Probing the mechanism of sulfoxide-catalyzed hemiacetal activation in dehydrative glycosylation

Article abstract of DOI:10.1021/jo050294c

The concept of sulfoxide-covalent catalysis has been established in the context of a versatile hemiacetal hydroxyl activation/substitution reaction for the formation of anomeric linkages. Mechanistic studies focused on the hemiacetal activation process sh

Full text of DOI:10.1021/jo050294c

Probing the Mechanism of Sulfoxide-Catalyzed Hemiacetal
Activation in Dehydrative Glycosylation
Timothy A. Boebel and David Y. Gin*
Department of Chemistry, University of Illinois, Urbana, Illinois 61801
gin@scs.uiuc.edu
Received February 16, 2005
The concept of sulfoxide-covalent catalysis has been established in the context of a versatile
hemiacetal hydroxyl activation/substitution reaction for the formation of anomeric linkages.
Mechanistic studies focused on the hemiacetal activation process show that this transformation
proceeds in the presence of a sulfonic anhydride and an acid scavenger through the intermediacy
of a glycosyl sulfonate species (10), which serves as a resting state prior to the addition of an external
nucleophile and subsequent glycosidic bond formation. Successful determination of the proportion
of 18O incorporation in 10 as a function of its formation, via the technique of dynamic monitoring
of 13C-16/18O isotopic chemical shift perturbations, provides strong evidence that hemiacetal
activation proceeds through initial nucleophilic addition of the hemiacetal hydroxyl to the S(IV)-
center of putative sulfonium sulfonate 6. Further confirmation was obtained through the
1
independent synthesis, structure verification, and H NMR detection of glycosyl oxosulfonium 11
during the sulfoxide-catalyzed conversion of hemiacetal 3 to glycosyl sulfonate 10.
Introduction
and shelf-stable. Despite these favorable characteristics,
the application of sulfoxide catalysis has received only
limited attention. Sulfoxides have been used as ligands
for metal-centered catalysts3 and as additives to effect
rate acceleration through phase-transfer catalysis,4 but
The development of catalytic systems to effect hydroxyl
group functionalization has been an area of active inter-
est, with particular recent emphasis on the identification
of nonmetallic catalysts. The majority of the work in this
field has focused on the discovery of catalysts for hydroxyl
group derivatization derived from either nitrogen or
phosphorus(III) species to enhance the rates of acylation
or alkylation through nucleophilic activation of an ap-
propriate electrophile.1 Potential alternatives to nitrogen-
and phosphorus(III)-based catalyst systems are sulfox-
ides, nonbasic nucleophiles2 which are readily available
(2) Olavi, P.; Virtanen, I.; Koppela, J. Suom. Kemistil. B 1968, 41,
326-330.
(3) For the use of sulfoxide ligands in metal-centered catalysis, see:
(a) James, B. R.; McMillan, R. S. Can. J. Chem. 1977, 55, 3927-3932.
(b) Grennberg, H.; Gogoll, A.; Ba¨ckvall, J.-E. J. Org. Chem. 1991, 56,
5808-5811. (c) Khiar, N.; Ferna´ndez, I.; Alcudia, F. Tetrahedron Lett.
1993, 34, 123-126. (d) Carren˜o, M. C.; Ruano, J. L. G.; Maestro, M.
C.; Cabrejas, L. M. M. Tetrahedron: Asymmetry 1993, 4, 727-734. (e)
Ordon˜ez, M.; Guerrero-de la Rosa, V.; Labastida, V.; Llera, J. M.
Tetrahedron: Asymmetry 1996, 7, 2675-2686. (f) Chelucci, G.; Berta,
D.; Saba, A. Tetrahedron 1997, 53, 3843-3848. (g) Hiroi, K.; Suzuki,
Y. Heterocycles 1997, 46, 77-81. (h) Petra, D. G. I.; Kamer, P. C. J.;
Spek, A. L.; Schoemaker, H. E.; van Leeuwen, P. W. N. M. J. Org.
Chem. 2000, 65, 3010-3017. (i) Hiroi, K.; Suzuki, Y.; Abe, I.; Kawag-
ishi, R. Tetrahedron 2000, 56, 4701-4710. (j) Hiroi, K.; Wantanabe,
K.; Abe, I.; Koseki, M. Tetrahedron Lett. 2001, 42, 7617-7619. (k)
Priego, J.; Manchen˜o, O. G.; Cabrera, S.; Carretero, J. C. J. Org. Chem.
2002, 67, 1346-1353. (l) Wantanabe, K.; Hirasawa, T.; Hiroi, K. Chem.
Pharm. Bull. 2002, 50, 372-379. (m) Rowlands, G. J. Synlett 2003,
236-240. (n) Ekegren, J. K.; Roth, P.; Ka¨llstro¨m, K.; Tarnai, T.;
Andersson, P. G. Org. Biomol. Chem. 2003, 1, 358-366. (o) Hiroi, K.;
Izawa, I.; Takizawa, T.; Kawai, K. Tetrahedron 2004, 60, 2155-2162.
(p) Chen, M. S.; White, M. C. J. Am. Chem. Soc. 2004, 126, 1346-
1347.
(1) (a) Scriven, E. F. V. Chem. Soc. Rev. 1983, 12, 129-161. (b)
Vedejs, E.; Diver, S. T. J. Am. Chem. Soc. 1993, 115, 3358-3359. (c)
Ruble, J. C.; Latham, H. A.; Fu, G. C. J. Am. Chem. Soc. 1997, 119,
1492-1493. (d) Kawabata, T.; Nagato, M.; Takasu, K.; Fuji, K. J. Am.
Chem. Soc. 1997, 119, 3169-3170. (e) Miller, S. J.; Copeland, G. T.;
Papaioannou, N.; Horstmann, T. E.; Ruel, E. M. J. Am. Chem. Soc.
1998, 120, 1629-1630. (f) Vedejs, E.; Daugulis, O. J. Am. Chem. Soc.
1999, 121, 5813-5814. (g) Fu, G. C. Acc. Chem. Res. 2000, 33, 412-
420. (h) Chen, Y.; Tian, S.-K.; Deng, L. J. Am. Chem. Soc. 2000, 122,
9542-9543. (i) Vedejs, E.; Daugulis, O.; MacKay, J. A.; Rozners, E.
Synlett 2001, 1499-1505. (j) Murugan, R.; Scriven, E. F. V. Aldrichim.
Acta 2003, 36, 21-27. (k) France, S.; Guerin, D. J.; Miller, S. J.; Lectka,
T. Chem. Rev. 2003, 103, 2985-3012.
10.1021/jo050294c CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/20/2005
5818
J. Org. Chem. 2005, 70, 5818-5826

Mechanism of Sulfoxide-Catalyzed Hemiacetal Activation
SCHEME 1
SCHEME 2
SCHEME 3
have not been exploited as substoichiometric stand-alone
covalent catalysts.5 Recently, we described an expansion
of the available methods for catalytic hydroxyl group
functionalization through the use of an alkyl sulfoxide
as a stand-alone covalent catalyst for sulfonylation of
hemiacetal hydroxyls in the context of dehydrative gly-
cosylation.6 Herein, we report our further investigations
into the mechanism of hemiacetal activation in this
unique sulfoxide-catalyzed hemiacetal hydroxyl function-
alization.
catalytic cycle prior to the addition of an external
nucleophilic glycosyl acceptor. Therefore, the key to the
successful achievement of sulfoxide turnover is the
identification of a sulfoxide and sulfonic anhydride
combination that provides the appropriate balance of
sulfonate nucleophilicity and glycosyl sulfonate stability
and reactivity.
Evaluation of a number of sulfoxide and sulfonic
anhydride reagent pairs led to the identification of
n-butyl sulfoxide (n-Bu2SO) and benzenesulfonic anhy-
dride ((PhSO2)2O) as a suitable system to effect the
formation of new glycosidic linkages with the catalytic
turnover of substoichiometric sulfoxide. In this simple
procedure, treatment a glycosyl hemiacetal (1, Scheme
2) with (PhSO2)2O (1.2 equiv), 2,4,6-tri-tert-butylpyridine
(TBP, 2.5 equiv), and n-Bu2SO catalyst (0.25 equiv) in
dichloromethane (0.1 M) at 23 °C, followed by addition
of an appropriate nucleophile (1.4 equiv), results in the
formation of new anomeric linkage in the glycoside 5.
This method was found to be amenable to the preparation
of a variety of different glycosidic linkages,6 although only
limited mechanistic investigations into this novel sul-
foxide-catalyzed process have been conducted.
Initial Investigations into the Mode of Catalytic
Hemiacetal Activation. A reasonable mechanism for
the conversion of hemiacetal 1 to the corresponding
glycosyl sulfonate with sulfoxide turnover involves initial
electrophilic activation of n-Bu2SO with (PhSO2)2O to give
the sulfonium sulfonate intermediate (6, Scheme 3).
Subsequent nucleophilic addition of the hemiacetal 1 into
the S(VI)-center of 6 would result in the generation of
glycosyl sulfonate 7 as the resting state with concomitant
regeneration of n-Bu2SO. In this mode of hemiacetal
activation, sulfoxide serves a dual role, first as a Lewis-
basic nucleophilic catalyst to activate (PhSO2)2O, and
then as a leaving group to facilitate the formation of the
glycosyl sulfonate. The function of sulfoxide is not unlike
that of 4-(dimethylamino)pyridine (DMAP) or phosphine
in the catalyzed acylation of nucleophiles.1
Results and Discussion
Some time ago we disclosed a method for the dehydra-
tive glycosylation of nucleophilic acceptors (Nu-H) with
glycosyl hemiacetals (1) employing the reagent combina-
tion of diphenyl sulfoxide (Ph2SO) and trifluoromethane-
sulfonic anhydride (Tf2O) (Scheme 1A) as the dehydrating
agent.7 This approach was found to be general and
applicable to the formation of a wide variety of glycosidic
linkages. In this procedure, the sulfoxide species, al-
though necessary for the reaction to proceed, was not
consumed in the transformation, suggesting the possibil-
ity that the reaction could be extended to employ a
substoichiometric amount of sulfoxide. However, subse-
quent studies demonstrated that when a deficit quantity
of sulfoxide (0.2 equiv) is used in a dehydrative glycosy-
lation with 2,3,4,6-tetra-O-methyl-D-mannopyranose (3,
Scheme 1B), only the product of hemiacetal dimerization
(4) is obtained.8 The formation of the 1,1′-R,R′-disac-
charide 4 in 49% arises from the rapid condensation of a
reactive glycosyl electrophile with unactivated hemiacetal
(3). The failure of sulfoxide turnover in this initial
reagent system can be rationalized by the in situ genera-
tion of a highly reactive glycosyl triflate intermediate,8,9
which cannot serve as a stable resting state in the
(4) Select sulfoxides have been employed as phase-transfer catalysts.
See: (a) Mikolajczyk, M.; Grzejszczak, S.; Zatorski, A.; Montanari, F.;
Cinquini, M. Tetrahedron Lett. 1975, 16, 3757-3760. (b) Furukawa,
N.; Kishimoto, K.; Ogawa, S.; Kawai, T.; Fujihara, H.; Oae, S.
Tetrahedron Lett. 1981, 22, 4409-4412.
(5) For the use of excess chiral sulfoxide for asymmetric allylation
of hydrazones and aldehydes through coordinative acceleration see:
(a) Kobayashi, S.; Ogawa, C.; Konishi, H.; Sugiura, M. J. Am. Chem.
Soc. 2003, 125, 6610-6611. (b) Massa, A.; Molkov, A. V.; Kocˇovsky´,
P.; Scettri, A. Tetrahedron Lett. 2003, 44, 7179-7181 (correction:
Tetrahedron Lett. 2003, 44, 9067). (c) Rowlands, G. J.; Barnes, W. K.
Chem. Commun. 2003, 2712-2713.
In an alternative reaction pathway (Scheme 4), sulfo-
nium sulfonate 6 undergoes nucleophilic attack at S(IV)
by the hemiacetal 1 to provide glycosyl oxosulfonium 8.
Subsequent expulsion/regeneration of the sulfoxide cata-
lyst from oxosulfonium 8 by anomeric nucleophilic sub-
(6) Boebel, T. A.; Gin, D. Y. Angew. Chem., Int. Ed. 2003, 42, 5874-
5877.
(7) Garcia, B. A.; Poole, J. L.; Gin D. Y. J. Am. Chem. Soc. 1997,
119, 7597-7598.
-
stitution with PhSO3 results in the generation of
(8) Garcia, B. A.; Gin, D. Y. J. Am. Chem. Soc. 2000, 122, 4269-
4279.
glycosyl sulfonate 7. This mechanism is notably distinct
from that of traditional nucleophilic catalysis in that the
(9) Crich, D.; Sun, S. J. Am. Chem. Soc. 1997, 119, 11217-11223.
J. Org. Chem, Vol. 70, No. 15, 2005 5819

Boebel and Gin
TABLE 1.
SCHEME 4
SCHEME 5
Conversely, if nucleophilic addition of the [1-18O]-
labeled hemiacetal 1[C1-18O] into the S(IV)-center of
sulfonium sulfonate 6 is operative, a net transfer of 18O
to sulfoxide catalyst as well as the sulfonate anion would
be observed (Scheme 6A). However, it is worth noting
that the degree of 18O-incorporation into sulfoxide at the
conclusion of the reaction can be depleted (Scheme 6B)
depending on the relative rates of consumption versus
regeneration of n-Bu2SO in the catalytic cycle. If the rate
of sulfoxide regeneration (k3, 8[C1-18O] f 7, Scheme 6B)
is faster than that of initial sulfoxide sulfonylation (k1,
n-Bu2SO f 6), a mixture of both labeled and unlabeled
sulfoxide would accumulate during the course of the
reaction, both of which could undergo electrophilic acti-
vation with (PhSO2)2O. Although each successive catalyst
turnover would increase the 18O/16O-isotope proportion
within n-Bu2SO, there could remain a statistical fraction
of the sulfoxide that might undergo zero turnovers and
therefore remain unlabeled.
SCHEME 6
Despite these potential modes of 18O-scrambling
(Schemes 5B and 6B), the hemiacetal 18O-transfer ex-
periment was performed (Table 1). [1-18O]-Labeled 2,3,4,6-
tetra-O-methyl-D-mannopyranose (3[C1-18O], 87% 18O-
incorporation)10 was activated with (PhSO2)2O (1.2 equiv)
and n-Bu2SO catalyst (0.25 equiv) (Table 1, Entry 1) at
23 °C in CH2Cl2, in a reaction procedure mimicking that
established for optimal yield of the glycoside product.6
After 1 h, the reaction was quenched with an excess of
2-propanol to provide 91% isopropyl mannoside 9 (R:â 1.2:
1) and 3% 1,1′-R,R′-linked disaccharide 4.11 Interestingly,
mass spectral analysis of recovered sulfoxide catalyst
demonstrated essentially no 18O-incorporation. While this
initial result suggests a S(VI)-addition pathway (Schemes
3 and 5), one must also consider the possibility of artificial
late-stage 18O-wash-out from newly labeled sulfoxide due
to the g0.2 equiv of excess (PhSO2)2O present at the end
of the process. This could result in sulfonylation of the
newly generated 18O-labeled catalyst, leading to some
degree of 18O-wash-out from the sulfoxide upon addition
of an external unlabeled O-nucleophile or hydrolysis at
the end of the reaction.12 Indeed, when the experiment
sulfoxide catalyst plays a 3-fold role, first as an O-
nucleophile, then as a S(IV)-electrophile, and finally as
a leaving group to allow for catalyst turnover and glycosyl
sulfonate formation.
A straightforward method to gain insight into the
pathway for hemiacetal activation is to perform a reaction
employing [1-18O]-labeled hemiacetal 1[C1-18O] (Schemes
5 and 6). If the S(VI)-addition pathway is operative
(Scheme 5A), nucleophilic addition of hemiacetal (1[C1-18O]
)
into sulfonium sulfonate 6 would result in immediate
transfer of 18O to the sulfonate anion (PhSO218O-) fol-
lowing the addition of the external nucleophile (Nu-H).
However, this mode of hemiacetal activation is not
restricted to exclusive transfer of 18O to sulfonate anion,
especially if sulfonate exchange on sulfonium sulfonate
6 (Scheme 5B) is rapid during the course of the reaction,
which may lead to a small amount of 18O-incorporation
in the sulfoxide catalyst. Nevertheless, even if this were
(10) Prepared through dehydrative glycosylation of H218O (95% 18O-
incorp) to 2,3,4,6-tetra-O-methyl-D-mannopyranose.
(11) Product ratios determined by 1H NMR.
the case, the level of 18O-incorporation into sulfoxide by
1
this mechanism should not exceed that of
/ of the
6
(12) Additional 18O-wash-out can arise as a consequence of 1,1′-R,R′-
linked disaccharide (4) formation, which provides for an even greater
excess of unconsumed sulfonic anhydride at the conclusion of the
process.
hemiacetal starting material based solely on the propor-
tion of 18O/16O of all reactive oxygens present in the
activation process.
5820 J. Org. Chem., Vol. 70, No. 15, 2005

Mechanism of Sulfoxide-Catalyzed Hemiacetal Activation
SCHEME 7
SCHEME 8
is performed under modified conditions with only 1.0
equiv of (PhSO2)2O (Table 1, entry 2), mass spectral
analysis of the regenerated sulfoxide catalyst revealed
65% 18O-incorporation, a significant yet nonquantitative
level of 18O-transfer from hemiacetal to sulfoxide.13 These
apparently disparate results highlight the dramatic
sensitivity of this type of 18O-transfer experiment, based
solely on product analysis, to trace quantities of excess
(PhSO2)2O. Thus, a novel strategy was envisioned to
provide an unambiguous distinction between the two
hemiacetal activation pathways via the dynamic moni-
toring of 18O-incorporation into the glycosyl sulfonate
intermediate 7.
Unambiguous Determination of the Mode of
Hemiacetal Activation. Previous 18O-tracer experi-
ments (Table 1), based solely on analysis of regenerated
sulfoxide catalyst, were complicated by factors such as
sulfoxide turnover, competitive hemiacetal self-condensa-
tion, and excess quantities of (PhSO2)2O. This precluded
the acquisition of accurate 18O-transfer data for mecha-
nistic evaluation under conditions typically used in the
catalytic dehydrative glycosylation process,6 which em-
ploy a small excess of (PhSO2)2O (1.2 equiv) to minimize
self-condensation of the hemiacetal donor.
1
1
at approximately
/ or /
that of the [1-18O]-labeled
3 6
hemiacetal 3[C1-18O] (rapid intramolecular or intermo-
lecular sulfonate exchange, respectively) or decrease to
those values during the reaction (slow intramolecular or
intermolecular sulfonate exchange, respectively).
In contrast, consideration of the pathway of hemiacetal
addition to the S(IV)-center of sulfonium sulfonate 6
would provide a different profile for the 18O/16O isotope
ratio as a function of conversion to glycosyl sulfonate. In
this scenario (Scheme 8), there can be no 18O-transfer to
the anomeric position of the glycosyl sulfonate 10 until
after the first turnover of sulfoxide (i.e., Scheme 8A,
Bu2SO f 6 f 11[C1-18O] f 10). At that point, each
successive sulfonylation of newly generated n-Bu2S18O
with (PhSO2)2O would result in the increase of the 18O/
16O isotope ratio of anionic sulfonate in solution, which
ultimately serves as a nucleophile to effect additional
However, the course of hemiacetal activation, be it
through S(VI)-addition or S(IV)-addition, also has a direct
consequence on the proportion of anomeric 18O-incorpora-
tion in the glycosyl sulfonate as the activation proceeds.
For example, if the activation of 3[C1-18O] proceeds
through hemiacetal addition at the S(VI)-center of sul-
fonium sulfonate 6 (Scheme 7), the initial product formed
would be the 18O-labeled glycosyl sulfonate (10[C1-18O]
)
whose 18O/16O isotope ratio would directly reflect that of
the hemiacetal (3[C1-18O]) starting material. In the (un-
likely) event that no anomeric exchange of sulfonate in
10[C1-18O] occurs, this 18O/16O ratio would remain at this
initial level throughout the course of glycosyl sulfonate
formation. Alternatively, if intramolecular or intermo-
lecular exchange of the anomeric sulfonate oxygens on
mannosyl sulfonate 10[C1-18O] is active, then the observed
18O/16O isotope ratio of 10[C1-18O] would either be constant
sulfoxide turnover (i.e., Scheme 8B, Bu2S18O f 6[18O]
f
11[C1-18O] f 10[C1-18O]). A direct consequence would be
an increase in the 18O/16O isotope ratio in the glycosyl
sulfonate 10 as it is being formed, reaching a maximum
level corresponding to the final statistical 18O/16O ratio
in the sulfonate anion.
It is important to note that for the S(VI)-addition
pathway (Scheme 7), there is no scenario that would
afford an increase in the anomeric 18O/16O ratio of the
glycosyl sulfonate 10 as a function of its extent of
formation, unlike that which is expected in the S(IV)-
addition pathway (Scheme 8).14 This would hold true
irrespective of the relative rates of the various sulfonate
anion exchange processes in the activation process.
(13) Lowering the amount of (PhSO2)2O from 1.2 to 1.0 equiv in
entry 2 also led to the formation of the 1,1′-R,R-linked disaccharide 4
in marginally enhanced yield (5%). This in turn would contribute to
incomplete consumption of (PhSO2)2O and account for some degree of
18O-wash-out from newly generated 18O-sulfoxide This is in contrast
to the near-quantitative 18O-transfer observed when 2,3,4,6-tetra-O-
methyl-D-glucopyranose was employed as the hemiacetal donor using
a deficit quantity (0.95 equiv) of (PhSO2)2O (ref 6). In a further
modification of the glycosylation procedure, treatment of 2 equiv of
(14) Small perturbations arising from C-16/18O equilibrium isotope
effects are likely to be negligible and should not significantly influence
the trends to be observed in the 18O-incorporation into 11 as a function
of its formation. See: Rishavy, M. A.; Cleland, W. W. Can. J. Chem.
1999, 77, 967-977.
3
[C1-18O] (2.2 equiv, 87% 18O-incorporation) with (PhSO2)2O (1.0 equiv)
and n-Bu2SO (0.25 equiv) leads to the formation of the 1,1′-R,R-linked
disaccharide 4 (91%) and recovered sulfoxide catalyst with 84% 18O-
incorporation.
J. Org. Chem, Vol. 70, No. 15, 2005 5821

Boebel and Gin
SCHEME 9
SCHEME 10a
a Reagents and conditions: (a) Dowex-50W, MeOH, 65 °C; (b)
50% NaOH, DMSO; MeI, 0 to 23 °C; (c) 8 N HCl, 60 °C; (d) Ph2SO,
Tf2O, TBP, -78 to -45 °C; 1% H218O in THF, -78 to 22 °C.
SCHEME 11
Measurement of the changes in the 18O/16O isotope
ratio of mannosyl sulfonate 10 during the course of its
formation can be achieved through the use of 13C NMR
spectroscopy as it is known that 13C NMR resonances
experience a small but significant upfield chemical shift
perturbation (∼0.01-0.05 ppm) when directly bound to
18O relative to that of 16O.15 This would provide a means
by which carbons bound to 16O and 18O could be dif-
ferentiated and their relative amounts quantified via 13
NMR integration.16,17
C
To conduct this experiment, the mannosyl sulfonate
10 first was independently synthesized and characterized
(Scheme 9A) so that its presence could be unambiguously
identified in the sulfoxide-catalyzed activation process.
The synthesis of mannosyl sulfonate 10 is relatively
straightforward (Scheme 9A), analogous to our previous
protocol in the generation of glycosyl triflates.8,9,18 Treat-
ment of mannosyl fluoride 1219 with trimethylsilyl ben-
zenesulfonate (PhSO3TMS)20 in CD2Cl2 provided man-
nosyl sulfonate 10 (85% conversion) after 4 h at 0 °C.
The structure of the R-glycosyl sulfonate 10 was verified
by a battery of NMR techniques (1H, 13C, JC1H1, HMQC,
nOe). For comparison (Scheme 9B), a solution of 3,
(PhSO2)2O, and TBP in CD2Cl2 was treated with n-Bu2-
The successful characterization of mannosyl sulfonate
10 therefore set the stage for the tracking of its 18O-
incorporation profile during the hemiacetal activation
process. In the interest of obtaining accurate 13C NMR
integral data with the substrate concentrations typical
for hemiacetal activation, a doubly labeled 13C1-labeled
hemiacetal donor with an anomeric 18OH group was
prepared (Scheme 10). [1-13C,18O]-Labeled 2,3,4,6-tetra-
O-methyl-D-mannopyranose (3[13C1-18O], >99% 13C-incor-
poration, 92% 18O-incorporation) was synthesized in a
four-step sequence starting from [1-13C]-labeled D-man-
nose (13). Fischer glycosylation of methanol with 13,
followed by permethylation, anomeric deprotection, and
C1-hydroxy exchange with H218O (95% 18O-incorporation)
1
SO at 20 °C. After 35 min, H NMR analysis revealed
1
>95% conversion of hemiacetal 3 to a species with H
NMR resonances identical to that of the previously
synthesized R-mannosyl sulfonate 10.
provides 3[13C1-18O]
.
(15) See, inter alia: (a) Jameson, C. J. J. Chem. Phys. 1977, 66,
4983-4988. (b) Risley, J. M.; Van Etten, R. L. J. Am. Chem. Soc. 1979,
101, 252-253. (c) Darensbourg, D. J. J. Organomet. Chem. 1979, 174,
C70-C76. (d) Vederas, J. C. J. Am. Chem. Soc. 1980, 102, 374-376.
(e) Vederas, J. C. Can. J. Chem. 1982, 60, 1637-1642. (f) Risley, J.
M.; Van Etten, R. L. In NMR: Basic Principles and Progress; Gunther,
H., Ed.; Springer-Verlag: Berlin, Germany, 1990; Vol. 22, pp 81-168.
(16) Measurement of the changes in 16O/18O-incorporation over time
by 13C NMR has been used to study the mechanism of sulfonate
exchange. See: (a) Fujio, M.; Sanematsu, F.; Tsuno, Y.; Sawada, M.;
Takai, Y. Tetrahedron Lett. 1988, 29, 93-96. (b) Tsuji, Y.; Kim, S. H.;
Saeki, Y.; Yatsugi, K.; Fujio, M.; Tsuno, Y. Tetrahedron Lett. 1995,
36, 1465-1468. (c) Tsuji, Y.; Toteva, M. M.; Amyes, T. L.; Richard, J.
P. Org. Lett. 2004, 6, 3633-3636.
(17) This method has also been used to monitor positional isotope
exchange in carbohydrates. See: (a) Mega, T. L.; Cortes, S.; Van Etten,
R. L. J. Org. Chem. 1990, 55, 522-528. (b) Barlow, J. N.; Girvin, M.
E.; Blanchard, J. S. J. Am. Chem. Soc. 1999, 121, 6968-6969.
(18) Callam, C. S.; Gadikota, R. R.; Krein, D. M.; Lowary, T. L. J.
Am. Chem. Soc. 2003, 125, 13112-13119.
(19) Prepared from the treatment of 2,3,4,6-tetra-O-methyl-D-man-
nopyranose with excess hydrogen fluoride-pyridine complex, following
a known procedure for the conversion of glycosyl hemiacetals to glycosyl
fluorides. See: Szarek, W. A.; Grynkiewicz, G.; Doboszewski, B.; Hay,
G. W. Chem. Lett. 1984, 1751-1754.
(20) Prepared from the treatment of benzene with trimethylsilyl
chlorosulfonate following a known procedure. See: Hofman, K.; Sim-
chen, G. Liebigs Ann. Chem. 1982, 282-297.
Sulfoxide-catalyzed hemiacetal activation was then
performed on [1-13C,18O]-labeled mannose hemiacetal
3[13C1-18O] (Scheme 11). A solution of 3[13C1-18O], (PhSO2)2O,
and TBP in CD2Cl2 was treated with n-Bu2SO at 20 °C
and monitored by 13C NMR.21-23 Examination of the
13C1 resonance of the mannosyl sulfonate 10 (δC1 99.75)24
revealed the emergence of a minor 13C resonance (δC1
(21) In the interest of maximizing signal to noise, individual spectra
were acquired without waiting for complete relaxation between pulses.
However, it was anticipated that this would not present a problem in
the determination of accurate 13C-16O/13C-18O ratios as the substitu-
tion of an 16O with an 18O (neither of which is NMR active) should not
result in a significant change in the rate of longitudinal relaxation of
the attached 13C atom. Indeed, measurement of the longitudinal
relaxation time (T1) of both [1-13C,16O]-labeled mannosyl sulfonate
(10[13C1],
Scheme 11) and [1-13C,18O]-labeled mannosyl sulfonate
(10[13C1-18O]) at the conclusion of hemiacetal 3[13C1-18O] activation by
the inversion recovery method demonstrated T1 values within 5% of
one another (1.64 ( 0.03 and 1.72 ( 0.04 s, respectively). For an
excellent review of 13C-relaxation mechanisms, see: Levy, G. C.;
Lichter, R. L.; Nelson, G. L. Introduction, Theory and Methods. In
Carbon-13 Nuclear Magnetic Resonance Spectroscopy, 2nd ed.; John
Wiley & Sons: New York, 1980; pp 1-28.
5822 J. Org. Chem., Vol. 70, No. 15, 2005

Mechanism of Sulfoxide-Catalyzed Hemiacetal Activation
CHART 1
SCHEME 12
99.71) during the course of the reaction, characteristic
of the upfield chemical shift perturbation of 13C-18O vs
13C-16O. Careful integration of the relative proportion
of these peaks as a function of time revealed an increase
in the percent 18O-incorporation of the glycosyl sulfonate
10 during the course of its formation (Chart 1) approach-
ing a limit at ∼12%. In comparison with the predicted
trends for both the S(VI)- versus S(IV)-pathways (Schemes
7 vs 8), this result clearly reinforces initial hemiacetal
nucleophilic addition at S(IV) of 6 (Schemes 4 and 8) as
the activation pathway, even when probed under typical
reaction conditions involving excess (PhSO2)2O dehydrat-
ing agent.25 Indeed, hemiacetal addition into the S(IV)-
center of 6 is consistent with the previously demonstrated
proclivity for sulfonium sulfonates to accept oxygen
nucleophiles at S(IV).26
Detection and Analysis of Mannosyl Oxosulfo-
nium 11. The evidence in support of hemiacetal activa-
tion through the S(IV)-pathway implies that the glycosyl
oxosulfonium species 11 (Scheme 8) must be a transient
intermediate in the hemiacetal activation process. If this
species is of sufficient stability to be detected, one would
expect it to be initially formed via the S(IV)-addition
pathway in a substoichiometric quantity and eventually
diminish as the conversion to the glycosyl sulfonate
concludes.
purpose of initial identification (Scheme 12A), was sig-
nificantly more challenging than its anomeric sulfonate
counterpart 10. Using an adaptation of a procedure
previously developed within our group,8 mannosyl fluo-
ride 1219 was treated with excess TMSOTf (6 equiv) in
CD2Cl2 at -78 °C to provide the glycosyl triflate inter-
mediate. After 30 min at -45 °C, the reaction was cooled
to -78 °C whereupon excess n-Bu2SO (7.0 equiv) was
added to give, after warming to 23 °C, the putative
1
mannosyl oxosulfonium 11 (66% conversion). H NMR
characterization of 11 revealed the presence of an ano-
meric doublet at δ 5.50 ppm (3JHH ) 4.2 Hz) which
exhibited nOe enhancement upon irradiation of the C(2)-
OCH3 resonance (1.3%). The anomeric carbon at δ 108.27
was unambiguously identified by correlation with the
anomeric proton and found to exhibit a 1JCH of 178.5 Hz,
indicative of an equatorial anomeric proton. Although
these data are consistent with the structure of 11, the
paucity of data for characterization of anomeric oxosul-
fonium intermediates in the literature necessitated the
further structure verification of 11 by confirming its
anomeric C-O connectivity (Scheme 12B). Following the
previously described procedure for the synthesis of 11,
[1-13C]-labeled mannosyl fluoride 12[13C1] was sequentially
treated with TMSOTf and n-Bu2S16/18O (49% 18O-incor-
poration).27 The resulting 13C NMR of the product (50%
conversion) showed the presence of two anomeric carbon
resonances, at δ 108.33 and δ 108.29, with a ∆δ of 0.04
ppm. The presence of two C1 resonances is indicative of
18O-induced isotope chemical shift perturbation by the
partially labeled sulfoxide, providing clear evidence that
Due to its enhanced reactivity, independent prepara-
tion of the putative mannosyl oxosulfonium 11, for the
(22) Individual spectra were acquired with an acquisition time of
2.0 s utilizing a 72° pulse. The optimum pulse angle was determined
using the Ernst equation, cos(R) ) e∧(-tr/T1), where R is the pulse
angle and tr is the relaxation time between pulses. See: Ernst, R. R.;
Anderson, W. A. Rev. Sci. Instr. 1966, 37, 93-102.
(23) Subsequent averaging of every 40 spectra provided one data
point every 80 s. Baseline resolution between the closely spaced 13C-
16O and 13C-18O resonances was achieved by the application of a
resolution enhancement of -0.8 Hz and a Gaussian apodization of 0.5
s. The application of resolution enhancement in FT NMR has been
shown to have only a limited effect on the accuracy of integration.
See: Kupka, T.; Dzie¸gielewski, J. O. Magn. Reson. Chem. 1988, 26,
353-357.
1
the species observed by H and 13C NMR indeed incor-
porates a C-O linkage between the anomeric carbon and
the sulfoxide oxygen, consistent with the structure of
mannosyl oxosulfonium 11.
With the successful preparation of both the mannosyl
oxosulfonium 11 and the glycosyl sulfonate 10, the entire
hemiacetal activation process was then monitored by 1H
NMR. Hemiacetal activation was performed on 2,3,4,6-
(24) δC1 is shifted slightly downfield (0.17 ppm) from the corre-
sponding resonance measured in the independent synthesis and
characterization of 10. This is likely the result of differing ionic
strengths for the two reaction solutions, as the synthesis of 10 from
mannosyl fluoride 12 (Scheme 9A) should yield no ionic products
whereas the conversion of hemiacetal 3[13C1-18O] to 10[13C1-18O] (Scheme
11) would also generate pyridinium sulfonate as a byproduct.
(25) It should be noted that the advent of turnover in sulfoxide
during the reaction precludes complete exclusion of the S(VI)-addition
pathway (Scheme 3) from these data alone.
(27) Partially labeled n-Bu2S18O was prepared by sequential treat-
ment of n-Bu2SO with 0.5 equiv of Tf2O at -78 °C in CH2Cl2 followed
by the addition of excess H218O (95% 18O-incorp) in THF. The percent
18O-incorporation was determined by using field ionization mass
spectrometry.
(26) Tidwell, T. T. Synthesis 1990, 857-870.
J. Org. Chem, Vol. 70, No. 15, 2005 5823

Boebel and Gin
SCHEME 13
SCHEME 14
SCHEME 15
a Reagents and conditions: (PhSO2)2O (1.2 equiv), TBP (2.5
equiv), CD2Cl2 (0.05 M); n-Bu2SO (0.24 equiv).
tetra-O-methyl-D-mannopyranose (3) in CD2Cl2 (Scheme
13). A solution of 3 (13:1 R:â), (PhSO2)2O (1.2 equiv), and
TBP (2.5 equiv) in CD2Cl2 was treated with n-Bu2SO
(0.24 equiv) at 20 °C, leading to complete consumption
of both the R-hemiacetal 3R (δ 5.22) and â-hemiacetal
3â (δ 4.61) within 30 min to give >95% of R-mannosyl
sulfonate 10 (δH1 5.91).28,29 A second species, with a
characteristic anomeric resonance at δ 5.48 (3JHH ) 3.9
Hz), was assigned as the transient mannosyl oxosulfo-
nium 11, which exhibited an initial rapid increase in the
formation followed by gradual decline over the period of
30 min. It should be noted that this δH1 is shifted slightly
upfield (0.02 ppm) from the corresponding resonance
measured in the independent synthesis and characteriza-
tion of 11 (Scheme 12A), although this is likely due to
the variability of oxosulfonium counterion in each of the
experiments (i.e., TfO- in Scheme 12, PhSO3- in Scheme
13). The observation of this reaction profile of the
transient R-mannosyl oxosulfonium 11 during hemiacetal
activation is again consistent with hemiacetal activation
through the S(IV)-addition pathway (Scheme 4).30,31
1H NMR Analysis of Glycosylation of 2-Propanol.
Following the 1H NMR tracking of the hemiacetal activa-
(δ 4.91 and δ 4.41, respectively) over time indicates that
the two products are formed at comparable rates and
maintain a steady ratio throughout the reaction.
The observed kinetic product ratio of 9 (1:1 R:â)32 in
the glycosylation of 2-propanol with the single R-man-
nosyl sulfonate 10 (Scheme 14) highlights the differing
mechanistic pathways by which the glycosides 9 can be
formed (Scheme 15). If glycosylation is an SN2 process
(ANDN), then the R- and â-sulfonates (10) (or R- and
â-oxosulfoniums (11)) must be in rapid equilibration,
leading to the formation of R-mannoside 9R at a rate
comparable to that of the â-mannoside 9â through
anomeric displacement with inversion. On the other
hand, if glycosylation is a dissociative process (DN + AN
or DN*AN),33 then the active glycosylating species would
be glycosyl oxocarbenium 14, formed through dissociation
of the anomeric leaving group. Anomeric substitution to
form the R- and â-products then occurs at similar rates
on 14 to give the observed product ratio (Scheme 15).
1
tion process, H NMR analysis of the subsequent glyco-
sylation stage of the reaction was then monitored. The
newly generated R-mannosyl sulfonate 10 (vide supra)
was treated with an excess of 2-propanol (4 equiv), and
the process of glycosylation was followed via 1H NMR at
1
20 °C (Scheme 14). Examination of the H NMR data
revealed near complete conversion of glycosyl sulfonate
10 to isopropyl 2,3,4,6-tetra-O-methyl-D-mannopyrano-
side (9) as a 1:1 R:â mixture of anomers after 100 min.
Inspection of the R- and â-mannoside (9Râ) resonances
(32) A solution of 9 (1:5.7 R:â) in CD2Cl2 was treated with (PhSO2)2O
(0.18 equiv), n-Bu2SO (0.25 equiv), PhSO3H (2.10 equiv), TBP
(2.53 equiv), and 2-propanol (2.03 equiv). After 100 min, 1H NMR
showed no change in the anomeric ratio of 9, indicating that the
glycoside product did not equilibrate under the reaction conditions and
that the observed 1: 1 R:â mixture of glycosidic products in the
glycosylation of 2-propanol to mannose hemiacetal 3 was under kinetic
control.
(28) 1H NMR spectra were acquired at a rate of one spectrum every
10 s.
(29) The 1H NMR resonances at δ 5.32 and δ 5.34 are due to the
presence of residual CDHCl2 and CH2Cl2.
(30) The species corresponding to δH1 5.14 ppm, formed slowly over
the course of the reaction to the extent of 2%, was determined to be
1,1′-R,R-linked disaccharide 4 (Scheme 1).
(31) The minor transient resonance at δ 5.40 may correspond to the
anomeric proton of the â-anomer of mannosyl sulfonate 11, although
its rigorous structural verification has yet to be achieved.
(33) The intermediacy of an oxocarbenium species is consistent
with previous observations. See: (a) Stubbs, J. M.; Marx, D. J. Am.
Chem. Soc. 2003, 125, 10960-10962. (b) Abdel-Rahman, A. A.-H.;
Jonke, S.; El Ashry, E. S. H.; Schmidt, R. R. Angew. Chem., Int. Ed.
2002, 41, 2972-2974. (c) Crich, D.; Chandrasekera, N. S. Angew.
Chem., Int. Ed. 2004, 43, 5386-5389.
5824 J. Org. Chem., Vol. 70, No. 15, 2005

Mechanism of Sulfoxide-Catalyzed Hemiacetal Activation
toluene in vacuo and subsequently handled and stored under
a nitrogen atmosphere. Trimethylsilyl trifluoromethane-
sulfonate was distilled at reduced pressure prior to use.
Infrared (IR) spectra were obtrained using a Perkin-Elmer
Spectrum BX spectrophotometer referenced to a polystyrene
standard. Data are presented as the frequency of absorption
(cm-1) and intensity of absorption (s ) strong, m ) medium,
w ) weak). Proton and carbon-13 nuclear magnetic resonance
(1H NMR or 13C NMR) spectra were recorded on a Varian 500
or a Varian Inova 500 NMR spectrometer; chemical shifts are
expressed in parts per million (δ scale) downfield from tet-
ramethylsilane and are referenced to the residual protium in
However, the data present in Scheme 14 are by them-
selves insufficient to distinguish between the distinct
glycosylation pathways. Future detailed studies on the
glycosylation event of this reaction will examine the
hybridization of the anomeric carbon in the rate deter-
mining transition state for the formation of both R- and
â-mannoside (15) products through the measurement of
kinetic isotope effects.33c,34
Conclusion
the NMR solvent (CHCl3: δ 7.26 for 1H NMR, δ 77.00 for 13
C
The concept of sulfoxide-covalent catalysis has been
established in the context of a versatile hemiacetal
hydroxyl activation/substitution reaction for the forma-
tion of anomeric linkages. Mechanistic studies focused
on the hemiacetal activation process show that this
transformation proceeds through the intermediacy of a
glycosyl sulfonate species (10), which serves as a resting
state prior to the addition of an external nucleophile and
subsequent glycosidic bond formation. Successful deter-
mination of the proportion of 18O-incorporation in 10 as
a function of its formation, via the technique of dynamic
monitoring of 13C-16/18O isotopic chemical shift perturba-
tions, provides strong evidence that hemiacetal activation
proceeds through initial nucleophilic addition of the
hemiacetal hydroxyl to the S(IV)-center of putative
sulfonium sulfonate 6. Further confirmation was ob-
tained through the independent synthesis, structure
verification, and 1H NMR detection of glycosyl oxosulfo-
nium 11 during the sulfoxide-catalyzed conversion of
hemiacetal 3 to glycosyl sulfonate 10. Future studies will
focus on the mechanism of anomeric bond formation
following the addition of an external nucleophile to
glycosyl sulfonate 10. The results of these studies should
allow for the expansion of the scope of sulfoxide-covalent
catalysis to more general hydroxyl functionalization
reactions.
NMR).
Benzenesulfonyl 2,3,4,6-Tetra-O-methyl-r-D-mannopy-
ranose (10). To a solution of 2,3,4,6-tetra-O-methyl-R-D-
mannopyranosyl fluoride (12)19 (15.7 mg, 0.066 mmol, 1.0
equiv) in CD2Cl2 (660 µL) at 0 °C was added trimethylsilyl
benzenesulfonate20 (41 µL, 0.20 mmol, 3.0 equiv). After stirring
for 5 h at -2 °C, the reaction was warmed to 22 °C and
transferred via cannula to a dry 5 mm NMR tube at 22 °C
fitted with a rubber septum whereupon it was placed in the
NMR probe at 20 °C. The mannosyl sulfonate 10 (85%
conversion) was then characterized by 1H and 13C NMR
(Scheme 9).
Bis(n-butyl)sulfonium 2,3,4,6-Tetra-O-methyl-r-D-man-
nopyranoside (11). To a solution of 2,3,4,6-tetra-O-methyl-
R-D-mannopyranosyl fluoride (12)19 (15.2 mg, 0.058 mmol, 1.0
equiv) in CD2Cl2 (585 µL) at -78 °C was added trimethylsilyl
trifluoromethanesulfonate (63 µL, 0.35 mmol, 6.0 equiv). After
2 min, the solution was warmed to -45 °C for 23 min, and
then cooled again to -78 °C at which point n-butyl sulfoxide
(80 µL, 0.41 mmol, 7.0 equiv) was added via syringe. The
reaction was warmed to 21 °C and transferred via cannula to
a dry 5 mm NMR tube at 21 °C fitted with a rubber septum
whereupon it was placed in the NMR probe at 20 °C. The
mannosyl oxosulfonium 11 (66% conversion) was then char-
1
acterized by H NMR (Scheme 12).
Bis(n-butyl)sulfonium 2,3,4,6-Tetra-O-methyl-r-D-[1-
13C, 16/18O]-mannopyranoside (11[13C1-18O]). To a solution of
2,3,4,6-tetra-O-methyl-R-D-[1-13C]-mannopyranosyl fluoride
(12[13C1])
19 (16.3 mg, 0.068 mmol, 1.0 equiv) in CD2Cl2 (680 µL)
at -78 °C was added trimethylsilyl trifluoromethanesulfonate
(75 µL, 0.41 mmol, 6.0 equiv). After 3 min, the solution was
warmed to -45 °C for 30 min, and then cooled again to -78
°C at which point n-butyl sulfoxide (49% 18O-incorp) (94 µL,
0.48 mmol, 7.0 equiv) was added via syringe. The reaction was
warmed to 21 °C, and transferred via cannula to a dry 5 mm
NMR tube at 22 °C fitted with a rubber septum whereupon it
was placed in the NMR probe at 20 °C. The mannosyl
oxosulfonium 11[13C1-18O] (53% conversion) was then character-
ized by 13C NMR (Scheme 12).
Experimental Section
General Procedures. All reactions were performed in
flame-dried modified Schlenk (Kjeldahl shape) flasks fitted
with a glass stopper under a positive pressure of argon unless
otherwise noted. NMR tubes were dried over calcium sulfate
at high vacuum. Dichloromethane, acetonitrile, and toluene
were purified by passage through two packed columns of
neutral alumina under an argon atmosphere. d2-Dichlo-
romethane was dried by vacuum transfer from calcium hy-
dride. 2-Propanol was distilled from calcium hydride at 760
Torr onto magnesium from which it was freshly distilled at
760 Torr as needed. Benzenesulfonic anhydride was recrystal-
lized35 from 10:1 diethyl ether:benzene, dried by azeotropic
removal of water with acetonitrile followed by dichloromethane
in vacuo, and subsequently handled and stored under a
nitrogen atmosphere. n-Butyl sulfoxide was dried over 4 Å
molecular sieves at 50 °C, stored under argon, and melted
under argon prior to use. 18O-Labeled n-butyl sulfoxide was
dried by azeotropic removal of water with toluene in vacuo,
and then melted under argon prior to use. 2,4,6-Tri-tert-
butylpyridine was dried by azeotropic removal of water with
Activation of 2,3,4,6-Tetra-O-methyl-r-D-mannopyra-
nose (3) To Form Benzenesulfonyl 2,3,4,6-tetra-O-meth-
yl-r-D-mannopyranose (10). 2,3,4,6-Tetra-O-methyl-R-D-
mannopyranoside (3) (7.5 mg, 0.032 mmol, 1.0 equiv) in a 5
mm NMR tube was dried by azeotropic removal of water with
CH3CN (250 µL, ×2) and then CH2Cl2 (250 µL) in vacuo. To
the dried mannose hemiacetal was added 2,4,6-tri-tert-bu-
tylpyridine (19.6 mg, 0.079 mmol, 2.5 equiv) and benzene-
sulfonic anhydride (11.5 mg, 0.038 mmol, 1.2 equiv) whereupon
the NMR tube was fitted with a rubber septum. The resulting
solid mixture was dissolved in CD2Cl2 (615 µL) and inserted
into the NMR probe equilibrated at 20 °C. After acquisition
1
of an H NMR spectrum, the reaction was removed from the
(34) For kinetic isotope effects associated with glycosidic bond
formation/hydrolysis, see: Schramm, V. L. In Enzyme Mechanism from
Isotope Effects; Cook, P. F., Ed.; CRC Press: Ann Arbor, MI, 1991; pp
367-388. (b) Bennet, A. J.; Sinnot, M. L. J. Am. Chem. Soc. 1986, 108,
7287-7294. (c) Lee, J. K.; Bain, A. D.; Berti, P. J. J. Am. Chem. Soc.
2004, 126, 3769-3776.
NMR probe and a solution of n-butyl sulfoxide (1.5 µL, 0.0077
mmol, 0.24 equiv) in CD2Cl2 (20 µL) was added via syringe.
The reaction was briefly agitated (∼5 s using a vortex
apparatus, holding the NMR tube at a 45° angle) and then
1
replaced in the NMR probe. H NMR spectra were acquired
(35) Field, L. J. Am. Chem. Soc. 1952, 74, 394-398.
at a rate of once every 10 s. Consumption of 3 was complete
J. Org. Chem, Vol. 70, No. 15, 2005 5825

Boebel and Gin
within 30 min to give benzenesulfonyl 2,3,4,6-tetra-O-methyl-
R-D-mannopyranoside (10) (>95% conversion) (Scheme 13).
Glycosylation with Benzenesulfonyl 2,3,4,6-Tetra-O-
methyl-r-D-mannopyranose (10) To Form Isopropyl
2,3,4,6-Tetra-O-methyl-D-mannopyranoside (9). 2-Pro-
panol (9.8 µL, 0.13 mmol, 4.0 equiv) was added to a solution
of benzenesulfonyl 2,3,4,6-tetra-O-methyl-R-D-mannopyrano-
side (10) (0.032 mmol, 1.0 equiv) in CD2Cl2 (635 µL) at 20 °C
in a 5 mm NMR tube (vide supra). The reaction was briefly
agitated (∼5 s using a vortex apparatus, holding the NMR tube
Acknowledgment. This research was supported by
the NIH (GM58833) and the Camille and Henry Dreyfus
Foundation. A Pharmacia Graduate Fellowship awarded
to T.A.B. is gratefully acknowledged. NMR spectra were
obtained in the Varian Oxford Instrument Center for
Excellence in NMR Laboratory, funded in part by the
W. M. Keck Foundation, the National Institutes of
Health (Grant PHS 1 S10 RR10444), and the National
Science Foundation (Grant NSF CHE 96-10502). The
assistance of Drs. Vera Mainz, Feng Lin, and Paul
Molitor of the VOICE NMR Spectroscopy Lab at the
University of Illinois is recognized.
1
at a 45° angle) and then placed in the NMR probe. H NMR
spectra were acquired at a rate of once every 20 s. Consump-
tion of 10 was complete within 100 min to give isopropyl
2,3,4,6-tetra-O-methyl-D-mannopyranoside (9) as a 1:1 mixture
of anomers (Scheme 14).
JO050294C
5826 J. Org. Chem., Vol. 70, No. 15, 2005