J. Am. Chem. Soc. 1997, 119, 11341-11342
11341
Scheme 1a
Chiral Triarylcarbenium Ions in Asymmetric
Mukaiyama Aldol Additions
Chien-Tien Chen,* Shi-Deh Chao, Kuao-Chung Yen,
Chung-Horng Chen, Iy-Chen Chou, and Sang-Weng Hon
Department of Chemistry, National Taiwan
Normal UniVersity, Taipei, Taiwan 117
ReceiVed March 18, 1997
The Mukaiyama aldol reaction is one of the most versatile
synthetic methods for stereoselective carbon-carbon bond
formation. Asymmetric catalysis of this category utilizing chiral
complexes derived from B, Al, Sn(II), Ti(IV), Cu(II), Pd(II),
and Ln(III) has been explored in the past 10 years with
significant breakthroughs.1 Mukaiyama and co-workers have
documented novel uses of various trityl salts serving as efficient
catalysts in various aldol type transformations,2 highlighting their
potential in asymmetric variations. Nevertheless, several in-
trinsic and pending problems still hinder their practical design
in that context. First, the reacting carbenium ion center is sp2-
hybridized. Placement of the three flanking aryl groups in a
chiral environment is so far impossible due to the extremely
low barrier to racemization of chiral carbenium ions.3 Second,
in sharp contrast to most existing chiral Lewis acids generated
from chiral natural sources (e.g., diols, diamines, amino acids,
and tartrates),1,4 no natural skeleton has been found that is
relevant to the triarylmethyl scaffold. Third, the precise nature
of the catalytic species in these transformations remains elusive
in view of the recent elegant mechanistic study by Bosnich.5
Apparently, development of new types of chiral Lewis acids
with reactive carbenium-based centers are essential in view of
their potential impact on both mechanistic and synthetic utility
aspects. We describe herein our preliminary findings toward
this end.
Platzek and Snatzke have reported the synthesis of C2-
symmetric diol 1, a common skeleton in various anti-inflam-
matory drugs, in scalemic form.6 This resolved (10R,11R)-1,
was utilized as a conceivable trityl ion precursor, whose
enantiomeric purity was determined to be >99% enantiomeric
excess (ee) by HPLC analysis on a chiral support (Chiralcel
OJ). We have so far accessed two different 10,11-dialkyl
(dimethyl and diethyl) substituted C2-symmetric trityl salts. To
convert the alcohol moieties into methyl appendages, the diol-1
was mesylated with methanesulfonyl chloride (MsCl) in CH2-
Cl2 in the presence of Et3N (6 equiv). Reduction of the resultant
dimesylate with LiEt3BH (3 equiv) in anhydrous THF provided
10,11-dimethyldibenzosuberane (2) in essentially quantitative
yield (Scheme 1). In a similar manner, the scalemic diol 1 was
transformed in 96% yield to the corresponding ditosylate by
treatment with TsCl in the presence of Et3N and catalytic
4-(dimethylamino)pyridine (DMAP). Double SN2 displacement
(a) Key: (a) (i) MsCl/Et3N/CH2Cl2, (ii) LiBEt3H/THF, 0 °C f rt;
(b) (i) TsCl/Et3/N/CH2Cl2, cat. DMAP, (ii) Me2CuLi/ether, -10 °C.
Scheme 2
of the ditosylate with (CH3)2CuLi at -10 °C provided the diethyl
analog 3 quantitatively.7 These two dialkyldibenzosuberanes
were readily oxidized at ambient temperature to the respective
ketones, 4 and 5, by KMnO4 (2.5 equiv) in benzene using
dicyclohexano-18-crown-6 as a phase transfer catalyst.8 Both
ketones were obtained in 91% yields.
To probe the stereoelectronic influence of the 5-aryl group
on the structure, reactivity, and selectivity of triarylcarbenium
ions in the aldol process, two representative aryl appendages
(tert-BuC6H4 and 2,6-(MeO)2C6H3) were selected in addition
to the parent phenyl group.9 The requisite trityl alcohols, 6 and
7a-c, were prepared in 84-98% yields by aryllithium addition
to the respective dibenzosuberones.10
Independent treatment of 1-aryldibenzosuberols, 6 and 7a,b
(Ar ) Ph and 4-tert-butylphenyl), with an appropriate acid (HX)
in the presence of a water scavenger (acetic anhydride) allowed
the preparation of the corresponding chiral trityl salts, 9a,b and
10a-c, with three different counter ions (TfO-, ClO4-, and
PF6-).11 In all cases, the reddish orange salts could be obtained
in good yields (80-98%) by gradual addition of cold diethyl
ether into the reaction media at 0 °C to induce crystallization
(Scheme 2). In some instances, these highly moisture- and heat-
sensitive materials can even be obtained in analytically pure
form (i.e., 9b and 10a), allowing for unambiguous determination
of composition. The 2,6-dimethoxy substituted trityl alcohol
7c was not amenable to the Dauben procedure11 (HX/Ac2O)
due to the extreme solubility of the resulting trityl salts in acetic
anhydride. The final targeted hexachloroantimonates, 10d (Ar
) 4-tert-BuC6H4) and 10e (Ar ) 2,6-(MeO)2C6H3), could only
be generated in situ by a Meerwein salt-promoted ionization of
the appropriate trityl methyl ethers (8b and 8c), recently
developed in our laboratories.9 The requisite methyl ether
precursors were formed in 93% (8b) and 82% (8c) yields,
respectively, by the standard Williamson etherification (NaH/
CH3I) of the corresponding alcohols (Scheme 1).12
(1) For leading references for each case, see: (B and Al) (a) Kiyooka,
S.-I.; Kaneko, Y.; Kume, K.-I. Tetrahedron Lett. 1992, 33, 4927. (b) Parmee,
E. R.; Tempkin, O.; Masamune, S. J. Am. Chem. Soc. 1991, 113, 9365. (c)
Corey, E. J.; Cywin, C. L.; Roper, T. D. Tetrahedron Lett. 1992, 33, 6907.
(d) Hattori, K.; Yamamoto, H. Tetrahedron 1994, 50, 2785. (Sn(II)) (e)
Kobayashi, S.; Hayashi, T. J. Org. Chem. 1995, 60, 1098. (Ti(IV)) (f) Singer,
R. A.; Carreira, E. M. J. Am. Chem. Soc. 1995, 117, 12360. (g) Mikami,
K.; Takasaki, T.; Matsukawa, S.; Maruta, M. Synlett 1995, 1057. (h)
Mukaiyama, T.; Inubushi, A.; Suda, S.; Hara, R.; Kobayashi, S. Chem. Lett.
1990, 1015. (Cu(II)) (i) Evans, D. A.; Murry, J. A.; Kozlowski, M. C. J.
Am. Chem. Soc. 1996, 118, 5814. (Pd(II)) (j) Sodeoka, M.; Ohrai, K.;
Shibasaki, M. J. Org. Chem. 1995, 60, 2648. (Ln(III)) (k) Uotsu, K.; Sasai,
H.; Shibasaki, M. Tetrahedron: Asymmetry 1995, 6, 71.
(7) Posner G. H. Org. React. 1975, 22, 253.
(2) (a) Kobayashi, S. Kagaku Kogyo 1989, 42, 245 and references cited
therein. (b) Denmark, S. E.; Chen, C.-T. Tetrahedron Lett. 1994, 35, 4327.
(3) (a) Wallis, E. J. J. Am. Chem. Soc. 1931, 53, 2253. (b) Gomberg,
M.; Gorden, W. E. J. Am. Chem. Soc. 1935, 73, 119. (c) Wallis, E. S.;
Adams, F. H. J. Am. Chem. Soc. 1933, 55, 3338.
(8) Sam, D. J.; Simmons, H. E. J. Am. Chem. Soc. 1972, 94, 4024.
(9) A total of six different aryl groups were surveyed in model studies.
Chen, C.-T.; Chao, S.-D; Yen, K.-C. Manuscript in preparation.
(10) (a) McEwen, W. E.; Janes, A. B.; Knapczyk, J. W.; Kyllingstad,
V. L.; Shiau, W.-I.; Shore, S.; Smith, J. H. J. Am. Chem. Soc. 1978, 100,
7304. (b) Borhan, B.; Wilson, J. A.; Gasch, M. J.; Ko, Y.; Kurth, D. M.;
Kurth, M. J. J. Org. Chem. 1995, 60, 7375.
(4) Blaser, A. P. Chem. ReV. 1992, 92, 935.
(5) Hollis, T. K.; Bosnish, B. J. Am. Chem. Soc. 1995, 117, 4570.
(6) Platzek, J.; Snatzke, G. Tetrahedron 1987, 43, 4947 and references
cited therein.
(11) Dauben, H. J., Jr.; Honnen, L. R.; Harmon, K. M. J. Org. Chem.
1960, 25, 1442.
S0002-7863(97)00900-1 CCC: $14.00 © 1997 American Chemical Society