Tris(4-bromophenyl)aminium hexachloroantimonate-mediated glycosylations of selenoglycosides and thioglycosides. Evidence for single electron transfer?

Article abstract of DOI:10.1016/S0008-6215(98)00163-3

Radical cation-initiated glycosylation reactions of phenyl selenoglycosides are described. Glycosylations of phenyl selenoglycosides effected by the single-electron-transfer (SET) reagent, tris(4-bromophenyl)aminium hexachloroantimonate (BAHA), are examined with primary and secondary hydroxyl acceptors. The corresponding reaction of an ethyl thioglycoside with a primary hydroxyl acceptor is also examined. Reactions are performed in the presence of the SET quenching reagent, 1,2,4,5-tetramethoxybenzene, to assess whether BAHA-mediated glycosylation reactions involve SET. These experiments indicate that the reactions are completely quenched in dichloromethane but only partially in acetonitrile. The results provide support for the SET mechanism but an alternative mechanism involving electrophilic activation cannot be discounted. The oxidation potentials of various selenoglycosides are determined by cyclic voltammetry. Copyright (C) 1998 Elsevier Science Ltd.

Full text of DOI:10.1016/S0008-6215(98)00163-3

Carbohydrate Research 310 (1998) 43±51
Tris(4-bromophenyl)aminium
hexachloroantimonate-mediated glycosylations
of selenoglycosides and thioglycosides.
y
Evidence for single electron transfer?
Seema Mehta, B. Mario Pinto *
Department of Chemistry, Simon Fraser University, Burnaby, British Columbia, Canada V5A 1S6
Received 30 January 1998; accepted 8 June 1998
Abstract
Radical cation-initiated glycosylation reactions of phenyl selenoglycosides are described.
Glycosylations of phenyl selenoglycosides e€ected by the single-electron-transfer (SET) reagent,
tris(4-bromophenyl)aminium hexachloroantimonate (BAHA), are examined with primary and
secondary hydroxyl acceptors. The corresponding reaction of an ethyl thioglycoside with a pri-
mary hydroxyl acceptor is also examined. Reactions are performed in the presence of the SET
quenching reagent, 1,2,4,5-tetramethoxybenzene, to assess whether BAHA-mediated glycosylation
reactions involve SET. These experiments indicate that the reactions are completely quenched in
dichloromethane but only partially in acetonitrile. The results provide support for the SET
mechanism but an alternative mechanism involving electrophilic activation cannot be discounted.
The oxidation potentials of various selenoglycosides are determined by cyclic voltammetry. # 1998
Elsevier Science Ltd. All rights reserved
1
. Introduction
In contrast to conventional Lewis acid-mediated
chalcogen to a suitable single-electron acceptor.
Subsequent cleavage of the C ±X bond results in
1
the formation of the oxocarbenium ion which
undergoes nucleophilic attack by an alcohol to
generate a new glycosidic linkage. The single elec-
tron transfer from the glycosyl donor can be e€ec-
ted either electrochemically [1], photochemically
[2], or chemically [3].
The electrochemical glycosylation method was
introduced by Noyori and Kurimoto [1a] with aryl
O-glycosides. This novel concept was extended by
Balavoine et al. [1b] and Amatore et al. [1c], inde-
pendently, to phenyl S-glycosides. The lower oxi-
dation potentials of the phenyl S-glycosides
glycosylations that are two-electron processes,
radical cation-initiated glycosylations involve sin-
gle-electron-transfer (SET). Radical-cation initi-
ated (RCI) glycosylations can be represented as
shown in Scheme 1. These reactions require the
initial generation of a radical cation by the
transfer of a single electron from the aglyconic
* Corresponding author. Tel.: 604-291-4327; Fax: 604-
91-3765; e-mail: bpinto@sfu.ca
2
y
This paper is dedicated, with respect and gratitude, to
the memory of Margaret A. Clark.
0008-6215/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII S0008-6215(98)00163-3

44
S. Mehta, B.M. Pinto/Carbohydrate Research 310 (1998) 43±51
scan showed one anodic wave having a peak
potential corresponding to the oxidation potential
of the substrate. The reverse scan did not show any
cathodic peaks. This result is typical of an `EC
type' mechanism in which the initial electro-
chemical reaction is followed by a chemical reac-
tion, and is consistent with the disproportionation
of the radical-cation to the oxocarbocenium ion
(
see Scheme 1). The oxidation potentials of the
selenoglycosides investigated fall in the range of
.35±1.5 V (Table 1). These values are lower than
1
those of S-glycosides and O-glycosides [10] and are
re¯ective of the ionization potentials of oxygen,
sulfur, and selenium lone pairs that decrease in the
same order [11].
The SET reagent chosen for RCI glycosylations
of selenoglycosides was tris(4-bromophenyl)ami-
nium hexachloroantimonate (BAHA) whose
reduction potential is reported to be 1.05 V versus
SCE [12]. Since the oxidation potentials of the
selenoglycosides are higher than that of the ami-
nium salt, one might have expected that the oxida-
tion of selenoglycosides by BAHA would be
Scheme 1. Radical cation-initiated glycosylation.
compared to those of the corresponding phenyl
O-glycosides o€ered the advantage of conducting
glycosylations at a lower oxidation potential, pro-
viding compatibility with a wider variety of pro-
tecting groups.
We and others have demonstrated the versatility
of phenyl selenoglycosides in selective activation
strategies for oligosaccharide synthesis [4±7]. Acti-
vation has been achieved with various promoters
such as silver tri¯ate, methyl tri¯ate, phenylselene-
nyl tri¯ate, and NIS/tri¯ic acid. We now describe
full details of the chemically-induced, radical
cation-initiated glycosylation reactions with phe-
nyl selenoglycosides, a novel method of e€ecting
selenoglycoside activation, using tris(4-bromo-
phenyl)aminium hexachloroantimonate, (p-BrC H )
thermodynamically
unfavorable.
However,
Kamata et al. [13] had reported the radical-cation-
mediated desulfurization of thiiranes with BAHA
as the SET reagent, although the oxidation
potentials of their substrates were in the range of
1.44±1.84 V versus SCE in acetonitrile. Thereafter,
Marra et al. [3] claimed to e€ect RCI glycosyla-
tions of thioglycosides with BAHA, in spite of the
higher oxidation potentials of the thioglycosides.
The results described in the following section
establish the validity of the SET mechanism in
BAHA-mediated glycosylation reactions.
RCI glycosylations of primary and secondary
hydroxyl acceptors with armed and disarmed sele-
noglycosides were investigated with BAHA as the
SET reagent. The glycosylation of the methyl gly-
coside 12 [14] with the selenoglycoside 1 yielded the
disaccharide 13 [4b] in 76% yield (Table 2, entry 1).
Glycosylation of the above acceptor with a more
reactive armed selenoglycoside donor 5 proceeded
in 94% yield producing an ꢀ:ꢁ mixture of disac-
charides 14 [15] and 15 [16] in a 1.7:1 ratio (Table 2,
entry 2). These glycosylations were e€ected at
ambient temperature.
6
4 3
+
Á
N
[
(BAHA), as the single electron transfer reagent
8]. Marra et al. [3a] have reported similar glycosy-
lations with ethyl thioglycosides and this group has
subsequently applied the method to the synthesis of
oligosaccharides [3b,c]. We also provide evidence to
support the existence of single electron transfer in
BAHA-mediated glycosylation reactions.
2
. Results and discussion
In order to determine a suitable single electron
transfer reagent to e€ect radical cation-initiated
glycosylations, the oxidation potentials of various
selenoglycosides [4b,9] were determined by cyclic
voltammetry [10]. The voltage was scanned from 0
to 2 V in the anodic direction at which point the
direction of the scan was reversed. The forward
Initial attempts to extend this methodology to
secondary alcohol acceptors proved only partly
successful. Glycosylation of the methyl glycoside
16 [16] with the selenoglycoside 1 a€orded a
mixture of products. In addition to a mixture of

S. Mehta, B.M. Pinto/Carbohydrate Research 310 (1998) 43±51
45
Table 1
Oxidation potentials of selenoglycosides
a
Selenoglycoside
Oxidation potential (V)
Selenoglycoside
1.35
1.41
1.42
1.50
1.33
1.35
Oxidation potential (V)
1.39
1.41
1.52
1.47
1.48
a
Oxidation potentials measured versus a standard saturated calomel electrode in dry acetonitrile with 0.1 M tetra-
ethylammonium perchlorate as supporting electrolyte at a scan rate of 200 mV/s.
Table 2
BAHA-Mediated glycosylations of a primary hydroxyl acceptor with selenoglycoside donors
Entry
Donor
Acceptor
Molar ratioa Time (h)
Product
Yield (%)
a
Donor : acceptor : BAHA.
disaccharides 17 [16] and 18 [16], obtained in
2% yield, 2,3,4-tri-O-acetyl-ꢀ-l-rhamnopyranosyl
ꢀ=ꢁ mixture of the disaccharides 24 [21] and 25 [21]
in a yield of 50% (Table 3, entry 3). The reaction
did not reach completion and the unreacted accep-
tor was recovered. The yield of the reaction based
on the recovered acceptor was 90%. Based on
these results, we deduced that glycosylations of
acceptors containing secondary hydroxyl groups
were not very ecient and might require additional
amounts of the promoter to reach completion. In
the following glycosylation of the same acceptor
with the selenoglycoside 1, the concentration of the
promoter BAHA was increased from 1.5 to 3 mol
equiv with respect to the selenoglycoside. As
observed in earlier experiments, this glycosyl donor
had a tendency to form the corresponding chloride
under the reaction conditions. Therefore, a two-
fold excess of the donor 1 was used to ensure
that rhamnopyranosyl chloride formation did not
5
chloride 19 [17], formed by the reaction of the
donor 1 with chloride ions in solution, was also
isolated in 36% yield (Table 3, entry 1). The
recovered acceptor 16 and trans-esteri®cation pro-
ducts 20 [17] and 21 [18] were also isolated. A
similar glycosylation of the unreactive, tribenzoyl
acceptor 22 [19] did not provide any of the desired
disaccharide. Although the selenoglycoside activa-
tion had been e€ected, as suggested by the isolation
of the chloride 19, 90% of the glycosyl acceptor 22
was recovered from the reaction mixture (Table 3,
entry 2). We surmised that modulation of the
reactivity of the acceptor was required. Hence, the
reaction of the benzylated acceptor 23 was exam-
ined. Glycosylation of the methyl glycoside accep-
tor 23 [20] with the selenoglycoside 5 a€orded an

46
S. Mehta, B.M. Pinto/Carbohydrate Research 310 (1998) 43±51
Table 3
BAHA-Mediated glycosylations of secondary hydroxyl acceptors with selenogylcoside donors
Entry Donor Acceptor Product
Molar ratioa Time (h)
Yield (%)
a
Donor : acceptor : BAHA.
Yield based on acceptor recovered.
b
reduce the yield of the reaction. The disaccharide
6 was obtained in 73% yield (Table 3, entry 4)
quenching reagents. 1,2,4,5-Tetramethoxybenzene
(TMB) has been used as an e€ective SET quench-
ing reagent for the evaluation of BAHA-mediated
SET reactions [13,22]. Since the oxidation potential
of TMB (0.81 V) [23] is lower than the oxidation
potentials of the selenoglycosides, it should be
capable of interfering with the BAHA-mediated
glycosylations if SET was the dominant mechan-
ism. In the case of acid catalysis, the use of TMB as
a quencher should produce little or no change in
the outcome of the reaction. The possibility of
electrophilic activation is dicult to assess since
the addition of quencher will consume the aminium
radical cation and will quench the reaction. Thus,
proof that the aminium salt is required for the
reaction does not necessarily give clear evidence of
the mechanism of the reaction.
2
and 20% of the acceptor remained unreacted.
RCI Glycosylations or alternative mechanisms?Ð
We realized that the aminium salt BAHA had the
potential of e€ecting glycosylations via four possi-
ble mechanisms. The ®rst mechanism, as discussed
thus far, involved the intermediacy of radical-
cations generated via single electron transfer from
the selenoglycoside to the aminium radical cation.
Owing to the strongly acidic nature of the reaction
medium, a second possibility of acid catalysis was
also considered. A third possibility involves direct
electrophilic activation of the selenium or sulfur
atom, or direct polar reaction of the aminium ion
at the p-position of the phenyl moiety of the phenyl
selenoglycosides. A ®nal possibility involves the
intermediacy of glycosyl chlorides.
The BAHA-mediated glycosylations of the
methyl glycoside acceptor 12 and the selenoglyco-
side 5 were performed in dichloromethane and in
An e€ective method of establishing the existence
of electron-transfer reactions is by the use of SET

S. Mehta, B.M. Pinto/Carbohydrate Research 310 (1998) 43±51
47
acetonitrile as the solvents, in the absence and pre-
sence of the quencher, TMB (Table 4, entries 1±4).
The reactions in a given solvent were performed
simultaneously under identical conditions and
concentrations of reactants. The reactions were
closely monitored by TLC. A TLC comparison of
the reactions with and without quencher in acet-
onitrile (Table 4, entries 4 and 3) revealed that,
after 1 h, the glycosyl acceptor 12 had completely
reacted in the latter reaction but was still detectable
in the former reaction. The yield of the dis-
accharides isolated in the reaction with quencher
indicated that the reaction had been suppressed in
the presence of the quencher (Table 4, entry 8). In
the latter case, a substantial amount of the
unreacted acceptor was detected by TLC in addi-
tion to the expected disaccharide. The unreacted
acceptor was recovered from the reaction in a yield
of 25%.
We also decided to probe the nature of the
BAHA-mediated glycosylations with the ethyl
thioglycoside 27 [24] (Table 4). The glycosylation
in dichloromethane a€orded an ꢀ=ꢁ mixture of the
disaccharides 14 and 15 (Table 4, entry 9). This
result contrasts with that of Marra et al. [3] who
reported that no reaction occurs in dichloro-
methane. An analogous reaction in the presence of
the quencher was completely quenched (Table 4,
entry 10). In acetonitrile, the quencher was e€ective
in suppressing the reaction (Table 4, entries 11 and
12); the disaccharides were isolated in a low yield
of 28% and the unreacted acceptor was isolated in
a yield of 63% (Table 4, entry 12).
Analysis of these results suggests that, in
dichloromethane, SET may be the dominant path-
way, since the reactions were completely quenched
in the presence of TMB in all three cases investi-
gated. In acetonitrile, however, the reactions
continued to proceed in spite of the presence of
TMB. We speculated that, in spite of having
quenched the SET (or electrophilically activated)
reaction, an alternative acid-catalyzed pathway
was available to the reactants. The acid-catalyzed
(78%) was lower than for the reaction conducted
in the absence of the quencher (89%). The reaction
in the presence of TMB conducted in dichlor-
omethane (Table 4, entries 1, 2) was completely
quenched and the reactants were recovered almost
quantitatively.
The results of a similar investigation performed
for the glycosylation of the acceptor 12 with the
selenoglycoside 1 indicate that, as in the above
case, the reaction was completely quenched in the
presence of the quencher TMB when dichloro-
methane was used as the solvent, and the reactants
were recovered quantitatively (Table 4, entries 5
and 6). In acetonitrile, the disaccharide 13 was
obtained in 60% yield in the absence of the
quencher (Table 4, entry 7). No unreacted acceptor
was present at the completion of the reaction. A
TLC comparison of this reaction after 1 h with
an analogous reaction in the presence of TMB
Table 4
Quenching experiments
Entry
Reaction
Conditionsa,b,c
Results
1
2
3
4
5
6
5+12
5+12
5+12
5+12
1+12
1+12
3 h, CH Cl , without quencher
2
Disaccharides 14, 15, 94%; ꢀ : ꢁ ˆ 1:8 : 1
Acceptor 12 recovered, 96%
Disaccharides 14, 15, 89%; ꢀ : ꢁ ˆ 1:7 : 1
Disaccharides 14, 15, 78%; ꢀ : ꢁ ˆ 1 : 7
Disaccharide 13, 76%
2
3 h, CH Cl , with quencher
2
3
3
2
1 h, CH
1 h, CH
CN, without quencher
CN, with quencher
2 h, CH Cl , without quencher
2
2
2 h, CH
Cl
2 2
, with quencher
Acceptor 12 recovered, 96%
Donor 1 recovered, 95%
7
8
1+12
1+12
1 h, CH
1 h, CH CN, with quencher
3
CN, without quencher
Disaccharide 13, 60%
Disaccharide 13, 51%
3
Acceptor 12 recovered, 25%
Disaccharides 14, 15, 88%; ꢀ : ꢁ ˆ 1:5 : 1
d
9
1
27 +12
d
2 h, CH
2
Cl
2
, without quencher
2 h, CH Cl , with quencher
0
27 +12
Acceptor 12 recovered, 94%
Donor 27 recovered, 96%
2
2
d
1
1
1
2
27 +12
d
0.5 h, CH
0.5 h, CH
3
CN, without quencher
CN, with quencher
Disaccharides 14, 15, 81%; ꢀ : ꢁ ˆ 1 : 1
Disaccharides 14, 15, 28%; ꢀ : ꢁ ˆ 1 : 4:5
Acceptor 12 recovered, 63%
27 +12
3
a
b
c
Donor : acceptor : BAHA=1 : 0.8 : 1.5.
Quencher=1,2,4,5-tetramethoxybenzene.
Donor : acceptor : BAHA : quencher=1 : 0.8 : 1.5 : 6.
Ethyl 2,3,4,6-tetra-O-benzyl-1-thio-ꢁ-d-glucopyranoside 27.
d

48
S. Mehta, B.M. Pinto/Carbohydrate Research 310 (1998) 43±51
reaction was less ecient than the RCI reaction,
and, therefore, an incomplete reaction was
observed. However, this reasoning did not explain
why a similar result was not observed in dichloro-
methane, since Lewis acid-catalyzed glycosylations
are generally conducted in the latter solvent.
reactions in acetonitrile is noteworthy. Glycosyla-
tions in acetonitrile are reported to occur via the
intermediacy of a glycosyl-acetonitrilium ion. Sev-
eral workers have proposed that the acetonitrilium
ion forms at the ꢀ-face [26]. This intermediate then
undergoes an inversion of con®guration in the
presence of a glycosyl acceptor. Thus, acetoni-
trile has been the solvent of choice to promote
ꢁ-glycosylations with substrates possessing a non-
participating substituent at the C-2 position.
However, other workers have proposed an
equatorial orientation for the acetonitrilium ion [27],
whose stability was attributed to the reverse
anomeric e€ect. Subsequent inversion by the accep-
tor was postulated to account for the preferential
formation of the ꢀ-disaccharide in these cases.
In our glycosylations with selenoglycoside 5, the
results in dichloromethane and acetonitrile are com-
parable and the ꢀ-disaccharide 14 is only slightly
favored. These results are inconsistent, therefore,
with either of the above proposals. We suggest
that the reaction is under thermodynamic control.
In contrast to the lack of stereoselectivity of this
glycosylation reaction in acetonitrile in the absence
of the quencher TMB, reactions in the presence of
TMB a€orded predominantly the ꢁ-disaccharide
15. We speculate that this is due to the formation
of an encounter complex of the TMB and the
oxocarbenium ion in which the ꢀ-face is shielded;
nucleophilic attack then occurs at the ꢁ-face.
We suggest, therefore, the following explanation
for our observations. Since the oxidation potential
of BAHA is lower than that of the selenoglyco-
sides, the direct oxidation of the selenoglycosides
by BAHA is thermodynamically unfavorable. We
postulate that the oxidation is preceded by the
formation of a complex between the selenium atom
of the selenoglycoside and the aminium radical-
cation. This adduct has a modi®ed oxidation
potential and is amenable to oxidation. Similar
adducts have been proposed by Gilbert and Mar-
riott [25] for the reaction of NH3+ with various
sul®des and sulfoxides. We propose that the polar-
ity of acetonitrile facilitates the initial ion-pair dis-
sociation of BAHA and, in turn, the complexation
of the selenium and the aminium radical-cation. If
trapping of the aminium radical cation by TMB is
slower than the electron transfer from selenium,
this would lead to a more facile SET (or electro-
philically activated) reaction and to an incomplete
quenching of the glycosylations by TMB in acet-
onitrile. Hence, SET (or elecrophilic activation)
appears to be the dominant mechanism of the
reaction in both dichloromethane and acetonitrile
and the acid-catalyzed pathway is of lesser impor-
tance. This conclusion is further corroborated by
the result of a glycosylation reaction in the pre-
sence of anisole as the SET quenching reagent. The
oxidation potential of anisole is +1.76 V [22].
Since this value is higher than those of the sub-
strates, the quencher should have no appreciable
e€ect on the SET glycosylations. As predicted, the
reaction of the methyl glycoside acceptor 12 and
the selenoglycoside 1 in the presence of anisole,
with BAHA as the promoter, was not quenched,
and the disaccharide 13 was isolated in 79% yield.
The question of the stereochemical outcome of
the reactions remains to be addressed. The glyco-
sylations with the phenyl selenorhamnoside 1 pro-
ceed in a 1,2-trans fashion, as expected due to the
anchimeric assistance provided by the acetate
substituent at the C-2 position of the glycosyl
donor. The glycosylation with the perbenzylated
phenyl selenoglucoside 5 in dichloromethane favors
the formation of the thermodynamically more
stable ꢀ-disaccharide 14. The stereoselectivity of the
3. Conclusions
O-Substituted phenyl selenoglycosides are
amenable to radical-cation-initiated glycosylations
e€ected by the single electron transfer reagent,
tris(4-bromophenyl)aminium hexachloroantimon-
ate (BAHA). The glycosylations of reactive
primary alcohol acceptors are promising, although
the glycosylations of unreactive acceptors are less
ecient. Acceptors of moderate reactivity require a
greater concentration of the promoter to achieve
high yielding reactions.
The oxidation potentials of various phenyl sele-
noglycosides have been determined. The oxidation
potentials of BAHA (1.05 V) and the phenyl
selenoglycosides (1.35±1.5 V) indicate that the
direct oxidation of the selenoglycosides via single-
electron transfer by BAHA is not probable. We
suggest a prior complexation of the selenoglycoside
and BAHA that produces a complex of modi®ed

S. Mehta, B.M. Pinto/Carbohydrate Research 310 (1998) 43±51
49
Ê
the acceptor, and 4 A molecular sieves was stirred
in anhydrous solvent (4 mL) under N for 1 h.
2
oxidation potential, which then undergoes a single-
electron-transfer reaction. Results based on
quenching experiments are consistent with the
existence of the SET mechanism in dichlor-
omethane and acetonitrile, although the possibility
of electrophilic activation of the selenoglycosides
cannot be discounted.
BAHA was added and the reaction mixture was
stirred until the reaction was complete. The reaction
ꢀ
mixture was cooled to 0 C and neutralized with
Et N. The mixture was ®ltered through Celite with
3
dichloromethane, the ®ltrate was concentrated, and
the residue was puri®ed by column chromato-
graphy with hexane-ethyl acetate as eluant.
4
. Experimental
Typical procedure for BAHA-mediated glycosyl-
ations with phenyl 2,3,4,6-tetra-O-benzyl-1-seleno-
b-d-glucopyranoside (5) and ethyl 2,3,4,6-tetra-O-
benzyl-1-thio-b-d-glucopyranoside (27) in dichloro-
methane or acetonitrile.ÐA mixture of phenyl
2,3,4,6-tetra-O-benzyl-1-seleno-ꢁ-d-glucopyrano-
side (5) or ethyl 2,3,4,6-tetra-O-benzyl-1-thio-ꢁ-d-
glucopyranoside (27) (0.1 mmol), the acceptor
General methods.ÐMelting points were deter-
mined on a Fisher±Johns melting-point apparatus
and are uncorrected. Optical rotations were
measured with a Rudolph Research Autopol II
automatic polarimeter. Analytical thin-layer chro-
matography (TLC) was performed on aluminum
plates precoated with Merck silica gel 60F-254 as
the adsorbent. The developed plates were air-dried,
exposed to UV light, and/or sprayed with 5% sul-
Ê
(0.08 mmol), and 4 A molecular sieves was stirred
in anhydrous solvent (2 mL) under N for 1 h.
2
BAHA (0.15 mmol) was added and the reaction
mixture was stirred until the reaction was complete.
ꢀ
furic acid in ethanol and heated at 150 C. All
ꢀ
compounds were puri®ed by medium-pressure col-
umn chromatography on Kieselgel 60 (230±400
mesh) according to a published procedure. Sol-
vents were distilled before use and were dried, as
necessary, by literature procedures. Solvents were
The reaction mixture was cooled to 0 C and neu-
tralized with Et N. The mixture was ®ltered through
3
Celite with dichloromethane, the ®ltrate was con-
centrated, and the residue was puri®ed by column
chromatography with hexane-ethyl acetate as eluant.
Methyl 2,3,4-tri-O-benzyl-6-O-(2,3,4-tri-O-acetyl-
evaporated under reduced pressure and below
ꢀ
4
0 C. Reactions performed under nitrogen were
a-l-rhamnopyranosyl)-a-d-glucopyranoside
[4b].ÐThe product was puri®ed by column chro-
(13)
also carried out in deoxygenated solvents. Trans-
fers under nitrogen were e€ected by means of
matography with hexane-ethyl acetate (2.5:1) as
eluant, R =0.37. The NMR spectral data were
identical to the literature data.
1
13
standard Schlenk-tube techniques. H and
NMR spectra were recorded on a Bruker AMX-
00 NMR spectrometer at 400.13 and 100.6 MHz,
C
f
4
Methyl 2,3,4-tri-O-benzyl-6-O-(2,3,4,6-tetra-O-
benzyl-a=b-d-glucopyranosyl)-b-d-glucopyranoside
(14, 15) [15].ÐThe product was puri®ed by col-
umn chromatography with hexane±ethyl acetate
for proton and carbon, respectively. The spectra
were recorded in deuterochloroform. All new
compounds were characterized by either micro-
analysis or electrospray mass spectrometry. The
microanalyses were performed by Mr. Mickey
Yang at Simon Fraser University.
(3:1) as eluant, [R =0.4] to give an ꢀ=ꢁ mixture of
f
disaccharides. The NMR spectral data were iden-
tical to the literature data.
Cyclic voltammetry.ÐCyclic voltammetry was
performed on 1 mM solutions of the substrates in
dry acetonitrile. Tetraethylammonium perchlorate
Methyl 2,3,6-tri-O-benzyl-4-O-(2,3,4,6-tetra-O-
benzyl-a; b-d-glucopyranosyl)-a-d-glucopyranoside
(24, 25) [21].ÐThe products were puri®ed by col-
umn chromatography with hexane±ethyl acetate
(4.5:1) as eluant, [R of 24 and 25=0.4, R of
recovered 23=0.2], to give a 1:1 ꢀ=ꢁ mixture of
disaccharides (43 mg, 50%), and 23 (18 mg, 43%).
The NMR spectral data were identical to the lit-
erature data.
Methyl 2,3,6-tri-O-benzyl-4-O-(2,3,4-tri-O-acetyl-
a-l-rhamnopyranosyl)-a-d-glucopyranoside (26).Ð
The product was puri®ed by column chromato-
graphy with hexane±ethyl acetate (2:1) as eluant,
(0.1 M) was used as the supporting electrolyte. The
cyclic voltammograms were measured at a plati-
num electrode with reference to a standard satu-
rated calomel electrode. The instrument was
calibrated with ferrocene as the standard.
Typical procedure for BAHA-mediated glycosyl-
ations with phenyl 2,3,4-tri-O-acetyl-1-seleno-a-l-
rhamnopyranoside (1) in dichloromethane or aceto-
nitrile.ÐA mixture of phenyl 2,3,4-tri-O-acetyl-1-
seleno-ꢀ-l-rhamnopyranoside (1) (87 mg, 0.2 mmol),
f
f

50
S. Mehta, B.M. Pinto/Carbohydrate Research 310 (1998) 43±51
[
2
(
Rf of disaccharide 26=0.25, Rf of acceptor
3=0.33], to give the disaccharide 26 as a syrup
(27) [24] (0.1 mmol) and methyl 2,3,4-tri-O-benzyl-
ꢀ-d-glucopyranoside (12) (40 mg, 0.08 mmol),
1,2,4,5-tetramethoxybenzene (115 mg, 0.6 mmol)
53 mg, 73%), and 23 (10 mg, 20%). The yield
ꢀ
22
Ê
based on reacted 23 is 91%. [ꢀ]
CH Cl ); C NMR (CDCl ): ꢂ 17.0 (C-6 ), 20.6,
2
5
7
7
1
J5
3
J =3.8 Hz, J =9.2 Hz, H-2), 3.65 (1H, dd,
J5,6a=2.8 Hz, J6a,6b=9.2 Hz, H-6a), 3.70±3.77 (2H,
m, H-4, H-6), 3.82±3.94 (2H, m, H-3, H-5), 4.04
+32 (c 1.0 in
0
and 4 A molecular sieves was stirred in anhydrous
solvent (2 mL) under N for 1 h. BAHA (115 mg,
d
13
2
2
3
2
0
0.7 (3COCH ), 55.2 (OCH ), 66.7 (C-5 ), 68.6 (C-
3
0.15 mmol) was added and the reaction mixture
was stirred under N for as long as the corre-
3
0
0
), 69.1 (C-6), 69.9 (C-3 ), 70.2 (C-2 ), 71.1 (C-4),
3.2 (2CH C H ), 75.1 (C-3), 75.3 (CH C H ),
2
sponding experiment in the absence of the
ꢀ
2
6
5
2
6
5
0
9.8 (C-3), 80.5 (C-2), 97.2 (C-1 ), 97.9 (C-1),
1
quencher. The reaction mixture was cooled to 0 C
and neutralized with Et N. The mixture was ®l-
27.2±137.9 (Ar); H NMR (CDCl ): ꢂ 1.17 (3H, d,
3
3
0
0
0
=6.0 Hz, H-6 ), 1.98, 1.99, 2.06 (9H, 3s,
tered through Celite with dichloromethane. The
®ltrate was concentrated, and the residue was pur-
i®ed by column chromatography with hexane-ethyl
acetate as eluant.
,6
COCH ), 3.36 (3H, s, OCH ), 3.58 (1H, dd,
3 3
1
,2
2,3
0
(
1H, m, H-5 ), 4.51 (2H, AB pattern, CH C H ),
2
6
5
4
J=12.0 Hz,
.58 (1H, d, J =3.8 Hz, H-1), 4.60 (1H, d,
Acknowledgements
1,2
CHHC H ),
4.69±4.76
0
=1.8 Hz, H-1 ), 4.98
1 ,2
(2H,
6
5
0
The authors are grateful to the Natural Sciences
and Engineering Research Council of Canada for
CHHC H ), 4.95 (1H, d, J
0
6
5
0
0
=20.0 Hz, H-4 ), 5.11 (1H, d,
,4 +4 ,5
(
1H, t, J3
0
0
0
®
nancial support.
J=11.0 Hz,
CHHC H ),
0
5.15
(1H,
0
=3.3 Hz, H-2 ), 5.25 (1H, dd,
2 ,3
dd,
6
5
0
J1
0
0
=1.8 Hz, J
=3.3 Hz,
,2
,3
0
0
=10.1 Hz, H-3 ), 7.30±7.50
3 ,4
J2
0
0
J
0
References
(15H, Ar); Anal. Calcd for C H O : C, 65.21; H,
40 48 13
6
.57; Found: C, 65.11; H, 6.69.
Typical procedure for BAHA-mediated glycosyl-
[
1] (a) R. Noyori and I. Kurimoto, J. Org. Chem.,
1 (1986) 4320±4322; (b) G. Balavoine, A. Greg,
5
ations with phenyl 2,3,4-tri-O-acetyl-1-seleno-a-l-
rhamnopyranoside (1) in dichloromethane or
acetonitrile in the presence of the quencher.ÐA
mixture of phenyl 2,3,4-tri-O-acetyl-1-seleno-ꢀ-l-
rhamnopyranoside (1) (87 mg, 0.2 mmol), methyl
J.-C. Fischer, and A. Lubineau, Tetrahedron Lett.,
31 (1990) 5761±5764; (c) C. Amatore, A. Jutand,
J.-M. Mallet, G. Meyer, and P. Sinay, J. Chem. Soc.,
Chem. Commun., (1990) 718±719; (d) J.-M. Mallet,
G. Meyer, F. Yvelin, A. Jutand, C. Amatore, and P.
Sinay, Carbohydr. Res., 244 (1993) 237±246.
2
0
1
,3,4-tri-O-benzyl-ꢀ-d-glucopyranoside (12) (46 mg,
.1 mmol), 1,2,4,5-tetramethoxybenzene (240 mg,
Ê
.2 mmol), and 4 A molecular sieves was stirred in
[
2] (a) J.D. Timpa and G.W. Grin, Carbohydr. Res.,
1
31 (1984) 185±196; (b) G.W. Grin,
N.C. Bandara, M.A. Clarke, W.-S. Tsang, P.J.
Garegg, S. Oscarson, and B.A. Silwanis, Hetero-
cycles, 30 (1990) 939±947; (c) T. Furuta, K.
Takeuchi, and M. Iwamura, J. Chem. Soc., Chem.
Commun., (1996) 157±158.
anhydrous solvent (4 mL) under N for 1 h. BAHA
2
(245 mg, 0.3 mmol) was added and the reaction
mixture was stirred under N for as long as the
2
corresponding experiment in the absence of the
ꢀ
quencher. The reaction mixture was cooled to 0 C
and neutralized with Et N. The mixture was ®l-
[3] (a) A. Marra, J.-M. Mallet, C. Amatore, and
P. Sinay, Synlett., (1990) 572±574; (b) P.
Sinay, Pure Appl. Chem., 63 (1991) 519±529;
(c) Y.-M. Zhang, J.-M. Mallet, and P. Sinay,
Carbohydr. Res., 236 (1992) 73±88.
4] (a) S. Mehta and B.M. Pinto, Tetrahedron. Lett.,
3
tered through Celite with dichloromethane, the ®l-
trate was concentrated, and the residue was
puri®ed by column chromatography with hexane-
ethyl acetate as eluant.
Typical procedure for BAHA-mediated glycosyl-
ations with phenyl 2,3,4,6-tetra-O-benzyl-1-seleno-
b-d-glucopyranoside (5) and ethyl 2,3,4,6-tetra-O-
benzyl-1-thio-b-d-glucopyranoside (27) [24] in
dichloromethane or acetonitrile in the presence of
the quencher.ÐA mixture of phenyl 2,3,4,6-tetra-
O-benzyl-1-seleno-ꢁ-d-glucopyranoside (5) or ethyl
[
32 (1991) 4435±4438; (b) S. Mehta and B.M. Pinto,
J. Org. Chem., 58 (1993) 3269±3276; (c) S. Mehta
and B.M. Pinto, in S.H. Khan and R.A. O'Neill
(
Eds.), Modern Methods in Carbohydrate Synthesis,
Harwood Academic Publishers, Amsterdam, 1996,
pp. 107±129.
[
5] (a) H.M. Zuurmond, P.H. van der Meer, P.A.M. van
der Klein, G.A. van der Marel, and J.H. van Boom,
J. Carbohydr. Chem., 12 (1993) 1091±1103; (b)
2
,3,4,6-tetra-O-benzyl-1-thio-ꢁ-d-glucopyranoside

S. Mehta, B.M. Pinto/Carbohydrate Research 310 (1998) 43±51
51
H.M. Zuurmond, P.A.M. van der Klein, J. de
Wildt, G.A. van der Marel, and J.H. van Boom, J.
Carbohydr. Chem., 13 (1994) 323±339.
[15] (a) J.-R. Pougny, J.-C. Jacquinet, M. Nassr, and
P. Sinay, J. Am. Chem. Soc., 99 (1977) 6762±6763;
(b) Y. Ito, T. Ogawa, M. Numata, and
M. Sugimoto, Carbohydr. Res., 202 (1990) 165±175.
[16] S. Kamiya, S. Esaki, and R. Ito-Tanaka, Agric.
Biol. Chem., 49 (1985) 2351±2358.
[
6] (a) P. Grice, S.V. Ley, J. Pietruszka,
H.W.M. Priepke, and E.P.E. Walther, Synlett.,
(
1995) 781±784; (b) P. Grice, S.V. Ley, J. Pie-
truszka, and H.W.M. Priepke, Angew. Chem. Int.
Ed. Engl, 35 (1996) 197±200; (c) P. Grice, S.V. Ley,
J. Pietruska, H.M.I. Osborn, H.W.M. Priepke, and
S.L. Warriner, Chem. Eur. J., 3 (1997) 431±440.
[17] C. Late, A.-M. Nguyen Phuoc Du, F. Winternitz,
and R. Wylde, Carbohydr. Res., 67 (1978) 91±
103.
[18] T. Otake, T. Sonobe, and T. Suami, Bull. Chem.
Soc. Jpn., 52 (1979) 3109±3110.
[
[
[
7] L.A.J.M.
Sliedregt,
H.J.G.
Broxterman,
G.A. van der Marel, and J.H. van Boom, Carbo-
hydr. Lett., 1 (1994) 61±69.
8] B.M. Pinto, Presented at The Sixth New Orleans
Carbohydrate Symposium, New Orleans, LA,
April, 1990.
9] S. Mehta, K.L. Jordan, T. Weimar, U.C. Kreis,
R.J. Batchelor, F.W.B. Einstein, and B.M. Pinto,
Tetrahedron Asymmetry, 5 (1994) 2367±2396.
[19] I.F. Pelyvas, T.K. Lindhorst, H. Streicher, and
J. Thiem, Synthesis, (1991) 1015±1018.
[20] J.-M. Kuster and I. Dyong, Justus Leibigs Ann.
Chem., (1975) 2179±2189.
[21] S. Koto, N. Morishima, M. Owa, and S. Zen,
Carbohydr. Res., 130 (1984) 73±83.
[22] (a) E. Hasegawa and T. Mukai, Bull. Chem. Soc.
Jpn., 58 (1985) 3391±3392; (b) M. Kamata,
K. Murayama, T. Suzuki, and T. Miyashi, J.
Chem. Soc., Chem. Commun., (1990) 827±829.
[23] A. Zweig, W.G. Hodgson, and W.H. Jura, J. Am.
Chem. Soc., 86 (1964) 4124±4129.
[24] F. Weygand and H. Ziemann, Justus Liebigs Ann.
Chem., 657 (1962) 179±198.
[25] B.C. Gilbert and P.R. Marriott, J. Chem. Soc.
Perkin Trans. 2, (1979) 1425±1431.
[
10] A.J. Bard and L.R. Faulkner, Electrochemical
Methods. Fundamentals and Applications, John
Wiley & Sons, New York, 1980.
11] S. Cradock and R.A. Whiteford, J. C. S. Faraday
II, 68 (1972) 281±288.
12] N.L. Bauld, D.J. Bellville, B. Harirchian,
K.T. Lorenze, R.A. Pabon, Jr., D.W. Reynolds,
D.D. Wirth, H.-S. Chiou, and B.K. Marsh, Acc.
Chem. Res., 20 (1987) 371±378.
13] M. Kamata, K. Murayama, and T. Miyashi, Tet-
rahedron Lett., 30 (1989) 4129±4132.
[
[
[26] (a) A.J. Ratcli€e and B. Fraser-Reid, J. Chem. Soc.
Perkin Trans. 1, (1990) 747±750; (b) Y. Ito and
T. Ogawa, Tetrahedron Lett., 28 (1987) 4701±4704.
[27] R.R. Schmidt and E. Rucker, Tetrahedron Lett., 21
(1980) 1421±1424.
[
[
14] P.J. Garegg, T. Iversen, and S. Oscarson, Carbo-
hydr. Res., 50 (1976) c12±c14.