellent regio- and enantioselec-
tivities, this reaction has a drawback: The use of tin enolates is
unfavorable for the environment. Enantioselective O-nitroso
aldol reactions in the presence of an organocatalyst or with
enamines have also been developed. However, the regiose-
lective formation of highly substituted enamines is a problem-
atic issue. As the silyl group is environmentally benign and the
regioselective formation of silyl enol ethers is well known, we
focused on the use of silyl enol ethers as nucleophiles. Herein,
we describe the enantioselective O-NA reaction of silyl enol
ethers (Scheme 1).
Entry
Silyl enol ether
3a/4a
Yield [%][b]
ee [%][c]
3a/4a
1[d]
2
3
4
5
2aa: Si=TMS
2aa: Si=TMS
2ab: Si=TBS
2ac: Si=TIPS
2ad: Si=SiMe2TMS
–
NR[e]
23–44
NR[e]
NR[e]
69
–
44:56
–
–
76/17
–
–
61:39
73/19
[a] Nitrosobenzene (1): 1 equivalent, 2: 1 equivalent. [b] Combined yield
of 3a and 4a. [c] The ee values were determined by HPLC analysis on a
chiral phase. [d] The reaction was conducted without CsF. [e] No
reaction. Difluorphos = [4,4’-bi(2,2-difluoro-1,3-benzodioxol)-5,5’-diyl]-
bis(diphenylphosphane), TBS = tert-butyldimethylsilyl, TIPS = triiso-
propylsilyl, TMS = trimethylsilyl.
with 2ad afforded 3a and 4a in 69% combined yield (Table 1,
entry 5).[5] The disilanyl enol ether was used for further
studies as a result of its high reactivity and stability.[6]
Various achiral ligands were surveyed in an attempt to
create a silver catalyst that would promote the selective
formation of 3a over 4a in the reaction of PhNO (1) with 2ad
(Table 2). A series of reactions with phosphine ligands
Scheme 1. The O-nitroso aldol reaction.
On the basis of the successful NA reaction of tin
enolates,[3c,e] we first attempted the reaction of PhNO (1)
with the silyl enol ether 2aa in the presence of a chiral silver
catalyst (Table 1). In the absence of a fluoride source, the
reaction did not proceed at all (Table 1, entry 1). To improve
the reactivity of the silyl enol ether, CsF was used as an
additive. Under these conditions, 3a and 4a were obtained as
an approximately 1:1 mixture in up to 44% yield (Table 1,
entry 2). When 1 was treated with the bulky silyl enol ethers
2ab and 2ac under the same conditions, the desired products
were not formed (Table 1, entries 3 and 4). However, we
Table 2: Effect of the ligand on N/O selectivity.[a]
Entry
Ligand
3a/4a[b]
Entry
Ligand
3a/4a[b]
1
2
3
4
5
PPh3
62:38
55:45
66:34
84:16
87:13
6
7[c]
8
PPh(OPh)2
PPh(OPh)2
P(OPh)3
P(OPh)3
P(OEt)3
93:7
95:5
90:10
97:3
84:16
P(c-Hex)3
P(2-furyl)3
P(C6F5)3
PPh2(OPh)
À
found that the disilanyl enol ether 2ad, which has a Si Si
bond, was an excellent nucleophile.[4] Thus, the reaction of 1
9[c]
10
[*] Dr. M. Kawasaki, P. Li, Prof. Dr. H. Yamamoto
Department of Chemistry
[a] Nitrosobenzene (1): 1 equivalent, 2: 1 equivalent. [b] The ratio was
determined by 1H NMR spectroscopy of the crude product mixture.
[c] Ligand: 10 mol%.
University of Chicago
5735 South Ellis Ave, Chicago, IL 60637 (USA)
Fax: (+1)773-702-0805
E-mail: yamamoto@uchicago.edu
Homepage: http://yamamotogroup.uchicago.edu
showed clearly that electron-deficient phosphine ligands
gave the O adduct 3a preferentially (Table 2, entries 1–4).
Encouraged by these results, we tested other electron-
deficient ligands and found that PPh(OPh)2 and P(OPh)3
gave 3a with good selectivity (Table 2, entries 5, 6, and 8).
Moreover, when 10 mol% of P(OPh)3 was used, 3a was
obtained almost exclusively (Table 2, entry 9). On the other
[**] Support of this research was provided by National Institutes of
Health (NIH) (Grant No. GM068433-01) and Merck Research
Laboratories. M.K. thanks the Uehara Memorial Foundation for
financial support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 3795 –3797
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3795

Communications
Table 4: The O-nitroso aldol reaction of 1 with various silyl enol ethers.[a]
hand, the selectivity of the reaction decreased significantly
with P(OEt)3 (Table 2, entry 10).
On the basis of these results, which clearly indicate that
electron-deficient ligands provide highly active catalysts with
high O selectivity, we examined the chiral phosphite ligands
5a–g derived from (S)-1,1’-binaphthalene-2,2’-diol ((S)-binol;
Table 3).[7] Generally, the reaction afforded 3a selectively in
Entry
Silyl enol ether
2ad: R=H
Yield [%]
ee [%][b]
1
2
3
85
72
85
95
98
96
2b: R=Me
2c: R=OCH2CH2O
Table 3: Effect of the ligand effect on the O-nitroso aldol reaction.[a]
4
2d
6
97
5
6
2e: n=1
2 f: n=3
41[c]
37[d]
90
64
7
2g
84
92
8
9
2h: Ar=Ph
2i: Ar=2-naphthyl
99
80
79
76
Entry
Ligand
Yield of 3a [%]
ee (3a) [%][b]
3a/4a[c]
1
2
3
4
5a
5b
5c
5d
5e
5 f
63
40
38
59
52
49
49
44
35
18
90
86
28
94
95:5
>99:1
93:7
97:3
>99:1
97:3
[a] Nitrosobenzene (1): 1 equivalent, 2: 1 equivalent. [b] The ee value was
determined by HPLC analysis on a chiral phase. [c] The corresponding
N adduct was obtained in 20% yield. [d] The corresponding N adduct
was obtained in 27% yield.
5
6[d]
7
5g
98:2
corresponding adduct with 97% ee in 66% yield (Table 4,
entry 4). The reaction of the five-membered-ring silyl enol
ether 2e gave the adduct 3e with 90% ee in 41% yield;
however, the ee value of the seven-membered-ring adduct 3 f
was significantly lower (Table 4, entries 5 and 6). The
a-tetralone derivative 2g reacted with 1 to afford the
corresponding adduct with 92% ee in 84% yield (Table 4,
entry 7). Furthermore, 2-substituted cyclohexanone deriva-
tives can be used: The reactions of 2h and 2i afforded the
corresponding adducts, which contain a tert-aminoxy group,
with 79 and 76% ee, respectively (Table 4, entries 8 and 9).
We examined the diastereoselectivity of this reaction with
the chiral silyl enol ether substrates (S)- and (R)-2j, which can
be prepared selectively from cyclohexenone.[8] The reaction
of (S)-2j with nitrosobenzene (1) in the presence of the
catalyst afforded (2R,3R)-3j as a single diastereomer in 91%
yield (Scheme 2). The enantiomeric substrate (R)-2j reacted
with 1 under same conditions to give (2R,3S)-3j with high
diastereoselectivity.[9] These results show that the stereo-
chemical outcome of the O-NA reaction can be controlled by
the catalyst regardless of the configuration of the silyl enol
ether substrate at C3. Thus, our process provides a new
synthetic strategy for controlling 1,2-stereoselectivity in
organic synthesis.
[a] Nitrosobenzene (1): 1 equivalent, 2: 1 equivalent. [b] The ee value was
determined by HPLC analysis on a chiral phase. [c] The ratio was
determined by 1H NMR spectroscopy of the crude product mixture.
[d] CsF was used instead of Me4NF.
the presence of the silver–phosphite catalysts. First, we tested
the simple binol-derived ligands 5a–c. The reactions in the
presence of 5a and 5b gave 3a with moderate enantioselec-
tivity (Table 3, entries 1 and 2). However, when 5c, which has
a bulky OAr moiety, was used as the ligand, the ee value of 3a
decreased significantly (Table 3, entry 3). Encouraged by
these results, we prepared the 3,3’-disubstituted binol-derived
ligands 5d–g. When 5d was used, the ee value of 3a increased
dramatically (Table 3, entry 4). Although the reaction with 5e
gave 3a with 86% ee, the enantioselectivity decreased with 5 f
(Table 3, entries 5 and 6). Finally, we found that 5g was the
optimal ligand: In the presence of this binol derivative, 3a was
formed with 94% ee (Table 3, entry 7).
Next, we optimized the reaction conditions (see the
Supporting Information for details). We found that the
reaction in the presence of AgBF4 (10 mol%) and 5g
afforded 3a with 95% ee in 85% yield when CsF was used
as the fluoride source. Furthermore, when the quantity of
AgBF4 was decreased to 5 mol%, 3a was formed with 95% ee
in 76% yield in the presence of 5g and CsF.
In conclusion, we have developed a silver-catalyzed regio-
and enantioselective O-nitroso aldol reaction of silyl enol
ethers with a new silver–phosphite catalyst system. Disilanyl
enol ethers were shown to be excellent nucleophiles, as they
are not only more stable but also more reactive than
trimethylsilyl enol ethers. Further studies on the use of the
silver–phosphite catalyst and synthetic applications of this
reaction are currently underway in our laboratory.
The generality of this reaction was studied with various
silyl enol ethers (Table 4). PhNO (1) reacted with the six-
membered-ring enol ethers 2ad, 2b, and 2c to provide the
corresponding products 3a–c with 95–98% ee (Table 4,
entries 1–3). Moreover, the tetrahydropyran derivative 2d is
a suitable substrate, the reaction of which with 1 gave the
3796 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3795 –3797

Angewandte
Chemie
Weyl: Methods of Organic Chemistry, Vol. E21 (Eds.: G. Helm-
chen, R. W. Hoffmann, J. Mulzer, E. Schaumann), Georg Thieme,
Stuttgart, 1995, p. 4497; c) P. Zhou, B. C. Chen, F. A. Davis in
Asymmetric Oxidation Reactions (Ed.: T. Katsuki), Oxford
University Press, Oxford, 2001, p. 128.
[2] For general reviews of nitroso chemistry, see: a) H. Yamamoto,
M. Kawasaki, Bull. Chem. Soc. Jpn. 2007, 80, 595; b) H.
Yamamoto, N. Momiyama, Chem. Commun. 2005, 3514; c) P.
Merino, T. Tejero, Angew. Chem. 2004, 116, 3055; Angew. Chem.
Int. Ed. 2004, 43, 2995.
[3] a) N. Momiyama, Y. Yamamoto, H. Yamamoto, J. Am. Chem.
Soc. 2007, 129, 1190; b) N. Momiyama, H. Yamamoto, J. Am.
Chem. Soc. 2005, 127, 1080; c) N. Momiyama, H. Yamamoto, J.
Am. Chem. Soc. 2004, 126, 5360; d) N. Momiyama, H. Torii, S.
Saito, H. Yamamoto, Proc. Natl. Acad. Sci. USA 2004, 101, 5374;
e) N. Momiyama, H. Yamamoto, J. Am. Chem. Soc. 2003, 125,
6038.
Scheme 2. Diastereoselective O-nitroso aldol reaction.
[4] For the synthesis of pentamethyldisilanyl enol ethers, see: a) R. F.
Cunico, Synth. Inorg. Met.-Org. Chem. 1985, 290, 33; b) K. K.
Laali, G. Maas, M. Gimmy, J. Chem. Soc. Perkin Trans. 2 1993,
1387.
[5] Several chiral phosphine ligands, including binap (2,2’-bis(diphe-
nylphosphanyl)-1,1’-binaphthyl), segphos ((4,4’-bi-1,3-benzodiox-
ole)-5,5’-diylbis(diphenylphosphane)), tol-binap (2,2’-bis(di-p-
tolylphosphanyl)-1,1’-binaphthyl), and xylyl-solphos (7,7’-bis(di-
3,5-xylylphosphanyl)-3,3’,4,4’-tetrahydro-4,4’-dimethyl-8,8’-
bi(2H-1,4-benzoxazine)) were tested; however, mixtures of
products were obtained with moderate ee values.
Experimental Section
General procedure: AgBF4 (9.7 mg, 50 mmol) and 5g (28 mg,
50 mmol) were placed in a Schlenk tube, and the mixture was dried
under vacuum for 10 min. Anhydrous THF (2 mL) was then added,
and the mixture was stirred under argon at room temperature for
0.5 h. The clear solution was then cooled to À788C, and PhNO
(53.5 mg, 0.50 mmol) dissolved in anhydrous THF (1 mL) was added
dropwise. The resulting blue solution was stirred for 5 min, and then
the silyl enol ether (0.5 mmol) was added dropwise. CsF (151.9 mg,
1 mmol) dissolved in anhydrous MeOH (1 mL) was added slowly to
the resulting solution at À788C over a period of 16 h. After
completion of the addition, the reaction mixture was diluted with
hexane and AcOEt (3:1), filtered through short pad of silica gel
(elution with hexane/ethyl acetate 3:1), and concentrated under
reduced pressure. The residue was purified by flash column chroma-
tography on silica gel to give the O-nitroso aldol adduct.
[6] Although the pentamethyldisilanyl group is bulky, it makes enol
À
ethers more reactive, because the Si Si bond donates electrons to
the p system by hyperconjugation; for similar examples, see:
a) M. Akakura, M. B. Boxer, H. Yamamoto, ARKIVOC 2007,
337; b) H. Yamamoto, M. B. Boxer, Chimia 2007, 61, 279; c) M. B.
Boxer, H. Yamamoto, J. Am. Chem. Soc. 2007, 129, 2762; d) M. B.
Boxer, H. Yamamoto, J. Am. Chem. Soc. 2006, 128, 48; e) M. B.
Boxer, H. Yamamoto, Org. Lett. 2005, 7, 3127.
[7] For the synthesis of chiral phosphite ligands, see: a) V. E. Albrow,
A. J. Blake, R. Fryatt, C. Wilson, S. Woodward, Eur. J. Org. Chem.
2006, 2549; b) M. Reetz, G. Mehler, Angew. Chem. 2000, 112,
4047; Angew. Chem. Int. Ed. 2000, 39, 3889.
[8] For the preparation of chiral silyl enol ethers, see: R. Shintani, N.
Tokunaga, H. Doi, T. Hayashi, J. Am. Chem. Soc. 2004, 126, 6240.
[9] The configuration of the products was confirmed by comparison
Received: December 11, 2007
Published online: February 26, 2008
Keywords: aldol reaction · asymmetric catalysis ·
.
enantioselectivity · hydroxylation · regioselectivity
À
with literature data after cleavage of the N O bond: E. Vedejs,
[1] For reviews on a hydroxylation, see: a) F. A. Davis, B. C. Chen,
Chem. Rev. 1992, 92, 919; b) F. A. Davis, B. C. Chen in Houben-
D. A. Engler, J. E. Telschow, J. Org. Chem. 1978, 43, 188.
Angew. Chem. Int. Ed. 2008, 47, 3795 –3797
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3797