13C NMR studies of isomerization of D-glucose in an aqueous solution of Ca(OH)2. The effect of molecular oxygen

Article abstract of DOI:10.1007/s11172-006-0065-x

Isomerization of D-glucose to fructose and mannose in aqueous solutions of Ca(OH)₂ with the initial pH 11.4 in a temperature interval of 20-90°C was studied by ¹³C NMR spectroscopy in the presence and absence of dissolved oxygen. In the presence of oxygen, the apparent equilibrium isomerization constant is much lower than that in the absence of oxygen. This is related to the oxidation of monosaccharides to formic and aldonic acids, a decrease in the pH of solutions, and cessation of isomerization at pH < 9.

Full text of DOI:10.1007/s11172-006-0065-x

Russian Chemical Bulletin, International Edition, Vol. 54, No. 8, pp. 1967—1972, August, 2005
1967
13C NMR studies of isomerization of Dꢀglucose
in an aqueous solution of Ca(OH)2.
The effect of molecular oxygen
A. N. Simonov, O. P. Pestunova,ꢀ L. G. Matvienko, and V. N. Parmon
G. K. Boreskov Institute of Catalysis, Siberian Branch of the Russian Academy of Sciences,
5 prosp. Akad. Lavrent´eva, 630090 Novosibirsk, Russian Federation.
Fax: +7 (383) 330 4719. Eꢀmail: oxanap@catalysis.ru
Isomerization of Dꢀglucose to fructose and mannose in aqueous solutions of Ca(OH)2 with
the initial pH 11.4 in a temperature interval of 20—90 °С was studied by 13C NMR spectrosꢀ
copy in the presence and absence of dissolved oxygen. In the presence of oxygen, the apparent
equilibrium isomerization constant is much lower than that in the absence of oxygen. This is
related to the oxidation of monosaccharides to formic and aldonic acids, a decrease in the pH
of solutions, and cessation of isomerization at pH < 9.
Key words: carbohydrates, glucose, fructose, rearrangement, 13C NMR spectroscopy, alkaꢀ
line solution, isomerization.
Epimerization of aldoses and their rearrangement to
ketoses, which occur in parallel in weakly alkaline aqueꢀ
ous solutions, the Lobry de Bruyn—Alberda van Ekenstein
rearrangement (hereinafter LdB—AvE) were discovered
in the late 1890s.1,2 For a long time, this reaction reꢀ
mained to be the simplest and most convenient method
for the synthesis of a series of rare monosaccharides. The
isomerization is known3,4 to proceed through the enediol
form (Scheme 1).
calcium ions. The study of the LdB—AvE reaction by
HPLC showed that the reaction is accompanied by side
processes and the composition of the products depends
on the nature of the starting monosaccharide.11,12
In the present work, we studied the transformation of
glucose in an aqueous solution of calcium hydroxide by
13С NMR spectroscopy. Isomerization was carried out in
a wide temperature interval from 20 to 90 °С in argon,
oxygen, and air.
Experimental
Scheme 1
DꢀGlucose (purity grade, Reakhim), Dꢀfructose (99%,
Sigma), Ca(OH)2 (analytical purity grade, Reakhim), NaOH
(analytical purity grade, Reakhim), and HCl (reagent grade,
Reakhim) were used. A suspension of Ca(OH)2 (0.09 g) in 50 mL
of water was prepared using a Bransonic 1510RꢀNT (USA) ultraꢀ
sonic bath for 25 min.
DꢀGlucose (9.0 g, 0.05 mol) and the suspension of Ca(OH)2
(50 mL) were placed in a threeꢀnecked reactor kept at constant
temperature and equipped with a thermometer, a tube for gas
supply, and a magnetic stirrer. If the reaction was carried out in
argon or oxygen, the reactor was preliminarily purged with the
corresponding gas, which was supplied through a temperatureꢀ
controlled saturator with distilled water during the reaction. The
system became homogeneous during the reaction. The temperaꢀ
ture interval of the reaction in air was 20—90 °С, and that for
argon and oxygen was 50—90 °С.
Samples for analysis were withdrawn every 24 h at room
temperature, and for other temperatures the intervals were the
following: at 50 °С, 12 h; at 60 and 70 °С, 20—40 min in air and
oxygen and 60—90 min in argon; at 80 and 90 °С, 10—20 min.
The samples were cooled to room temperature and analyzed (on
At the same time, the kinetics of transformation and
accumulation of monosaccharides remains unclear even
for the most studied reaction of this type, namely, isomerꢀ
ization of glucose to fructose and mannose.5—8 The use of
polarimetry5,6 for this purpose is not quite correct,9,10
because the change in the optical rotation is affected by
both isomerization in the presence of Ca(OH)2 as the
catalyst and formation of monosaccharide complexes with
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 8, pp. 1909—1913, August, 2005.
1066ꢀ5285/05/5408ꢀ1967 © 2005 Springer Science+Business Media, Inc.

1968
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 8, August, 2005
Simonov et al.
the same day) by 13С NMR spectroscopy on a Bruker DPXꢀ250
pulse FT NMR spectrometer (Germany) at room temperature.
Spectra were accumulated over 256 scans with a pulse delay of
3.088 s, which made it possible to obtain an acceptable signalꢀ
toꢀnoise ratio. To reveal oxidation products of saccharides and
other byꢀproducts in a reaction mixture, we recorded 13С NMR
spectra with a pulse delay of 6.088 s and a scan number of 8237.
Table 1. Time of apparent equilibration between glucose and
fructose in air, oxygen, and argon
T/°С
Time of apparent equilibration/h
Air, oxygen
Argon
20
50
60
70
80
90
>216
48
2.5
<1
*
1/ 4
—
71
6
2.5
1
Acetone (δ 30.89)13 was used as the internal standard.
Me
In experiments at 60 and 80 °C (in argon and air), samples
were additionally analyzed by UV spectroscopy. The spectra
were recorded in a 0.1ꢀcm cell with a UVIKON 923 spectromꢀ
eter (Kontron, Italy).
A solution of Dꢀglucose (0.5 mol L–1) in 0.5 М NaOH
(рН 12.5) was stored under argon for 18 h and acidified with 2 М
HCl to pH 11.4, 10.9, 10.5, 9.4, 8.0, and 6.2 (Iꢀ120 pHꢀmeter,
USSR), and absorption spectra were recorded.
3/4
* 25 min.
ditions studied using the obtained concentrations of difꢀ
ferent forms of saccharides (see Table 2). The temperaꢀ
ture plots of the resulting Ki values in the Arrhenius coorꢀ
dinates are presented in Fig. 1. These plots are very simiꢀ
lar for the isomerization in oxygen and air (see Fig. 1). In
the presence of oxygen, the increase in temperature is
accompanied by an increase in Ki. On the contrary, when
the isomerization occurs in argon, Ki decreases slightly
with temperature. At high temperatures (80—90 °С),
the Ki values for inert and oxygenꢀcontaining atmospheres
are almost the same, while at 50 °С they differ approxiꢀ
mately twofold.
The volume of absorbed oxygen was measured in a volumetꢀ
ric setup with a reactor, the temperature of which was mainꢀ
tained at 60 °С.
Results and Discussion
The chemical shifts of the C atoms of different isoꢀ
meric forms of glucose, fructose, and mannose in the
13C NMR spectra were assigned using published data.14,15
The relative content of each monosaccharide in a reꢀ
action mixture was determined from the integral intensiꢀ
ties of signals corresponding to the СН2 groups.
The intensities of signals for mannose at temperatures
below 80 °С corresponded to the mannose concentration
not higher than 1% of the initial concentration of glucose.
Only at 90 and 80 °C, the mannose content in the mixture
was 2—3% of the initial glucose concentration. Thereꢀ
fore, this concentration was neglected in further calculaꢀ
tions of concentrations of glucose and fructose.
To reveal a reason for that strong differences in the
isomerization parameters in the presence and absence of
oxygen at low temperatures, we studied the isomerization
lnKi
When the process was carried out in air at room temꢀ
perature, the apparent equilibrium was not attained even
after nine days (216 h). The calculated concentrations of
monosaccharides glucose, fructose, and mannose under
these conditions obtained by the extrapolation of the kiꢀ
netic curves were 86, 14, and <<1%, respectively. These
values of the apparent equilibrium concentrations differ
substantially from those determined by polarimetry under
identical conditions5,6: 70, 28.5, and 1.5%, respectively.
For the isomerization of Dꢀglucose in a solution of KOH
in an inert atmosphere,11,12 the content of these monosacꢀ
charides was 50.0, 33.5, and 8.4%, respectively (HPLC).
The mixture also contained 9.1% of other saccharides.
As should be expected, the increase in temperature is
accompanied by a considerable shortening of the time of
apparent equilibration. For the isomerization in argon,
this time is longer than in air and oxygen (Table 1).
The concentrations of monosaccharides at the end of
the reaction also differ substantially (Table 2).
1
2
–0.80
3
–1.20
–1.60
0.0029
0.0033
T
–1/K–1
Fig. 1. Temperature plots of the obtained apparent constants Ki
in the Arrhenius coordinates for Dꢀglucose isomerization in arꢀ
gon (1), oxygen (2), and air (3).
We calculated the apparent equilibrium constants (Ki)
of the isomerization of glucose to fructose under the conꢀ

13C NMR study of glucose isomerization
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 8, August, 2005
1969
Table 2. Calculated apparent equilibrium isomerization constants (Ki) and concentrations (C) of monosacꢀ
charides at the end of the reaction under different conditions
T/°С
Air
C/mol L–1
Oxygen
C/mol L–1
Argon
C/mol L–1
Ki
Ki
Ki
Glc
Fru
Glc
Fru
Glc
Fru
20
50
60
70
80
90
0.86
0.76
0.73
0.70
0.67
0.63
0.14
0.24
0.27
0.30
0.33
0.375
0.16
0.31
0.37
0.43
0.50
0.60
—
—
—
—
—
—
0.78
0.73
0.68
0.67
0.66
0.22
0.27
0.32
0.33
0.34
0.29
0.37
0.48
0.50
0.53
0.62
0.63
0.64
0.65
0.65
0.38
0.37
0.36
0.35
0.35
0.61
0.58
0.57
0.55
0.54
of glucose at 60 and 80 °С in more detail, including the
kinetics of oxygen absorption from air by the reaction
mixture at 60 °C (Fig. 2). In addition, changes in the
absorption spectra in the UV region (Fig. 3) and in the pH
of solutions (Fig. 4) at 60 and 80 °С in air and argon were
studied in parallel. The effect of pH is very important,
because the rate of the LdB—AvE reaction was shown11,12
to depend strongly on the pH of solutions, no interaction
occurring at pH < 9.
Oxygen absorption by the reaction mixture is described
by the zeroꢀorder kinetics with respect to oxygen and
stops after ~3 h (see Fig. 2). A small induction period in
the initial region can be explained by the time necessary
for complete dissolution of glucose. When the solution
that absorbed oxygen is acidified to pH 7, some amount of
oxygen is evolved, which is, most likely, reversibly bound
with monosaccharides. The amount of irreversibly abꢀ
sorbed oxygen was approximately 2.2% of the amount of
glucose. After completion of the reaction in the presence
of oxygen, the 13С NMR spectra also contained signals at
δ 170—220 corresponding to the carbonyl groups. Figꢀ
ure 5 shows the spectrum of the reaction mixture after
completion of the reaction in oxygen at 80 °С. The specꢀ
trum was detected with a pulse delay of 6.088 s and a scan
number of 8192 and exhibits signals for byꢀproducts that
A
7
a
6
2.0
1.5
1.0
0.5
5
4
3
2
1
250
300
350
λ/nm
A
V0 – Vt/mL
b
30
1.5
1.0
0.5
3
20
4
2
1
2
6
5
10
7
1
250
300
350
λ/nm
0
Fig. 3. Changes in the electronic absorption spectra of aqueous
solutions of glucose during isomerization at 60 °С in argon (a)
and in air (b); a: t = 30 (1), 60 (2), 90 (3), 120 (4), 240 (5),
300 (6), and 360 min (7); b: t = 30 (1), 60 (2), 90 (3), 120 (4),
150 (5), 180 (6), and 210 min (7).
0
50
100
150
τ/min
Fig. 2. Kinetics of oxygen absorption by an aqueous solution of
glucose at 60 °С under atmospheric pressure (1) and oxygen
evolution upon acidification of the final solution to pH 7 (2).

1970
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 8, August, 2005
Simonov et al.
pH
pH
a
b
1
2
10
9
10
9
8
8
0
100
200
300
t/min
0
25
50
75
100 t/min
A
A
c
d
1.6
1.2
0.8
0.4
2.0
1.5
1.0
0.5
0
25
50
75
100 t/min
0
100
200
300
t/min
Fig. 4. Changes in pH (a, b) and absorbance (c, d) of aqueous solutions of glucose at 290 nm in an inert atmosphere (1) and in air (2)
during isomerization at 60 (a, c) and 80 °С (b, d).
are present in the reaction mixture. Their amount is
about 2—3%.
Oxygen absorption by solutions of monosaccharides
can be explained16 by the oxidation of the latter to acids.
In this case, oxygen attacks the same enediol form of
monosaccharides, which is the intermediate in the monoꢀ
saccharide rearrangement. The formation of acids explains
completely the observed decrease in the pH of the soluꢀ
tion by more than three units (from 10.7 to 7.5, see
Fig. 4, a).
The kinetics of changes in pH of solutions in an inert
atmosphere and in oxygen differ significantly (see Fig. 4).
In the presence of oxygen, pH decreases much more rapꢀ
idly, and the sharp decrease in pH (see Fig. 4, a) occurs
simultaneously with the retardation of oxygen absorption
(see Fig. 2). In an inert atmosphere, pH of solutions also
decreases, but much more slowly (see Fig. 4). In this case,
the decrease in pH is related, most likely, to rearrangeꢀ
ments of carbohydrates due to which saccharinic acids are
formed. A comparison of the kinetic curves of pH changes
(see Fig. 4) and the times of apparent equilibration of the
system (see Table 1) shows that the isomerization ceases
approximately when the pH of the solutions becomes lower
than 9.
Changes in the concentration of the active intermediꢀ
ate enediol form of monosaccharides during the reaction
can be monitored by UV spectroscopy.17,18 As can be
seen from the data in Figs 3, a and b, the absorption band
appears at λ = 290 nm in the presence and absence of
oxygen. The absorbance of this band in argon increases
Oxidation
products
?
?
210
170
130
90
50
δ
Fig. 5. 13С NMR spectrum of the reaction mixture after the end
of Dꢀglucose isomerization in Ca(OH)2 solutions in an oxygen
atmosphere at 80 °С.

13C NMR study of glucose isomerization
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 8, August, 2005
1971
and reaches a plateau to the moment of equilibration. In
the presence of oxygen, the absorbance of this band first
increases and then decreases but a new band with a maxiꢀ
mum at λ = 270 nm appears. The absorbance maximum
at λ = 290 nm corresponds to the transition from the
slow region of the kinetic curve of pH change to the fast
region.
saccharinic acids. Since the first reaction is of zero order
with respect to the oxygen pressure over the solution, we
believe that its rate is mainly determined by dissolution of
oxygen in water. At high temperatures, the solubility of
oxygen in water decreases and, hence, this reaction rate
should decrease with temperature. At the same time, rearꢀ
rangements of monosaccharides in the aqueous phase inꢀ
volve no gaseous oxygen and, therefore, their rates inꢀ
crease and become higher than the oxidation rate. As a
result, the pH of the solution at an elevated temperature
has time to decrease to a critical value of 9 more rapidly
than noticeable oxidation of monosaccharides with oxyꢀ
gen occurs.
Our study of isomerization shows that the data obꢀ
tained in the early XX century on the equilibrium (as it
was believed) concentrations of monosaccharides formed
by the LdB—AvE rearrangement should be corrected subꢀ
stantially using the modern direct methods of investigaꢀ
tion. The ratio of concentrations of isomeric monosacꢀ
charides at the end of isomerization in aqueous solutions
is substantially affected by both temperature and pH. In
addition, it should be taken into account that the presꢀ
ence of oxygen in a solution and, as a consequence, the
formation of an even insignificant amount of oxidation
products of monosaccharides can result in a substantial
and usually uncontrolled decrease in pH of the solution.
As a result, the isomerization can be retarded strongly
followed by the apparent equilibration, which does not
correspond to the real thermodynamic properties of sacꢀ
charides. Therefore, the data on isomerization constants
of saccharides can be considered as correct only when the
reaction occurs in preꢀdeaerated solutions under pHꢀstatic
conditions.
The behavior of monosaccharides in aqueousꢀalkaline
media (pH ≥ 12.5) was studied by UV spectroscopy.17,18 It
was shown17 that in an inert atmosphere the spectrum of
glucose contains an absorption band with λ = 310 nm,
which shifts to 340 nm in the presence of СаCl2. After
saturation of an alkaline aqueous solution of a sugar with
oxygen, an intense band appears at λ = 265 nm. This band
can be ascribed17 to oxidation products of carbohydrates.
A comparison of the kinetics of changes in the absorbance
of bands in the UV spectra with the kinetics of H/D
exchange and degradation of sugars suggests that the
band with λ = 310 nm is caused by the absorbance of the
enediol anion, the band at λ = 340 nm is attributed to
calcium complexes with the enediol form of carbohydrate,
and that with λ = 265 nm corresponds to the oxidation
products of monosaccharides. Similar conclusions about
the nature of bands with λ = 310 and 340 nm are reꢀ
ported.18
The spectra observed in the present work somewhat
differ from those described in literature. In an oxygenꢀ
containing atmosphere, they contain two intense bands
with λ = 270 and 290 nm and a weak band at λ = 340 nm.
Low intensity of the latter can be explained by low conꢀ
centration of calcium as compared with that of glucose.
The shift of the band with λ = 310 nm, which corresponds
to the enediol anion, to λ = 290 nm is related to lower
pH values of the solutions under study. In fact, the UV
spectrum of a 0.5 М solution of Dꢀglucose in 0.5 М NaOH
(pH 12.5) stored under argon at room temperature for
18 h contains a single band with λmax = 310 nm. When
this solution is acidified to pH 11.4, 10.5, 9.4, and 8.0, the
position of the band maximum shifts to 290 nm and the
absorbance decreases slightly, which can be related, probꢀ
ably, to the transformation of the enediol anion into the
protonated form.
The authors thank A. V. Golovin for help in registraꢀ
tion of NMR spectra.
This work was financially supported by the Division of
Chemistry and Materials Science of the Russian Acadꢀ
emy of Sciences (Grant 9.3), Program "Integration
Projects of the Siberian Branch of the Russian Academy
of Sciences" (Grant 148), and Council on Grants of the
President of the Russian Federation (Program of State
Support for Leading Scientific Schools of the Russian
Federation, Grant NSh 1484.2003.3).
Differences in the absorption spectra and dynamics of
pH changes in aqueous solutions of glucose in an inert
and oxygenꢀcontaining atmosphere are much more proꢀ
nounced at 60 °С than at higher (80 °С) temperature (see
Figs 3 and 4). At the same time, the concentrations of
glucose and fructose at the end of the reaction in an inert
and oxygenꢀcontaining atmosphere at 80 and 90 °С are
almost the same, which can be explained as follows. At
least two reactions contribute to a decrease in pH of soluꢀ
tions in the presence of oxygen. These are, first, oxidation
of saccharides with atmospheric oxygen to form acids
and, second, rearrangement of carbohydrates to form
References
1. C. A. Lobry de Bruyn and W. A. Alberda van Ekenstein,
Recl. Trav. Chim. PaysꢀBas, 1895, 14, 195.
2. C. A. Lobry de Bruyn and W. A. Alberda van Ekenstein,
Recl. Trav. Chim. PaysꢀBas, 1897, 16, 256.
3. J. C. Speck, Jr., Adv. Carbohydr. Chem., 1958, 13, 63.
4. S. J. Angyal, in Glycoscience: Epimerization, Isomerizaꢀ
tion, and Rearrangement Reactions of Carbohydrates,

1972
Russ.Chem.Bull., Int.Ed., Vol. 54, No. 8, August, 2005
Simonov et al.
Ed. A. F. Stütz, SpringerꢀVerlag, Berlin—Heidelberg,
2001, p. 1.
13. H. E. Gottlieb, V. Kotlyar, and A. Nudelman, J. Org. Chem.,
1997, 62, 7512.
5. J. C. Sowden and R. Schaffer, J. Am. Chem. Soc., 1952,
74, 505.
6. M. L. Wolform and W. L. Lewis, J. Am. Chem. Soc., 1928,
50, 837.
7. A. M. Kuzin, Zh. Obshch. Khim., 1935, 5, 1373.
8. R. Yanagihara, K. Saeda, S. Osana, and S. Yshikawa, Bull.
Chem. Soc. Jpn, 1997, 66, 2268.
14. The Sadtler Standard Spectra: Cꢀ13 Nuclear Magnetic Resoꢀ
nance Spectra, Sadtler Research Lab., Philadelphia, 1978,
No. 822c; No. 4681c.
15. K. Bock and R. Thoegersen, Ann. Repts NMR Spectroscopy,
1982, 13, 1.
16. W. B. Gleason and R. Barker, Can. J. Chem., 1971, 49,
1425; 1433.
9. K. Fujino, J. Kobayashi, and I. Higuchi, Nippon Kagaku
Kaishi, 1972, 2287.
17. G. de Wit, A. P. G. Kieboom, and H. van Bekkum,
Carbohydr. Res., 1979, 74, 157.
10. J. A. Rendelman, Jr., Adv. Carbohydr. Chem., 1966, 21, 209.
11. H. S. El Khadem, S. Ennifar, and H. S. Isbell, Carbohydr.
Res., 1987, 169, 13.
18. T. I. Khomenko and O. V. Krylov, Kinet. Katal., 1974, 15,
625 [Kinet. Catal., 1974, 15 (Engl. Transl.)].
12. H. S. El Khadem, S. Ennifar, and H. S. Isbell, Carbohydr.
Res., 1989, 185, 51.
Received January 30, 2004;
in revised form January 19, 2005