Reaction kinetics and mechanism of sulfuric acid-catalyzed acetolysis of acylated methyl l-ribofuranosides

Article abstract of DOI:10.1002/ejoc.200900889

The mechanism of the sulfuric acid-catalyzed acetolysis of methyl 2,3,5-tri-O-acetyl- and methyl 2,3,5-tri-O-benzoyl-Lribofuranosides and the accompanying anomerization of both the starting material and the 1,2,3,5-tetra-O-acetyl- and 1-O-acetyl-2,3,5-tri-O-benzoyl-L-ribofuranoses formed was investigated. The progress of the reactions was followed by ¹H NMR spectroscopy and the rate constants for the reactions were determined for a proposed kinetic model. The role of H⁺ and Ac ⁺ as the catalytically active species was clarified, proving that the anomerization of the acylated methyl furanosides is activated by protonation, while, on the contrary, the anomerization of the 1-O-acetyl ribofuranoses is activated by the acetyl cation. The anomerization of the acylated methyl furanosides was verified to be activated on the ring oxygen leading to endocyclic CO-bond rupture while the 1O-acetyl ribofuranoses are activated on the acetyloxy group on C(1) leading to exocyclic cleavage.

Full text of DOI:10.1002/ejoc.200900889

FULL PAPER
DOI: 10.1002/ejoc.200900889
Reaction Kinetics and Mechanism of Sulfuric Acid-Catalyzed Acetolysis of
Acylated Methyl -Ribofuranosides
L
Jonas J. Forsman,[a] Johan Wärnå,[b] Dmitry Yu. Murzin,[b] and Reko Leino*[a]
Keywords: Carbohydrates / Acetolysis / Kinetics / Reaction mechanisms / NMR spectroscopy
The mechanism of the sulfuric acid-catalyzed acetolysis of
methyl 2,3,5-tri-O-acetyl- and methyl 2,3,5-tri-O-benzoyl-
ribofuranosides and the accompanying anomerization of
both the starting material and the 1,2,3,5-tetra-O-acetyl- and
1-O-acetyl-2,3,5-tri-O-benzoyl-L-ribofuranoses formed was
investigated. The progress of the reactions was followed by
1H NMR spectroscopy and the rate constants for the reactions
were determined for a proposed kinetic model. The role of
H+ and Ac+ as the catalytically active species was clarified,
proving that the anomerization of the acylated methyl fur-
anosides is activated by protonation, while, on the contrary,
the anomerization of the 1-O-acetyl ribofuranoses is acti-
vated by the acetyl cation. The anomerization of the acylated
methyl furanosides was verified to be activated on the ring
oxygen leading to endocyclic CO-bond rupture while the 1-
O-acetyl ribofuranoses are activated on the acetyloxy group
on C(1) leading to exocyclic cleavage.
L-
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
been carried out on furanoses and most of them cover the
acid-catalyzed hydrolysis of alkyl furanosides.[9] Very few
investigations exist on the acetolysis of furanoses[10] and
even fewer targeting the mechanistical issues.[11]
In the current work, we report our results on the mecha-
nism of the acetolysis and concurrent anomerization of
1,2,3,5-tetra-O-acetyl- and 1-O-acetyl-2,3,5-tri-O-benzoyl-
-ribofuranoses as followed by in situ 1H NMR spec-
troscopy. The reactions were carried out in a mixture of
acetic acid and acetic anhydride with a catalytic amount of
sulfuric acid. The rate constants on the reactions involved
have been calculated and are in good agreement with the
mechanisms proposed in the present work.
Introduction
Acetolysis, as one of the most fundamental chemical re-
actions in carbohydrate chemistry, was the focus of exten-
sive research from the early 1950s until the late 1960s. Dur-
ing this period, several mechanistic proposals related to the
acid-catalyzed acetolysis and the accompanying anomeri-
zation of different alkyl glycopyranosides, still accepted to-
day, were established. The early work concerned both the
site of anomeric activation,[1,2] as well as the acid species
involved in the acetic acid/acetic anhydride mixtures
used.[3,4] Miljkovic´ has recently investigated the electronic
effects influencing the acetolysis of pyranosides.[5] Kaczma-
rek and co-workers have likewise published on the mecha-
nism of the acetolysis of gluco- and mannopyranosides as
a method for degradation of polysaccharides.[6]
During the past 30 years much research has been focused
on the synthesis of nucleoside analogues for the use in anti-
viral therapy and several examples on the modification of
both - as well as -furanoses have been reported.[7] Since
one of the most widely used protocols in nucleoside synthe-
sis involves the use of acylated 1-acetyloxy sugars (the
Vorbrüggen glycosylation),[8] the acetolysis of methyl fur-
anosides is a key step in the chemical manipulation of
sugars. Nevertheless, only a few mechanistical studies have
Results and Discussion
Elucidation of Reaction Mechanisms
The acetolysis reactions of methyl 2,3,5-tri-O-acetyl-α-
(1) and -β- (2) and the methyl 2,3,5-tri-O-benzoyl-α- (7)
and -β--ribofuranosides (8) were carried out at 25 °C in a
mixture of acetic acid and acetic anhydride (5:4 v/v) con-
taining 0.75% sulfuric acid. The progress of the reactions
was monitored by 1H NMR spectroscopy. The chemical
shifts of the anomeric signals arising from the different
compounds present in the reaction mixture clearly deviate
from each other, thus allowing the determination of the rel-
ative product concentrations by simple integration of the
[a] Laboratory of Organic Chemistry, Åbo Akademi University,
20500 Åbo, Finland
Fax: +358-2-2154866
E-mail: reko.leino@abo.fi
1
anomeric signals in the H NMR spectra.
[b] Laboratory of Industrial Chemistry and Reaction Engineering,
Åbo Akademi University,
Figure 1 adequately illustrates the results obtained dur-
ing a 48 h run using methyl 2,3,5-tri-O-acetyl-α--ribofur-
anoside (1) as starting material. An analogous mixture of
20500 Åbo, Finland
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900889.
5666
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5666–5676

Kinetics of the Acid-Catalyzed Acetolysis of Acylated -Ribofuranosides
Figure 1. Monitoring of the acetolysis of methyl 2,3,5-tri-O-acetyl-α--ribofuranoside (1) during 48 h at 25 °C in a mixture of acetic acid
and acetic anhydride (5:4 v/v) containing 0.75% sulfuric acid. For structures of the compounds, see Scheme 1.
products was obtained when the reactions were carried out
starting from compounds 2, 7 and 8 as well.
All compounds present in the acetolysis reactions were
synthesized following slightly modified literature pro-
cedures (Scheme 1). First, the methyl α- and β--ribofur-
anosides (13 and 14) were synthesized from -ribose in 26%
and 69% yields, respectively.[12] Next, the acetylated methyl
-ribofuranosides 1 and 2,[13] and their benzoylated ana-
logues 7 and 8 were prepared from the corresponding
methyl ribofuranosides by conventional esterification meth-
ods.[14] The 1,2,3,5-tetra-O-acetyl-β--ribofuranose (4) and
the 1-O-acetyl-2,3,5-tri-O-benzoyl-β--ribofuranose (10)
were prepared using standard acetolysis conditions and
were crystallized from EtOH and iPrOH, respectively. Fur-
ther, portions of compounds 4 and 10 were converted into
Scheme 1. Reagents and conditions: (i) MeOH, H2SO4, 13 (26%)
the corresponding α-anomers 3 and 9 by the use of zinc and 14 (69%); (ii) CH2Cl2, pyridine, AcCl, 2 (69%), CH2Cl2, pyr-
chloride in acetic anhydride.[15] The isomeric mixture of the
idine, BzCl, 8 (89%); (iii) Ac2O, I2, 1 (72%), CH2Cl2, pyridine,
BzCl, 7 (86%); (iv) AcOH, Ac2O, H2SO4, 4 (43%), 10 (49%); (v)
acyclic (1R)- and (1S)-1,2,3,4,5-penta-O-acetyl--ribose
Ac2O, ZnCl2, 3 (8%), 9 (28%); (vi) AcOH, Ac2O, ZnCl2, 5 (92%),
methyl hemiacetals [(1R)-5 and (1S)-5] and acyclic (1R)-
11 (71%); (vii) AcOH, Ac2O, H2SO4, 6 (77%), 12 (74%).
and
(1S)-1,4-di-O-acetyl-2,3,5-tri-O-benzoyl--ribose
methyl hemiacetals [(1R)-11 and (1S)-11] (henceforth lab-
eled as 5 and 11 in the text and Figures) were synthesized 1-O-acetyl-α- (3) and β--ribofuranose (4) is accompanied
from 2 and 8, respectively. Finally, 5 and 11 were converted by a preceding anomerization of the starting material. Since
into the acyclic 1,1,2,3,4,5-hexa-O-acetyl--ribose hydrate the anomerization is considerably faster than the subse-
(6) and 1,1,4-tri-O-acetyl-2,3,5-tri-O-benzoyl--ribose hy- quent acetolysis, an equilibrium mixture containing 25% of
drate (12) as presented in Scheme 1.[16] All compounds pre- the α-anomer (1) and 75% of the β-anomer (2) is formed,
pared were characterized by 1H and 2D NMR spectro- regardless of whether 1 or 2 is used as starting material. It
scopic techniques (for complete characterization, see the seems likely that the anomerization of the methyl ribofur-
Exp. Section).
anosides 1 and 2 in an acetic acid-acetic anhydride mixture
Recently, we reported on the acid-catalyzed anomeri- takes place via a ring-opening/ring-closing process, as the
zation of 1-O-acetyl-2,3,5-tri-O-benzoyl--ribofuranose[17] cleavage of the glycosidic bond would cause the methanol
and hence decided to use the same reaction conditions in released to compete with acetic acid for the cyclic oxocarb-
the present study in order to enable the utilization of our enium ion for the anomerization to proceed (Scheme 3).
earlier results in the current experiments. Both the reactions This assumption is also in agreement with the results ob-
from the acetylated (1 and 2) as well as the benzoylated tained on the acid-catalyzed anomerization and hydrolysis
derivatives (7 and 8) gave a reaction mixture containing of alkyl furanosides suggesting a pathway via acyclic inter-
seven compounds: the anomeric methyl furanosides, the mediate for methyl furanosides.[9]
anomeric 1-O-acetyl furanoses, the isomeric methyl hemi-
The formation of the isomeric acyclic methyl hemiacetals
acetals and the fully acylated ribose hydrate. No pyranoses 5 also proceeds via the same acyclic oxocarbenium ion in-
were detected which is in agreement with the kinetically termediate (C). Lindberg proposed that the anomerization
favored formation of five-membered rings.
and ring-opening is catalyzed by an acetyl cation, from the
The time-dependent product distribution for the aceto- reaction of acetic anhydride with sulfuric acid, which adds
lysis of the acetylated methyl furanosides 1 and 2 is plotted to the ring oxygen and activates the cleavage of the endo-
in Figure 2. As seen in the Figure, the acetolysis to afford cyclic C–O bond followed by ring-closing through nucleo-
Eur. J. Org. Chem. 2009, 5666–5676
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5667

J. J. Forsman, J. Wärnå, D. Yu. Murzin, R. Leino
FULL PAPER
Figure 2. Time-dependent product distribution for the acetolysis of a) methyl 2,3,5-tri-O-acetyl-α- (1) and b) -β--ribofuranoside (2) in a
5:4 (v/v) AcOH-Ac2O mixture catalyzed by 0.75% of H2SO4. Experimental and modeled results are presented as symbols and solid lines,
respectively.
philic attack by the acetylated oxygen.[3,18] Our previous fast enough even when the oxygen at C(4) is acetylated. The
study evidenced that the ring-opening process takes place fact that the anomerization of the acetylated methyl ribofu-
via protonation of the ring oxygen and that this acyclic in- ranosides 1 and 2 is much faster than the ring-closing of
termediate (C) with a free hydroxyl group at C(4) can either the acyclic intermediate in the acetolysis of 5 brings further
ring-close or be acetylated to form the acyclic compound evidence for the proposal that the ring oxygen is protonated
5.[17] This proposal is in accordance with the results of prior to endocyclic cleavage in the anomerization reaction.
Painter[19] and Miljkovic´[5a] indicating that the acetylated Ring-closing of the fully acetylated acyclic oxocarbenium
oxygen at C(4) is not a strong enough nucleophile for ring- ion cannot, however, be entirely ruled out, since some for-
closing.
mation of furanoses in the acetolysis of 5 was detected (see
When 5 was used as starting material for the acetolysis Supporting Information Figure S2). The rapid anomeri-
reaction under the same conditions, the acyclic 1,1,2,3,4,5- zation of 1 and 2 nevertheless excludes this from being the
hexa-O-acetyl--ribose hydrate (6) was formed as the main main mechanistical pathway.
product while 3 and 4 were obtained in only 1 and 4%
A widely accepted mechanism for the acid-catalyzed ace-
yields, respectively. The slow conversion of 5 into 6 under tolysis of alkyl glycosides in acetic acid-acetic anhydride
these conditions is also observed in the acetolysis of 1 and mixtures presented by Rosenfeld and Ballou suggests that
2. Kaczmarek et al. obtained analogous results from the the aglycon part is cleaved as the corresponding acetate.[20]
acetolysis of the fully acetylated -glucose ethyl and -mann- On the contrary, in our present study, the formation of
ose methyl hemi-acetals under similar conditions.[6] As the methyl hydrogen sulfate was detected in the early stage of
formation of 6 probably proceeds via an acyclic oxocarb- the acetolysis prior to the formation of methyl acetate. This
enium ion formed after cleavage of methanol, 3 and 4 are observation suggests that the glycosidic methoxy group is
expected as the main products if the ring closing would be protonated and cleaved off as methanol which immediately
5668
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5666–5676

Kinetics of the Acid-Catalyzed Acetolysis of Acylated -Ribofuranosides
reacts with sulfuric acid to form a sulfate ester. The sulfate
ester is then in turn slowly converted to methyl acetate as
illustrated in Figure 3.
Scheme 2. Formation of methyl hydrogen sulfate and methyl ace-
tate from the acylated methyl -ribofuranosides under standard
acetolysis conditions.
cyclic oxocarbenium ion intermediate (I), which in turn is
attacked by acetic acid to give the peracetylated ribofur-
anoses 3 and 4. Nevertheless, we do not question the pres-
ence of the acetyl cation in this type of acetylating mixtures,
neither do we question its role as a catalytically active
source. In our previous study,[17] we reported that the acetyl
cation is the source of catalytic activity in the anomeri-
zation of the benzoylated 1-O-acetyl ribofuranoses 9 and
10. We showed that under various reaction conditions there
is no change in the overall reaction path of the anomeriz-
Figure 3. 1H NMR monitoring of methoxy containing compounds
during acetolysis of methyl 2,3,5-tri-O-acetyl-α--ribofuranoside
(1) during 18 h at 25 °C in a mixture of acetic acid and acetic anhy-
dride (5:4 v/v) containing 0.75% sulfuric acid.
As the concentration of sulfuric acid in the reaction mix- ation as long as acetic anhydride, the source of the acetyl
ture is fairly low, it appears probable that the protonated cation, is present. Furthermore, we showed that the ano-
methoxy group forms an ion pair with sulfuric acid prior to merization of 9 and 10 exclusively proceeds via exocyclic
cleavage. The two ions then react to form methyl hydrogen cleavage of the glycosidic bond and introduction of a new
sulfate. Without this kind of coordination it seems rather acetoxy group to the cyclic oxocarbenium ion intermediate.
unlikely that the methanol released would predominantly
As seen in Figure 2, the β-anomer 4 is formed much
react with sulfuric acid in the presence of such an excess of faster than the α-anomer 3 while being formed via the same
acetic anhydride. The amount of methyl sulfate in the reac- cyclic intermediate (I). Further, it is observed that com-
tion mixture remains at a constant level slightly below the pound 4 reaches a maximum concentration before equilib-
total concentration of sulfuric acid as long as unreacted rium between the two anomers 3 and 4 is obtained in a 1:3
methyl furanosides are present. When all methyl furanos- ratio. This observation can be explained by the formation of
ides are consumed the concentration of methyl hydrogen a cis-fused dioxonium ion intermediate (J) via neighboring
sulfate starts to decrease and is converted to methyl acetate. group participation by the acetoxy group at C(2) after
In the 1940s already, Malm reported that, under similar dissociation of the glycosidic bond.[1,23] This results in the
conditions, sulfuric acid forms sulfate esters with the pri- intermediate being 1,2-trans directing, considering the nu-
mary hydroxyl groups of cellulose during the preparation of cleophilic attack by acetic acid leading to the fast formation
cellulose acetate and that the sulfate esters then are slowly of the β-anomer 4. It is also very likely that the trans-1,2-
converted into the final acetate via sulfate acetolysis in the configuration in the acetylated methyl β-ribofuranoside 2
same step.[21] This mechanism was more recently verified by activates the cleavage of the glycosidic bond through neigh-
Hyatt and Tindall on monosaccharide model com- boring group participation, but as the anomerization of
pounds.[22] The formation of sulfate esters of methanol has compounds 1 and 2 is evidently faster than the formation
not earlier been reported in the acetolysis of alkyl glycos- of 3 and 4, this phenomenon is undetectable. When the per-
ides, probably due to the need to follow the reaction in situ acetylated ribofuranoses 3 and 4 are treated under the same
in order to be able to detect the methyl hydrogen sulfate acetolysis conditions, anomerization of the starting material
formation. A proposed mechanism for the formation of the is observed giving the same final equilibrium as when the
methyl hydrogen sulfate and the subsequent formation of reaction was carried out from 1 or 2, as presented in the
methyl acetate is hereby presented in Scheme 2.
Supporting Information (Figure S1). Further, some forma-
Based on our findings on the formation of the methyl tion of the per-acetylated ribose hydrate (6) was observed
hydrogen sulfate, we suggest that the acetolysis of the which is in good agreement with the results on the anomeri-
methyl furanosides 1 and 2 occurs via protonation of the zation of the benzoylated ribofuranose acetates 9 and 10
exocyclic oxygen followed by methanol cleavage to form a under these conditions.[17]
Eur. J. Org. Chem. 2009, 5666–5676
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5669

J. J. Forsman, J. Wärnå, D. Yu. Murzin, R. Leino
FULL PAPER
When the acetolysis reactions were carried out starting even more distinct in the reactions of the benzoylated
from the methyl 2,3,5-tri-O-benzoyl-α- (7) and β--ribofur- methyl furanosides (7 and 8). Further, the difference in the
anoside (8), the reaction followed a similar pattern as for formation rate of the 1-acetoxy α-anomer and β-anomer is
the acetylated counterparts 1 and 2. The time-dependent larger in the benzoylated furanoses 9 and 10 than in the
product distribution for the acetolysis reaction is plotted in acetylated ones 3 and 4. These differences could be ex-
Figure 4.
plained by the bulkiness of the benzoyl group making nu-
The anomerization of the starting material 7 and 8 leads cleophilic attack from the α-side more sterically hindered
to an equilibrium mixture with 23% of the α-anomer 7 and and hence the formation of the β-anomer even more fa-
77% of the β-anomer 8 which stays fairly constant during vored. It is, however, not possible to rule out the differences
the subsequent acetolysis to compounds 9 and 10. The ano- in electronic and solvent effects between the acetyl and ben-
merization of the benzoylated methyl furanosides 7 and 8 zoyl groups when trying to understand their role as a par-
is faster than the analogous anomerization of the acetylated ticipating group.
methyl furanosides 1 and 2. As the anomerization is faster,
Based on these results, we propose the following mecha-
also the ring-closing of the acyclic oxocarbenium ion inter- nism for the anomerization and subsequent acetolysis of the
mediate (C) has to be faster. This again can be seen as a acylated methyl -ribofuranosides 1, 2, 7 and 8 (Scheme 3).
minor decrease in the formation of the corresponding iso- We suggest that the anomerization of the starting material
meric acyclic methyl hemiacetals 11 and 1,1-di-O-acetyl ri- is activated by protonation of the ring oxygen leading to
bose hydrate 12. The same trends seen in the formation of endo C–O cleavage. When acetylated (as proposed by Lind-
the fully acetylated β-anomer 4 are as well seen in the for- berg),[18] the C(4) oxygen is not nucleophilic enough for the
mation of the benzoylated counterpart 10. The maximum subsequent ring-closing of C to take place. The formation
in the concentration of the β-anomer 10 observed before of the acyclic acetyl methyl hemiacetals (5 and 11) is a com-
the anomeric equilibrium between 9 and 10 is established is peting reaction with ring-closing and proceeds via acety-
Figure 4. Time-dependent product distribution for the acetolysis of a) methyl 2,3,5-tri-O-benzoyl-α- (7) and b) -β--ribofuranoside (8) in
a 5:4 (v/v) AcOH/Ac2O mixture catalyzed by 0.75% of H2SO4. Experimental data and calculations are presented as symbols and solid
lines, respectively.
5670
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5666–5676

Kinetics of the Acid-Catalyzed Acetolysis of Acylated -Ribofuranosides
lation of the free hydroxyl group at C(4) followed by nucleo- mediate (J) via neighboring group participation of the acyl
philic attack by acetic acid on the acyclic carbenium ion group at C(2). The cis-fused intermediate blocks the α-side
intermediate (C). Based on our findings on the formation of the ring and makes the nuclophilic attack by acetic acid
of methyl hydrogen sulfate, we propose that protonation ac- from the β-side more feasible.
tivates the anomeric methoxy group for exo C–O cleavage.
This participation is possibly also activating the cleavage
The cyclic oxocarbenium intermediate (I) formed is stabi- of methanol from the 1,2-trans configured methyl β-ribo-
lized by the formation of a cis-fused dioxonium ion inter- furanosides. However, due to fast anomerization, it is im-
Scheme 3. Proposed mechanism for the acid-catalyzed acetolysis of the acetylated (1 and 2) and benzoylated (7 and 8) methyl -ribofur-
anosides.
Eur. J. Org. Chem. 2009, 5666–5676
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5671

J. J. Forsman, J. Wärnå, D. Yu. Murzin, R. Leino
FULL PAPER
possible to distinguish which one of the anomers is more iments were done starting from the methyl furanosides 1 or
prone towards exo-cyclic cleavage. 2 and 7 or 8, the 1-O-acetyl furanoses 3 or 4 and 9 or 10
Based on our earlier results,[17] we propose a mechanism and then finally from the acyclic methyl hemiacetals 5 and
for the anomerization of 1-O-acetylated ribofuranoses (3, 4 11. When starting from the 1-O-acetyl furanoses the equi-
and 9,10), where the acetyl cation formed in the acetylation librium between the α- and β-anomer is produced and the
media activates the anomeric acetyl group for exo-cyclic acyclic 1,1-O-diacetyl ribose hydrates (6 and 12) are formed.
cleavage. The main pathway for the formation of the fully In this case, reaction rates r3–4 and r3–6 for the acetylated
acylated acyclic -ribose hydrates (6 and 12) proceeds via sugars and reaction rates r9–10 and r9–12 for the benzoylated
acetolysis of the acetyl methyl hemiacetals (5 and 11), while ones can be determined. Using the acyclic methyl hemiacet-
being also to some extent formed via ring-opening of the 1- als as starting material produces the 1,1-O-diacetyl ribose
O-acetylated ribofuranoses (3, 4 and 9,10).
hydrates (6 and 12) and to some extent the α- (3 and 9) and
The site of activation is, supposedly, an issue of the rela- β-anomers (4 and 10) of the 1-O-acetyl furanoses, now
tive basicities of the exo and endo oxygen atoms.[24] The r5–6 and r5–4 for the acetylated and r11–12 and r11–10 for the
interrelationship of the basicity of the ring oxygen and gly- benzoylated compounds can be determined. Finally, start-
cosidic oxygen probably changes when moving from the ing from the acylated methyl ribofuranosides 1 or 2 and 7
methyl glycoside to the acetate and this could explain the or 8 gives the remaining r1–2, r1–3, r1–4, r2–3, r2–4, r1–5 and
change in site of activation. This could furthermore explain r2–5 for the acetylated sugars and the corresponding reac-
the differences in H+ vs. Ac+ affinities for the oxygens dur- tion rates r7–8, r7–9, r7–10, r8–9, r8–10, r7–11 and r8–11 for the
ing different stages of the acetolysis reaction. These rather benzoylated counterparts (for complete kinetic data see the
speculative suggestions need, however, further investigation Supporting Information).
to be confirmed.
From the kinetic modeling it was found that the reaction
rates r1–3, r2–3 and r2–5 for the acetylated derivatives 1 and
2 as well as the reaction rates r7–9, r8–9 and r8–11 for the
benzoylated ones 7 and 8 were very low, hence they were
set to zero. These results further support that ring-opening
only takes place from the α-anomers and that acetolysis of
the starting material first gives the 1-O-acetylated β-ano-
mers 4 and 10 which subsequently anomerizes to reach the
final equilibria. The equilibrium constants for the reactions
between the acetylated compounds 3 and 4 and the benzo-
ylated counterparts 9 and 10 were calculated from the ex-
perimental data as K3–4 = 3.22 and K9–10 = 2.33. For the
anomerization of the acylated methyl -ribofuranosides the
Kinetic Calculations
The reactions needed to describe the kinetics of the ace-
tolysis of the acetylated methyl -ribofuranosides 1 and 2,
as well as for the benzoylated counterparts 7 and 8 are given
in Schemes 4 and 5, respectively.
equilibrium constants were estimated giving K1–2
=
3.2Ϯ0.1 and K7–8 = 3.5Ϯ0.1 for the acetylated and benzo-
ylated sugars, respectively. The anomerization of the benzo-
ylated methyl furanosides 7 and 8 is so fast that equilibrium
is immediately reached; therefore the rate constant k7–8 was
fixed to a large value for the kinetic modeling. The rate
constants obtained for the reactions involved in the acetoly-
sis of the acylated methyl -ribofuranosides 1, 2 and 7,8 are
given in Table 1. When the acetylated (3 and 4) or benzoyl-
ated (9 and 10) 1-acetoxy furanoses are treated under the
acetolysis conditions we see anomerization of the starting
material giving the same final equilibrium as when it is car-
ried out from the acylated methyl furanosides. The rate con-
stants for the formation of the peracetylated ribofuranoses
Scheme 4.
3 and 4 from these reactions were determined as k4–3
=
2.376Ϯ0.008 and k3–4 = 7.67Ϯ0.03, respectively. The cor-
responding rate constants for the benzoylated sugars 9 and
10 were estimated as k10–9 = 0.918Ϯ0.005 and k9–10
=
2.14Ϯ0.01. These rate constants are a tenfold higher than
the corresponding constants obtained from the reactions
carried out from the acylated methyl furanosides given in
Table 1. This further verifies that the sulfuric acid in the
Scheme 5.
In order to determine the rate constants for the reactions acetolysis reactions is converted into the less acidic methyl
in Schemes 4 and 5, each containing 11 reactions, the pa- hydrogen sulfate and which slows down the formation of
rameter estimation was separated into three parts. Exper- the 1-acetoxy furanoses 3, 4 and 9, 10.
5672
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5666–5676

Kinetics of the Acid-Catalyzed Acetolysis of Acylated -Ribofuranosides
points were determined in open glass capillaries and are uncor-
Table 1. First-order rate constants for the reactions involved in the
acid-catalyzed acetolysis of the acetylated (1 and 2) and benzoyl-
ated (7 and 8) methyl -ribofuranosides at 25 °C in AcOH/Ac2O
(5:4 v/v) with 0.75% of H2SO4.
rected. Optical rotations were measured with a Perkin–Elmer 241
polarimeter equipped with a Na lamp (589 nm) at 24 °C. TLC
analyses were performed using silica gel F254 precoated aluminum
sheets visualized by charring with 25% H2SO4 in methanol or UV.
Column chromatography was performed using silica gel 60 Å
(Merck, 230–400 mesh).
Acetolysis of compounds 1 and 2
k [h–1]
Acetolysis of compounds 7 and 8
k [h–1]
k1–2
k1–3
k1–4
k1–5
k2–1
k2–3
k2–4
k2–5
k3–4
k3–6
k4–3
k4–6
k5–4
k5–6
12.7Ϯ0.5
0[a]
K7–8
k7–9
k7–10
k7–11
k8–7
3.5Ϯ0.1[b]
0[a]
Kinetic Calculations: The kinetics of the acetolysis reactions of the
acetylated methyl -ribofuranosides were described with a first or-
der reversible model for reactions r1–2 and r3–4, for the other reac-
tions a first order irreversible model was used. However, it is worth
notifying that the rate constants are lumped and do not take into
account formation of the possible intermediates. Furthermore, the
rate constants k1–2, k1–5, k2–4, k3–4, k3–6 and k5–6 are apparent,
hence implicitly including the concentrations of the catalytically
active species which were kept constant during the experiments.
0[a]
0[a]
0.023Ϯ0.009
4.1Ϯ0.4
0.0172Ϯ0.0008
[b]
0[a]
k8–9
0[a]
0.87Ϯ0.02
k8–10
k8–11
k9–10
k9–12
k10–9
k10–12
k11–10
k11–12
1.002Ϯ0.004
0[a]
0[a]
0.490Ϯ0.002
0.0005Ϯ0.00007
0.1518Ϯ0.0006
0[a]
0.0023Ϯ0.0002
0.0183Ϯ0.0002
0.249Ϯ0.002
0.0008Ϯ0.0001
0.1074Ϯ0.0008
0[a]
0.00172Ϯ0.00001
0.0148Ϯ0.0001
[a] Rate constants set to 0 based on kinetic modelling. [b] Rate
constants k7–8 and k8–7 were set to a large value due to instan-
taneous equilibrium between 7 and 8.
The reactor was described with a batch reactor model and the mass
balances for each component were as follows
Conclusions
To conclude, new mechanistic and kinetic data on the
acid-catalyzed acetolysis for a set of acylated methyl -ri-
bofuranosides has been obtained. Furthermore, the role of
H+ vs. Ac+ as the catalytically active source as well as the
site of anomeric activation has been discussed. Based on
the observed formation of methyl hydrogen sulfate during
the acetolysis process it can be deduced that the methyl fur-
anosides are protonated on the exocyclic oxygen and that
the aglycon is cleaved off as methanol. The anomerization
of the acylated methyl ribofuranosides taking place via a
ring-opening/ring-closing process is much faster for the
benzoylated compounds than for the acetylated counter-
parts. Thus, less time is left for the attack by acetic acid on
the acyclic oxocarbenium ion intermediate leading to lower
formation of acylic side products in the reactions of the
benzoylated methyl ribofuranosides. The results reported
herein contribute to a better understanding of fundamental
reaction mechanisms in carbohydrate chemistry in particu-
lar and in organic chemistry in general.
For the parameter estimation the following objective function was
minimized:
where ci,t,exp and ci,t,model are the experimentally recorded concen-
trations and the concentrations predicted by the model, respec-
tively. The weight factors wi were set to 1 for all experimental
points. The software “Modest” was used to estimate the rate con-
stants and to solve the reactor mass balances. The software mini-
mizes the objective function with the Levenberg–Marquardt
method and solves the ODEs describing the reactor model by the
backward difference method.[25] The same models were used for the
estimation of the kinetics for the benzoylated compounds 7–12 as
well.
Experimental Section
The fit of the model to experimental data of the reactions is pre-
sented in Figures 2 and 4 and in the Supporting Information (Fig-
ures S7–S16). It can be seen from the figures that the model can
describe the experimental data very well. The estimated rate con-
stants are listed in Table 1 and in the Supporting Information
(Tables S11–S16).
General Experimental Details: Chemicals were purchased from
commercial sources and were used without further purification.
The NMR spectra were recorded using a Bruker Avance 600 MHz
spectrometer equipped with a 5 mm inverse z-axis fg probe op-
1
erating at 600.13 MHz for H and 150.92 MHz for 13C. All prod-
ucts were fully characterized with 1D 1H and 13C NMR spec-
troscopy in combination with 2D COSY, HMBC and HSQC by
using pulse sequences provided by the instrument manufacturer.
The NMR spectra were referenced against TMS or the solvent sig-
nal. Signal multiplicities and coupling constants are given in paren-
theses [br = broad unresolved multiplet (1H NMR)].
To evaluate the accuracy of the estimated parameters the MCMC
(Markov Chain Monte Carlo) method was used. The method is
based on the Bayesian approach. In this method all the uncertain-
ties in the data as well as the modeling results are treated as statisti-
cal distributions showing that the parameters were well identified
(data given in the Supporting Information).[26]
HRMS were recorded on a Bruker micrOTOF-Q and EIMS were
obtained with Fisons ZabSpec mass spectrometer at 70 eV. Melting
General Procedures for the Acetolysis Reactions: Compounds 1–12
(0.04 mmol) were dissolved in 550 µL of CD2Cl2. To this solution
Eur. J. Org. Chem. 2009, 5666–5676
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5673

J. J. Forsman, J. Wärnå, D. Yu. Murzin, R. Leino
FULL PAPER
63 µL of a freshly prepared 5:4 (v/v) mixture of AcOH and Ac2O
containing 0.75% H2SO4 was added. The reactions were carried
out in sealed NMR tubes inside the magnet thermostatted to 25 °C
by a Bruker variable temperature unit. H NMR spectra were re-
corded at different time intervals and the molar concentrations of
the products were determined from the integral ratios.
at ambient temperature for 18 h. The excess of AcCl was decom-
posed by the addition of crushed ice and the mixture was extracted
with CH2Cl2 (2ϫ30 mL). The organic phases were combined and
washed with water (2ϫ50 mL), 1  HCl (50 mL), saturated
NaHCO3 (50 mL), brine (50 mL) dried with Na2SO4 and the sol-
vents evaporated to dryness. The evaporation residue was purified
by column chromatography (silica gel, EtOAc/petroleum ether, 4:5)
which gave pure 2 (2.4 g, 69%), as a colorless oil: Rf = 0.73 (1:1
1
Methyl α- (13) and β-L-Ribofuranoside (14): A solution of -ribose
(5.0 g, 33.3 mmol) in MeOH (100 mL) was treated with H2SO4
(0.5 mL). After 4.5 h of stirring at room temperature the reaction
was neutralized with Na2CO3 (10 g), filtered and evaporated to ob-
tain a mixture of 13 and 14 as a yellowish syrup (5.4 g) in a 3:8
ratio. The anomers were separated by means of column chromatog-
raphy on a Dowex Monosphere 99 Ca2+ column which gave pure
13 (1.4 g, 26%) as a colorless oil: Rf = 0.36 (6:1 CHCl3/MeOH).
[α]2D4 = –89.9 (c = 1.01, H2O) [ref.[27] [α]2D0 = +99 (c = 1.8, H2O) for
-isomer]. 1H NMR (600.13 MHz, CD3OD, 25 °C): δ = 4.85 (d,
J1,2 = 4.5 Hz, 1 H, 1-H), 4.01 (dd, J2,1 = 4.5, J2,3 = 6.6 Hz, 1 H, 2-
H), 3.98 (ddd, J4,3 = 3.1, J4,5a = 3.5, J4,5b = 4.2 Hz, 1 H, 4-H), 3.94
(dd, J3,4 = 3.1, J3,2 = 6.6 Hz, 1 H, 3-H), 3.66 (dd, J5a,4 = 3.5, J5a,5b
= 12.0 Hz, 1 H, 5a-H), 3.60 (dd, J5b,4 = 4.2, J5b,5a = 12.0 Hz, 1 H,
5b-H), 3.43 (s, 3 H, 1-OMe) ppm. 13C NMR (150.9 MHz, CD3OD,
25 °C): δ = 104.8 (C-1), 86.9 (C-4), 73.2 (C-2), 71.4 (C-3), 63.4 (C-
5), 55.7 (OCH3) ppm. EIMS calcd. for C15H36O5Si3 [M+] 380.1870,
found 380.1849.
EtOAc/hexane). [α]2D4 = +16.9 (c = 1.05, MeOH) [ref.[29] [α]2D2
=
1
+14.6 (c = 2.3, MeOH)]. H NMR (600.13 MHz, CDCl3, 25 °C): δ
= 5.32 (dd, J3,2 = 4.8, J3,4 = 6.8 Hz, 1 H, 3-H), 5.22 (d, J2,3
4.8 Hz, 1 H, 2-H), 4.90 (s, 1 H, 1-H), 4.36 (dd, J5a,4 = 3.7, J5a,5b
=
=
11.8 Hz, 1 H, 5a-H), 4.30 (ddd, J4,5a = 3.7, J4,5b = 5.9, J4,3 = 6.8 Hz,
1 H, 4-H), 4.10 (dd, J5b,4 = 6.9, J5b,5a = 11.8 Hz, 1 H, 5b-H), 3.37
(s, 3 H, 1-OMe), 2.11 (s, 3 H, 2-OAc), 2.09 (s, 3 H, 5-OAc), 2.05
(s, 3 H, 3-OAc) ppm. 13C NMR (150.9 MHz, CDCl3, 25 °C): δ =
170.8 (C=O-5), 169.8 (C=O-3), 169.8 (C=O-2), 106.4 (C-1), 78.7
(C-4), 74.8 (C-2), 71.7 (C-3), 64.6 (C-5), 55.4 (OCH3-1), 20.9
(CCH3-5), 20.8 (CCH3-2), 20.7 (CCH3-3) ppm. EIMS calcd. for
C12H17O8 [M+ – H] 289.0923, found 289.0930.
1,2,3,5-Tetra-O-acetyl-β-L-ribofuranose (4): A solution of 14 (2.9 g,
17.7 mmol) in glacial AcOH (7.5 mL) and Ac2O (10 mL) was
cooled (ice bath) and H2SO4 (0.3 mL) was added. The reaction was
stirred at room temperature for 1 h and then again placed on ice
bath and a second portion of H2SO4 (0.5 mL) was added. The reac-
tion was stirred for two more hours at room temperature and then
quenched by the addition of NaOAc (2.5 g). The reaction mixture
was evaporated and the brownish residue was dissolved in CHCl3
(50 mL), washed with water (3ϫ50 mL), dried with Na2SO4 and
the solvents evaporated to dryness. The yellow oil (4.2 g) was crys-
tallized from EtOH to afford 5 (2.4 g, 43%) as bright white crystals:
Rf = 0.58 (1:1 EtOAc/hexane); m.p. 80–83 °C [ref.[30] 81–83 °C].
[α]2D4 = +12.4 (c = 1.03, CHCl3) [ref.[31] [α]2D0 = +12.1 (c = 2.47,
The pure β-anomer obtained by chromatographic separation was
subsequently crystallized from iPrOH to give 14 (3.7 g, 69%) as
white needles: Rf = 0.36 (6:1 CHCl3/MeOH); m.p. 76–80 °C [ref.[28]
79–80 °C]. [α]2D4 = +51.2 (c = 1.02, H2O) [ref.[28] [α]2D0 = –50 (c =
1
2.0, H2O) for -isomer]. H NMR (600.13 MHz, CD3OD, 25 °C):
δ = 4.75 (s, 1 H, 1-H), 4.03 (dd, J3,2 = 4.7, J3,4 = 6.9 Hz, 1 H, 3-
H), 3.94 (ddd, J4,5a = 3.6, J4,5b = 6.5, J4,3 = 6.9 Hz, 1 H, 4-H), 3.87
(d, J2,3 = 4.7 Hz, 1 H, 2-H), 3.72 (dd, J5a,4 = 3.6, J5a,5b = 11.8 Hz,
1 H, 5a-H), 3.54 (dd, J5b,4 = 6.5, J5b,5a = 11.8 Hz, 1 H, 5b-H), 3.35
(s, 3 H, 1-OMe) ppm. 13C NMR (150.9 MHz, CD3OD, 25 °C): δ =
110.0 (C-1), 85.0 (C-4), 76.3 (C-2), 72.8 (C-3), 65.2 (C-5), 55.5
(OCH3) ppm. EIMS calcd. for C15H36O5Si3 [M+] 380.1870, found
380.1849.
1
CHCl3)]. H NMR (600.13 MHz, CDCl3, 25 °C): δ = 6.17 (br. s, 1
H, 1-H), 5.36 (dd, J3,2 = 4.8, J3,4 = 6.2 Hz, 1 H, 3-H), 5.34 (dd,
J2,1 = 0.9, J2,3 = 4.8 Hz, 1 H, 2-H), 4.38 (ddd, J4,5a = 3.6, J4,5b
=
5.5, J4,3 = 6.2 Hz, 1 H, 4-H), 4.34 (dd, J5a,4 = 3.6, J5a,5b = 12.1 Hz,
1 H, 5a-H), 4.15 (dd, J5b,4 = 5.5, J5b,5a = 12.1 Hz, 1 H, 5b-H), 2.13
(s, 3 H, 2-OAc), 2.10 (s, 3 H, 1-OAc), 2.09 (s, 3 H, 5-OAc), 2.08 (s,
3 H, 3-OAc) ppm. 13C NMR (150.9 MHz, CDCl3, 25 °C): δ = 170.5
(C=O-5), 169.7 (C=O-3), 169.4 (C=O-2), 169.0 (C=O-1), 98.2 (C-
1), 79.3 (C-4), 74.1 (C-2), 70.5 (C-3), 63.7 (C-5), 21.0 (CCH3-1),
20.7 (CCH3-5), 20.5 (CCH3-2), 20.5 (CCH3-3) ppm. EIMS calcd.
for C11H15O8 [M+ – CH3CO] 275.0767, found 275.0755.
Methyl 2,3,5-Tri-O-acetyl-α-L-ribofuranoside (1): To a suspension
of 13 (1.0 g, 6.1 mmol) in Ac2O (5 mL) was added iodine (200 mg,
0.78 mmol). The reaction was stirred at room temperature for 24 h,
then poured into a mixture of crushed ice and saturated aqueous
NaS2O3 (50 mL) and the mixture was extracted with CH2Cl2
(3ϫ50 mL). The combined organic fractions were washed with sat-
urated aqueous Na2SO4 (2ϫ100 mL), dried with Na2SO4 and
1,2,3,5-Tetra-O-acetyl-α-L-ribofuranose (3): To a solution of ZnCl2
evaporated to obtain
a yellowish oil (1.4 g). Flash column
(1.0 g, 7.3 mmol) in Ac2O (15 mL) 1,2,3,5-tetra-O-acetyl-β--ribo-
furanose (5) (6.0 g, 18.9 mmol) was added. After 2.5 h stirring at
room temperature the reaction was quenched by the addition of
0.7  NaOAc solution (50 mL). The reaction mixture was extracted
with CHCl3 (2ϫ25 mL) and the combined organic layers were
washed with saturated NaHCO3 (50 mL) and water (2ϫ50 mL),
dried with Na2SO4, concentrated and co-evaporated with EtOH
(2ϫ10 mL). The obtained syrup was dissolved in EtOH and most
of the remaining β-anomer was crystallized and filtered off. The
filtrate was evaporated and the yellow oil (1.4 g) was purified by
column chromatography (silica gel, EtOAc/petroleum ether, 1:3) to
give 6 (500 mg, 8%) as a colorless oil: Rf = 0.58 (1:1 EtOAc/hex-
ane). [α]2D4 = –79.2 (c = 1.00, CHCl3);. 1H NMR (600.13 MHz,
CDCl3, 25 °C): δ = 6.43 (d, J1,2 = 4.4 Hz, 1 H, 1-H), 5.26 (dd, J3,4
= 2.8, J3,2 = 6.7 Hz, 1 H, 3-H), 5.23 (dd, J2,1 = 4.4, J2,3 = 6.7 Hz,
1 H, 2-H), 4.44 (ddd, J4,3 = 2.8, J4,5a = 3.2, J4,5b = 4.0 Hz, 1 H, 4-
H), 4.33 (dd, J5a,4 = 3.2, J5a,5b = 12.2 Hz, 1 H, 5a-H), 4.21 (dd,
J5b,4 = 4.0, J5b,5a = 12.2 Hz, 1 H, 5b-H), 2.14 (s, 3 H, 3-OAc), 2.13
chromatography (silica gel, EtOAc/petroleum ether, 1:5) gave 1
(1.25 g, 72%), as a colorless oil: Rf = 0.46 (1:1 EtOAc/hexane).
1
[α]2D4 = –114.9 (c = 0.99, MeOH). H NMR (600.13 MHz, CDCl3,
25 °C): δ = 5.16 (dd, J3,4 = 3.7, J3,2 = 7.4 Hz, 1 H, 3-H), 5.12 (d,
J1,2 = 4.6 Hz, 1 H, 1-H), 4.96 (dd, J2,1 = 4.6, J2,3 = 7.4 Hz, 1 H, 2-
H), 4.37 (dd, J5a,4 = 3.0, J5a,5b = 11.9 Hz, 1 H, 5a-H), 4.26 (ddd,
J4,5a = 3.0, J4,3 = 3.7, J4,5b = 4.1 Hz, 1 H, 4-H), 4.20 (dd, J5b,4
=
4.1, J5b,5a = 11.9 Hz, 1 H, 5b-H), 3.43 (s, 3 H, 1-OMe), 2.12 (s, 6 H,
2-OAc, 5-OAc), 2.08 (s, 3 H, 3-OAc) ppm. 13C NMR (150.9 MHz,
CDCl3, 25 °C): δ = 170.7 (C=O-5), 170.6 (C=O-3), 170.0 (C=O-2),
101.7 (C-1), 79.6 (C-4), 70.9 (C-2), 69.9 (C-3), 63.6 (C-5), 55.7
(OCH3-1), 20.9 (CCH3-3), 20.7 (CCH3-5), 20.7 (CCH3-2) ppm.
EIMS calcd. for C12H17O8 [M+ – H] 289.0923, found 289.0930.
Methyl 2,3,5-Tri-O-acetyl-β-L-ribofuranoside (2): To a cooled (ice
bath) solution of compound 14 (2.0 g, 12.2 mmol) in pyridine
(7 mL) and CH2Cl2 (5 mL) AcCl (3.5 mL, 48.8 mmol) was added
drop wise. The ice bath was removed and the reaction was stirred
5674
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5666–5676

Kinetics of the Acid-Catalyzed Acetolysis of Acylated -Ribofuranosides
(s, 3 H, 1-OAc), 2.11 (s, 3 H, 5-OAc), 2.09 (s, 3 H, 2-OAc) ppm.
NMR (150.9 MHz, CDCl3, 25 °C): δ = 170.5 (C=O-5), 169.9
13C NMR (150.9 MHz, CDCl3, 25 °C): δ = 170.4 (C=O-5), 170.1 (C=O-4), 169.5 (C=O-3), 169.4 (C=O-2), 168.3 (C=O-1), 168.3
(C=O-3), 169.7 (C=O-1), 169.4 (C=O-2), 94.0 (C-1), 81.6 (C-4), (C=O-1), 86.0 (C-1), 69.5 (C-4), 69.3 (C-2), 68.5 (C-3), 61.6 (C-5),
70.0 (C-2), 69.8 (C-3), 63.3 (C-5), 21.0 (CCH3-1), 20.8 (CCH3-5), 20.8 (CCH3-2), 20.7 (CCH3-1), 20.7 (CCH3-4), 20.6 (CCH3-1), 20.6
20.6 (CCH3-3), 20.3 (CCH3-2) ppm. EIMS calcd. for C11H15O8
(CCH3-3), 20.6 (CCH3-5) ppm. EIMS calcd. for C17H24O12 [M+]
420.1268, found 420.1278.
[M+ – CH3CO] 275.0767, found 275.0755.
(1R,1S)-1,2,3,4,5-Penta-O-acetyl-L-ribose Methyl Hemiacetals (5):
Methyl 2,3,5-Tri-O-benzoyl-α-L-ribofuranoside (7): Compound 13
To a solution of ZnCl2 (1.6 g, 11.7 mmol) in Ac2O (14 mL) and
AcOH (6 mL), methyl 2,3,5-tri-O-acetyl--ribofuranoside (2) (1 g,
3.45 mmol) was added. The reaction was stirred at ambient tem-
perature for 48 h, followed by the addition of ice water (100 mL)
to decompose the excess of Ac2O. The mixture was extracted with
CHCl3 (50 mL), the organic layer was dried with Na2SO4. Evapora-
tion gave an isomeric mixture of the (1R,1S)-1,2,3,4,5-penta-O-ace-
tyl--ribose methyl hemi-acetals (5) (1.25 g, 92%, isomeric ratio
(2.0 g, 12.2 mmol) was dissolved in pyridine (6 mL) and CH2Cl2
(10 mL), the solution was cooled down to 0 °C and BzCl (5.7 mL,
48.8 mmol) was added dropwise. The ice bath was removed and the
reaction was stirred at ambient temperature for 20 h. The excess of
BzCl was decomposed by the addition of crushed ice and the mix-
ture was extracted with CH2Cl2 (3ϫ50 mL). The combined organic
phases were washed with water (2ϫ200 mL), dried with Na2SO4
and the solvents evaporated. The evaporation residue was purified
by column chromatography (silica gel, hexane/EtOAc, 10:1) to give
7 (4.99 g, 86%) as a colorless oil: Rf = 0.51 (4:1 petroleum ether/
4:5), as a pale yellow oil: Rf = 0.58 (1:1 EtOAc/hexane); (major
1
isomer) H NMR (600.13 MHz, CDCl3, 25 °C): δ = 5.92 (d, J1,2
=
EtOAc). [α]2D4 = –85.6 (c = 0.97, CHCl3). H NMR (600.13 MHz,
1
6.0 Hz, 1 H, 1-H), 5.41 (dd, J3,2 = 4.9, J3,4 = 5.9 Hz, 1 H, 3-H),
5.29 (ddd, J4,5a = 2.8, J4,3 = 5.9, J4,5b = 6.2 Hz, 1 H, 4-H), 5.26
(dd, J2,3 = 4.9, J2,1 = 6.0 Hz, 1 H, 2-H), 4.35 (dd, J5a,4 = 2.8, J5a,5b
= 12.3 Hz, 1 H, 5a-H), 4.14 (dd, J5b,4 = 6.2, J5b,5a = 12.3 Hz, 1 H,
5b-H), 3.45 (s, 3 H, 1-OMe), 2.10 (s, 3 H, 4-OAc), 2.10 (s, 3 H, 2-
OAc), 2.09 (s, 3 H, 1-OAc), 2.08 (s, 3 H, 3-OAc), 2.06 (s, 3 H, 5-
OAc) ppm. 13C NMR (150.9 MHz, CDCl3, 25 °C): δ = 170.7
(C=O-1), 170.6 (C=O-5), 169.8 (C=O-4), 169.4 (C=O-3), 169.4
(C=O-2), 94.9 (C-1), 70.5 (C-2), 69.5 (C-4), 69.1 (C-3), 61.8 (C-5),
57.4 (OCH3-1), 20.9 (CCH3-1), 20.9 (CCH3-4), 20.7 (CCH3-2), 20.7
(CCH3-3), 20.7 (CCH3-5) ppm. (minor isomer) 1H NMR
(600.13 MHz, CDCl3, 25 °C): δ = 5.78 (d, J1,2 = 4.3 Hz, 1 H, 1-H),
5.43 (dd, J3,4 = 4.6, J3,2 = 5.8 Hz, 1 H, 3-H), 5.33 (ddd, J4,5a = 3.2,
J4,3 = 4.6, J4,5b = 7.2 Hz, 1 H, 4-H), 5.29 (covered, 1 H, 2-H), 4.38
(dd, J5a,4 = 3.2, J5a,5b = 12.2 Hz, 1 H, 5a-H), 4.12 (dd, J5b,4 = 7.2,
J5b,5a = 12.2 Hz, 1 H, 5b-H), 3.43 (s, 3 H, 1-OMe), 2.15 (s, 3 H, 2-
OAc), 2.13 (s, 3 H, 1-OAc), 2.09 (s, 3 H, 3-OAc), 2.08 (s, 3 H, 4-
OAc), 2.04 (s, 3 H, 5-OAc) ppm. 13C NMR (150.9 MHz, CDCl3,
25 °C): δ = 170.6 (C=O-1), 170.4 (C=O-5), 169.9 (C=O-4), 169.7
(C=O-2), 169.3 (C=O-3), 95.4 (C-1), 71.2 (C-2), 69.8 (C-4), 69.2
(C-3), 61.7 (C-5), 57.7 (OCH3-1), 20.9 (CCH3-1), 20.9 (CCH3-2),
20.7 (CCH3-3), 20.7 (CCH3-5) ppm. EIMS calcd. for C16H24O11
[M+] 392.1319, found 392.1320.
CDCl3, 25 °C): δ = 8.10–7.29 (m, 15 H, aromatic), 5.73 (dd, J3,4
3.6, J3,2 = 7.1 Hz, 1 H, 3-H), 5.39 (d, J1,2 = 4.5 Hz, 1 H, 1-H), 5.33
(dd, J2,1 = 4.5, J2,3 = 7.1 Hz, 1 H, 2-H), 4.75 (dd, J5a,4 = 3.0, J5a,5b
=
= 11.7 Hz, 1 H, 5a-H), 4.65 (ddd, J4,5a = 3.0, J4,3 = 3.6, J4,5b
=
3.8 Hz, 1 H, 4-H), 4.62 (dd, J5b,4 = 3.8, J5b,5a = 11.7 Hz, 1 H, 5b-
H), 3.49 (s, 3 H, 1-OMe) ppm. 13C NMR (150.9 MHz, CDCl3,
25 °C): δ = 166.2 (C=O-5), 166.0 (C=O-3), 165.5 (C=O-2), 133.4–
128.3 (18C, aromatic), 101.9 (C-1), 79.4 (C-4), 71.8 (C-2), 70.8 (C-
3), 64.1 (C-5), 55.8 (OCH3-1) ppm. EIMS calcd. for C27H23O8
[M+ – H] 475.1393, found 475.1390.
Methyl 2,3,5-Tri-O-benzoyl-β-L-ribofuranoside (8): Compound 8
was prepared from 14 following the same procedure as for the prep-
aration of compound 7. Purification by column chromatography
(silica gel, hexane/EtOAc, gradient elution) gave 8 (5.2 g, 89%).
[α]2D4 = –58.6 (c = 1.07, CHCl3) [ref.[23] [α]2D5 = +58 (c = 0.94, CHCl3)
for -isomer]. 1H NMR (600.13 MHz, CDCl3, 25 °C): δ = 8.10–
7.29 (m, 15 H, aromatic), 5.87 (dd, J3,2 = 4.9, J3,4 = 6.8 Hz, 1 H,
3-H), 5.68 (d, J2,1 = 4.9 Hz, 1 H, 2-H), 5.16 (s, 1 H, 1-H), 4.73
(ddd, J4,5a = 4.2, J4,5b = 5.1, J4,3 = 6.9 Hz, 1 H, 4-H), 4.72 (dd,
J5a,4 = 4.2, J5a,5b = 11.8 Hz, 1 H, 5a-H), 4.53 (dd, J5b,4 = 5.1, J5b,5a
= 11.8 Hz, 1 H, 5b-H), 3.42 (s, 3 H, 1-OMe) ppm. 13C NMR
(150.9 MHz, CDCl3, 25 °C): δ = 166.3 (C=O-5), 165.4 (C=O-3),
165.3 (C=O-2), 133.4–128.4 (aromatic), 106.4 (C-1), 79.0 (C-4),
75.5 (C-2), 72.4 (C-3), 64.8 (C-5), 55.4 (OCH3-1) ppm. EIMS calcd.
for C27H23O8 [M+ – H] 475.1393, found 475.1390.
1,1,2,3,4,5-Hexa-O-acetyl-aldehydo-L-ribose (6): Compound
5
(1.0 g, 2.5 mmol) was dissolved in a mixture of Ac2O (17.5 mL)
and AcOH (7.5 mL) followed by the addition of concd. H2SO4
(0.6 mL). The reaction was left to proceed for 24 h at ambient tem-
perature and subsequently poured on crushed ice to decompose
remaining Ac2O. The mixture was extracted with CH2Cl2
(2ϫ60 mL), the combined organic layers were successively washed
with water (4ϫ30 mL) and saturated NaHCO3 (20 mL), dried with
Na2SO4 and concentrated to give a yellow oil (0.99 g). The oil was
analyzed by 1H NMR (250 MHz) in CDCl3 showing the desired
1-O-Acetyl-2,3,5-tri-O-benzoyl-β-L-ribose (10): To a –15 °C (NaCl/
ice) solution of 8 (4 g, 8.4 mmol) in AcOH (7 mL) and Ac2O
(3 mL) concd. H2SO4 (450 µL) was slowly added. The ice bath was
removed and the reaction stirred at ambient temperature for 4 h.
The reaction was quenched by the addition of NaOAc (3 g) fol-
lowed by the addition of water (5 mL) to decompose remaining
Ac2O. The reaction mixture was concentrated and the residue was
dissolved in CH2Cl2 (50 mL), washed with water (50 mL) and satu-
rated NaHCO3 (50 mL), dried with Na2SO4 and the solvents evap-
orated. The resulting syrup was crystallized from iPrOH to obtain
2 (2.08 g, 49%) as white crystals. The analytical data were in agree-
ment with those reported earlier.[17]
product plus 17% of compounds
3 and 4. Flash column
chromatography (EtOAc/petroleum ether, gradient elution) did not
improve the purity.
Since the product (6) was obtained in no more than 82% purity,
only the 1H and 13C NMR resonances and EIMS are reported: 1H
NMR (600.13 MHz, CDCl3, 25 °C): δ = 6.97 (d, J1,2 = 3.9 Hz, 1
H, 1-H), 5.41 (dd, J2,1 = 3.9, J2,3 = 6.7 Hz, 1 H, 2-H), 5.38 (dd,
1-O-Acetyl-2,3,5-tri-O-benzoyl-α-
tri-O-benzoyl-α--ribose was prepared by conversion of the β-an-
4.3, J4,5b = 6.8 Hz, 1 H, 4-H), 4.34 (dd, J5a,4 = 3.5, J5a,5b = 12.1 Hz, omer (10) following the same procedure as for the preparation of
L-ribose (9): The 1-O-acetyl-2,3,5-
J3,4 = 4.3, J3,2 = 6.7 Hz, 1 H, 3-H), 5.28 (ddd, J4,5a = 3.5, J4,3
=
1 H, 5a-H), 4.12 (dd, J5b,4 = 6.8, J5b,5a = 12.1 Hz, 1 H, 5b-H), 2.15
(s, 3 H, 2-OAc), 2.10 (s, 3 H, 4-OAc), 2.09 (s, 3 H, 1-OAc), 2.09 (s,
3 H, 3-OAc), 2.08 (s, 3 H, 1-OAc), 2.05 (s, 3 H, 5-OAc) ppm. 13C
compound 3 but with additional CH2Cl2 du to lower solubility in
Ac2O.[15] NMR and EIMS analysis of 9 were in agreement with
literature data.[17]
Eur. J. Org. Chem. 2009, 5666–5676
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5675

J. J. Forsman, J. Wärnå, D. Yu. Murzin, R. Leino
FULL PAPER
Chem. 1974, 39, 3654–3660; c) U. Niedballa, H. Vorbrüggen,
J. Org. Chem. 1974, 39, 3660–3663; d) U. Niedballa, H.
Vorbrüggen, J. Org. Chem. 1974, 39, 3664–3667; e) U. Nied-
balla, H. Vorbrüggen, J. Org. Chem. 1974, 39, 3668–3671; f) U.
Niedballa, H. Vorbrüggen, J. Org. Chem. 1974, 39, 3672–3674.
a) J. Kaczmarek, J. Szafranek, K. C. B. Wilkie, H. Lönnberg,
Finn. Chem. Lett. 1987, 14, 171–177; b) A. J. Bennet, M. L.
Sinnot, W. S. S. Wijesundera, J. Chem. Soc. Perkin Trans. 2
1985, 1233–1236; c) H. Lönnberg, A. Kankaanperä, K. Haa-
pakka, Carbohydr. Res. 1977, 56, 277–287; d) H. Lönnberg, A.
Kulonpää, Acta Chem. Scand. A 1977, 31, 306–312.
V. Ferrières, M. Gelin, R. Boulch, L. Toupet, D. Plusquellec,
Carbohydr. Res. 1998, 314, 79–83.
P. Zhang, Z. E. Dong, T. P. Cleary, Org. Process Res. Dev. 2005,
9, 583–592.
P. C. Kline, A. S. Serianni, J. Am. Chem. Soc. 1990, 112, 7373–
7381.
K. P. R. Kartha, R. A. Field, Tetrahedron 1997, 53, 11753–
11766.
H. G. Fletcher Jr., The Anomeric Tri-O-benzoyl- -arabinofur-
anosyl Bromides. In Methods in Carbohydrate Chemistry (Eds.:
R. L. Whistler, M. L. Wolfrom, J. N. BeMiller); Academic
Press, New York, 1963, vol. 2, p. 228.
(1R,1S)-1,4-Di-O-acetyl-2,3,5-tri-O-benzoyl-L-ribose Methyl Hemi-
acetals (11): Compound 11 was prepared from 8 following the same
procedure as for the preparation of compound 5.[16] The crude
product was purified by column chromatography (silica gel, hex-
ane/EtOAc, gradient elution) which gave 11 (0.78 g, 71%, isomeric
ratio 3:1) as a colorless oil. The NMR and EIMS data were in
agreement with those reported earlier.[17]
[9]
1,1,4-Tri-O-acetyl-2,3,5-tri-O-benzoyl-L-ribose Hydrate (12): Com-
pound 11 (2.5 g, 4.3 mmol) was dissolved in a mixture of Ac2O
(35 mL) and AcOH (15 mL) followed by the addition of concd.
H2SO4 (1.2 mL). The reaction was left to proceed for 20 h at ambi-
ent temperature before being neutralized with NaOAc (6 g). The
remaining Ac2O was decomposed by the addition of water (80 mL)
and the mixture was extracted with CHCl3 (60 mL). The organic
layer was successively washed with saturated NaHCO3 (50 mL) and
brine (50 mL), dried with Na2SO4 and concentrated. The crude
product was purified by column chromatography (silica gel, hex-
ane/EtOAc, 5:1) which gave 3 (1.94 g, 74%) as a colorless oil. NMR
and HRMS analysis of 12 were in agreement with earlier reported
data.[17]
[10]
[11]
[12]
[13]
[14]
Supporting Information (see also the footnote on the first page of
this article): Copies of 1H and 13C NMR spectra of all compounds
and complete kinetic data on the different acetolysis experiments.
[15]
[16]
H. Zinner, Chem. Ber. 1953, 86, 817–824.
E. D. Montgomery, R. M. Hann, C. S. Hudson, J. Am. Chem.
Soc. 1937, 59, 1124–1129.
[17]
J. J. Forsman, J. Wärnå, D. Yu. Murzin, R. Leino, Carbohydr.
Res. 2009, 344, 1102–1109.
Acknowledgments
[18]
[19]
[20]
[21]
J. Janson, B. Lindberg, Acta Chem. Scand. 1960, 14, 877–881.
E. P. Painter, Chem. Ind. (London) 1965, 1380–1381.
L. Rosenfeld, C. E. Ballou, Carbohydr. Res. 1974, 32, 287–298.
a) C. Malm, L. Tanghe, Ind. Eng. Chem. Anal. Ed. 1942, 14,
940–942; b) C. Malm, L. Tanghe, B. C. Laird, Ind. Eng. Chem.
1946, 38, 77–82; c) C. Malm, L. Tanghe, Ind. Eng. Chem. 1955,
47, 995–999.
Financial support to J. F. from the Finnish National Glycoscience
Graduate School is gratefully acknowledged. The authors also wish
to thank Prof. Harri Lönnberg and Dr. Patrik Eklund for the fruit-
ful discussions. We also thank Mr. Markku Reunanen for the
EIMS and HRMS analysis.
[22]
[23]
[24]
J. A. Hyatt, G. W. Tindall, Heterocycles 1993, 35, 227–234.
ˇ
A. Stimac, J. Kobe, Carbohydr. Res. 2000, 324, 149–160.
[1] R. U. Lemieux, W. P. Shyluk, G. Huber, Can. J. Chem. 1954,
33, 148–162.
[2] B. Capon, Chem. Rev. 1969, 69, 407–498.
D. R. McPhail, J. R. Lee, B. Fraser-Reid, J. Am. Chem. Soc.
1992, 114, 1905–1906.
H. Haario, Modest 6.0 – A User’s Guide, ProfMath, Helsinki,
2001.
H. Haario, E. Saksman, J. Tamminen, Bernoulli 2001, 7, 223–
242.
T. E. Walker, H. P. C. Hogenkamp, T. E. Needham, N. A. Mat-
wiyoff, Biochemistry 1974, 13, 2650–2655.
R. Barker, H. G. Fletcher Jr., J. Org. Chem. 1961, 26, 4605–
4609.
T. Sato, T. Simadate, Y. Isido, Nippon Kagaku Zasshi 1960, 81,
1442–1444.
R. D. Guthrie, S. C. Smith, Chem. Ind. (London) 1968, 547–
548.
K. S. Ramasamy, R. C. Tam, J. Bard, D. R. Averett, J. Med.
Chem. 2000, 43, 1019–1028.
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[3] B. Lindberg, Acta Chem. Scand. 1949, 3, 1153–1169.
[4] E. P. Painter, J. Am. Chem. Soc. 1953, 75, 1137–1146.
[5] a) M. Miljkovic´, M. Marino-Habash, J. Org. Chem. 1983, 48,
855–860; b) M. Miljkovic´, D. Yeagley, P. Deslongchamps, Y. L.
Dory, J. Org. Chem. 1997, 62, 7597–7604.
[6] a) J. Kaczmarek, M. Preyss, H. Lönnberg, J. Szafranek, Car-
bohydr. Res. 1995, 279, 107–116; b) J. Kaczmarek, Z. Kaczyn´-
ski, Z. Trumpakaj, J. Szafranek, M. Bogalecka, H. Lönnberg,
Carbohydr. Res. 2000, 325, 16–29.
[7] a) L. A. Agrofoglio, Curr. Org. Chem. 2006, 10, 333–362; b) G.
Gumina, Y. Chong, G.-Y. Song, C. K. Chu, Curr. Top. Med.
Chem. 2002, 2, 1065–1086; c) E. De Clercq, Mini-Rev. Med.
Chem. 2002, 2, 163–175; d) G. Gumina, Y. Chong, G.-Y. Song,
C. K. Chu, FEMS Microbiol. Lett. 2001, 202, 9–15.
[8] a) U. Niedballa, H. Vorbrüggen, Angew. Chem. Int. Ed. Engl.
1970, 9, 461–462; b) U. Niedballa, H. Vorbrüggen, J. Org.
Received: August 4, 2009
Published Online: September 24, 2009
5676
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 5666–5676