Chiral silver phosphate-catalyzed cycloisomeric kinetic resolution of α-allenic alcohols

Article abstract of DOI:10.1021/ja300453u

A kinetic resolution of α-allenic alcohols is realized through chiral silver phosphate-catalyzed cycloisomerization with high stereoselectivity (selectivity factor up to 189) and tolerance of a variety of functional groups. A mechanistic model is proposed to interpret the origin of the high stereoselectivity and broad substrate scope.

Full text of DOI:10.1021/ja300453u

Communication
pubs.acs.org/JACS
Chiral Silver Phosphate-Catalyzed Cycloisomeric Kinetic Resolution
of α-Allenic Alcohols
Yan Wang, Kuan Zheng, and Ran Hong*
Key Laboratory of Synthetic Chemistry of Natural Substances, Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, 345 Lingling Road, Shanghai 200032, China
S
* Supporting Information
Encouraged by this phenomenon, we envisioned that a
ABSTRACT: A kinetic resolution of α-allenic alcohols is
silver−chiral ligand system may preferentially promote the
realized through chiral silver phosphate-catalyzed cyclo-
isomerization with high stereoselectivity (selectivity factor
up to 189) and tolerance of a variety of functional groups.
A mechanistic model is proposed to interpret the origin of
the high stereoselectivity and broad substrate scope.
cyclization of one stereoisomer of an α-allenic alcohol through
cycloisomerization in a general sense (Scheme 1). According to
Scheme 1. Kinetic Resolution of α-Allenic Alcohols
mong transition metals involved in chemical reactions
A
with allenes, gold has shown incredible reactivity in the
past decade.1 A seminal work on gold-catalyzed asymmetric
cycloisomerization of allenes was recently orchestrated by the
Toste group.2 The concept of anion metathesis was particularly
remarkable in utilizing chiral silver phosphate to synthesize the
gold catalyst that exhibits excellent enantioselectivity in the
cyclization of allenic alcohols or amines. However, the silver salt
itself could not promote the transformation with considerable
enantioselectivity. Despite the success of silver-catalyzed
cyclization reactions of allenic alcohols and decades of research3
since the first report in 1974,4 enantioselective silver-catalyzed
cycloisomerization remains elusive.
this strategy, the cycloisomeric kinetic resolution of α-allenic
alcohols (rac-1) would deliver 2,5-dihydrofurans (2)15 and
recovered starting material 1 with excellent enantioselectivities.
Herein we report experimental results to prove the concept.
To begin the study, phenyl-2,3-allen-1-ol (rac-1a) was
chosen as a model substrate for the asymmetric silver-catalyzed
kinetic resolution, since the resulting products 1a and 2-phenyl-
2,5-dihydrofuran (2a) are well-documented in a variety of
synthetic methods in the literature.6−8,9a,13 Although Marshall’s
protocol (AgNO3/CaCO3)12 could promote the cyclization of
1a, when bisoxazoline 4a was added, no product was detected
even after 3 days at room temperature (Table 1, entry 1). The
gold triflate expected after anion metathesis did not give any
conversion in the presence of ligand 4a with the combination of
gold chloride and silver triflate (entry 2). When we turned to
silver complexes, representative chiral ligands such as chiral
bisoxazolines 4a and 4b, salen-type diimines 5 and 6, BINAP
(7), and monodentate phosphoramidite (MonoPhos) 8 were
screened with different silver salts (AgNO3 and AgOTf), all of
which gave low levels of enantioselectivity for dihydrofuran 2a
and recovered allenic alcohol 1a (entries 3−8). Recently,
sporadic examples of chiral silver catalysts that promote novel
reactions have appeared,14 but the catalytic power of chiral
silver phosphate is largely underestimated. To our delight,
when chiral silver phosphate 9a was screened, 1a was recovered
with 66% ee and 2a was isolated with 35% ee (entry 9). Next,
several 3/3′-substituted chiral phosphates were examined.
Biphenyl-substituted phosphate 9c gave a selectivity factor (s)
of 43.4 (entry 11). Although 9d gave a similar level of
selectivity as 9c, further optimization was abandoned because of
the lower conversion and ee (entry 12). The low efficiency of in
situ-generated gold phosphate2 indicates the exclusive role of
silver phosphate (entry 13). A variety of solvents (tetrahy-
Recent progress on the chemistry of α-allenic alcohols
underlines that they can be readily and stereoselectively
converted into versatile functional groups.5 Methods for the
construction of chiral α-allenic alcohols by coupling of
propargylic reagents with carbonyl groups have been limited
to three categories: (1) stoichiometric chiral reagents, such as
chiral tin and borane reagents;6 (2) catalytic chiral Lewis base-
activated silane reagents;7 and (3) transition-metal-catalyzed
asymmetric allenylation.8 These conditions always suffer from
limited substrate scope, complicated reaction systems, and
multiple-step syntheses needed for the preparation of chiral
ligands. Enzymatic kinetic resolution has given excellent
enantioselectivities for the substrates that have been tested,
but the special steric and electronic requirements during
preorganization in the active site of an enzyme leave problems
for late-stage transformations.9 Moreover, the straightforward
asymmetric reduction of allenone seems far from general for
practical use in organic synthesis.10,11 A practical method with
broad compatibility and excellent enantioselectivity for a variety
of functional groups is still needed.
During the biomimetic synthesis of stemoamide, the α-allenic
alcohol derived from the iminium ion-initiated cyclization was
eventually converted into the desired butenolide.12 We applied
a silver-promoted cyclization of the α-allenic alcohol to deplete
one stereoisomer on the basis of conformational strain.
Received: January 15, 2012
Published: February 13, 2012
© 2012 American Chemical Society
4096
dx.doi.org/10.1021/ja300453u | J. Am. Chem. Soc. 2012, 134, 4096−4099

Journal of the American Chemical Society
Communication
Table 1. Ligand Screening for Catalytic Kinetic Resolution
of α-Allenic Alcohol rac-1a
Table 2. Substrate Scope for Aryl-Substituted α-Allenic
Alcohols
a
b
The ee’s were determined by chiral HPLC. 1 equiv of chiral ligand
c
c
was used. 5 mol % ligand was used. PhMe was used as the solvent.
e
a
The selectivity factor (s) given in parentheses was calculated as s =
Isolated yields are given in parentheses; the ee’s were determined by
b
ln[(1 − C)(1 − ee)]/ln[(1 − C)(1 + ee)]. C represents conversion.
chiral HPLC. n.d. = not determined.
drofuran, acetone, dimethoxyethane, acetonitrile, and tert-butyl
methyl ether) were screened, but none of these gave better
results than nonpolar solvents such as dichloromethane, 1,2-
dichloroethane, and toluene [Table S2 in the Supporting
Information (SI)].
Although several recovered starting materials did not reach 90%
ee (1b, 1c, 1d, 1g, and 1h), given by higher s factors, high ee’s
of α-allenic alcohol and 2,5-dihydrofuran could be obtained
when the conversion is slightly above or below 50%, depending
on the desired product. For instance, when the conversion of
1c was only 44%, the allenic alcohol was recovered in 54% yield
with 75% ee, while dihydrofuran 2c was isolated in 43% yield
with 90% ee; for this substrate, s = 97. Moreover, ortho-
substituted 1i and 1j gave impressive selectivities (s = 55), and
the allenic alcohols were recovered in 44−45% yield with 97%
ee, although the substrate bearing an electron-deficient
substituent required a longer reaction time (144 h vs 47 h).
Finally, naphthalene-bearing substrate 1k could be accessed
with high enantiopurity (93% ee, 47% yield, s = 35). Its
corresponding cyclized product 2k was obtained with 83% ee in
53% yield.
With a proper chiral catalyst found in the preliminary
screening (Table 1, entry 11), the catalyst loading and reaction
temperature were also examined. Decreasing the catalyst
loading dramatically affected the reaction rate, and a significant
drop of selectivity was observed (Table S4, entries 2 and 3 vs
entry 1). Decreasing the reaction temperature to −10 °C clearly
gave better selectivity (s increased from 43.4 to 66.3; entry 5 vs
1), but a further decrease to −20 °C gave a sluggish reaction
and low conversion within a comparable time (entry 6).
We next investigated the scope of substrates for cyclo-
isomeric kinetic resolution by the aforementioned optimal
reaction conditions of chiral silver phosphate (S)-dipPAg (9c).
First, a variety of substituents on the aromatic ring were
investigated (Table 2). Both electron-deficient and electron-
rich substituents gave excellent selectivities (s = 17−97).
The uniformly excellent levels of asymmetric induction with
aryl-substituted substrates encouraged us to examine alkyl- and
alkenyl-substituted allenic alcohols (Table 3). To our delight, a
4097
dx.doi.org/10.1021/ja300453u | J. Am. Chem. Soc. 2012, 134, 4096−4099

Journal of the American Chemical Society
Communication
several groups have revealed that metal salt contaminations
have a profound effect in some cases.16 We thus designed
several experiments to ascertain the catalytic species in our
catalytic system. Chiral phosphoric acid alone preferentially
promoted the isomerization to give enone 3a as the sole
product in 17% isolated yield, and the recovered 1a was isolated
in 75% yield with 4.7% ee (s = 1.4; Table 4, entry 3). Silver
Table 3. Substrate Scope for Alkyl- and Alkenyl-Substituted
α-Allenic Alcohols
Table 4. Verification of Chiral Silver Phosphate
a
The mixture of Ag2CO3 and 10 was stirred for 1 h at rt before rac-1a
b
c
was added. Reaction time = 6 h. n.d. = not determined.
carbonate cannot mediate the cyclization (entry 2). Interest-
ingly, when a mixture of phosphoric acid 10 and silver
carbonate in CH2Cl2 was stirred at room temperature for 1 h,
the resulting catalyst system was found to be capable of
catalyzing the kinetic resolution. The excellent enantioselectiv-
ity (98.2% ee) of the recovered starting material 1a was
achieved when the conversion was 61% (s = 19; entry 4). The
selectivity was comparable to that for the experiment in which
10 mol % chiral silver phosphate 9c was used (s = 27; entry 5).
The considerably lower s factor indicated that other species
(e.g., coexisting carbonate anions in the reaction system) may
deteriorate the catalytic efficiency of 9c.
Although there is a lack of conclusive evidence to establish
the coordination mode of the silver−π-allene complex,17 we
tentatively propose an η2-allene complex with silver on the basis
of the structure and reactivity of other π-allene complexes of
group-11 metals and related computational studies.18 To
interpret the excellent enantioselectivity achieved in our system,
a quadrant diagram in which the shaded quadrants represent
the hindered side (Scheme 2) can be used. When the racemic
α-allenic alcohol approaches the metal center, the R enantiomer
of 1 can preferentially coordinate the metal center while the S
enantiomer remains in the reaction mixture, since the bulky R
group in the former would preferentially lie in the unhindered
area. The bonding R isomer subsequently undergoes cyclization
to form a dihydrofuranylsilver species. After the subsequent
protodemetalation and ligand exchange with the allenic alcohol,
the corresponding enantiomerically enriched 2,5-dihydrofuran
is formed to regenerate the catalytic species.
a
Isolated yields are given in parentheses; the ee’s were determined by
b
chiral HPLC (see the SI for details). n.d. = not determined;
compound 2r is extremely volatile. The reaction was run at room
temperature.
c
majority of these substrates underwent the kinetic resolution
with a higher reaction rate and equally high enantioselectivities.
For typical alkyl substituents such as benzyl (1l), n-heptyl
(1m), and cyclohexyl (1n) groups, excellent selectivities were
observed, and the alkylated allenic alcohols were recovered in
good yields (35−50%) with outstanding asymmetric induction
(92.0−99.8% ee). More interestingly, although resolution of 1o
bearing a functionalized alkyl chain gave moderate selectivity (s
= 10), compound 1o was still isolated with 96.8% ee after 48 h.
The enantiomerically enriched 1o would allow further
elaboration to afford some key intermediates in natural product
synthesis. Alkenylated substances 1p−s exhibited even better
selectivity than alkylated ones. For 1p and 1q, both products
were obtained with over 90% ee, and 1q gave the highest
selectivity factor in this study (s = 189). The sterically bulky
tert-butyl group was tolerated, and 1r was obtained with 99.6%
ee, although the reaction was little slow. Particularly note-
worthy is the kinetic resolution of tertiary alcohol 1t. In view of
its steric crowdedness at C(1) and the fact that it is inaccessible
by known enzymatic resolution and asymmetric catalytic
alkenylation, the impressive selectivity (s = 6.8) shows the
unprecedented power of this approach for asymmetric
induction.
On the basis of the proposed model, the substituent at C(2)
would certainly interact with the backbone of the quadrant.
This was indeed the case when a methyl group was introduced
at C(2), as shown in 1u (Scheme 3A), for which s dramatically
dropped to 2.8 with a sluggish reaction. For substrate 1t bearing
methyl and phenyl groups at C(1) (Scheme 3B), raising the
reaction temperature to room temperature pushed the reaction
Chiral phosphoric acid has been rapidly recognized as a
privileged organic catalyst in the past several years since its
renaissance by the Akiyama and Terada groups.15 Very recently,
4098
dx.doi.org/10.1021/ja300453u | J. Am. Chem. Soc. 2012, 134, 4096−4099

Journal of the American Chemical Society
Communication
Scheme 2. Proposed Mechanism for Kinetic Resolution
AUTHOR INFORMATION
Corresponding Author
rhong@sioc.ac.cn
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support for this work was generously provided by the
NSFC (20872157 and 21172236), CAS, and the 973 Program
(2010CB833200 and 2011CB710800). We thank Dr. Rob
Hoen for helpful discussions.
REFERENCES
■
(1) (a) Gorin, D. J.; Toste, F. D. Nature 2007, 446, 395. (b) Corma,
A.; Leyva-Per
́
ez, A.; Sabater, M. J. Chem. Rev. 2011, 111, 1657.
(c) Krause, N.; Winter, N. Chem. Rev. 2011, 111, 1994. (d) Pradal, A.;
Toullec, P. Y.; Michelet, V. Synthesis 2011, 1501.
(2) Hamilton, G. L.; Kang, E. J.; Mba, M.; Toste, F. D. Science 2007,
317, 496.
(3) Silver in Organic Chemistry; Harmata, M., Ed.; Wiley: Hoboken,
NJ, 2010.
Scheme 3. Sterically Bulky Substrates
(4) Leandri, G.; Monti, H.; Bertrand, M. Tetrahedron 1974, 30, 289.
(5) (a) Modern Allene Chemistry; Krause, N., Hashimi, A. S. K., Eds.;
Wiley-VCH: Weinheim, Germany, 2004. (b) Deng, Y.-Q.; Gu, Z.-H.;
Ma, S.-M. Chin. J. Org. Chem. 2006, 26, 1468. (c) Yu, S.; Ma, S. Chem.
Commun. 2011, 47, 5384.
(6) (a) Boldrini, G. P.; Lodi, L.; Tagliavini, E.; Tarasco, C.; Trombini,
C.; Umani-Ronchi, A. J. Org. Chem. 1987, 52, 5447. (b) Corey, E. J.;
Yu, C.-M.; Lee, D.-H. J. Am. Chem. Soc. 1990, 112, 878. (c) Brown, H.
C.; Khire, U. R.; Narla, G. J. Org. Chem. 1995, 60, 8130.
(7) (a) Iseki, K.; Kuroki, Y.; Kobayashi, Y. Tetrahedron: Asymmetry
1998, 9, 2889. (b) Nakajima, M.; Saito, M.; Hashimoto, S.
Tetrahedron: Asymmetry 2002, 13, 2449.
(8) (a) Yu, C.-M.; Yoon, S.-K.; Baek, K.; Lee, J.-Y. Angew. Chem., Int.
Ed. 1998, 37, 2392. (b) Inoue, M.; Nakada, M. Angew. Chem., Int. Ed.
2006, 45, 252. (c) Xia, G.; Yamamoto, H. J. Am. Chem. Soc. 2007, 129,
496.
(9) (a) Burgess, K.; Jennings, L. D. J. Am. Chem. Soc. 1990, 112,
6129. (b) Xu, D.; Li, Z.; Ma, S. Chem.Eur. J. 2002, 8, 5012.
(10) Yu, C.-M.; Kim, C.; Kweon, J. H. Chem. Commun. 2004, 2494.
(11) A reductive SN2′-type reaction to prepare chiral α-allenic
alcohols was reported with two examples. See: Kang, S.-K.; Kim, D.-
Y.; Rho, H.-S.; Yoon, S.-H.; Ho, P.-S. Synth. Commun. 1996, 26, 1485.
(12) Wang, Y.; Zhu, L.; Zhang, Y.; Hong, R. Angew. Chem., Int. Ed.
2011, 50, 2787.
(13) For leading references, see: (a) Ozawa, F.; Kubo, A.; Hayashi,
T. J. Am. Chem. Soc. 1991, 113, 1417. (b) Loiseleur, O.; Meier, P.;
Pfaltz, A. Angew. Chem., Int. Ed. Engl. 1996, 35, 200.
(14) (a) Yoshida, S.; Fukui, K.; Kikuchi, S.; Yamada, T. J. Am. Chem.
Soc. 2010, 132, 4072. (b) Rauniyar, V.; Wang, Z. J.; Burks, H. E.;
Toste, F. D. J. Am. Chem. Soc. 2011, 133, 8486.
(15) (a) Akiyama, T.; Itoh, J.; Yokota, K.; Fuchibe, K. Angew. Chem.,
Int. Ed. 2004, 43, 1566. (b) Uraguchi, D.; Terada, M. J. Am. Chem. Soc.
2004, 126, 5356. (c) Terada, M. Synthesis 2010, 1929.
(16) (a) Hatano, M.; Moriyama, K.; Maki, T.; Ishihara, K. Angew.
Chem., Int. Ed. 2010, 49, 3823. (b) Klussmann, M.; Ratjen, L.;
Hoffmann, S.; Wakchaure, V.; Goddard, R.; List, B. Synlett 2010, 2189.
(c) Terada, M.; Kanomata, K. Synlett 2011, 1255.
(17) (a) Nagendrappa, G.; Joshi, G. C.; Devaprabhakara, D. J.
Organomet. Chem. 1971, 27, 421. (b) Chenier, J. H. B.; Howard, J. A.;
Mile, B. J. Am. Chem. Soc. 1985, 107, 4190.
to reach a high level of conversion (69%) and furnish excellent
enantioselectivity of the recovered starting material (s = 7.2 and
92.3% ee for 1t, vs s = 6.8 in Table 3) (see the SI for the
determination of the absolute configuration). The bent distal
CC bond and the extra C(1)−C(2) σ bond allow the
sterically bulky quaternary carbon center in 1t to move away
from the hindered quadrant region.
In summary, for the first time, an excellent level of
enantioselectivity has been achieved in the kinetic resolution
of α-allenic alcohols bearing a variety of functional groups
through chiral silver phosphate-catalyzed cycloisomerization.
The proposed mechanism shows that a silver−allene complex
and the structure in the chiral phosphate are responsible for the
success of high stereoselectivity. This efficient approach leading
to both enantiomerically enriched α-allenic alcohols and 2,5-
dihydrofurans opens an avenue for future method development
and synthesis of biologically interesting natural products.19
Further investigations of other silver-catalyzed enantioselective
cycloisomerizations involving allenes in the context of natural
product synthesis are currently underway and will be reported
in due course.
ASSOCIATED CONTENT
(18) (a) Brown, T. J.; Sugie, A.; Dickens, M. G.; Widenhoefer, R. A.
■
̀
Organometallics 2010, 29, 4207. (b) Gandon, V.; Lemiere, G.; Hours,
A.; Fensterbank, L.; Malacria, M. Angew. Chem., Int. Ed. 2008, 47,
7534.
(19) The gram-scale syntheses of 1a and 1s and their synthetic
applications are exemplified in the SI.
S
* Supporting Information
Experimental procedures, synthetic applications, and character-
ization of new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
4099
dx.doi.org/10.1021/ja300453u | J. Am. Chem. Soc. 2012, 134, 4096−4099