Mechanism of the transition-metal-catalyzed mutarotation reaction of N-(p-chlorophenyl)-β-D-glucopyranosylamine in methanol

Article abstract of DOI:10.1016/S0008-6215(03)00023-5

Rate constants for the mutarotation reaction of N-(p-chlorophenyl)-β-D-glucopyranosylamine (NGlc) in methanol have been determined in the presence of transition metal chlorides (MCl₂), at 25°C. The activity of the metal ions catalyzing the α-py

Full text of DOI:10.1016/S0008-6215(03)00023-5

Carbohydrate Research 338 (2003) 969–975
www.elsevier.com/locate/carres
Mechanism of the transition-metal-catalyzed mutarotation reaction
of N-(p-chlorophenyl)-b-
-glucopyranosylamine in methanolꢀ
D
Kazimiera Smiataczowa, Jarosław Kosmalski, Teresa Widernik, Zygmunt Warnke*
Faculty of Chemistry, Uni6ersity of Gdan´sk, Sobieskiego 18, PL-80-952 Gdan´sk, Poland
Received 3 October 2002; accepted 6 January 2003
Abstract
Rate constants for the mutarotation reaction of N-(p-chlorophenyl)-b-
D-glucopyranosylamine (NGlc) in methanol have been
determined in the presence of transition metal chlorides (MCl2), at 25 °C. The activity of the metal ions catalyzing the
a-pyranosidelb-pyranoside interconversion has been found to increase in the following series: Mn2+ BCo2+ BNi2+ BZn2+
B
Cu2+. The pHs of the methanolic solutions of the chlorides were measured and acidity constants of the [MCl(CH3OH)5]+ ions
and NGlc were determined in this solvent. Addition of NGlc to the salt solutions resulted in lowering their pH. Raising the
methyloxonium ion concentration in the solutions resulted in rapid increase in the rate of mutarotation in the presence of MCl2.
It is suggested that in solutions of NGlc and MCl2, the CH3OH2+ ions are generated by solvolysis of the salts and additionally
by dissociation of the hydroxyl group at C-6 of the glucosylamine molecule taking place during complexation of the metal ions.
A scheme has been derived for interaction of deprotonated NGlc molecules with the transition metal ions. © 2003 Elsevier Science
Ltd. All rights reserved.
Keywords: N-(p-Chlorophenyl)-b-D-glucopyranosylamine; Methanol; Metal chlorides
1. Introduction
of sugars has been surveyed by Mitzner and Behrend-
wald. They studied the kinetics of mutarotation of
The a-pyranosidelb-pyranoside interconversion is a
characteristic feature of glycopyranosylamines. The rate
of attaining equilibrium of this reaction is strongly
catalyzed by acids and proceeds through an acyclic
immonium ion1–3 intermediate as schematically shown
in Scheme 1. In addition the rate of a weak-acid-cata-
lyzed mutarotation reaction has been found to be af-
fected by salts. Neutral salts, such as alkali metal
chlorides enhance the reaction rate by virtue of the
secondary salt effect.4 Salts of weak acids, for instance,
sodium benzoate, slow down the rate on account of the
Law of Mass Action effect resulting in suppressing
methyloxonium ion concentration in solution.5,6 The
influence of the transition metal salts on the mutarota-
tion of glycosylamines has not been studied to date.
This type of interactions in the area of mutarotation
b-D-mannose in water in the presence of the cobalt(II)
and copper(II)7 and also iron(II), zinc(II) and mangane-
se(II) ions.8 Based on the results of further investiga-
tions, they derived the following rank order of catalytic
activity of metal ions:9 Cu2+ \Zn2+ \Fe2+ \Co2+
\Ni2+ \Mn2+. Moreover, they found that catalytic
activity of Ni(NO3)2 in the mutarotation reaction de-
pended on the nature of sugar.10 They also found that
the initial and final optical rotations of sugar solutions
were identical in the presence and absence of a salt and
did not depend on salt concentration, thus excluding
the formation of stable complexes in solution. Accord-
ing to the authors, only labile transition products are
formed whose composition was determined on the basis
of a relationship between activation parameters and salt
concentration. Generally, the results have shown that
the metal ion is attached to the ring oxygen atom of the
sugar molecule and to the oxygen of the ꢀCH2OH
group to form a five-membered ring.10 The formation
of compounds of this type apparently accelerates the
mutarotation of monosaccharides and explains the en-
ꢀ The experiments were partly carried out by Katarzyna
Maj and Anna Urban´czyk.
* Corresponding author. Fax: +48-58-3410357
E-mail address: warnke@chemik.univ.gda.pl (Z. Warnke).
0008-6215/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0008-6215(03)00023-5

970
K. Smiataczowa et al. / Carbohydrate Research 338 (2003) 969–975
Scheme 1. Mechanism of anomerization of glycopyranosylamines.
hydroxyl groups, can bind metal ions in different ways.
Metal ions can co-ordinate one, two or even three sugar
molecules.13,14,18 The numerous hydroxyls can also form
chelate-like structures. In more concentrated solutions
also polynuclear species can emerge.12,13
The stability of complexes is considerably enhanced
after substitution of a carboxyl, phosphate, amine and
other groups to the sugar molecule. Extremely strong
tendency towards complexation of diamines, in particu-
lar ethylenediamine, has been utilized to the synthesis
of solid nickel(II) and cobalt(III) complexes in which
glycosylamine products of condensation of diamines
with monosaccharides were used as ligands. These
products were characterized by crystallography.19
In continuation of our studies concerned with the
influence of acids and salts on the rate of mutarotation
hanced acidity of solutions, because it facilitates the
release of proton from the anomeric ꢀOH group re-
sponsible for the acidity of sugars.
The influence of transition metal ions on the mutaro-
tation of a-D-glucose in water has also been investi-
gated.11,12 An extremely strong catalytic activity of
copper(II) salts has been attributed to small amounts of
binuclear [Cu(OH)2Cu]2+ ions present in the solutions.
Just these ions, apart from the OH− ones, have been
recognized as the strongest catalysts of the mutarota-
tion reaction of a-D-glucose.
All in all, the metal–sugar interactions in aqueous
solutions are weak, because metal ions co-ordinate
water molecules more strongly than alcoholic hydrox-
yls. Stability constants of the metal complexes of sugars
are invariably small.13,14 Stronger electron-donating
properties are exhibited by sugars after deprotonation
of the hydroxyl groups.13,15,16 The associations of metal
ions with sugars are freely soluble in water and it is
difficult to obtain them in the solid state. Nonetheless,
a number of sugar complexes with transition metal ions
have recently been isolated from strongly alkaline solu-
tions. Products precipitated from aqueous solutions
after addition of methanol contained also the sodium
and potassium ions.15–17 Additional obstacles in study-
ing metal–sugar interactions arise from the finding that
in solutions of monosaccharides, different anomeric
and conformational species occur at equilibrium. Fur-
ther, sugar isomers, depending on the sequence of
of
D-glucosylamines and bearing in mind the foregoing
literature reports, we embarked on investigation of
catalytic activity of selected transition metal chlorides.
Further, the purpose of this contribution was to eluci-
date the mechanism of interaction of the transition
metal ions with the molecule of glycosylamine during
its mutarotation in methanol.
2. Results and discussion
The rate of the mutarotation reaction of glucosylamines
has been found to depend on the strength and concen-
tration of the acid catalyst.1–3 The strongest catalyst in
methanolic solutions is the methyloxonium ion,
CH3OH+2 (referred to as the H+ ion). Divalent transi-
tion metal chlorides are weak Bro¨nsted acids as shown
by their acidity constants listed in Table 1. They are,
however, much stronger catalysts of the mutarotation
reaction as compared with other weak acids.20 For
instance, CuCl2 gives almost instantaneous attainment
of equilibrium of the reaction. For this reason, very
dilute (3.3×10−6 M) salt solutions were used for the
kinetic experiments. The reaction proceeds according to
the first kinetic order. The rate constants, k, listed in
Table 1 show that MnCl2 accelerates the rate of mu-
tarotation only slightly as compared to the reaction
proceeding in methanol. The effectiveness of CoCl2
matches that of NiCl2, both enhancing the rate almost
twofold, whereas ZnCl2 more than fivefold and CuCl2
more than 20-fold. The catalytic influence of the transi-
Table 1
Rate constants, k, of the mutarotation of a 0.02 M methano-
lic NGlc solution catalyzed with 3.3×10−6 M transition
metal chlorides and rate constants, kb, of this reaction deter-
mined in the benzoate buffer (2:1)
22
Compound pKa(H O)
k (10−3 min−1
)
kb (10−3 min−1
)
2
CuCl2
ZnCl2
NiCl2
CoCl2
MnCl2
[Cu(acac)2]
Methanol
7.29 a
8.96
41.390.2
8.8090.15
8.9790.17
9.0790.10
8.9590.16
8.9090.15
9.7390.11
4.3990.06
3.9390.06
2.3990.02
1.8990.04
1.7890.05
9.86
10.00 a
10.59
8.9190.17
a Data obtained in this work.

K. Smiataczowa et al. / Carbohydrate Research 338 (2003) 969–975
971
Fig. 1. Plot of the logarithm of the mutarotation rate constant of NGlc vs. pKa(H O) of the transition metal chlorides.
2
tion metal chlorides on the mutarotation of the gluco-
sylamine can be due to the presence of H+ ions in
solutions of the transition metal salts. In order to check
whether the reaction is catalyzed solely by the H+ ions
derived from solvolysis of the salts, or by the metal ions
themselves, the rate measurements were run in a
methanolic benzoate buffer solution. The rate con-
stants, kb, (Table 1) oscillated around 9×10−3 min−1
and were independent of the kind of cation, this show-
ing that the H+ ions were the sole catalysts of this
reaction.
[MCl(CH3OH)5]++CH3OH
l[MCl(CH3O)(CH3OH)4]+CH3OH2+
The H+ ion concentration in methanolic solutions of
the transition metal chlorides was determined by pH
measurements using salicylate and succinate buffer so-
lutions as standards.25 To achieve better accuracy of
results of the pH metric measurements, solutions of
higher concentrations were used than those prepared
for kinetic measurements. Table 2 lists results of the pH
measurements of the 4.0×10−5 M methanolic solu-
tions of the transition metal chlorides. There is a strictly
linear relationship between H+ ion concentration and
the rate of mutarotation of NGlc which can be ex-
pressed as:
Table 1 lists also the rate constant of the mutarota-
tion catalyzed by copper(II) acetylacetonate. This is
almost identical with that determined in methanol and
shows that the very stable complex (log i(H O)
=
18.40)21 does not undergo solvolysis and does not2 gen-
erate H+ ions catalyzing the reaction.
k=1.71×106[H+]+1.94×10−4 (r=0.9997)
It is also remarkable that there is a linear relationship
between ln k and the literature acidity constants,
22
pKa(H O)
.
With manganese(II), nickel(II) and zinc
chlori2des it is so accurate that the acidity constants of
the hydrated Cu2+ and Co2+ ions (7.29 and 10.00,
respectively) for which the literature quantities vary
within broad limits23 can be determined more accu-
Table 2
Acidity constants of the transition metal chlorides,
pKa(CH OH)
,
determined from pH measurements in two
3
batches of methanol (pH(I) and pH(II)
)
Compound
pH(I)
pH(II)
pKa(CH OH)
rately from the equation: ln k= −0.866 pKa(H O)+3.12
3
2
(r= −0.9999). The relationship is presented graphi-
cally in Fig. 1.
CuCl2
ZnCl2
NiCl2
CoCl2
MnCl2
Methanol
NGlc
7.62
8.24
8.62
8.73
8.82
8.91
8.19
8.92
7.72
8.60
8.66
8.82
8.94
9.42
10.9490.14
12.4690.50
12.9690.02
13.2990.03
13.6290.13
Anhydrous transition metal chlorides undergo disso-
ciation in methanol, and in dilute solutions an equi-
librium CoCl2lCoCl++Cl− is set up.24 Cations of
the remaining metals also retain the chloride ion in the
co-ordination sphere and after solvation assume usually
an octahedral symmetry.15–17 Solvolysis of these com-
plexes results in acidification of the solution due to the
following equilibrium:
O-Ac-NGlc
9.41
Salt concentration, cs=4×10−5 M.

972
K. Smiataczowa et al. / Carbohydrate Research 338 (2003) 969–975
Table 3
the pH of a tetra-O-acetylated NGlc equal to 8.92 (run
I), a value identical with that of the methanol used in
this run.
pH values of 3.3×10−6 M salt solutions with and without
0.02 M NGlc in methanol
Similar results were obtained in run II. These results
reveal that NGlc is an amphiprotolyte, because on one
hand the ring oxygen atom readily attracts the
proton3,29 and on the other hand, at least one of the
four hydroxyl groups undergoes deprotonation. The
pKa and pKb quantities of the glucosylamine in
methanol indicate that its acid and basic properties are
almost equal.
In the presence of the transition metal chlorides, the
pH of the NGlc solutions is further depressed. In Table
3 presented are the pH values of 3.3×10−6 M solu-
tions of the salts (pHs) used for the kinetic measure-
Compound
pH(s)
pH(NGlc+s)
Found
Found
Calcd
Calcd
CuCl2
ZnCl2
NiCl2
CoCl2
MnCl2
8.18
8.26
8.45
8.86
8.75
8.20
8.79
8.86
8.89
8.90
6.70
7.60
7.83
7.99
8.24
8.05
8.18
8.20
8.19
8.19
The differences in pH between the two runs of the
measurements (I and II) are due to different pH values
of two batches of the solvent used (8.91 and 9.42). The
different and slightly higher than the literature pH
values of absolute methanol (8.35)24 can be due to trace
amounts of water in both batches of the solvent.26
The dissociation constant of CoCl2 in methanol is
0.1.24 It has been assumed that the remaining metal
chlorides are equally well dissociated, and consequently
it can be assumed that in the 4.0×10−5 M solutions
the concentration of MCl+ is equal to the salt concen-
tration. Having in hand the measured pHs and taking
into account the concentration of the CH3OH+2 ions
derived from autodissociation of methanol, the acidity
constants of the transition metal chlorides in methanol
were calculated from the equation:
ments and those containing also 0.02
M NGlc
(pH(NGlc+s)). The H+ ion concentration in a solution
containing weak acids (NGlc+MCl2) in a solvent of
autoprotolysis constant Ki can be calculated from the
equation:
[H+]=[(Ka×c)MCl +(Ka×c)NGlc+Ki]1/2
2
Hence, for instance, the pH of a methanolic CuCl2 and
NGlc solution at the concentrations indicated above is
8.05, while the measured value is 6.70 (Table 3). With
the remaining salts, the measured pHs are also lower
than those calculated. It can be suggested that the
increase in acidity of the metal chloride solutions fol-
lowing addition of NGlc can be due to the influence of
the cationic forms of the metals on deprotonation of
the glucosylamine molecule.
[H+]=[(Ka×c)MCl +Ki]1/2
2
The rank order of the rate constants of mutarotation
of NGlc in methanol in the presence of the transition
metal chlorides, associated with acidity of the NGlc–
MCl2 system, can be useful for estimation of the rela-
tive stability of compounds being formed. The
transition metal ions, arranged in this rank order, form
The calculations were performed separately for either
of the two batches of the solvent. In Table 2 collected
are average pKa(CH OH) quantities for particular salts.
With CuCl2, CoCl23, NiCl2 and MnCl2 the deviations
from the average value do not exceed 1.5%. The acidity
constant of ZnCl2 is burdened with the greatest error
probably due to very strong hygroscopicity of the salt.
Similar to other weak acids of this same type, there is
the following series: Cu2+ \Zn2+ \Ni2+ \Co2+
\
Mn2+ consistent with the Irving–Williams series. The
values for Cu2+ distinctly stand out. The stability
constants of the CuCl+, NiCl+ and CoCl+ ions in
methanol, determined by Khan and Bouet,30 arrange
themselves in the same sequence, the stability constants
for CuCl+ being much higher than for the remaining
ones. This is compatible with our results and reveals the
stronger complexation capacity of the Cu2+ ion.
As mentioned above, the literature reports show that
interactions between metals and sugars are weak and
depend much on the structure of a sugar molecule.
Among six-membered monosaccharides, the most con-
venient ligands are conformers containing either axial–
equatorial–axial or 1,3,5-triaxial hydroxyl groups.18
Just these groups form complexes with metal ions.
The sugar residue of NGlc has all hydroxyls arranged
equatorially and thus it does not obey any of the
foregoing conditions. By analogy to the finding that in
20
a linear relationship between pKa(H O) and pKa(CH OH)
with the slope of 0.81 and correlation coefficient3, r=
2
0.995.
N-(p-Chlorophenyl)-b-D-glucopyranosylamine is a
very weak base in methanol with pKb(CH OH) of 14.65 as
3
determined by titration with hydrochloric acid.27 Dur-
ing these experiments, it has unexpectedly been found
that the pH of a 0.02 M solution of the base in
methanol is lower (8.19) than that of methanol (8.91)
(run I, Table 2). This means that NGlc is a very weak
acid with pKa(CH OH) of 14.70 as determined by us.
Monosaccharides 3in water are also very weak acids
with the protolysis constants of the order of 10−13 28
.
The H+ ions in solutions of NGlc are derived from
deprotonation of one of the hydroxyl groups, as is the
case with free sugars. This assumption is supported by

K. Smiataczowa et al. / Carbohydrate Research 338 (2003) 969–975
973
the literature evidence31 and an approximately 6000-
fold excess of NGlc in respect to the transition metal
chlorides used allow to conclude that in the solutions
considered a complex is formed with involvement of
two deprotonated glucosylamine molecules as shown in
Scheme 2.
acid solutions solvated proton is readily bound to the
ring oxygen atom of the NGlc molecule,1–3 it can be
assumed that also a metal cation would co-ordinate to
the oxygen. Subsequently, following simultaneous in-
volvement of the ꢀOH group at C-6 a chelate is formed
whose five-membered ring has a structure similar to
that suggested by Mitzner and Behrenwald10 which
facilitates deprotonation of the ꢀCH2OH group. Both
Interactions taking place during complex formation
with involvement of the sugar residue of NGlc manifest
themselves in the absorption spectra presented in Figs.
2 and 3. In both figures there is an increase in absorp-
tion intensity of solutions containing CoCl2 or NiCl2
and NGlc relative to that of salt solutions without
NGlc. Both the shape of the spectra and location of the
absorption bands reveal complexes of a distorted octa-
hedral symmetry. Only a slight displacement of the
bands towards shorter wavelengths, noted after addi-
tion of NGlc to the methanolic solution of the metal
chlorides, shows that both in the case of methanol and
the deprotonated NGlc molecules, the oxygen atoms
are electron donors. The nitrogen atom of the amino
group of NGlc is virtually not involved in complexa-
tion, this being consistent with our previous
calculations3 revealing much higher electron density on
the oxygen atom of the pyranoside ring as compared to
that of the amine nitrogen. An increase in the absorp-
tion bands can be explained in terms of chelate forma-
tion of with deprotonated NGlc molecules exhibiting
much lower symmetry as compared to that of the
MCl(CH3OH)+5 ions.
Scheme 2. Suggested structure of the complex formed.
3. Experimental
3.1. Reagents
Methanol was first dried over Na2SO4, then with
iodine-activated magnesium metal and distilled. To re-
move basic impurities, the tartaric acid was added to
the solvent and distillation was repeated. Finally, the
solvent was distilled through a Widmer column and a
fraction boiling strictly at the bp of pure MeOH, i.e.,
65 °C/1013 hPa32 was collected. The pHs of the two
batches of MeOH were 8.91 and 9.42. Zinc chloride
was dried at 180 °C. The remaining chlorides were dried
at 130 °C for 3 h. The salts were stored over P4O10 in a
desiccator. The N-glucosyl-p-chloroaniline was synthe-
sized by methanolic method,32 whereby 0.06 mol of
glucose and 0.066 mol of p-chloroaniline were dissolved
in 70 cm3 of freshly purified MeOH and refluxed for
several hours. The reaction was monitored by thin-layer
chromatography (TLC) using aluminium-supported
plates coated with Silica Gel 60 (0.2 mm, E. Merck,
Darmstadt, Germany) in a CCl4–acetone solvent sys-
tem 3:1 (v/v). The crude product was purified by crys-
talization from 96% EtOH; mp 147–148 °C (Ref. 32
146 °C); initial specific rotation [h]25456= −130° (c=
Fig. 2. Absorption spectra of methanolic solutions of 0.01 M
cobalt(II) chloride (----) and 0.01 M cobalt(II) chloride after
addition of a 0.02 M NGlc solution (—).
Fig. 3. Absorption spectra of methanolic solutions of 0.01 M
nickel(II) chloride (----) and 0.01 M nickel(II) chloride after
addition of a 0.02 M NGlc solution (—).

974
K. Smiataczowa et al. / Carbohydrate Research 338 (2003) 969–975
0.579, MeOH) (Ref. 32 [h]25456= −137°); 1H NMR
(CD3OD): l 4.49 (d, 1 H, J1,2 8.8 Hz, H-1), 3.45 (t, 1 H,
J4,5 8.8 Hz, H-4), 3.36–3.28 (m, 2 H, H-2, H-3), 3.37
(m, 1 H, J5,6 2.4 Hz, H-5), 3.85 (dd, 1 H, J6,6% 12 Hz,
H-6), 3.66 (dd, 1 H, J5,6% 5.2 Hz, H-6%).
References
1. Isbell, H. S.; Pigman, W. Ad6. Carbohydr. Chem.
Biochem. 1969, 24, 13–65.
2. Paulsen, H.; Pflughaupt, K. W. Carbohydr. Chem.
Biochem. 1980, IB, 881–927.
3. Smiataczowa, K.; Maj, K.; Skurski, P. Eur. J. Org. Chem.
2001, 4269–4274.
3.2. Measurements of the rate constants
4. Smiataczowa, K.; Maj, K. Polish J. Chem. 2000, 74,
429–437.
5. Smiataczowa, K.; Wasielewska, M.; Jasinski, T. Polish J.
Chem. 1981, 55, 179–186.
6. Smiataczowa, K.; Maj, K.; Korewa, R. Polish J. Chem.
1997, 71, 831–839.
7. Mitzner, R.; Behrenwald, E. Z. Phys. Chem. 1972, 249,
The rate constants of mutarotation of N-glucosyl-p-
chloroaniline were determined polarimetrically by mea-
suring the angle of rotation of the plane of polarized
light at 546 nm over time with an accuracy of 90.01°.
The measurements were carried out at 25 °C. Solutions
were placed in 2 dm long water-jacketed polarimetric
tubes thermostated to 90.1 °C. The concentration of
the N-glucoside in MeOH was 2×10−2 M. The con-
centration of the salts was 3.3×10−6 M. The reaction
started immediately after addition of a metal chloride.
The rates were also measured in a benzoate buffer of
pH 8.38 obtained by mixing together 0.2 M benzoic
acid with 0.1 M sodium benzoate solutions.6
56–62.
8. Mitzner, R.; Behrenwald, E. Z. Phys. Chem. 1972, 249,
236–242.
9. Mitzner, R.; Behrenwald, E. Z. Chem. 1971, 11, 64–65.
10. (a) Mitzner, R.; Behrenwald, E. Z. Phys. Chem. 1971,
246, 25–32;
(b) Mitzner, R.; Behrenwald, E. Z. Phys. Chem. 1971,
247, 78–84.
11. O’Connor, C. J.; Odell, A. L.; Bailey, A. A. T. Aust. J.
Chem. 1982, 35, 951–960.
12. Angyal, S. J. Carbohydr. Res. 1990, 200, 181–188.
13. Gyurcsik, B.; Nagy, L. Coord. Chem. Re6. 2000, 203,
81–149.
14. Whitfield, D. M.; Stojkovski, S.; Sarkar, B. Coord. Chem.
Re6. 1993, 122, 171–225.
15. (a) Rao, C. P.; Geetha, K.; Raghavan, M. S. S.; Sreed-
hara, A.; Tokunaga, K.; Yamaguchi, T.; Jadhav, V.;
Ganesh, K. N.; Krishnamoorthy, T.; Ramaiah, K. V. A.;
Bhattacharyya, R. K. Inorg. Chim. Acta 2000, 297, 373–
382;
(b) Bandwar, R. P.; Rao, C. P. Carbohydr. Res. 1997,
297, 341–346.
16. (a) Bandwar, R. P.; Giralt, M.; Hidalgo, J.; Rao, C. P.
Carbohydr. Res. 1996, 284, 73–84;
3.3. pH measurements
The pH of the solutions were measured at 25 °C using
a prototype of an in house designed high quality pH
controller linked to a PC running Windows operating
system. The pH unit consisted of a high impedance
input circuitry, precision 918 bit ADC, a microcom-
puter and a standard RS232C interface. Results of the
measurements could be presented either graphically (on
the screen or paper) or transferred into other pro-
grammes for further processing. A combination pH
electrode filled with a satd methanolic KCl solution was
used. For calibration of the instrument, buffer solutions
were used, whose pHs in MeOH were: 7.53 (0.01 M
salicylic acid+0.01 M sodium salicylate) and 8.75 (0.01
M succinic acid+0.01 M lithium hydrogen succi-
nate).25 The pH measurements were run over the salt
(b) Bandwar, R. P.; Rao, C. P. Carbohydr. Res. 1996,
287, 157–168.
17. (a) Bandwar, R. P.; Rao, C. P.; Giralt, M.; Hidalgo, J.;
Kulkarni, G. U. J. Inorg. Biochem. 1997, 66, 37–44;
(b) Bandwar, R. P.; Sastry, M. D.; Kadam, R. M.; Rao,
C. P. Carbohydr. Res. 1997, 297, 333–339.
18. Angyal, S. J. Ad6. Carbohydr. Chem. 1989, 47, 1–47.
19. (a) Ishida, K.; Yano, S.; Yoshikawa, S. Inorg. Chem.
1986, 25, 3552–3554;
concentration ranges of 4×10−5 and 3.3×10−6
M
and for a 3.3×10−6 M salt solution containing 2×
10−2 M NGlc.
(b) Yano, S. Coord. Chem. Re6. 1988, 92, 113–156;
(c) Burger, K. Biocoordination Chemistry: Coordination
Equilibria in Biologically Acti6e Systems; Ellis: Morwood,
NY, 1990; pp 259–263.
3.4. Spectra
20. Smiataczowa, K.; Wawrzyno´w, A.; Korewa, R. Polish J.
Chem. 1995, 69, 1306–1314.
21. Perrin, D. D. Stability Constants of Metal-Ion Com-
plexes, Part B, Organic Ligands; Pergamon Press: Ox-
ford, New York, Toronto, Sydney, Paris, Frankfurt,
1979; p 274.
The NMR spectra were taken on a Varian Unity Plus
500 spectrometer at 500 MHz in CD3OD, and the
UV–Vis spectra on a Perkin–Elmer Lambda 18
instrument.
22. Baes, C. F.; Mesmer, R. E. The Hydrolysis of Cations;
Wiley-Interscience Publication, John Wiley and Sons:
New York, London, Sydney, Toronto, 1976; pp 241–293.
23. Sillen, L.G.; Martell, A.E. Stability Constants of Metal-
ion Complexes, Chemical Society Special Publications 17
and 25 (Supplement 1); The Chemical Society; London,
1964, 1971.
Acknowledgements
This research was supported by the Polish State
Committee for Scientific Research under grant BW
8000-5-0336-2.

K. Smiataczowa et al. / Carbohydrate Research 338 (2003) 969–975
975
24. Charlot, G.; Tremillon, B. Les reactions chemiques dans
les sol6ants; PWN; Warszawa, 1968 (in Polish).
25. De Ligny, C. L.; Luykx, P. F. M.; Rehbach, M.;
Wieneke, A. A. Rec. Tra6. Chim. 1960, 79, 713–726.
26. Kortu¨m, G. Lehrbuch der Elektrochemie; PWN;
Warszawa, 1970; p. 417 (in Polish).
27. Smiataczowa, K.; Maj, K.; Widernik, T.; Nesterowicz,
M. Polish J. Chem. 1998, 72, 587–594.
28. Christensen, J. J.; Rytting, J. H.; Izatt, R. M. J. Chem.
Soc., Sect. B 1970, 1646–1648.
29. Imamura, A.; Okajima, T.; Kanda, K. Bull. Chem. Soc.
Jpn. 1986, 59, 943–945.
30. (a) Khan, M.; Bouet, G. Transition Met. Chem. 1996, 21,
231–234;
(b) Khan, M. A.; Meullemeestre, J.; Schwing, M. J.;
Vierling, F. Inorg. Chem. 1989, 28, 3306–3309.
31. Nagy, L.; Yamashita, S.; Yamaguchi, T.; Sipos, P.; Wak-
ita, H.; Nomura, M. J. Inorg. Biochem. 1998, 72, 49–55.
32. Smiataczowa, K. J. Chem. Soc., Perkin Trans. II 1989,
1593–1598.