Investigations into the chemistry of some 1,6-epithio 1,6-episeleno β-D-glucopyranoses

Article abstract of DOI:10.1071/ch99164

Derivatives of 1,6-dideoxy-1,6-epithio-β-D-glucopyranose have been shown to undergo oxidation reactions to afford the corresponding sulfoxides and sulfones. The sulfoxides participate in Pummerer reactions to afford the corresponding α-acetoxy sulfides wh

Full text of DOI:10.1071/ch99164

C S I R O P U B L I S H I N G
Australian Journal
of Chemistry
Volume 53, 2000
© CSIRO 2000
A journal for the publication of original research
in all branches of chemistry and chemical technology
w w w. p u b l i s h . c s i ro . a u / j o u rn a l s / a j c
All enquiries and manuscripts should be directed to
The Managing Editor
Australian Journal of Chemistry
CSIRO PUBLISHING
PO Box 1139 (150 Oxford St)
Collingwood
Vic. 3066
Australia
Telephone: 61 3 9662 7630
Facsimile: 61 3 9662 7611
Email: john.zdysiewicz@publish.csiro.au
Published by CSIRO PUBLISHING
for CSIRO and
the Australian Academy of Science

Aust. J. Chem., 2000, 53, 389–397
Investigations into the Chemistry of Some
ꢀ
1,6-Epithio and 1,6-Episeleno -D-Glucopyranoses
Brian W. Skelton,A Robert V. Stick,A,B D. Matthew G. Tilbrook,A
Allan H. WhiteA and Spencer J. WilliamsA
A Department of Chemistry, The University of Western Australia,
Nedlands, W.A. 6907.
B Author to whom correspondence should be addressed.
Derivatives of 1,6-dideoxy-1,6-epithio-ꢀ-D-glucopyranose have been shown to undergo oxidation reactions to
afford the corresponding sulfoxides and sulfones. The sulfoxides participate in Pummerer reactions to afford the
corresponding ꢁ-acetoxy sulfides which were then oxidized further. None of the sulfoxides, sulfones or ꢁ-acetoxy
sulfides prepared were particularly efficient glycosyl donors. Also presented are crystal structures of 1,6-dideoxy-
1,6-epithio-ꢀ-D-glucopyranose S,S-dioxide and 1,6-dideoxy-1,6-episeleno-ꢀ-D-glucopyranose, interesting ana-
logues of 1,6-anhydro-ꢀ-D-glucopyranose.
Keywords. Episeleno; epithio; glucopyranose; oxidation; sulfoxides; sulfones; Pummerer reactions; X-ray crystal
structures.
in ethanol5) provided the novel ‘tri-O-acetyl selenolevoglu-
cosan’ (4) in 56% yield. 4 We also attempted to prepare the
1,6-epidithio sugar (5) by treatment of the bromide (3) with
the ‘disulfide transfer’ reagent, benzyltriethylammonium
tetrathiomolybdate. This combination gave not the desired
1,6-epidithio sugar (5) but rather the 1,6-epithio sugar (2), in
a very surprising yield (88%), presumably through a redox
process. 4 This method now appears to constitute the most
reliable and simplest preparation of thiolevoglucosans.
Recently, Lowary and Bundle reported the unanticipated
preparation of a thioglucosan by treatment of the 6-O-tosyl
derivative of an ethylthio ꢀ-D-glucoside with sodium
iodide.6 Finally, the exo-sulfoxide (6) has been prepared
stereoselectively by the action of peracetic acid on the
sulfide (1). X-ray crystallography was used to assign the
stereochemistry of the resultant sulfoxide.7
The limited investigations into the oxidation chemistry of
thioglucosan (1) warranted further studies into the prepara-
tion of derivatives containing the thioglucosan core. In addi-
tion, the novel preparation of a thioglucosan by Lowary and
Bundle merited consideration of their method as an alterna-
tive preparation of thioglucosan.
To this end, the tosylate (7) was prepared by treatment of
ethyl 1-thio-ꢀ-D-glucopyranoside with tosyl chloride, fol-
lowed by acetylation of the crude material after workup.
Treatment of the tosylate (7) with sodium iodide in refluxing
butanone provided the triacetate (2) in a disappointing yield
of 31%, with no improvement on the alternative syntheses
outlined in the literature. 1–4
1,6-Anhydro sugars are routinely employed in the con-
struction of a variety of carbohydrate-based molecules. Of
particular importance in this class of sugars is the conforma-
tional constraint enforced on the sugar by the 1,6-anhydro
bridge and the protection afforded to the two most reactive
hydroxyls, at C 1 and C 6 which, together, provide 1,6-
anhydro sugars with a good deal of synthetic potential not
present in other, simple sugar derivatives, such as glycosides.
The analogous molecules containing sulfur or selenium
have, particularly in the case of sulfur, been known for some
time. Nevertheless, investigations into the reactions and
properties of these molecules have been limited, owing in
part to the difficulty in preparing these compounds.
The first synthesis of ‘thiolevoglucosan’(1) was achieved
in 1963 by Akagi and coworkers.1 The method relied on the
early introduction of sulfur at C 1 of a hexopyranose, via an
anomeric thioacetate, and then exposure of the derived thio-
late ion (sodium methoxide) to a leaving group conveniently
located at C 6, thus installing the 1,6-epithio bridge.
Acetylation of the crude material then provided ‘tri-O-acetyl
thioglucosan’ (2) in a moderate yield. A number of related
syntheses has since been reported that similarly rely on base
treatment of molecules having a thioester at C 1 or C 6 and a
good leaving group at C 6 or C 1, respectively.2,3
Recently, we reported a synthesis of thiolevoglucosan
which relied on the treatment of a doubly activated hexopy-
ranose, such as (3), having leaving groups at both C 1 and
C 6, with sulfide ion prepared by treatment of hydrogen
sulfide with triethylamine. 4 In related work, treatment of the
same bromide (3) with sodium hydrogen selenide (prepared
by treatment of elemental selenium with sodium borohydride
Dimethyldioxiran (DMDO) is a potent, yet mild, oxidant,
the attractiveness of which is enhanced owing to the volatil-
Manuscript received 18 November 1999
© CSIRO 2000
10.1071/CH99164
0004-9425/00/050389

390
B. W. Skelton et al.
ity of the only by-product of its use, acetone.8 The use of
DMDO for the oxidation of sulfur in thioglycosides has been
demonstrated to a limited extent, showing its utility for the
preparation of sulfoxides and sulfones. The preparation of
sulfoxides is, in general, less efficient than that of the corre-
sponding sulfone owing to competing oxidation to the
sulfone. Thus, Gervay and coworkers have reported that
treatment of a C-glycoside sulfide with an excess of DMDO
afforded the corresponding sulfone in quantitative yield,
whereas treatment of the same C-glycoside with an equiva-
lent of meta-chloroperbenzoic acid afforded a mixture of the
corresponding sulfoxides in only 34% yield.9 Similarly,
Berkowitz and coworkers found that treatment of a thiogly-
coside with DMDO afforded the corresponding sulfoxides in
77% yield; interestingly, however, a cautious treatment of
the same thioglycoside with a slight excess of DMDO afforded
only the major sulfoxide diastereoisomer and the corre-
sponding sulfone.10
Here, the phenyl thioglycoside (8) was treated with 1.2
equiv. of DMDO in acetone to afford, after evaporation of the
solvent and chromatography, the sulfoxides (9) in 64% yield.
By comparison, oxidation of the same thioglycoside with
meta-chloroperbenzoic acid afforded the same sulfoxides in
85% yield.11 In a similar approach, the benzoylated sulfox-
ides (10) were prepared in 84% yield by treatment of the
thioglycoside (11) with meta-chloroperbenzoic acid.
It has recently been suggested in the case of some 1-thio
ꢁ-D-mannopyranosides that the observed stereoselectivity in
the oxidation to the corresponding sulfoxide is a result of the
oxidation of the least hindered lone-pair of the preferred con-
formation as predicted by the exo-anomeric effect.12 In the
case of a 1-thio-ꢀ-D-glucopyranoside, the exo-anomeric
effect allows for the prediction of the preferred conforma-
tion, but it is not obvious which lone-pair on the sulfur will
be less sterically hindered. Presumably, small differences in
the steric environment of each lone-pair on the sulfur atoms
in (8) and (11) give rise to the observed, small diastereo-
selectivities. That one lone-pair on the sulfur in a 1-thio-ꢀ-D-
glucopyranoside is more crowded than the other could
explain the results observed by Berkowitz and coworkers
above.10 An alternative explanation for the observed
diastereoselectivity in the oxidation of some 1-thio-ꢀ-D-
galactopyranosides has been proposed.13
Whilst the reagent of choice for the conversion of thio-
glycosides into sulfoxides appeared to be meta-chloroper-
benzoic acid, treatment of the two thioglycosides, (8) and
(11), with an excess of DMDO rapidly provided the sulfones
(12) and (13), respectively, in almost quantitative yields.
With these results in hand, the oxidation of the triacetate
(2) was examined. Treatment of (2) with meta-chloroperben-
zoic acid in dichloromethane at 0° resulted in the rapid con-
version to two new, high-polarity compounds (t.l.c.). These
compounds were separated and identified as the endo-sulf-
oxide (14) (7%) and the exo-sulfoxide (15) (84%). The endo-
stereochemistry of (14) was assigned on the basis of the large
downfield shift of H 2 relative to that of the sulfide (2),
attributed to deshielding by the anisotropic S–O bond.14
Deacetylation of (15) provided the triol (6), previously pre-
pared by Novotny and coworkers but characterized only by
X-ray crystallography.7
The tribenzoate (16) was treated with meta-chloroperben-
zoic acid to afford only the exo-sulfoxide (17), the increased
stereoselectivity presumably resulting from the increased
bulk of the benzoate group at C 3, compared with the acetate
group in (2). In the two cases above, the predominant sulf-
oxide obtained was that of the exo-configuration.
Presumably, the oxidant reacts with the most exposed lone-
pair on sulfur to afford the corresponding sulfoxide.
The next compounds to be investigated were the sulfones.
The triacetate (2) was treated with an excess of meta-
chloroperbenzoic acid at room temperature for two days,
resulting in, initially, a fast conversion to the sulfoxides (14)
and (15) (t.l.c.), followed by a slower conversion into the
sulfone (18), eventually isolated in a yield of 95%. Whilst

391
1,6-Epithio and 1,6-Episeleno ꢀ-D-Glucopyranoses
the yield of the reaction was excellent, the use of a large
excess of meta-chloroperbenzoic acid and the prolonged
reaction time were not ideal, prompting an attempt at the oxi-
dation using DMDO. Indeed, treatment of the triacetate (2)
with an excess of DMDO rapidly provided the sulfone (18) in
a quantitative yield, with no requirement for purification.
Similarly, the tribenzoate (16), upon treatment with DMDO,
was converted in good yield into the sulfone (19), although
at a lower rate. The tri-O-benzoyl sulfone (19) was a remark-
ably refractory material, possessing only a slight solubility in
chloroform and only marginally better solubility in dimethyl
sulfoxide. Deacetylation of the tri-O-acetyl sulfone (18) with
sodium methoxide in methanol provided the triol (20), the
structure being confirmed by a single-crystal X-ray diffrac-
tion analysis (Fig. 1).
X-Ray crystal structures have already been determined for
glucosan (21),15 thioglucosan (1)16 and its exo-sulfoxide (6).7
To complete the glycosan series, the crystal structure of
selenoglucosan (22) was determined (Fig. 1).
Table 1 summarizes the C 1–X and C 6–X bond lengths
from all of the known crystal structures for simple glucosan
derivatives. It is of interest to note that the unit cell of the
sulfoxide (6) contains two crystallographically independent
molecules, possessing almost the same conformation. On the
other hand, the unit cell of the selenoglucosan (22), the only
crystallographically characterized selenium-containing
member of the array, contains two independent molecules
having quite different conformations, namely one in a
‘twisted’ B3,O conformation and the other in a slightly
1
deformed C4 conformation. The series containing oxygen,
sulfur and selenium exhibits an increase in the C 1–X and
C 6–X bond lengths, as would be expected, the greatest
increase being from oxygen to sulfur [ꢂ(C 1–X) = 0.388 Å,
ꢂ(C 6–X) = 0.383 Å)], with a smaller increase from sulfur to
selenium [ꢂ(C 1–X) = 0.125, 0.183 Å; ꢂ(C 6–X) = 0.123,
0.122 Å]; the latter, nevertheless, may be sufficient to begin
to tip the balance of conformational stability in favour of the
new form.
The trend in bond lengths for the sequence containing
thioglucosan (1), the sulfoxide (6) and the sulfone (20) is less
obvious; however, there appears to be a slight increase in the
length of the C 1–S bond. The C 6–S bond through the series
shows no such trend; however, it is only in the sulfone (20)
that the C 6–S bond is shorter than the C 1–S bond.
Table 1. Selected bond lengths and conformations in glucosan
derivatives
Deriv-
ative
Atom
X
C 1–X
(Å)A
C 6–X
(Å)A
Conforma-
tion
(21)15
(1)16
(22)
O
S
Se
1.427(4)
1.815B
1.94(1),
1.444(5)
1.827B
1.95(1),
1.94(1)
1.835(4),
1.868(4)
1.779(2)
1C4
1C4
1C4, B3,O
1.998(9)
(6)7
exo-SO
1.828(3),
1.821(4)
1.825(3)
1C4
1C4
Fig. 1. Projections of (20) and (22) (two molecules) in the orientation
of the schema (approximately), showing the two conformers of (22).
20% thermal ellipsoids are shown for the non-hydrogen atoms, hydro-
gen atoms having arbitrary radii of 0.1 Å.
(20)
SO2
A Error in value shown in parentheses.
B Errors in measurement are not quoted in ref. 16.

392
B. W. Skelton et al.
intermediates.25 A number of alternative promoters for the
reaction has since been developed.24,26–30
Sulfoxides with ꢁ-protons, upon treatment with acylating
reagents such as acetic anhydride, are prone to rearrange-
ment to ꢁ-acetoxy sulfides.17 This reaction, known as the
Pummerer rearrangement, effectively achieves an oxidation
adjacent to sulfur, thus providing a route to thioacetals.18,19
The benzoylated sulfoxide (17) was treated with sodium
acetate and acetic anhydride at reflux, resulting in a slow but
smooth conversion into a new compound. Workup, followed
by flash chromatography, provided the ꢁ-acetoxy sulfide
(23) (91%). The stereochemistry of the newly formed centre
at C 6 was assigned on the basis of the very small magnitude
of the coupling constant between H 5 and H6 (J Ϸ 0 Hz)
which, by an inspection of models, could only result from the
acetoxy group being present in the exo-configuration.20 In a
similar fashion, the triacetate (15) was treated with acetic
anhydride and sodium acetate, resulting in a slightly faster
conversion into the tetraacetate (24).
Both of the Pummerer rearrangements on the exo-sulf-
oxides (15) and (17) shown above were stereoselective,
affording products with the same exo-configuration as the
starting sulfoxide. This result is possibly a consequence of
the migration of the acetate group through a close-contact
ion pair. Alternatively, the stereoselectivity of this reaction
may result from the addition of acetate to the least hindered
side of an intermediate cation.17
The established reliability of the sulfoxide glycosylation
protocol prompted an investigation into the activation of the
sulfoxide (17) as a glycosyl donor. Initially, however, the
model sulfoxide (10) was treated with the alcohol (28) and
triflic anhydride in the presence of 2,6-di-t-butylpyridine,
resulting in a complex mixture of products. Repetition of this
reaction, but with the omission of the base, resulted in the
clean formation of the disaccharide (29) in good yield (78%).
Glycosylation under Kahne’s conditions with base, with
benzoylated donors such as (10), has been shown to afford
equimolar mixtures of orthoesters and glycosides; however,
it has been shown that without base the reaction proceeds
cleanly to afford glycosides in good yield.23,26 Presumably,
the sulfenic acid liberated during the reaction causes rear-
rangement of intermediate orthoesters to afford the desired
glycosides.
Treatment of a suspension of the sulfoxide (17) and the
alcohol (28) in ether with triflic anhydride afforded none of
The oxidation of these ꢁ-acetoxy sulfides was the next
reaction to be investigated. Thus, the ꢁ-acetoxy sulfide (23)
was treated briefly with meta-chloroperbenzoic acid, afford-
ing two sulfoxides, (25) and (26) (a combined yield of 87%),
in the ratio of 1 : 5. Again, the stereochemistry of the newly
formed stereogenic centres was assigned on the basis of a
downfield shift of H 2 caused by the endo-sulfoxide of the
(S)-S-oxide (25). In these cases, the reduced stereoselectivity
in the oxidation of (23) compared with that of (16) is most
likely a result of the increased hindrance of the exo-face of
(23) by the newly installed acetoxy group.
Finally, the ꢁ-acetoxy sulfide (23) was treated with an
excess of DMDO, resulting in a quantitative conversion into
the sulfone (27). This work provided access to a number of
potential glycosyl donors, namely the sulfoxide (17), the
sulfone (19), the ꢁ-acetoxy sulfide (23), and the ꢁ-acetoxy
sulfone (23). Our next piece of work details investigations
into the utility of these compounds as glycosyl donors.
The construction of the glycosidic bond through the use of
glycosyl sulfoxides has gained a great deal of popularity
since the original disclosure by Kahne and coworkers that
such glycosyl donors may be activated with triflic anhydride
in the presence of a hindered base; even sterically hindered
acceptor alcohols gave the reaction in good yield.21 The out-
standing utility of this glycosylation protocol has been
demonstrated in a synthesis of the Lea, Leb and Lex blood
group antigenic determinants through the singular use of the
sulfoxide glycosylation protocol to generate both ꢁ- and ꢀ-
linkages.22 Mechanistic studies have been performed that
indicate that the actual reactive intermediate may, in fact, be
a glycosyl triflate and that the reaction may be conducted
with only half an equivalent of triflic anhydride.23,24 More
recently, glycosyl sulfenates have also been implicated as

393
1,6-Epithio and 1,6-Episeleno ꢀ-D-Glucopyranoses
the desired disaccharide, but only the triflate (30).31
Undoubtedly, the low solubility of the sulfoxide (17) in ether
prevented glycosylation and the alternative sulfonylation
reaction occurred. The reaction was repeated using 1,2-
dichloroethane as the solvent; however, only polar products
appeared to be formed (t.l.c). Under the assumption that this
polar material arose from the decompostion of an intermedi-
ate sulfenyl triflate, e.g. (31), the reaction was attempted
again but with tetrabutylammonium iodide added in order to
promote the formation of the disulfide (32) by way of the
sulfenyl iodide (33). Encouragingly, the disulfide (32) was
formed, although in a poor yield (29%).
With these only partially successful results available, the
sulfoxide method was set aside in favour of investigations
into glycosylations using the sulfone (19). Again, the model
sulfone (13) was the initial focus of the investigation.
Ferrier and coworkers32 and, more recently, Ley and
coworkers33,34 have reported preliminary investigations
into the use of sulfones as glycosyl donors. The reaction of
sulfones with alcohols is characterized by the very mild
conditions required for activation, namely magnesium
bromide etherate and sodium bicarbonate. Thus, a mixture
of the sulfone (13), the alcohol (28), magnesium bromide
etherate and sodium bicarbonate was ultrasonicated at 40°
for 3 h. After this time, there appeared to be no change
(t.l.c.) and the reaction was abandoned. The lack of reaction
shown by the donor (13) is not overly surprising—Ley has
only been able to activate 2-deoxy glycosyl sulfones, a
class of sugars well known for their lability toward acid
hydrolysis.33,34 As a result, the glycosylation of the alcohol
(28) with the sulfones (19) and (27) was not attempted.
The last compound to be investigated for its ability to
function as a glycosyl donor was the ꢁ-acetoxy sulfide
(23). By analogy with the successful use of thioglucosans
as glycosyl donors, it was hoped that the thioacetal of (23)
could act as a glycosyl donor. In the event, a solution of
the ꢁ-acetoxy sulfide (23) and the alcohol (28) was
treated with N-iodosuccinimide/triflic acid. After a pro-
longed treatment, no reaction other than degradation of
the starting materials was evident and the reaction was
abandoned.
We still maintained an interest in the unusual disulfide (5)
and made an attempt at its synthesis. Initially, however, we
treated the thiocyanate (34) with benzyltriethylammonium
tetrathiomolybdate, according to the procedure of Prabhu
and coworkers, to give an easily separated mixture of the
thiol (35) and the disulfide (36).35 It was hoped that an anal-
ogous treatment of the dithiocyanate (37) with the same
reagent would provide the desired disulfide (5).
Unexpectedly, treatment of the bromide (38) with potassium
thiocyanate gave not the dithiocyanate (37), but rather the
isothiocyanate (39). The change in linkage was demonstrated
by the application of infrared spectroscopy and 13C n.m.r.
spectroscopy. At this stage, our investigations drew to a
close; however, it has been reported that glycosyl thio-
cyanates may be readily prepared from benzoylated glycosyl
bromides, with little formation of the isomeric isothio-
cyanates.36
Experimental
General experimental procedures have been given previously.37
Ethyl 2,3,4-Tri-O-acetyl-6-O-(p-tolylsulfonyl)-1-thio-ꢀ-D-
glucoside (7)
Ethyl 2,3,4,6-tetra-O-acetyl-1-thio-ꢀ-D-glucoside38,39 (2.00 g,
5.10 mmol) was suspended in dry MeOH (20 ml) and treated with a
small piece of sodium metal at room temperature. The suspension
rapidly dissolved and, after 20 min, t.l.c. indicated complete conversion
to a higher polarity compound. The solvent was evaporated and the
residue was taken up in dry pyridine (10 ml) and TsCl (1.4 g, 7.3 mmol)
was added to the solution at 0°. The mixture was stirred overnight at
room temperature under N2. Ac2O (3 ml) was added and the solution
was allowed to stand for 10 min. Water (5 ml) was added and the
mixture, after 10 min, was subjected to the usual workup (EtOAc). The
residue, after evaporation of the solvent, was taken up in EtOH and
allowed to crystallize to give the tosylate (7) as colourless needles
(1.42 g, 55%), m.p. 116–117° (EtOH), [ꢁ]D –5.9° (Found: C, 50.2; H,
1
5.5. C21H28O10S2 requires C, 50.0; H, 5.6%). H n.m.r. (300 MHz) ꢃ
1.19, t, J 7.5 Hz, CH2CH3; 1.96, 1.96, 2.01, 3s, 9H, Ac; 2.42, s, ArMe;
2.54–2.66, m, CH2CH3; 3.72, ddd, J4,5 10.1, J5,6 3.0, 5.6 Hz, H5; 4.02,
dd, J6,6 11.0 Hz, H 6; 4.09, dd, H6; 4.41, d, J1,2 10.1 Hz, H 1; 4.89, dd,
J3,4 9.5 Hz, H 4; 4.92, dd, J2,3 9.3 Hz, H 2; 5.15, dd, H3; 7.30–7.32,
7.72–7.75, 2m, 4H, Ar. 13C n.m.r. (75.5 MHz) ꢃ 14.81, CH2CH3; 20.49,
20.53, 20.64, 3C, COMe; 21.62, ArMe; 24.06, CH2CH3; 67.73, C 6;
68.52, 69.60, 73.57, 75.43, C 2,3,4,5; 83.37, C 1; 128.04, 129.85,
132.36, 145.12, Ar; 169.30, 169.42, 170.09, 3C, CO.
2,3,4-Tri-O-acetyl-1,6-dideoxy-1,6-epithio-ꢀ-D-glucose (2)
Asolution of the tosylate (7) (516 mg, 1.02 mmol) and NaI (270 mg,
1.80 mmol) in butanone (10 ml) was heated under reflux for 2 h. The
solvent was evaporated and the residue partitioned between EtOAc and
water. The organic phase was separated and washed with aqueous
sodium thiosulfate solution, then brine and dried. The residue was puri-
fied by flash chromatography (30% EtOAc/petrol) to afford the tri-
acetate (2) as a clear oil (98 mg, 31%). The 1H n.m.r. (300 MHz)
spectrum of this material agreed with that reported in the literature.4
(R)- and (S)-Phenylsulfinyl Tetra-O-acetyl-ꢀ-D-glucopyranoside (9)
A solution of phenyl tetra-O-acetyl-1-thio-ꢀ-D-glucopyranoside
(8)38,39 (220 mg, 0.500 mmol) in acetone (3 ml) was treated with DMDO
(6 ml of 0.1 M in acetone, 0.6 mmol; –78°, 3 min). The solvent was
evaporated under reduced pressure to afford a white, crystalline solid
that was purified by flash chromatography (50–75% EtOAc/petrol) to
afford the sulfoxides (9) as a white, crystalline powder (147 mg, 64%,
1: 3). The 1H n.m.r. spectrum was largely in accordance with that given
in the literature.40 Partial 1H n.m.r. (300 MHz) ꢃ 3.57, ddd, J4,5 10.1, J5,6
2.3, 5.8 Hz, H 5 minor; 3.67, ddd, J4,5 10.2, J5,6 Ϸ J5,6 3.6 Hz, H 5 major.
13C n.m.r. (75.5 MHz) ꢃ 20.32, 20.36, 20.42, 4C, Me; 61.16, 61.53, C 6;
67.16, 67.22, 67.54, 73.40, 73.60, 76.05, 76.30, C 2,3,4,5; 89.64, 91.99,
C 1; 125.51–138.77, Ph; 168.76, 169.06, 169.13, 169.92, 170.15, 4C,
CO.
(R)- and (S)-Phenylsulfinyl Tetra-O-benzoyl-ꢀ-D-glucopyranoside (10)
meta-Chloroperbenzoic acid (80%, 270 mg, 1.25 mmol) was added
to a solution of phenyl tetra-O-benzoyl-1-thio-ꢀ-D-glucopyranoside
(11)41 (688 mg, 1.00 mmol) in dry CH2Cl2 (20 ml) at 0° under N2. After
5 min, t.l.c. indicated a trace of starting material remaining so further
meta-chloroperbenzoic acid (20 mg, 0.09 mmol) was added and the
mixture was stirred for another 5 min. Aqueous KI solution (10%, 2 ml)
and saturated aqueous NaHCO3 solution (6 ml) were added and the
mixture was stirred for 20 min. The mixture was washed with aqueous
sodium thiosulfate solution (0.5 M) and the organic phase was sepa-
rated, dried and the solvent evaporated. The residue was purified by
flash chromatography (0–50% EtOAc/toluene) to give the sulfoxides
(10) as a clear foam (593 mg, 84%, 2 : 3). 1H n.m.r. (300 MHz) for the
major diastereoisomer: ꢃ 4.21, ddd, J4,5 9.8, J5,6 2.7, 4.4 Hz, H5; 4.40,
dd, J6,6 12.4 Hz, H6; 4.66, dd, H6; 4.87, d, J1,2 9.8 Hz, H1; 5.54, dd, J3,4

394
B. W. Skelton et al.
m.p. 231° (MeOH), [ꢁ]D –117° (H2O) (Found: C, 36.9; H, 5.0.
C6H10O5S requires C, 37.1; H, 5.2%). 1H n.m.r. (300 MHz, D2O) ꢃ 2.63,
br dd, J5,6 8.6, J6,6 13.2 Hz, H6; 3.40, 3.60, 3.78, 3 br s, H 2,3,4; 3.99,
br d, H 6; 5.13, br d, H5; 5.19, br s, H1. 13C n.m.r. (75.5 MHz, D2O) ꢃ
58.73, C 6; 67.44, 68.48, 69.15, 81.51, C 2,3,4,5; 102.35, C 1.
Ϸ J4,5 9.8 Hz, H 4; 5.70, dd, J2,3 9.3 Hz, H2; 5.95, dd, H 3; 7.07–8.05,
m, Ph. 1H n.m.r. (300 MHz) for the minor diastereoisomer: ꢃ 4.12, ddd,
J4,5 9.7, J5,6 2.8, 6.4 Hz, H 5; 4.43, dd, J6,6 12.4 Hz, H 6; 4.53, dd, H 6;
4.59, d, J1,2 9.6 Hz, H1; 5.61, dd, J3,4 Ϸ J4,5 9.7 Hz, H 4; 5.91, dd, J1,2
Ϸ J2,3 9.6 Hz, H2; 6.01, dd, H3; 7.07–8.05, m, Ph. 13C n.m.r.
(75.5 MHz) ꢃ 62.11, 62.91, C 6; 67.91, 68.11, 68.34, 73.72, 73.81,
76.73, 77.08, C 2,3,4,5; 90.72, 92.93, C 1; 125.46–138.22, Ph; 164.91,
165.78, 4C, CO.
2,3,4-Tri-O-benzoyl-1,6-dideoxy-1,6-epithio-
ꢀ-D-glucose (S)-S-Oxide (17)
meta-Chloroperbenzoic acid (85%, 265 mg, 1.30 mmol) was added
to an ice-cold solution of the tribenzoate (16) (491 mg, 1.00 mmol) in
dry CH2Cl2 (20 ml). The mixture was stirred under argon for 5 min. The
mixture was treated to a workup (CH2Cl2) as for the preparation of (10)
above to afford a pale yellow oil. The oil was taken up in CH2Cl2 and
crystallized upon addition of petrol to give the sulfoxide (17) as pale
yellow needles (498 mg, 98%), m.p. 126–128° (CH2Cl2/petrol), [ꢁ]D
–50.5°. 1H n.m.r. (300 MHz) ꢃ 3.06, ddd, J4,6 0.93, J5,6 8.5, J6,6 13.8 Hz,
H6; 4.14, d, H6; 4.93, br s, 1H, 5.36–5.39, m, H2,3,4; 5.50, br d, H5;
5.66, br s, H 1; 7.36–7.69, 8.04–8.12, 2m, 15H, Ph. 13C n.m.r.
(75.5 MHz) ꢃ 59.12, C 6; 65.68, 67.46, 68.02, 78.64, C 2,3,4,5; 100.29,
C 1; 127.94–130.08, 133.77, 134.33, Ph; 163.66, 164.84, 165.05, 3C,
CO. High-resolution mass spectrum (f.a.b.) m/z 507.1163 (C27H23O8S
[(M+H)+•] requires 507.1114).
Phenylsulfonyl Tetra-O-acetyl-ꢀ-D-glucopyranoside (12)
A solution of phenyl tetra-O-acetyl-1-thio-ꢀ-D-glucopyranoside
(8)38,39 (220 mg, 0.500 mmol) in acetone (2 ml) was treated with DMDO
(12 ml of 0.1 M in acetone, 1.2 mmol; 0°, 5 min). The solvent was evap-
orated under reduced pressure to afford a white, crystalline solid. This
solid was purified by flash chromatography (50% EtOAc/petrol) to
give the sulfone (12) as white needles (224 mg, 95%), m.p. 189–190°
(PriOH; lit.42 188–189°), [ꢁ]D –27.9° (lit.43 –26.9°).
Phenylsulfonyl Tetra-O-benzoyl-ꢀ-D-glucopyranoside (13)
The 1-thio-ꢀ-D-glucoside (11) (505 mg, 734 ꢄmol) was treated with
a solution of DMDO (30 ml of 0.1 M in acetone, 3 mmol; 0°, 10 min). At
this stage, t.l.c. indicated the formation of a new compound of slightly
higher polarity. The solvent was evaporated and the residue was dried
under vacuum to give a white, crystalline solid (535 mg). This material
was taken and recrystallized to give the sulfone (13) as fine needles,
m.p. 180–181.5° (CH2Cl2/petrol), [ꢁ]D +1.9° (Found: C, 66.9; H, 4.6.
2,3,4-Tri-O-acetyl-1,6-dideoxy-1,6-epithio-
ꢀ-D-glucose S,S-Dioxide (18)
(A) A solution of the triacetate (2) (304 mg, 1.00 mmol) and meta-
chloroperbenzoic acid (85%, 510 mg, 2.5 mmol) in dry CH2Cl2 (5 ml)
was allowed to stand at room temperature for 2 days. T.l.c. analysis
indicated a fast reaction, initially, to give the sulfoxide, followed by a
slow reaction to give a much less polar product. The mixture was
treated to a workup (CH2Cl2) as for the preparation of (10) above to
afford a solid residue that was purified by flash chromatography (75%
EtOAc/petrol) to give the sulfone (18) as a white, crystalline residue
(320 mg, 95%). A sample was recrystallized to give fine needles, m.p.
181–182° (EtOAc/petrol), [ꢁ]D –21.7° (Found: C, 43.1; H, 4.8; S, 9.9.
1
C40H32O11S requires C, 66.7; H, 4.5%). H n.m.r. (500 MHz) ꢃ 4.19,
ddd, J4,5 9.9, J5,6 2.7, 4.9 Hz, H 5; 4.38, dd, J6,6 12.4 Hz, H 6; 4.64, dd,
H6; 4.88, dd, J1,2 9.8 Hz, H 1; 5.50, dd, J2,3 9.7 Hz, H2; 5.78, dd, J3,4
9.8 Hz, H4; 5.92, dd, H3; 7.26–7.61, 7.77–8.00, 2m, Ph. 13C n.m.r.
(125.8 MHz) ꢃ 61.99, C 6; 67.63, 68.21, 73.56, 76.75, C 2,3,4,5; 89.21,
C 1; 128.32–130.44, 133.27–134.49, Ph; 164.93, 165.03, 165.72, 4C,
CO.
2,3,4-Tri-O-acetyl-1,6-dideoxy-1,6-epithio-ꢀ-D-glucose (R)-S-Oxide
(14) and 2,3,4-Tri-O-acetyl-1,6-dideoxy-1,6-epithio-
ꢀ-D-glucose (S)-S-Oxide (15)
1
C12H16O9S requires C, 42.9; H, 4.8; S, 9.5%). H n.m.r. (300 MHz) ꢃ
2.12, 2.14, 2.15, 3s, 9H, Me; 3.42–3.44, m, 2H, H6; 4.72, dd, J3,4 4.7,
J4,5 2.2 Hz, H 4; 4.80, br s, H1; 4.88–4.93, m, H 5; 5.08, br dd, J2,3 3.7
Hz, H3; 5.27, br d, H2. 13C n.m.r. (75.5 MHz) ꢃ 20.67, 20.81, 3C, Me;
50.43, C 6; 67.25, 67.96, 70.41, 76.63, C 2,3,4,5; 86.63, C 1; 169.08,
169.26, 170.11, 3C, CO. Mass spectrum (f.a.b.) m/z 337.40 (C12H17O9S
[(M+H)+•] requires 337.05).
(B) The triacetate (2) (76 mg, 0.25 mmol) was treated with DMDO
(20 ml of 0.1 M in acetone, 2 mmol; 0°, 20 min). The solvent was evap-
orated under reduced pressure to afford the sulfone (18) as a white,
crystalline solid (83 mg), identical by 1H n.m.r. (200 MHz) spec-
troscopy to the material prepared above.
meta-Chloroperbenzoic acid (85%, 530 mg, 2.6 mmol) was added in
two portions to a solution of the triacetate (2) (608 mg, 2.00 mmol) in
dry CH2Cl2 (10 ml) at 0° under N2. The mixture was stirred for 10 min,
then more meta-chloroperbenzoic acid (85%, 50 mg, 0.25 mmol) was
added. The mixture was stirred (10 min) and then treated to a workup
(CH2Cl2) as for the preparation of (10) above. Flash chromatography
(60–100% EtOAc, then 10% EtOH/EtOAc) of the residue, after the
removal of solvent, gave, firstly, the endo-sulfoxide (14) as needles
(44 mg, 7%), m.p. 119–120° (EtOAc/petrol), [ꢁ]D –149° (Found: C,
44.8; H, 5.3. C12H16O8S requires C, 45.0; H, 5.0%). 1H n.m.r.
(300 MHz) ꢃ 2.06, 2.09, 2.10, 3s, 9H, Me; 3.09, dd, J5,6 6.8, J6,6 13.5
Hz, H 6; 3.21, br d, H 6; 4.63, br d, H5; 4.78, dd, J3,4 8.5, J4,5 2.2 Hz,
H4; 5.44, dd, J2,3 7.3 Hz, H 3; 5.52, br s, H 1; 5.60, dd, J1,2 0.85 Hz, H 2.
13C n.m.r. (75.5 MHz) ꢃ 20.63, 20.78, 20.83, 3C, Me; 55.51, C 6; 65.47,
68.26, 73.38, 81.54, C 2,3,4,5; 92.95, C 1; 169.89, 169.93, 170.58, 3C,
CO.
1,6-Dideoxy-1,6-epithio-ꢀ-D-glucopyranose S,S-Dioxide (20)
The sulfone (18) (180 mg) was suspended in dry MeOH (5 ml) at
room temperature and a small piece of sodium metal was added. After
30 min, t.l.c. analysis indicated conversion into a high polarity com-
pound. Resin (Dowex 50, H+) was added until the solution became
neutral. The mixture was filtered and the residue evaporated to afford
the sulfone (20) as a white, crystalline solid (109 mg). A portion of this
solid was crystallized as hexagonal prisms, m.p. 208–209° (MeOH),
[ꢁ]D –40.6° (H2O) (Found: C, 34.7; H, 5.2. C6H10O6S requires C, 34.3;
H, 4.8%). 1H n.m.r. (300 MHz, D2O) ꢃ 3.34–3.35, m, 2H, H6;
3.52–3.93, m, H2,3,4; 4.77–4.80, m, H1,5. 13C n.m.r. (75.5 MHz, D2O)
ꢃ 49.80, C 6; 67.90, 70.28, 71.53, 78.70, C 2,3,4,5; 89.14, C 1. High-
resolution mass spectrum (f.a.b.) m/z 210.9755 (C6H11O6S [(M+H)+•]
requires 211.0276).
Next to elute was the exo-sulfoxide (15) as chunky crystals (538 mg,
84%), m.p. 122–123.5° (PriOH), [ꢁ]D –75.0° (Found: C, 45.0; H, 4.9.
1
C12H16O8S requires C, 45.0; H, 5.0%). H n.m.r. (300 MHz) ꢃ 2.02,
2.06, 2.07, 3s, 9H, Me; 2.78, ddd, J4,6 0.9, J5,6 8.5, J6,6 13.6 Hz, H 6;
3.85, d, H6; 4.39, br s, 1H, 4.67, m, 1H, 4.76, br s, H2,3,4; 5.14, br d,
H5; 5.26, br s, H 1. 13C n.m.r. (75.5 MHz) ꢃ 20.31, 20.38, 20.45, 3C,
Me; 58.46, C 6; 64.54, 66.47, 67.41, 78.33, C 2,3,4,5; 99.56, C 1;
167.44, 168.94, 169.26, 3C, CO.
1,6-Dideoxy-1,6-epithio-ꢀ-D-glucopyranose (S)-S-Oxide (6)
2,3,4-Tri-O-benzoyl-1,6-dideoxy-1,6-epithio-
ꢀ-D-glucose S,S-Dioxide (19)
A suspension of the triacetate (15) (34 mg) in dry MeOH (2 ml) was
treated with a small piece of sodium at room temperature for 1 h, afford-
ing a white precipitate. Water (2 ml) was added and the precipitate dis-
solved. Resin (Amberlite IR-120, H+) was added until the solution
became neutral and the solvent was evaporated to afford a white, crys-
talline solid (20 mg). Crystallization afforded the triol (6) as needles,
The tribenzoate (16) (49 mg, 100 ꢄmol) was treated with DMDO
(10 ml of 0.1 M in acetone, 1 mmol; 0°, 2 h). T.l.c. indicated a slow but
smooth conversion into a new product. The solvent was evaporated to
give a white, crystalline residue (52 mg). This material was recrystal-

395
1,6-Epithio and 1,6-Episeleno ꢀ-D-Glucopyranoses
lized to give the sulfone (19) as chunky prisms, m.p. 246–248°
(CHCl3/petrol), [ꢁ]D –26.4° (dimethyl sulfoxide, 5 ml cell) (Found: C,
62.0; H, 4.2. C27H22O9S requires C, 62.1; H, 4.2%). 1H n.m.r.
(300 MHz) ꢃ 3.55–3.66, m, 2H, H 6; 5.08, br s, H 1; 5.12, dd, J3,4 4.5,
J4,5 2.2 Hz, H 4; 5.16, m, H5; 5.73, br d, J2,3 4.0 Hz, H 2; 5.79, m, H 3;
7.39–8.13, m, Ph. 13C n.m.r. (75.5 MHz) ꢃ 50.53, C 6; 68.28, 68.39,
70.95, 76.63, C 2,3,4,5; 86.93, C 1; 128.60–130.15, 133.91, 133.99, Ph;
164.87, 164.95, 165.56, 3C, CO.
(6R)-6-O-Acetyl-2,3,4-tri-O-benzoyl-1,6-dideoxy-1,6-epithio-ꢀ-D-
gluco-hexodialdo-1,5-pyranose S,S-Dioxide (27)
The ꢁ-acetoxy sulfide (23) (55 mg, 100 ꢄmol) was treated with
DMDO (14 ml of 0.1 M in acetone, 1.4 mmol; room temperature, 90 min),
during which t.l.c. indicated a smooth conversion into one compound.
The solvent was evaporated to give a white, crystalline solid (56 mg).
Recrystallization gave the sulfone (27) as needles, m.p. 147–149°
(CH2Cl2/petrol), [ꢁ]D –80.8° (Found: C, 59.9; H, 4.4. C29H24O11S
requires C, 60.0; H, 4.2%). 1H n.m.r. (500 MHz) ꢃ 2.30, s, Me;
4.99–5.01, m, H5; 5.16, br s, H1; 5.25, dd, J3,4 5.9, J4,5 2.7 Hz, H 4; 5.77,
dd, J1,2 1.1, J2,3 4.8 Hz, H2; 5.93, br dd, H3; 5.95, br s, H6; 7.24–7.63,
8.03–8.16, 2m, 15H, Ph. 13C n.m.r. (125.8 MHz) ꢃ 20.02, Me; 67.84,
68.86, 69.54, 80.54, 80.91, 88.32, C1,2,3,4,5,6; 128.19–130.17, 133.96,
134.11, Ph; 164.93, 165.03, 165.60, 3C, COPh; 169.50, COMe.
(6R)-6-O-Acetyl-2,3,4-tri-O-benzoyl-1,6-dideoxy-1,6-epithio-ꢀ-D-
gluco-hexodialdo-1,5-pyranose (23)
A mixture of the sulfoxide (17) (1.01 g, 2.00 mmol) and sodium
acetate (492 mg, 6.00 mmol) in Ac2O (25 ml) was heated under reflux
under N2 for 12 h. The solvent was evaporated under reduced pressure
to give a dark oil that was subjected to the usual workup (EtOAc). The
residual oil was purified by flash chromatography (0–5%
EtOAc/toluene) to give the ꢁ-acetoxy sulfide (23) as a clear oil (1.00 g,
91%) which crystallized. A sample was recrystallized to give white
needles, m.p. 123–125° (CH2Cl2/petrol), [ꢁ]D –143.3° (Found: C, 63.3;
6-O-(Tetra-O-benzoyl-ꢀ-D-glucopyranosyl)-1,2:3,4-di-O-
isopropylidene-ꢁ-D-galactose (29)
Triflic anhydride (39 ꢄl, 240 ꢄmol) was added dropwise over 3 min
to a suspension of the sulfoxide (10) (169 mg, 240 ꢄmol) and the
alcohol (28) (52 mg, 200 ꢄmol) in dry Et2O (4 ml) at 0° under N2. After
5 min, t.l.c. indicated the absence of the starting materials, so the solu-
tion was quenched by the addition of saturated aqueous NaHCO3 solu-
tion (1 ml) and the mixture stirred for 5 min. The organic layer was
separated, washed with brine and dried. The solvent was evaporated
and the residue purified by flash chromatography (10–20%
EtOAc/toluene). The residue from chromatography was treated with
Ac2O (1 ml) and pyridine (2 ml) at room temperature for 1 h. The
solvent was then evaporated and the residue purified by flash chro-
matography (5–10% EtOAc/toluene) to give the disaccharide (29) as a
clear oil (130 mg, 78%), identical by 1H n.m.r. (300 MHz) spectroscopy
to that reported in the literature.44
1
H, 4.4. C29H25O9S requires C, 63.5; H, 4.4%). H n.m.r. (300 MHz) ꢃ
2.12, s, Me; 5.02, br d, J4,5 2.0 Hz, H 5; 5.09, dd, J3,4 5.1 Hz, H 4; 5.15,
br d, J2,3 3.8 Hz, H 2; 5.77, br t, H 3; 5.86, br s, H1; 6.50, br s, H6;
7.41–7.62, 8.07–8.11, 2m, Ph. 13C n.m.r. (75.5 MHz) ꢃ 20.92, Me;
69.16, 69.96, 74.07, 82.56, 83.13, 84.36, C 1,2,3,4,5,6; 128.55–130.01,
133.66–133.76, Ph; 164.95, 165.47, 165.61, 3C, COPh; 170.17,
COMe.
(6R)-2,3,4,6-Tetra-O-acetyl-1,6-dideoxy-1,6-epithio-ꢀ-D-gluco-
hexodialdo-1,5-pyranose (24)
The sulfoxide (15) (89 mg, 280 ꢄmol) was treated in the same
manner (reflux, 8 h) as for the sulfoxide (17) to give, after flash chro-
matography (30–40% EtOAc/petrol), the tetraacetate (24) as a pale
yellow oil (89 mg, 88%), [ꢁ]D –218° (Found: C, 46.6; H, 4.8.
1,2:3,4-Di-O-isopropylidene-6-O-trifluoromethylsulfonyl-ꢁ-D-
galactose (30)
1
Triflic anhydride (29 ꢄl, 280 ꢄmol) was added dropwise to an ice-
cold suspension of the sulfoxide (17) (122 mg, 240 ꢄmol) and the
alcohol (28) (52 mg, 200 ꢄmol) in dry Et2O (5 ml) at 0° under N2. After
5 min, the mixture was quenched by the addition of saturated aqueous
sodium bicarbonate solution (1 ml) and the mixture stirred for 5 min.
The mixture was diluted with CH2Cl2 and the organic layer separated
and washed with brine, then dried. The solvent was evaporated and the
residue crystallized from CH2Cl2/petrol to give the unreacted sulfoxide
(17) as needles (73 mg, m.p. 128–131°). The mother liquor was evapo-
rated and the residue was purified by flash chromatography (50%
toluene/petrol, then 0–10% EtOAc/toluene) to give the D-galacto tri-
C14H18O9S requires C, 46.4; H, 5.0%). H n.m.r. (500 MHz) ꢃ 2.06,
2.07, 2.11, 2.11, 4s, 12H, Me; 4.62, dd, J3,4 6.1, J4,5 2.1 Hz, H4;
4.69–4.71, m, H 2,5; 5.12, br dd, J2,3 4.3 Hz, H 3; 5.53, br s, H1; 6.24,
br s, H 6. 13C n.m.r. (125.8 MHz) ꢃ 20.70, 20.81, 20.84, 4C, Me; 69.38,
69.89, 74.75, 82.43, 83.02, 84.65, C 1,2,3,4,5,6; 169.38, 169.95,
170.00, 170.21, 4C, CO.
(6R)-6-O-Acetyl-2,3,4-tri-O-benzoyl-1,6-dideoxy-1,6-epithio-
ꢀ-D-gluco-hexodialdo-1,5-pyranose (S)-S-Oxide (25) and
(6R)-6-O-Acetyl-2,3,4-tri-O-benzoyl-1,6-dideoxy-1,6-epithio-
ꢀ-D-gluco-hexodialdo-1,5-pyranose (R)-S-Oxide (26)
flate (30) as a clear oil (54 mg, 69%), [ꢁ]D –46.1° (lit.31 –49.9°). H
1
meta-Chloroperbenzoic acid (80%, 58 mg, 280 ꢄmol) was added to
a solution of the ꢁ-acetoxy sulfide (23) (110 mg, 200 ꢄmol) in dry
CH2Cl2 (5 ml) at 0°. After 10 min, the mixture was treated to a workup
(CH2Cl2) as for the preparation of (10) above. The residue, after the
removal of solvent, was purified by flash chromatography (0–60%
EtOAc/toluene) to give, firstly, unreacted ꢁ-acetoxy sulfide (23)
(15 mg).
Next to elute was the endo-sulfoxide (25) as a glass (18 mg, 16%),
[ꢁ]D –204° (Found: C, 61.8; H, 4.5. C29H24O10S requires C, 61.7; H,
4.3%). 1H n.m.r. (500 MHz) ꢃ 2.22, s, Me; 4.76, br d, J4,5 2.8 Hz, H5;
5.25, dd, J3,4 8.9 Hz, H 4; 5.85, br s, H 1; 6.00, d, J2,3 7.5 Hz, H2; 6.24,
dd, H 3; 6.27, br s, H6; 7.13–7.61, 7.98-8.09, 2m, 15H, Ph. 13C n.m.r.
(125.8 MHz) ꢃ 20.53, Me; 66.17, 68.13, 71.04, 84.16, 93.11, 95.00,
C 1,2,3,4,5,6; 128.20–133.91, Ph; 165.55, 165.64, 166.08, 3C, COPh;
169.47, COMe.
n.m.r. ꢃ 1.33, 1.44, 1.52, 3s, 12H, Me; 4.11, ddd, J4,5 2.0, J5,6 4.6, 7.5
Hz, H5; 4.24, dd, J3,4 7.8 Hz, H 4; 4.35, dd, J1,2 5.0, J2,3 2.6 Hz, H2;
4.58, dd, J6,6 10.7 Hz, H 6; 4.63, dd, H6; 4.65, dd, H3; 5.53, d, H1. 13
C
n.m.r. ꢃ 24.34, 24.81, 25.81, 25.89, 4C, Me; 66.04, 70.20, 70.35, 70.62,
C 2,3,4,5; 74.64, C 6; 96.09, C 1; 109.10, 110.10, 2C, CMe2; 118.56, q,
JC,F 320 Hz, CF3.
2,3,4-Tri-O-benzoyl-1,6-dideoxy-1,6-epithio-ꢀ-D-glucose (16)
Triflic anhydride (39 ꢄl, 280 ꢄmol) was added dropwise over 4 min
to an ice-cold solution of the sulfoxide (17) (122 mg, 240 ꢄmol), the
alcohol (28) (52 mg, 200 ꢄmol) and Bu4NI (148 mg, 400 ꢄmol) in dry
1,2-dichloroethane (5 ml) at 0° under N2. The solution immediately
became very dark and was allowed to warm to room temperature. After
5 min, t.l.c. indicated a large amount of unreacted starting material.
More triflic anhydride (39 ꢄl, 280 ꢄmol) was added, causing the dis-
appearance of all starting material (t.l.c.). The mixture was quenched by
the addition of saturated aqueous NaHCO3 solution (2 ml) and aqueous
sodium thiosulfate solution (0.5 M, 5 ml). The mixture was stirred for 5
min, then the organic layer was separated and dried. The solvent was
evaporated and the residue purified by flash chromatography (0–15%
EtOAc/toluene) to give the tribenzoate (16) as a clear oil (30 mg, 26%
based on the sulfoxide), identical by 1H n.m.r. (200 MHz) spectroscopy
to that reported in the literature.45
Last to elute was the exo-sulfoxide (26) as needles (80 mg, 71%),
m.p. 200–202° (decomposition) (CH2Cl2/petrol), [ꢁ]D –129° (Found: C,
62.1; H, 4.4. C29H24O10S requires C, 61.7; H, 4.3%). 1H n.m.r.
(500 MHz) ꢃ 2.31, s, Me; 5.16, 5.25, 5.41, 5.54, 4 br s, 4H; 5.46–5.52,
m, H 1,2,3,4,5; 6.34, br s, H6; 7.14–7.70, 8.07–8.16, 2m, 15H, Ph. 13
C
n.m.r. (125.8 MHz) ꢃ 20.00, Me; 65.62, 66.89, 66.95, 80.81, 86.35,
99.06, C 1,2,3,4,5,6; 128.20–133.91, Ph; 163.54, 164.80, 165.02, 3C,
COPh; 169.75, COMe.

396
B. W. Skelton et al.
9.3 Hz, H2; 5.23, t, J2,3 Ϸ J3,4 9.3 Hz, H 3; 5.71, d, H 1. 13C n.m.r.
(125.8 MHz) ꢃ 20.55, 20.56, 20.69, 20.76, 4C, Me; 41.43, C 6; 70.29,
70.72, 72.67, 73.01, C 2,3,4,5; 91.60, C 1; 168.86, 169.24, 169.59,
170.06, 4C, CO.
6,6Ј-O-[6,6Ј-Dithiobis(2,3,4-tri-O-benzoyl-6-deoxy-ꢀ-D-
glucosyl)]bis(1,2:3,4-di-O-isopropylidene-ꢁ-D-galactose) (32)
Triflic anhydride (78 ꢄl, 560 ꢄmol) was added dropwise over 4 min
to an ice-cold suspension of the sulfoxide (17) (122 mg, 240 ꢄmol) and
Bu4NI (148 mg, 400 ꢄmol) in dry 1,2-dichloroethane (3 ml) at 0° under
N2. The solution immediately became very dark. After 5 min, an ice-
cold solution of the alcohol (28) (52 mg, 200 ꢄmol) in dry 1,2-
dichloroethane (4 ml) was added. After 5 min, t.l.c. indicated the
absence of the alcohol. The mixture was quenched by the addition of
saturated aqueous sodium bicarbonate solution (2 ml) and aqueous
sodium thiosulfate solution (0.5 M, 10 ml). The mixture was stirred for
5 min, then the organic layer was separated and dried. The solvent was
evaporated and the residue purified by flash chromatography (0–20%
EtOAc/toluene). The residue was treated with pyridine (2 ml) andAc2O
(1 ml) at room temperature for 2 h, then the solvent was evaporated
under reduced pressure to give an oil which was purified by flash chro-
matography (10–20% EtOAc/toluene) to give the disulfide (32) as a
clear oil (44 mg, 29%), identical by 1H n.m.r. (200 MHz) spectroscopy
to that reported in the literature.36
2,3,4-Tri-O-acetyl-6-deoxy-6-thiocyanato-ꢀ-D-glucosyl
Isothiocyanate (39)
A solution of 2,3,4-tri-O-acetyl-6-deoxy-6-thiocyanato-ꢁ-D-gluco-
syl bromide (38)46 (1.01 g, 2.46 mmol) and KSCN (748 mg, 7.71 mmol)
in N,N-dimethylformamide (20 ml) was heated at 65° under N2 for 2 h.
The solvent was evaporated and the residue was partitioned between
EtOAc and water. The organic phase was separated and washed sequen-
tially with water and brine then dried (MgSO4). The solvent was evap-
orated and the residue purified by flash chromatography (15%
EtOAc/toluene) to afford a pale yellow solid. The solid was recrystal-
lized to afford the isothiocyanate (39) as colourless plates (298 mg,
30%), m.p. 108–110° (EtOH), [ꢁ]D +91.8° (Found: C, 43.2; H, 4.4.
C14H16N2O7S2 requires C, 43.3; H, 4.2%). I.r. (NaCl) ꢅmax 2157 (sharp,
1
moderate SCN), 2023 (br, strong NCS), 1754 (strong CO) cm–1. H
n.m.r. (300 MHz) ꢃ 2.01, 2.07, 2.10, 3s, 9H, Me; 2.99, dd, J5,6 8.1, J6,6
14.1 Hz, H6; 3.23, dd, J5,6 2.3 Hz, H6; 3.83, ddd, J4,5 9.8 Hz, H5; 4.98,
t, J3,4 Ϸ J4,5 9.8 Hz, H4; 5.03–5.12, m, 2H, H1,2; 5.18–5.26, m, H3.
13C n.m.r. (75.5 MHz) ꢃ 20.44, 20.53, 4C, Me; 35.16, C 6; 70.30, 71.77,
71.91, 74.26, C 2,3,4,5; 83.36, C 1; 111.35, SCN; 145.12, NCS; 168.96,
169.54, 169.85, 3C, CO.
Tetra-O-acetyl-6-deoxy-6-thio-ꢀ-D-glucopyranose (35) and
6,6Ј-Dithiobis(1,2,3,4-tetra-O-acetyl-ꢀ-D-glucose) (36)
A
mixture of tetra-O-acetyl-6-deoxy-6-thiocyanato-ꢀ-D-gluco-
pyranose (34)46 (285 mg, 733 ꢄmol) and benzyltriethylammonium
tetrathiomolybdate47 (1.34 g, 2.20 mmol) in dry MeCN (10 ml) was
stirred at room temperature under N2 overnight. The solvent was evap-
orated and the dark residue was slurried with CH2Cl2 and filtered
through a plug of silica (50% EtOAc/petrol). The solvent was evapo-
rated and the residue purified by flash chromatography (15–40%
EtOAc/petrol) to give, firstly, the thiol (35) as a powder (71 mg, 27%),
Structure Determinations of (20) and (22)
Unique room-temperature single counter/‘four-circle’ diffractome-
ter data sets were measured (2ꢆ/ꢆ scan mode; monochromatic Mo Kꢁ
radiation, ꢇ 0.71073 Å; T c. 295 K) yielding N independent reflections,
No with I > 3(I) ‘observed’ absorption corrected reflections being used
in the full-matrix least-squares refinements, anisotropic thermal para-
m.p. 100–103° (MeOH; lit.46 108°), [ꢁ]D +9.3° (lit.46 +9°). H n.m.r.
1
(500 MHz) ꢃ 1.74, dd, J6,SH 7.4, 9.6 Hz, SH; 2.00, 2.02, 2.04, 2.11, 4s,
12H, Me; 2.61, ddd, J5,6 6.6, J6,6 14.3 Hz, H 6; 2.70, ddd, J5,6 3.0 Hz,
H6; 3.73, ddd, J4,5 9.7 Hz, H 5; 5.11, m, H 2,4; 5.24, t, J2,3 Ϸ J3,4 9.5 Hz,
H3; 5.70, d, H 1. 13C n.m.r. (125.8 MHz) ꢃ 20.54, 20.55, 20.63, 20.79,
4C, Me; 25.73, C 6; 70.15, 70.30, 72.76, 74.88, C 2,3,4,5; 91.64, C 1;
168.98, 169.23, 169.50, 170.10, 4C, CO.
meter forms being refined for the non-hydrogen atoms; (x, y, z, Uiso
)
H
were refined for (20) and constrained at estimated values (hydroxy H
located in difference maps) for (22), in which two molecules comprise
the asymmetric unit. Conventional residuals R, Rw [statistical weights,
derivative of ꢈ2(I) = ꢈ2(Idiff) + 0.0004ꢈ4(Idiff)] on |F| are quoted; neutral
atom complex scattering factors were employed, computation using the
Xtal 3.4 program system.49 Pertinent results are given in Fig. 1 and
Tables 1 and 2. Full details of atom coordinates, thermal parameters,
non-hydrogen geometries, and structure factor amplitudes have been
deposited.† Chiralities as determined crystallographically are concor-
dant with those expected from the chemistry.
Next to elute was the disulfide (36) as a microcrystalline powder
(105 mg, 39%), m.p. 162–164° (MeOH; lit.48 164–165°), [ꢁ]D +16.7°
1
(lit.48 +15.2°). H n.m.r. (500 MHz) ꢃ 2.00, 2.02, 2.05, 2.12, 4s, 12H,
Me; 2.86, dd, J5,6 7.2, J6,6 14.2 Hz, H 6; 2.98, dd, J5,6 3.4 Hz, H6; 3.88,
ddd, J4,5 9.3 Hz, H5; 5.02, t, J3,4 Ϸ J4,5 9.3 Hz, H4; 5.11, dd, J1,2 8.3, J2,3
Table 2. Ring torsion angles (degrees) of glucosan derivatives (1), (6), (20), (21), (21*) and (22)
Atoms are denoted by number only, heteroatoms are italicized. Compound (21*) is the triacetate of (21)
Atoms
(21)A,B
(21*)C
(1)D
(6)E
(20)F
(22)F
–25(1),
–37(1),
52(1),
–5(1),
–58.4(9),
75.4(8),
5–1–2–3
1–2–3–4
2–3–4–5
3–4–5–5
4–5–5–1
5–5–1–2
–57.5(3), –57.3(3)
35.2(3), 35.3(3)
–35.1(3), –35.2(3)
55.8(3), 55.8(2)
–75.3(3), –75.4(2)
75.8(3), 76.0(2)
–53.7(4)
30.3(4)
–32.1(4)
55.7(4)
–75.4(3)
75.3(3)
–53.5(2)
37.2(2)
–39.5(2)
57.5(2)
–73.5(2)
71.3(2)
–55.3(4), –57.6(3)
42.9(4), 43.1(4)
–44.6(4), –42.8(4)
59.2(4), 54.8(4)
–71.8(4), –68.8(4)
70.1(4), 70.8(4)
–47.9(3)
32.0(3)
–36.9(3)
55.9(3)
–71.1(3)
67.3(3)
–56(1)
37(1)
–37(1)
54(1)
–7(1)
73(1)
1–1–5–5
1–5–5–6
5–5–6–1
5–6–1–1
6–1–1–5
–43.6(7), –43.3(2)
43.9(3), 43.9(2)
–28.6(3), –28.7(2)
–43.2(3)
44.6(3)
–30.4(3)
4.7(3)
–50.7(2)
50.8(2)
–26.7(2)
–1.4(7)
29.4(2)
–51.2(3), –51.2(3)
51.6(4), 53.8(4)
–27.6(3), –30.8(4)
–54.2(2)
51.2(3)
–21.0(3)
–8.0(2)
35.4(2)
–44.8(8) –52.5(7)
63.5(9), 54.3(9)
–47.1(9), –29.4(9)
3.1(3),
3.3(3)
–0.7(3),
2.6(3)
18.5(8),
0.9(6)
24.8(3) 24.5(3)
23.4(3)
28.2(3), 26.8(3)
14.1(7), 28.7(6)
1–1–2–3
2–1–1–6
3–4–5–6
4–5–6–1
58.9(3), 58.5(3)
–94.1(3), –93.9(3)
–56.4(3), –56.3(3)
88.5(3), 88.1(3)
62.7(3)
–96.4(3)
–56.8(3)
86.6(3)
63.3(2)
–90.0(2)
–62.2(2)
93.5(2)
60.8(3), 59.8(4)
–91.0(3), –94.0(3)
–59.3(4), –62.9(4)
92.1(3), 89.6(3)
65.2(3)
–85.1(2)
–62.6(3)
98.8(3)
91.2(9), 64(1)
–105.3(8), –92.9(9)
–12.3(1),
74(1),
–67(1)
92(1)
A Two determinations. B Ref. 15 and Lindberg, K. B., Acta Chem. Scand., Ser. A, 1974, 28, 1181. C Leung, F., and Marchessault, R. H., Can. J. Chem.,
1974, 52, 2516. D Ref. 16. E Ref. 7; molecules 1 and 2. F This work; molecules 1 and 2 for compound (22).
† Copies are available, until 31 December 2005, on application to the Australian Journal of Chemistry, P.O. Box 1139, Collingwood, Vic. 3066.

397
1,6-Epithio and 1,6-Episeleno ꢀ-D-Glucopyranoses
18
19
20
21
Crystal/Refinement Data
Pummerer, R., Chem. Ber., 1909, 42, 2282.
Pummerer, R., Chem. Ber., 1910, 43, 1401.
Ferrier, R. J., and Furneaux, R. H., Aust. J. Chem., 1980, 33, 1025.
Kahne, D., Walker, S., Cheng, Y., and Van Engen, D., J. Am. Chem.
Soc., 1989, 111, 6881.
Yan, L., and Kahne, D., J. Am. Chem. Soc., 1996, 118, 9239.
Crich, D., and Sun, S., J. Am. Chem. Soc., 1997, 119, 11217.
Raghavan, S., and Kahne, D., J. Am. Chem. Soc., 1993, 115,
1580.
Gildersleeve, J., Pascal, R. A., Jr, and Kahne, D., J. Am. Chem. Soc.,
1998, 120, 5961.
Bamhaoud, T., Lancelin, J.-M., and Beau, J.-M., J. Chem. Soc.,
Chem. Commun., 1992, 1494.
Kim, S.-H., Augeri, D., Yang, D., and Kahne, D., J. Am. Chem. Soc.,
1994, 116, 1766.
Sliedregt, L. A. J. M., van der Marel, G. A., and van Boom, J. H.,
Tetrahedron Lett., 1994, 35, 4015.
Zhang, H., Wang, Y., and Voelter, W., Tetrahedron Lett., 1995, 36,
1243.
Alonso, I., Khiar, N., and Martín-Lomas, M., Tetrahedron Lett.,
1996, 37, 1477.
Barrette, E.-P., and Goodman, L., J. Org. Chem., 1984, 49, 176.
Ferrier, R. J., Furneaux, R. H., and Tyler, P. C., Carbohydr. Res.,
1977, 58, 397.
Brown, D. S., Ley, S. V., and Vile, S., Tetrahedron Lett., 1988, 29,
4873.
Brown, D. S., Ley, S. V., Vile, S., and Thompson, M., Tetrahedron,
1991, 47, 1329.
Prabhu, K. R., Ramesha, A. R., and Chandrasekaran, S., J. Org.
Chem., 1995, 60, 7142.
Marino, C., Varela, O., and de Lederkremer, R. M., Carbohydr. Res.,
1997, 304, 257.
McAuliffe, J. C., and Stick, R. V., Aust. J. Chem., 1997, 50, 193.
Dasgupta, F., and Garegg, P. J., Acta Chem. Scand., 1989, 43, 471.
Veeneman, G. H., van Leeuwen, S. H., and van Boom, J. H.,
Tetrahedron Lett., 1990, 31, 1331.
Kakarla, R., Dulina, R. G., Hatzenbuhler, N. T., Hui, Y. W., and
Sofia, M. J., J. Org. Chem., 1996, 61, 8347.
Ferrier, R. J., and Furneaux, R. H., Carbohydr. Res., 1976, 52, 63.
Collins, P. M., and Whitton, B. R., J. Chem. Soc., Perkin Trans. 1,
1974, 1069.
Compound (20). C6H10O6S, M 210.2. Orthorhombic, space group
P212121 (D42, No. 19), a 10.867(2), b 10.463(2), c 6.965(2) Å, V 792.0
Å3. Dc(Z = 4) 1.763 g cm–3; F(000) 440. ꢄMo 4.1 cm–1; specimen: 0.25
by 0.30 by 0.50 mm; A*
1.10, 1.14 (Gaussian correction). 2ꢆmax
min, max
22
23
24
55°; N 1030, No 976; R 0.028, Rw 0.036 (preferred hand).
Compound (22). C6H10O4Se, M 225.1. Monoclinic, space group
P21 (C22, No. 4), a 9.357(4), b 6.660(4), c 12.593(3) Å, ꢀ 106.28(3)°, V
792.3 Å3. Dc(Z = 4) 1.985 g cm–3; F(000) 448. ꢄMo 49 cm–1; specimen:
0.20 by 0.16 by 0.22 mm; A*min, max 2.4, 2.9 (analytical correction). 2ꢆmax
60°; N 2359, No 1518; R 0.050, Rw 0.046 (preferred hand).
25
26
27
28
29
30
Hydrogen Bonding
Hydroxyl assignments are complemented by observed/implied
hydrogen-bonding interactions: in (20), O(2),H0(2)···O(11) (x – ½, ½ –
y, 1 – z) are 2.814(3), 2.24(3); O(3)···O(4) (1 – x, 1½ + y, 1½ – z)
2.901(3), 2.26(4); O(4),H0(4)···O(2) (1½ – x, 1 – y, ½ + z) 2.800(3),
1.92(4); in (22), O(12),H0(12)···O(24) (x–1, 1½ + y, z) 2.86(1), 2.5 (est.);
O(14)···O(23) 2.80(1) (1.9); in (24), H0(24)···O(22) (1 – x, ½ + y, 1 – z)
2.798(8) (2.1) Å, for O···O < 3 Å.
-
-
31
32
References
1
33
34
35
36
Akagi, M., Tejima, S., and Haga, M., Chem. Pharm. Bull., 1963, 11,
58.
2
Whistler, R. L., and Seib, P. A., Carbohydr. Res., 1966, 2, 93.
Lundt, I., and Skelbæk-Pedersen, B., Acta Chem. Scand., Ser. B,
3
1981, 35, 637.
4
Driguez, H., McAuliffe, J. C., Stick, R. V., Tilbrook, D. M. G., and
Williams, S. J., Aust. J. Chem., 1996, 49, 343.
Klayman, D. L., and Griffin, T. S., J. Am. Chem. Soc., 1973, 95, 197.
Lowary, T. L., and Bundle, D. R., Tetrahedron: Asymmetry, 1994, 5,
5
6
37
38
39
2397.
7
Novotny, J., Podlaha, J., and Cerny, M., Collect. Czech. Chem.
Commun., 1991, 56, 2169.
Adam, W., and Hadjiarapoglou, L., Top. Curr. Chem., 1993, 164, 45.
Gervay, J., Flaherty, T. M., and Holmes, D., Tetrahedron, 1997, 53,
8
40
9
41
42
16355.
10
Berkowitz, D. B., Danishefsky, S. J., and Schulte, G. K., J. Am.
Chem. Soc., 1992, 114, 4518.
Praly, J.-P., and Descotes, G., Tetrahedron Lett., 1990, 31, 1133.
Crich, D., Mataka, J., Sun, S., Lam, K.-C., Rheingold, A. L., and
11
43
44
45
46
47
48
49
Bonner, W. A., and Drisko, R. W., J. Am. Chem. Soc., 1948, 70,
2435.
Garegg, P. J., and Norberg, T., Acta Chem. Scand., Ser. B,1979, 33,
116.
Stick, R. V., Tilbrook, D. M. G., and Williams, S. J., Aust. J. Chem.,
1999, 52, 685.
Stanek, J., and Tajmr, L., Collect. Czech. Chem. Commun., 1959, 24,
1013.
Ramesha, A. R., and Chandrasekaran, S., Synth. Commun., 1992,
22, 3277.
‘Beilsteins Handbuch der Organischen Chemie’, Vol. 17, IV, p. 3758
(Springer: Berlin 1975).
Hall, S. R., King, G. S. D., and Stewart, J. M., (Eds) ‘Xtal 3.4
Users Manual’ (The University of Western Australia, Lamb: Perth
1995).
12
Wink, D. J., J. Chem. Soc., Chem. Commun., 1998, 2763.
Khiar, N., Alonso, I., Rodriguez, N., Fernandez-Mayoralas, A.,
13
Jiménez-Barbero, J., Nieto, O., Cano, F., Foces-Foces, C., and
Martin-Lomas, M., Tetrahedron Lett., 1997, 38, 8267.
Foster, A. B., Duxbury, J. M., Inch, T. D., and Webber, J. M., J.
14
Chem. Soc., Chem. Commun., 1967, 881.
Park, Y. J., Kim, H. S., and Jeffrey, G. A., Acta Crystallogr., Sect. B,
15
1971, 27, 220.
16
Takagi, S., and Jeffrey, G. A., Acta Crystallogr., Sect. B, 1978, 34,
816.
17
Kennedy, M., and McKervey, M. A., ‘Oxidation Adjacent to Sulfur’
in ‘Comprehensive Organic Synthesis’ (Eds B. M. Trost and I.
Fleming) Vol. 7, p. 193 (Pergamon: Oxford 1991).