obtained for glycosyl donors 3r,21 4r,22 and 5r23 having
no anchimerically assisting neighboring groups in the
2-position; again, mainly the β-glycosides of 3A,24 4A,
and 5A were formed (entries 11, 12, and 13). Even the less
reactive fully O-acetylated glucosyl donor 2r19 could be
activated with PhBF2 in the presence of isopropanol as an
acceptor (entry 14) furnishing as expected the β-glucopyr-
anoside 2Aβ (Scheme 3).25
As PhBF2 gave mainly excellent glycosidation results at
temperatures as low as ꢀ78 °C, hence also the less Lewis
acidic Ph2BF was prepared18 and studied as a catalyst
under the same conditions. Entries 15ꢀ17 (Table 1) show
that this compound also fulfills the criteria for a good
catalyst.
furnish better results than PhBF2, for instance in the
formation of 1C or 1H (entries 3 and 9). Preferential 1,2-
trans product formation was also observed for glycosyl
donors 3rꢀ5r (entries 10ꢀ15), thus furnishing mainly
3Cβ,37 3Dβ,21 4Cβ, 4Dβ, 5Cβ, and 5Fβ.
Investigationswith Ph2BF asthe catalyst(entries 16ꢀ20,
formation of 1C, 1D, 3D, 4C, 5C) exhibited very good
glycosidation results. It was particularly pleasing that
Ph2BF provided in the reactions of 1r with D to 1D
(entry 17) and of 3r with D to 3D (entry 18) better results
than PhBF2 (see entries 4 and 11); thus, as expected, the
steric effect of the two phenyl groups supports the con-
certed donor activationꢀacceptor transfer. Hence, the
choice of catalyst is of great importance in these intramo-
lecular acidꢀbase catalyzed glycosidations.
Preliminary experiments were also performed with the
correspondingO-β-D-glucopyranosyltrichloroacetimidate
1β1a as a glycosyl donor. With isopropanol (A) as the
acceptor there is a clear preference for R-product (1Ar)26
formation (entry 18). Almost exclusive β-selectivity was
observed for allyl alcohol (B) as the acceptor affording
mainly 1Br27 (entry 19). Possibly, some steric hindrance in
the transition state leads to lower anomeric selectivity than
observed for 1r. Hence, other catalyst types may be
required for the selective generation of 1,2-cis-glycosides.
The glycosidation results with glycosyl donors 1r, 3r,
4r, and 5r with more demanding carbohydrate acceptors
C,26 D,29 E,30 F,30 G,31 and H32 in the presence of different
catalysts are compiledinTable 2. Reaction of1r with these
acceptors having unprotected hydroxy groups at 6-, 4-, 2-,
and 3-positions with PhBF2 as the catalyst (entries 1, 2,
4ꢀ8) exhibited preferential formation of the β-glucopyr-
anosides of 1C,28 1D,21 1E,33 1F,34 1G,35 and 1H. The re-
actions did not proceed at the same rate (entry 8); the ap-
pearance of some glycosyl fluoride (entries 4, 5, 7) showed
that the catalyst is partly consumed, thus leading to a
decreased reaction rate that also influenced the product
yield. However, no phenyl C-glycoside formation was
In conclusion, PhBF2 and Ph2BF, formally representing
BdC, basically fulfill the requirements for good cata-
lysts for O-glycosyl trichloroacetimidate (and related
systems) activation: Not the catalyst itself, but only the
adduct AꢀBꢀCꢀH with acceptor AꢀH is sufficiently
acidic to activate the glycosyl donor. Hence, the catalyst
carries the acceptor to the donor generating a tempora-
rily H-bonded noncovalent donor leaving groupꢀ
catalystꢀacceptor complex that permits via an intramo-
lecular reaction course proton transfer to the leaving
group and the nucleophilicity increase of the acceptor
facilitating the concomitant glycoside bond formation.
The results support the prevalence of this concerted
SN2-type mechanism between the glycosyl donor and
the AꢀBꢀCꢀH adduct, as preferentially the inversion
product is obtained. Competing reaction courses lead-
ing to R-product and glycosyl fluoride formation were
only effective in some cases. H-bonding as a means for
intramolecular acidꢀbase catalyzed glycosidation re-
sulting in concomitant glycosyl donor and glycosyl
acceptor activation obviously has the potential to be-
come a general and very efficient glycosidation method.
observed.36 TMSOTf or BF3 OEt2 as catalysts did not
3
Acknowledgment. This work was supported by the
University of Konstanz and the Fonds der Chemischen
Industrie. V.K. is particularly grateful for a fellowship
from the Alexander von Humboldt Foundation.
(22) Dere, R. T.; Kumar, A.; Kumar, V.; Zhu, X.; Schmidt, R. R.
J. Org. Chem., submitted.
(23) Knerr, L.; Schmidt, R. R. Synlett 1999, 1802–1804.
(24) Mereyala, H. B.; Reddy, G. V. Tetrahedron 1991, 47, 6435–6438.
(25) Lemieux, R. U.; Hindsgaul, O. Carbohydr. Res. 1980, 82, 195–
208.
(26) Tsvetkov, Y. E.; Klotz, W.; Schmidt, R. R. Liebigs Ann. Chem.
1992, 371–375.
(27) Ronchi, P.; Vignando, S.; Guglieri, S.; Polito, L.; Lay, L. Org.
Supporting Information Available. Experimental details
and NMR spectra of new compounds (1H, 4A, 4C, 4D,
5A, 5C, 5F) and NMR spectra of 1Aꢀ1G, 2A, 3A, 3C, 3D.
This material is available free of charge via the Internet
Biomol. Chem. 2009, 7, 2635–2644.
(28) Eby, R.; Schuerch, C. Carboyhdr. Res. 1974, 34, 79–90.
(29) Garegg, P. J.; Iversen, T.; Oscarson, S. Carbohydr. Res. 1976, 50,
12–11ꢀ14.
(30) Barrett, A. G. M.; Road, R. W.; Barton, D. H. R. J. Chem. Soc.,
Perkin Trans. 1 1980, 2184–2190.
(31) Westerlind, U.; Hagback, R.; Duk, M.; Norberg, T. Carbohydr.
(34) Ito, Y.; Ogawa, T. Carbohydr. Res. 1990, 202, 165–175.
(35) Cassel, S.; Plessis, I.; Wessel, H. P.; Rollin, P. Tetrahedron Lett.
1998, 39, 8097–8100.
Res. 2002, 337, 1517–1522.
(32) Roussel, F.; Knerr, L.; Schmidt, R. R. Eur. J. Org. Chem. 2001,
2067–2073.
(36) Mitchell, T. A.; Bode, J. W. J. Am. Chem. Soc. 2009, 131, 18057–
18059 and references therein.
(33) Nagai, H.; Sasaki, K.; Matsumura, S.; Toshima, K. Carbohydr.
Res. 2005, 340, 337–353.
(37) Vankar, Y. D.; Vankar, P. S.; Behrendt, M. E.; Schmidt, R. R.
Tetrahedron 1991, 47, 9985–9992.
Org. Lett., Vol. 13, No. 14, 2011
3615