Acid-catalyzed isomerization of methyl 2-deoxy-D-arabino-hexosides: Equilibria, kinetics and mechanism

Article abstract of DOI:10.1016/S0008-6215(01)00305-6

Four isomers of methyl 2-deoxy-D-arabino-hexosides were isolated by HPLC as chromatographically homogeneous compounds. The rates of pyranoside isomerization (α^p and β^p) at 40°C and of furanoside isomerization (α^f and β^f) at 26°C were determined. A mechanism has been suggested for transformations taking place during isomerization of methyl 2-deoxy-D-arabino-hexosides in methanolic solution catalyzed with hydrogen chloride.

Full text of DOI:10.1016/S0008-6215(01)00305-6

Carbohydrate Research 337 (2002) 265–272
www.elsevier.com/locate/carres
Acid-catalyzed isomerization of methyl
2-deoxy- -arabino-hexosides: equilibria, kinetics and mechanism
D
Andrzej Nowacki, Kazimiera Smiataczowa, Regina Kasprzykowska,
Barbara Dmochowska, Andrzej Wis´niewski*
Department of Chemistry, Sugar Chemistry Group, Uni6ersity of Gdan˜sk, 18 Sobieskiego, PL-80-952 Gdan˜sk, Poland
Received 19 July 2001; accepted 9 November 2001
Abstract
Four isomers of methyl 2-deoxy-
D-arabino-hexosides were isolated by HPLC as chromatographically homogeneous compounds.
The rates of pyranoside isomerization (ap and bp) at 40 °C and of furanoside isomerization (a f and b f) at 26 °C were determined.
A mechanism has been suggested for transformations taking place during isomerization of methyl 2-deoxy-
methanolic solution catalyzed with hydrogen chloride. © 2002 Elsevier Science Ltd. All rights reserved.
D-arabino-hexosides in
Keywords: Methyl 2-deoxy-D-arabino-hexosides; Kinetic studies; Isomerization; Anomerization
1. Introduction
this compound at different temperatures, based on
changes of optical rotation, was determined by them.
They also observed a slow change of optical rotation,
which was interpreted as being due to a secondary
reaction resulting in a change of the heterocyclic ring
size. The composition of the ‘furanoside–pyranoside’
mixture was quantified on the basis of formaldehyde
and formic acid concentrations arising after oxidation
with sodium periodate. They determined the overall
rate constant for the change in furanosidic ring size.
Extremely interesting research has recently been con-
ducted by Swedish scientists.5 They carried out isomer-
ization of methyl glycosides in the presence of
4-methylmorpholine hydroborane. This compound re-
duces the carboxonium ions to give alditols and/or
anhydroalditols. These results evidently indicated the
cyclic and/or chain structures of the transition state
formed during isomerization of the O-glycoside.
The first report on the synthesis of methyl 2-deoxy-a-
and b-
-glucopyranosides appeared in 1922.1 Hughes et
al. in 1949 prepared a mixture, named by them ‘a- and
b-methyl-2-deoxy- -glucofuranosides’ because its opti-
cal rotation suggested another isomeric structure than
that of methyl 2-deoxy-a- or b-
-glucopyranoside.2
Bhat and Zorbach3 reported the synthesis, isolation and
purification of methyl 2-deoxy-a- -arabino-hexofura-
D
D
D
D
noside. The configuration of the anomeric carbon atom
was established on the basis of optical rotation ([h]D24
+117.1°) only. The authors demonstrated also that the
‘a- and b-methyl-2-deoxy-
contained ca. 30% of a- and b-pyranosides in the 1:1
-arabino-hexofura-
D-glucofuranosides’ mixture
ratio, ca. 35% of methyl 2-deoxy-a-
D
noside and ca. 35% of the starting sugar.
Capon et al. in 1962 reported important results on
the isomerization of methyl b-D-glucoside in a
The isomerization reaction of the methyl 2-deoxy-a-
-arabino-hexofuranoside under conditions described
methanolic solution of methanesulfonic acid.4 The
D
overall rate constant (k1+k−1) for anomerization of
by the authors, in addition in the presence of N-methyl-
morpholine borohydride, affords two ‘capture’ prod-
ucts, a chain and a cyclic alditol.5
In this article we report conditions for the isolation
of methyl per-O-acetyl-2-deoxy-D-arabino-hexosides by
HPLC and the results of kinetic studies of acid-cata-
* Corresponding author. Tel.: +48-58-3415271; fax: +48-
58-3410357.
E-mail address: andrzejw@chemik.chem.univ.gda.pl (A.
Wis´niewski).
lyzed isomerization of O-deacetylated glycosides.
0008-6215/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(01)00305-6

266
A. Nowacki et al. / Carbohydrate Research 337 (2002) 265–272
2. Results and discussion
Glycosidation of 2-deoxy-
D
-arabino-hexose in abso-
lute methanol containing catalytic amounts of hydro-
gen chloride was performed under two procedures
described by Hughes et al.,2 which resulted in two
mixtures of products (labeled A and B). In mixture A,
apart from small quantities of unreacted sugar, we
detected two main products using the capillary gas
chromatography (CGC) method. These products were
isolated by HPLC as exhaustively O-acetylated deriva-
tives. Their NMR spectra enabled us to identify them
Fig. 1. Structures of methyl tri-O-acetyl-2-deoxy-D-arabino-
hexosides.
Table 1
as methyl 3,4,6-tri-O-acetyl-2-deoxy-a- and b-
D-ara-
Retention times of the methyl per-O-acetyl-2-deoxy-
bino-hexoside isomers in CGC and HPLC
D-ara-
bino-hexopyranosides (3 and 4).
Again, in mixture B four compounds were detected.
Two of these were identical with those present in sam-
ple A. Thus, we decided to isolate the remaining two,
tR (CGC) (min) tR (HPLC) (min)
a-Furanoside (a f, 1) 12.47
13.30
8.80
11.69
9.83
methyl 3,4,6-tri-O-acetyl-2-deoxy-a- and b-D-arabino-
b-Furanoside (b f, 2) 13.57
a-Pyranoside (ap, 3) 14.32
b-Pyranoside (bp, 4) 14.67
hexofuranosides (1 and 2). In order to determine the
configuration of the anomeric carbon atom, NOE,
COSY and 13C NMR spectra of these compounds were
taken. Their NMR spectra indicate that furanosides
have been isolated. A NOE spectrum of compound 1
showed an interaction across the space between H-1
and H-5 and unequivocally confirmed its a configura-
tion (Fig. 1).
Table 2
Percentage concentrations of kinetically controlled products
of the methyl 2-deoxy-a- and b-D-arabino-hexopyranosides
isomerization a
Retention indices in CGC and HPLC are summa-
rized in Table 1.
Time (min)
ap
bp
a f
bp
ap
a f
Four chromatographically homogeneous isomers af-
ter O-deacetylation were used for the kinetic studies of
the isomerization. The concentrations of kinetically
controlled products of the acid-catalyzed isomerization
of the pyranosides at 40 °C are summarized in Table 2.
Equilibrium mixtures contained three isomers, ap, bp
and a f, in concentrations of 88.0, 10.5 and 1.5%, re-
spectively. The concentrations of kinetically controlled
products of the acid-catalyzed isomerization of the
furanosides at 26 °C are shown in Table 3, and the
equilibrium concentrations of isomers ap, bp, a f and b f
were 84.0, 13.5, 2.2 and 0.3%, respectively. It is note-
worthy that the very low concentration of methyl 2-de-
0
1
3
100
98.0
92.7
92.0
90.9
0
2.0
7.3
7.2
8.0
0
0
0
0.8
1.1
100
52.2
0
0
46.4 1.4
64.0 1.8
62.4 6.5
72.3 3.9
82.3 1.9
83.3 1.5
88.3 1.4
88.5 1.3
34.2
30.1
23.8
15.5
15.3
10.2
11.4
5
10
15
30
60
120
86.9
88.6
88.0
11.5
9.4
10.3
1.6
2.0
1.7
a 0.143 M methanolic solution of HCl at 40 °C.
Fig. 4 explicitly shows that the isomerization of pyran-
osides is a classic two-step process with such a complex
transformation.
Following a short period of rapid variations in reac-
tant concentrations, a slow stage of attaining equi-
librium of the ap ? bp anomerization begins. The
straight-line portion of curve a (Fig. 4) describes a stage
of slow transformation rate of ap with the simultaneous
formation of the bp isomer. The slope of straight line a
(Fig. 4) is equal to rate constant k−p a(s)=kbp(s)=
0.0603 min−1. Again, the rate constant of ap formation
from bp (straight line b in Fig. 4) is close to the value of
k−p a(s), namely k−p b(s)=kap(s)=0.0609 min−1. Thus,
the mean rate constant for interconversion of the a,b-
pyranosides calculated from Eq. (1) (see Section 3) is
0.0606 min−1, and it can be assumed that it is simple
oxy-b-D-arabino-hexofuranoside was determined on the
basis of the fixed [b f]/[a f] concentration ratio during
the whole process of isomerization.
The kinetic curves in Figs. 2 and 3 describing tempo-
ral changes of the concentrations of products obey
first-order kinetics. However, the reactions are not as
simple as highlighted by the semilogarithmic relation-
ships presented in Fig. 4. Significant deviations from
the reaction of first-order are characteristic of complex
reactions, i.e., two parallel reactions of first-order but
running at different rates and leading to the same final
product.6 A similar pattern was observed with mutaro-
tation of some sugars7 and their derivatives.8 The py-
ranose–furanose isomerization is usually rapid, but
that of a-pyranose–b-pyranose is a slower process.9,10
k
1
k
−1

A. Nowacki et al. / Carbohydrate Research 337 (2002) 265–272
267
Table 3
Percentage concentrations of kinetically controlled products of isomerization of methyl 2-deoxy-a- and b-
D
-arabino-hexofura-
nosides a
Time (min)
a f
ap
bp
b f
b f
ap
bp
a f
0
1
3
5
7
10
15
20
30
40
60
120
240
100
76.5
0
7.5
0
3.5
4.4
7.2
9.3
0
100
17.9
0
2.8
6.8
9.9
11.8
14.9
22.7
28.8
44.0
48.4
62.3
82.1
86.1
0
2.2
4.2
5.4
5.9
8.5
10.3
13.0
16.0
17.2
18.7
14.5
12.0
0
12.4
12.0
13.4
11.8
12.4
9.6
77.1
79.2
74.9
72.4
66.2
58.8
51.5
35.0
30.0
16.2
3.4
73.9
68.5
64.7
54.6
56.1
9.7
10.9
14.2
19.4
19.5
9.9
9.8
9.9
9.4
8.2
6.7
5.0
4.4
2.7
0.0
0.0
13.6
14.8
24.1
48.3
24.2
3.4
14.9
2.9
62.9
81.9
20.1
15.2
2.1
0.0
2.2
a 0.143 M methanolic solution of HCl at 26 °C.
reversible reactions, k−p a(s)=k−p b(s)=k1+k−1
=
Both constant rates can be related to direct elemen-
0.0606 min−1 or 7.06×10−3 dm3/(mol×s), because
the reaction was catalyzed with hydrogen chloride (c=
0.143 M). Thus, since that k1/k−1=0.119 (Table 4),
then k1=7.51×10−4 dm3/(mol×s) and k−1=6.31×
10−3 dm3/(mol×s).
tary reactions (ap“bp and bp“ap) because product
concentrations are very low at the beginning, and the
reversible reaction does not attain sufficient rate in
accordance with the mass action law. However, since
k−p a+k−p bꢀk1+k−1 and the kbp/kpa ratio is 0.21, be-
ing greater than the equilibrium constant, the k−p a and
k−p b constants reveal another pathway of anomeriza-
tion of the pyranosides during the fast stage of the
reaction. Furthermore, data of Table 2 show that a
small quantity of the a f appears also during the isomer-
ization of both pyranosides. The furanoside formation
rate is approximately equal to the rate of pyranoside
isomerization, kaf=0.34 and 1.44 min−1 for the iso-
merization of the ap and bp isomer, respectively. This
suggests that the rapid stage of the pyranoside anomer-
ization proceeds via the furanoid form analogously to
The slow stage of the anomerization is preceded by a
rapid stage that also led to inversion of the anomeric
carbon atom configuration. Owing to the high rate and
small number of the data points for the first stage of the
reaction, k constants calculated from Eq. (2) (see Sec-
tion 3) are certainly of low precision. They are, how-
ever, highly different and allow one to reach
conclusions about the course of the reaction. Thus,
isomer bp is formed from isomer ap with the rate
expressed by k−p a=kpb=0.304 min−1, and more than a
half of the equilibrium quantity of ap is formed from bp
at a rate constant k−p b=kap =1.43 min−1
.
the methyl glycoside formation from D
-xylose.11 In our
Fig. 3. Isomerization of methyl 2-deoxy-b-
ranoside. ꢀ, ap; ꢁ, bp; ꢂ, a f.
D-arabino-hexopy-
Fig. 2. Isomerization of methyl 2-deoxy-a-
ranoside. ꢀ, ap; ꢁ, bp; ꢂ, a f.
D-arabino-hexopy-

268
A. Nowacki et al. / Carbohydrate Research 337 (2002) 265–272
A careful analysis of the results of these experiments
allows one to suggest the interconversions presented in
Scheme 1.
Accordingly to Scheme 1, the isomerization rate con-
stant of the labile isomer b f calculated from Eq. (2) for
the rapid stage of the reaction is:
k−f b=k3+k−3=2.62 min−1
Since k−3/k3=0.137, then k3=2.30 min−1 and k−3
0.315 min−1
=
.
Fig. 4. Deviations from the first order reaction rate for the
ap ? bp anomerization (ꢀ) and bp ? ap (ꢁ). Straight line a:
y=0.0603(90.0015)x+1.5144(90.0104),
r=0.9998;
straight line b: y=0.0609(90.0016)x+3.376(90.02556),
r=0.9993.
Table 4
Equilibrium constants K for the isomerization reactions
Equilibrium
K (40 °C)
K (26 °C)
K1
K2
K2K3
K−3
ap ? bp
bp ? b f
bp ? a f
a f ? b f
ap ? b f
ap ? a f
0.119
ꢁ0.015
0.143
0.161
0.022
0.163
0.137
0.0036
0.026
Fig. 5. Temporal variations of concentrations of the methyl
2-deoxy-a-
-arabino-hexofuranoside isomers. ꢀ, ap; ꢁ, bp;
D
ꢂ, a f; ꢃ, b f.
K−3K−4
K−4
0.017
case, the rapid stage of bp anomerization is inhibited
before completion of the anomerization process and
then in the curve of a f formation a small maximum
appears (Fig. 3). These findings lead to the assumption
that at the beginning of the rapid stage of isomeriza-
tion, the irreversible reaction bp“a f“ap occurs,8 and
as soon as the concentration of a f reaches a sufficient
level, the first stage of the reaction becomes reversible:
bp ? a f“ap. Formation of bp is quicker than ap, thus
further progress of the reaction slows down before
attaining the bp ? ap equilibrium.
Experiments with the isomerization of furanosides
should be performed at lower temperatures since fura-
nosides are unstable.
Fig. 6. Temporal variations of concentrations of the methyl
2-deoxy-b-
-arabino-hexofuranoside isomers. ꢀ, ap; ꢁ, bp;
D
Variations of concentrations of the reactants and
products during isomerization of the furanosides are
shown in Table 3 and Figs. 5 and 6. During the initial
rapid stage of the reaction, isomer b f is quickly con-
verted into isomer a f, followed by setting up the
b f ? a f equilibrium. On the other hand, during the
slow stage of isomerization of the furanosides, corre-
sponding pyranosides are formed: a f“ap and b f“bp.
ꢂ, a f; ꢃ, b f.
Scheme 1. Proposed anomerization pathway for methyl 2-de-
oxy-D-arabino-hexosides.

A. Nowacki et al. / Carbohydrate Research 337 (2002) 265–272
269
The rate constants of the slow stage of isomerization
of the furanosides, determined as slopes of straight lines
a and b in Fig. 7 are: k−f b(s)=0.0259 min−1 and
k−f a(s)=0.0300 min−1. Since kpa=0.0306 min−1, the
rate of decay of isomer a f is equal to the rate of
formation of pyranoside ap (Fig. 8). Thus it can be
assumed that k−f a=kap =k4+k−4=0.0303 min−1
.
Hence, k4=0.0295 and k−4=0.000768 min−1
.
The b-pyranoside is formed at a higher rate than its
a counterpart (Fig. 8). The rate of formation of bp is
linearly related to the isomerization rate of b f. How-
ever, the rates are unequal.
During the slow stage of isomerization, isomer b f
remains at equilibrium with isomer a f:
k
k
−3
a f ? b f “−2bp
k
3
Fig. 7. Deviations from the first-order reaction for isomeriza-
Since k4ꢂk−3, it can be assumed that −d[b f]/dt=
tion of methyl 2-deoxy-a-
D-arabino-hexofuranoside (ꢂ) and
methyl 2-deoxy-b- -arabino-hexofuranoside (ꢃ). Straight line
k
−2[b f]=k−2k−3/k3[a f]. Hence:
D
a: y=−0.030(90.001)x+4.329(90.020), r=0.9986; straight
line b: y= −0.026(90.001)x+2.441(90.024), r=0.9982.
k−f b(s)=k−2k−3/k3=0.0259 min−1, and then k−2
=0.189 min−1
.
Isomer b f can be converted into bp via the acyclic
cation and the rate of formation of bp is: d[bp]/dt=
12
k
k
−5[SiH+]. If k−5ꢀk−2
,
then kpb=k−2k−5/(k5+
−5)=0.0919 (Fig. 8) ik5=k−5 and hence, as
previously, k−2=0.184 min−1
.
Based on Scheme 1 we can assume, that the rapid
stage in the reaction of ap isomer (the most stable one),
could be expressed as k−p a=k4+k−4=0.304 min−1
.
Since
k
−4/k4=0.017, thus k4=0.299 min−1 and
k
−4=0.00508 min−1. In turn, for the bp isomer:
k−p b=k2+k−2=1.43 min−1. We were not able to
determine experimentally the value k2/k−2 because of
too little concentration of b f at 40 °C. However, if we
accept the analogous logarithmic dependence log k=
f(1/T) for the reactions bp ? b f and ap ? a f, then K2
(40 °C)$0.015, and k2 and k−2 are 0.02 and 1.41
min−1, respectively. If the slow stage of isomerization
of the pyranosides proceed via the same way of trans-
formation, i.e., in accordance with Scheme 1, the rate
constant k−p b(s) will be K2K3k4=0.043 min−1. This
calculated value is not markedly smaller than those
experimentally found (0.060 min−1).
Fig. 8. Rate constants of formation of methyl 2-deoxy-a-
arabino-hexopyranoside (ꢀ) and methyl 2-deoxy-b- -gluco-
pyranoside (ꢁ) from b-furanoside. Straight line a:
y= −0.0306(90.0004)x+4.4379(90.0183), r=0.9994;
straight line b: y= −0.0919(90.0013)x+2.5271(90.0145),
r=0.9996.
D-
D
Experimental rate constants and the calculated con-
stants of the elementary reactions are collected in
Tables 5 and 6, respectively. They determine the trans-
formation rates of particular isomers and decline in the
Table 5
Experimental rate constants (min−1) in 0.143 M methanolic
HCl
40 °C
26 °C
following series: k3\k−3\k−2\k4\k2\k−4
k−1\k1.
\
k−p b=1.43
k−f b=2.62
Based on data given in Table 6 we calculated approx-
imate values of activation parameters for some reaction
steps at 26 °C. Values of DHc2 , DH−c2, DHc4 and
DHc−4 are: 21.1, 26.3, 30.2 and 24.5 kcal/mol, respec-
tively. Approximate values of DSc2 , DS−c2, DSc4 and
DSc−4 are: −3.08, 21.9, 31.2 and 5.00 cal/(mol K),
k−p a=0.304
k−f b(s)=0.0259
k−f a(s)=0.0300
kpa=0.0306
k−p a(s)=0.0603
k−p b(s)=0.0609
kaf (from ap)=0.34
kaf (from bp)=1.44
kpb=0.0919

270
A. Nowacki et al. / Carbohydrate Research 337 (2002) 265–272
Table 6
Elementary rate constants
k
Constant
k
k
k
(40 °C)
(26 °C)
(40 °C)
(26 °C)
min−1
min−1
dm3/(mol×s)
dm3/(mol×s)
k1
k−1
k2
k−2
k3
k−3
k4
6.44×10−3
7.51×10−4
6.32×10−3
2.45×10−3
1.64×10−1
5.42×10−2
ꢁ2.1×10−2
4.10×10−3
1.86×10−1
2.30
4.78×10−4
2.17×10−2
2.68×10−1
3.67×10−2
3.44×10−3
8.95×10−5
1.41
3.15×10−1
2.95×10−2
7.68×10−4
2.99×10−1
5.08×10−3
3.48×10−2
5.92×10−4
k−4
respectively, and values of DGc given in the same order
are as follows: 22.0, 19.8, 20.9 and 23.0 kcal/mol. These
values correspond well to the stability of studied com-
pounds (ap\bp\a f\b f).
The fast process of anomerization of the pyranosides
is a consequence of the heterocyclic oxygen protona-
tion. Cleavage of the pyranoid ring needs enthalpy of
activation ꢁ23 kcal/mol, and very low values of the
DSc2 and DS−c4 (close to zero) could suggest that the
structures of transition states and substrates are similar.
Thus, one can suppose that the pseudoacyclic cations
keep their starting conformations and easily undergo
the following transformations: SaH+“a f and SbH+“
b f. However, the results obtained do not provide any
evidence for the SaH+ ? SbH+ interconversion.
The slow process of anomerization of the pyran-
osides could be caused by a cyclic carboxonium ion
formation. However, the transformation bp ? a f“ap
is more probable, i.e., after rapid preliminary equilibra-
tion, the final equilibrium state ap ? bp is slowly
reached.
A complete cycle of transformation of four methyl
2-deoxy-D-arabino-hexosides in methanol containing
hydrogen chloride is presented in Scheme 2.
In the presence of an excess of strong acid, the
isomerization reactions proceed via appropriate cations.
Anomerization of furanosides b f ? a f is the most rapid
and the easiest process. The rate of this reaction is more
than ten times faster than the rate of furanoside“py-
ranoside isomerization via an acyclic carboxonium
cation (k3/k−2=12.4 and k−3/k4=10.7). One can as-
sume that anomerization of furanosides goes via an
electrophilic attack of a solvated proton on the glyco-
sidic oxygen atom and subsequent cyclic carboxonium
ion formation. The presence of such cations was docu-
mented during isomerization of other glycosides.5,13 On
the other hand, protonation of the heterocyclic oxygen
atom in the furanoside molecules leads to ring cleavage.
Very high values of DSc−2 and DSc4 confirm the state-
ment mentioned above. Nevertheless, high values of
activation enthalpy (ꢁ28 kcal/mol) do not favor this
reaction.
3. Experimental
General methods.—All HPLC separations were per-
formed using a VISTA 5500 (Varian) HPLC chro-
matograph equipped with a Kromasil C8 column (5 mm
packing material, 4.6×250 mm) from Eka–Nobel
Scheme 2. Transformations of the methyl 2-deoxy-D-arabino-hexoside isomers in methanolic hydrogen chloride solution.

A. Nowacki et al. / Carbohydrate Research 337 (2002) 265–272
271
ln/ct−cꢃ/= −kt+ln/c0−cꢃ/
(1)
(Sweden) for analytical runs and an Ultrasphere ODS
column (5 mm packing material, 10×250 mm) from
Beckman (USA) for semipreparative runs. Constant
flowrates were 1 and 4.5 mL/min, respectively. The
products were detected at u=226 nm. The 8:17
MeCN–water solvent system for analytical separations
and another MeCN–water system (linear gradient from
18.4 to 32% run over 90 min) for semipreparative
separations were used. A small portion of the sample A
or B was injected into a Vega 6180 (Carlo Erba)
capillary gas chromatograph equipped with a cold ‘on
column’ injector and a flame ionization detector (FID).
Gas flowrates for FID were 50 and 90 mL/min for
hydrogen and air, respectively. Hydrogen (2 mL/min)
was used as the carrier gas. Separation of the mixture
was achieved with a DB-23 fused-silica capillary
column (60 m×0.258 mm i.d., 0.15 mm film thickness)
from J&W Scientific (Folsom, CA, USA) using a tem-
perature program from 140 to 160 °C at 4 °C/min, 160
to 200 °C at 6 °C/min, 200 to 240 °C at 8 °C/min (held
for 10 min). The FID system was held at 260 °C. All
where t is the reaction time in min, c0 is the starting
concentration of the solution, ct is the concentration
after t minutes, and cꢃ is the isomer concentration at
equilibrium. With complex reactions, the rate of the
slow stage(s) was calculated from Eq. (1), whereas that
of the fast reaction from Eq. (2)
1
k= ln
t
c0−c0(s)
c1−c1(s)
(2)
where c0 is the initial concentration, c1 is the concentra-
tion measured after 1 min, c0(s) and c1(s) are, respec-
tively, appropriate quantities calculated from Eq. (1)
back to t=0 and t=1 min of the slow stage of the
reaction.6–8 All rate constants were determined at the
fixed HCl concentration (0.143 M). The precision of
calculation of the rate constants from Eq. (1) was
determined by standard deviations calculated from the
following expression:
(S x)2
1
500 MHz H NMR spectra were recorded in CDCl3
D
S x2−
with Me4Si as an internal standard using a Varian
Unity Plus 500 spectrometer. Optical rotations were
measured with a JASCO J-20 spectropolarimeter.
Kinetics studies—Sol6ent purification.—Methanol
was first dried over Na2SO4 and then with magnesium
methoxide and distilled.14
Catalyst.—A methanolic solution of hydrogen chlo-
ride was prepared by bubbling dry hydrogen chloride
through purified MeOH. The concentration of the
stock solution was determined by titration.
Sample preparation.—Four starting solutions were
prepared by dissolution of compound ap or bp or a f or
b f (8 mg) in 4 mL of purified MeOH. Four sets of
ampoules with the same amounts of starting isomer (ap
or bp or a f or b f) were prepared by taking 50 mL of the
starting solution. Then the solvent was removed under
a N2 stream. To each ampoule containing 0.1 mg of
appropriate isomer, 200 mL of 0.143 M methanolic
solution of hydrogen chloride was added, and the vial
was tightly closed. Two sets of ampoules containing
pyranosides (ap and bp) were held at 4090.1 °C, and
two other sets containing furanosides (a f and b f) were
held at 2690.1 °C. The reaction, after appropriate
time, was quenched by adding one drop of 25% aq
ammonia (the solution was alkaline). Next, solvents
were removed under a N2 stream at rt. The dry residue
was exhaustively O-acetylated with 1:1 pyridine–Ac2O
at rt for 24 h. Then solvents were removed under a
nitrogen stream, and the dry residue was dissolved in
100 mL of CHCl3 and analyzed with CGC.
n
|n=
n−1
Syntheses.—The anomeric mixture of methyl 3,4,6-
tri-O-acetyl-2-deoxy-
D
-arabino-hexosides (sample A)
was obtained under the procedure of Hughes et al.2
(heating at 45 ° C during 1 h) from 1 g (6 mmol) of
2-deoxy-D-arabino-hexose. The resulting oil (1.3 g) was
exhaustively O-acetylated with 1:1 Ac2O–pyridine at
ambient temperature during 24 h. Then the volatile
components were removed under reduced pressure.
CGC analysis revealed two main components. The
crude product (1.59 g) was decolorized by passing its
solution in 9:1 CCl4–acetone through a short column
with Kieselgel MN 60, vB0.08. A sample of the solu-
tion, after concentration to a dense syrup under re-
duced pressure, was separated by semipreparative
HPLC.
Methyl tri-O-acetyl-2-deoxy-h-D-arabino-hexopyran-
1
oside 3.—741 mg, [h]2D2 +118.8° (c 0.8, MeOH); H
NMR (CDCl3) l 4.87 (dd, 1 H, J1,2 1.4, J1,2a 3.4 Hz,
H-1); 2.26 (m, 1 H, J2,2a 13.1, J2,3 5.4 Hz, H-2); 1.84 (m,
1 H, J2a,3 11.5 Hz, H-2a); 5.31 (m, 1 H, J3,4 9.7 Hz,
H-3); 5.02 (t, 1 H, J4,5 9.7 Hz, H-4); 3.96 (m, 1 H, J5,6
4.5, J5,6% 2.4 Hz, H-5); 4.33 and 4.09 (dd, 2 H, J6,6% 12.2
Hz, H-6 and H-6%); 3.37 (s, 3 H, OCH3); 2.04–2.12 (3s,
each 3 H, OAc); 13C NMR: l 98.00 (C-1), 34.86 (C-2),
69.01 (C-3), 67.65 (C-4), 69.20 (C-5), 62.33 (C-6), 54.87
(OCH3), 20.95–20.72 (OAc), 170.74–169.88 (CꢀO).
Methyl
tri-O-acetyl-2-deoxy-i-D-arabino-hexopy-
1
ranoside 4.—163 mg, [h]2D0 −27.8° (c 0.5, EtOH); H
NMR (CDCl3) l 4.50 (dd, 1 H, J1,2 2.0, J1,2a 9.8 Hz,
H-1); 2.32 (m, 1 H, J2,2a 12.7, J2,3 4.9 Hz, H-2); 1.75 (m,
1 H, J2a,3 11.2 Hz, H-2a); 5.04 (m, 1 H, J3,4 9.5 Hz,
Calculations.—Rate constants, k, were calculated
with the least-squares method from the following
equation:

272
A. Nowacki et al. / Carbohydrate Research 337 (2002) 265–272
H-3); 5.00 (t, 1 H, J4,5 9.5 Hz, H-4); 3.62 (m, 1 H, J5,6
4.6, J5,6% 2.4 Hz, H-5); 4.32 and 4.12 (dd, 2 H, J6,6% 12.2
Hz, H-6 and H-6%); 3.50 (s, 3 H, OCH3); 2.04–2.10 (3s,
each 3 H, OAc); 13C NMR: l 100.54 (C-1), 36.01 (C-2),
70.55 (C-3), 69.03 (C-4), 71.92 (C-5), 62.37 (C-6), 56.90
(OCH3), 20.92–20.73 (OAc), 170.70–169.85 (CꢀO).
69.59 (C-5), 63.78 (C-6), 55.77 (OCH3), 21.20–20.99
(OAc), 170.94–169.96 (CꢀO).
Acknowledgements
The mixture of four methyl tri-O-acetyl-2-deoxy-
D
-
This work was partially financed by grants: DS/8361-
4-0134-1, DS/8230-4-0088-1 and DS/8350-4-0131-1.
arabino-hexosides (sample B) was obtained under the
procedure of Hughes et al.2 (at rt, during 9 min) from
0.5 g (3 mmol) of 2-deoxy-D-arabino-hexose. The re-
sulting oil (0.529 g) was exhaustively O-acetylated as
described above. CGC analysis of the mixture of per-O-
acetylated compounds (0.895 g) showed the presence of
four components. The mixture was decolorized as de-
scribed above, concentrated and finally separated by
semipreparative HPLC.
References
1. (a) Bergmann, M.; Schotte, H.; Leschinsky, W. Ber. Dtsch.
Chem. Ges. 1922, 55, 158–172;
(b) Bergmann, M.; Schotte, H.; Leschinsky, W. Ber. Dtsch.
Chem. Ges. 1923, 56, 1052–1059.
2. Hughes, I. W.; Overend, W. G.; Stacey, M. J. Chem. Soc.
1949, 2846–2849.
Methyl tri-O-acetyl-2-deoxy-h-D-arabino-hexofura-
noside 1.—67 mg, [h]2D0 +92.8° (c 0.9, EtOH); 1H
NMR (CDCl3) l 5.20 (dd, 1 H, J1,2 3.4, J1,2% 5.4 Hz,
H-1); 2.28 (m, 1 H, J2,2% 15.1, J2,3 6.1 Hz, H-2); 2.21 (m,
1 H, J2%,3 1.5 Hz, H-2%); 5.48 (m, 1 H, J3,4 3.9 Hz, H-3);
4.21 (dd, 1 H, J4,5 9.2 Hz, H-4); 5.28 (m, 1 H, J5,6 2.2,
J5,6% 5.4 Hz, H-5); 4.57 and 4.21 (dd, 2 H, J6,6% 12.2 Hz,
H-6 and H-6%); 3.48 (s, 3 H, OCH3); 2.03–2.13 (3s,
each 3 H, OAc); 13C NMR: l 104.66 (C-1), 40.90
(C-2), 72.61 (C-3), 76.74 (C-4), 68.43 (C-5), 63.78 (C-6),
55.65 (OCH3), 21.11–20.94 (OAc), 170.89–169.99
(CꢀO).
3. Bhat, K. V.; Zorbach, W. W. Carbohydr. Res. 1968, 6,
63–74.
4. Capon, B.; Loveday, G. W.; Overend, W. G. Chem. Ind.
1962, 1537–1538.
5. Garegg, P. J.; Johansson, K.-J.; Konradsson, P.; Lindberg,
B. J. Carbohydr. Chem. 1999, 18, 31–40.
6. Frost, A. A.; Pearson, R. S. Kinetics and Mechanisms;
Wiley: New York, 1961; p. 162.
7. Pigman, W.; Isbell, H. S. Ad6. Carbohydr. Chem. Biochem.
1968, 23, 11–57.
8. Smiataczowa, K. Pol. J. Chem. 1999, 73, 1513–1522.
9. Isbell, H. S.; Pigman, W. Ad6. Carbohydr. Chem. Biochem.
1969, 24, 13–65.
10. Lemieux, R. U.; Anderson, L.; Conner, A. H. Carbohydr.
Res. 1971, 20, 59–72.
11. Bishop, C. T.; Cooper, F. P. Can. J. Chem. 1962, 40,
224–232.
12. Schwetlick, K. Kinetische Methoden zur Untersuchung 6on
Reactionmechanismen; PWN: Warszawa, 1975; p. 66 (in
Polish).
Methyl tri-O-acetyl-2-deoxy-i-D-arabino-hexofura-
noside 2.—24 mg, [h]2D0 −17.3° (c 0.4, EtOH); 1H
NMR (CDCl3) l 5.09 (d, 1 H, J1,2 5.4 Hz, H-1); 2.33
(m, 1 H, J2,2% 14.7, J2,3 5.9 Hz, H-2); 2.13 (1 H, H-2%);
5.47 (dd, 1 H, J3,4 4.9 Hz, H-3); 4.30 (dd, 1 H, J4,5 9.2
Hz, H-4); 5.32 (m, 1 H, J5,6 2.2, J5,6% 5.0 Hz, H-5); 4.67
and 4.21 (dd, 2 H, J6,6% 12.2 Hz, H-6 and H-6%); 3.40 (s,
3 H, OCH3); 2.01–2.12 (3s, each 3 H, OAc); 13C NMR:
l 105.25 (C-1), 39.86 (C-2), 71.48 (C-3), 78.90 (C-4),
13. Bunton, C. A.; Lewis, T. A.; Lewellyn, D. R.; Vernon, C.
A. J. Chem. Soc. 1955, 4419–4423.
14. Vogel, A. I. Textbook of Practical Organic Chemistry;
WNT: Warszawa, 1984; p. 219 (in Polish).