TEMPO-mediated oxidation of maltodextrins and D-glucose: Effect of pH on the selectivity and sequestering ability of the resulting polycarboxylates

Article abstract of DOI:10.1016/S0008-6215(00)00263-9

Maltodextrins were oxidized to polyglucuronic acids with the ternary oxidation system: NaOCl-NaBr-2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO). The chemoselective oxidation at the primary alcohol groups was shown to be strongly pH dependent. Oxidation of

Full text of DOI:10.1016/S0008-6215(00)00263-9

www.elsevier.nl/locate/carres
Carbohydrate Research 330 (2001) 21–29
TEMPO-mediated oxidation of maltodextrins and
D-glucose: effect of pH on the selectivity and
sequestering ability of the resulting polycarboxylates
Jean-Franc¸ois Thaburet, Nabyl Merbouh, Mathias Ibert, Francis Marsais,*
Guy Queguiner
UMR 6014 (CNRS), Institut de Recherche en Chimie Organique Fine (IRCOF), INSA of Rouen, BP 08,
F-76131 Mont-Saint-Aignan, France
Received 8 June 2000; received in revised form 6 September 2000; accepted 14 September 2000
Abstract
Maltodextrins were oxidized to polyglucuronic acids with the ternary oxidation system: NaOCl–NaBr–2,2,6,6-
tetramethylpiperidine-1-oxyl (TEMPO). The chemoselective oxidation at the primary alcohol groups was shown to
be strongly pH dependent. Oxidation of polysaccharides was best achieved at pH 9.5 in order to minimize
depolymerization, whereas oxidation of oligosaccharides required stronger alkaline conditions (pH 11–11.5). The
resulting sodium polyglucuronates present interesting sequestering properties, the best of which being obtained from
maltodextrins with the highest degrees of polymerization. The same oxidation process allowed the convenient
conversion of
D
-glucose to
D
-glucaric acid in high yield (\90%), under strongly basic conditions (pH\11.5).
© 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Primary alcohol group oxidation; TEMPO; Maltodextrins; Polyuronic acids; Water softeners
C-3 secondary positions, which induces car-
1. Introduction
bonꢀcarbon cleavage of glucose units to car-
boxy residues. In few cases, oxidation can be
limited to the primary hydroxyl groups, result-
ing in polyglucuronates. Such polyuronic
acids are often advantageous for applications
such as complexing agents or stabilizers. Some
regio-controlled oxidations at the C-6 hydrox-
yls have been reported, but they are not com-
pletely satisfactory. So, dioxygen in the
presence of platinum on carbon2–4 oxidizes
mono and disaccharides in moderate yields,
but only low yields were observed with
polysaccharides. The use of nitrogen dioxide
in the presence of sodium nitrite allows the
conversion of polysaccharides to poly-
glucuronic acids, but depolymerization com-
petitively occurs and secondary alcohol
Oxidation of carbohydrates such as starch
and maltodextrins is an important approach
to modify the physical and chemical proper-
ties of these compounds.1 Thus, oxidized
derivatives of starch can be used as gelling
agents, binders, and complexing agents (phos-
phate substitutes).
Many processes for the oxidation of carbo-
hydrate polymers are known, however most of
them result in some depolymerization. The
carbohydrate can be oxidized both at the C-6
primary hydroxyl groups and at the C-2 and
* Corresponding author. Tel.: +33-2-35522475.
E-mail address: francis.marsais@insa-rouen.fr (F. Marsais).
0008-6215/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(00)00263-9

22
J.-F. Thaburet et al. / Carbohydrate Research 330 (2001) 21–29
functionalities are partly converted to ketones.5
Similarly, autocatalytic oxidation of glucans
has been reported using sodium nitrate with
catalytic amount of sodium nitrite in phospho-
ric acid.6 Maltodextrins and starch have been
also oxidized by the sodium tungstate–hydro-
gen peroxide system leading to polycarboxylic
acids by cleavage of the C-2ꢀC-3 bonds.7
In the course of the 1980s, the new water-sol-
uble oxidation catalyst, 2,2,6,6-tetramethyl-1-
piperidinyloxy free radical (TEMPO) became
commercially available, and has been success-
fully used with cooxidants for the oxidation of
primary alcohols,8 and more recently of carbo-
hydrates. The TEMPO–NaBr–NaClO system
was first applied9 to the oxidation of partly
protected monosaccharides. The same condi-
tions allow to oxidize some high-molecular-
weight polysaccharides (starch, pullulan, …) to
polyuronic acids.10–12 Starch was selectively
oxidized to polyglucuronates with a selectivity
and yield of at least 95%, and a C-6 conversion
rate of 98%. Interesting results were recently
published concerning the TEMPO-mediated
oxidations of polysaccharides including starch,
maltodextrins, amylose, amylopectin and
hyaluronan.13 However, to our knowledge,
no TEMPO-mediated oxidations of small
oligosaccharides or unprotected monosaccha-
rides have still been described. Moreover, little
has been done to study the reaction parameters
such as pH, in the case of TEMPO-mediated
oxidation of poly- and oligosaccharides.
compared to reference products such as sodium
tripolyphosphate.
Our attention was mainly focused on the
oxidation of sugars with general structure 1
(n=1 to n\200) such as
D-glucose, glucose
syrups, maltodextrins and modified starch, as
well as hydrogenated oligosaccharides 2 such as
maltitol and maltotriitol (Scheme 1).
The average number of glucosidic units of
these poly-(1“4)-a- -glucopyranosides is
D
characterized by the degree of polymerization
(DP). The average number of lactol functions
(terminal hemiacetal group) is defined as the
dextrose equivalent (DE) and is a function of
the degree of polymerization: DE=100/DP.
The commercial maltodextrins (Entries 4–6;
DP\5) and glucose syrups (Entries 1–3; 5\
DP\1.25) are known as Glucidex™ (Table 1).
2. Experimental
Materials.—Water-soluble starch hydrolysis
products were supplied by Roquette Fre`res
(F-62080 Lestrem, France). Anhydrous a- -
D
glucose and TEMPO were acquired from
Aldrich. All other chemicals were analytical-
grade commercial products and were used with-
out purification.
Equipment.—Reactions were carried out in a
200 mL reactor, cooled in a salt–ice mixture,
and equipped with a thermometer and a pH
electrode. The reaction kinetics appeared to be
very fast for some pH values, and the temper-
ature can rise to 20 or 30 °C, if all the oxidant
is added at once. So, the reaction temperature
was kept about 5 °C, and addition of oxidant
was monitored by a Logilap™ system. Using
the same automatic system, the pH was main-
tained by titration with 1.5 M NaOH solution.
Therefore, in order to prepare sodium
polyglucuronates, and to characterize the prop-
erties of these potential sequestering agents
substitutes, we decided to investigate the
TEMPO-mediated oxidation of various
maltodextrins and glucose syrups, as well as
D
-glucose itself. The sequestering properties of
the resulting products were also evaluated and
Scheme 1. Maltodextrins (1) and malto-itols (2).

J.-F. Thaburet et al. / Carbohydrate Research 330 (2001) 21–29
23
Table 1
Oligo- and polysaccharides used for the oxidation experiments
Entry Oligosaccharide (trade name) DP Common
name
Entry Polysaccharides (trade name) DP
Common name
1
2
3
4
Glucidex 47
Glucidex 39
Glucidex 32
Glucidex 17
2.2 glucose syrup
2.6 glucose syrup
3.1 glucose syrup
5.9 maltodextrin
5
6
7
Glucidex 6
Glucidex 2
Pregeflo P250
17 maltodextrin
50 maltodextrin
\200 starch
precipitation as calcium salt form at pH 2.
Procedure.—Oxidation experiments were
Electrospray ionization mass spectrometry
(ESIMS) was performed at a flow of 5 mL/s
with samples dissolved in a 1:3 MeOH–water
solution (2.0 mg/L) containing 5 M NaOAc.
The GC-coupled mass spectrometry (DB1
column) used a Finnigan SSQ7000 spectrome-
ter, with samples (acid form) derivatized with
performed as follows: the carbohydrate (5 g
containing x mmol of primary alcohol and
hemiacetal functions), TEMPO (0.01 equiv),
and NaBr (0.15 equiv) were dissolved in water
(60 mL). A 1.8 M NaClO solution (2.2 mmol
NaClO/mmol primary alcohol and 1.1 mmol
NaClO/mmol of hemiacetal, which is a 10%
excess compared to the stoichiometric
amounts) was adjusted to the desired pH by
adding 4 M aq HCl. Both solutions were
cooled to 0 °C.
The oxidant solution was added at about
5 °C, and the appropriate pH was kept con-
stant by addition of NaOH. When the oxida-
tion was finished (stable pH value), the
reaction was quenched with EtOH, and the
pH adjusted to 8.5 by adding 1 M aq HCl.
Then, the solution was concentrated to 50
mL, and the oxidized carbohydrate was pre-
cipitated by adding 150 mL of EtOH. The
precipitate was dissolved in 50 mL water, and
the precipitation operation was repeated. The
resulting solid was washed with a 4:1 EtOH–
water mixture, and dried at 50 °C under re-
duced pressure.
a
mixture of bis(trimethylsilyl)trifluoro-
acetamide and trimethylsilylimidazole.
In GPC, products in the Na+ form were
fractionated with 3 columns (a precolumn OH
Pack SBG and two columns OH Pack
SB806HQ and OH Pack SB 804HQ) in series.
The products were detected by refractive index
and molar mass was determined on-line with a
DAWN-DSP-F (Wyatt Technology Co.)
MALLS detector (after pullulan calibration).
Samples (50 g/L) were dissolved in 0.1 M
LiNO3 solutions.
The calcium-complexing ability of the oxi-
dized carbohydrates was evaluated by titration
of a 100 mL solution containing 40 mg of
Ca(II) with known amounts of the substrate
(if necessary, the pH was adjusted to 10.5).
The activity of uncomplexed Ca(II) was mea-
sured using a XS530 Tacussel membrane elec-
trode reversible to calcium. The calcium-
sequestering capacity is defined as the number
of mg of Ca(II) bound by one gram of com-
plexant, until the concentration of the free
Ca(II) reaches 10−5 M (this value is generally
considered as the upper limit for Ca(II) during
a washing process14).
Analysis.—NMR spectra were recorded on
a Bruker AC 300F spectrometer, operating a
13C NMR frequency of 75 MHz. All oxidized
products as their carboxylate form were dis-
solved in D2O, and the carbon spectra
recorded in gate decoupling mode; chemical
shifts (l) are reported in ppm downfield from
Me4Si.
Carboxylate content of the oxidized samples
was measured by passing an aqueous product
solution (2 g) through a column of Dowex
(H+) resin to obtain the free-acid form. The
solutions were then titrated with 0.5 M
NaOH. Oxalate amounts were determined by
3. Results and discussion
Conversion of primary hydroxyl groups to
carboxylates by oxidation was estimated from

24
J.-F. Thaburet et al. / Carbohydrate Research 330 (2001) 21–29
the 13C NMR spectra (C-6 signal) (Table 2,
converted CH2OH). This overall conversion
rate is independent of the pH, and more
than 98% of the primary hydroxyl functions
were oxidized (the signals corresponding to
the primary alcohol functions at 61–62 ppm
have disappeared).
The oxidation selectivity was estimated by
comparison of the carboxylates amount of
the oxidized product (estimated by acid–
base titration) with the expected value calcu-
lated for a total and selective oxidation of
the primary hydroxyl and hemiacetal func-
tions. The oxidation yield (Table 2, molar
yield) was calculated as the weight ratio of
recovered polycarboxylates (mass of oxidized
product−mass of oxalate) to the theoretical
mass of sodium polyglucuronate.
ppm, Fig. 1) resulting from oxidative degra-
dation.
The high selectivity of the oxidation was
gained from the macroscopic analysis of the
recovered products (disappearance of the sig-
nal corresponding to the primary alcohol
groups in 13C NMR), but it lacked structural
information. So, a more accurate approach
could be achieved by a precise study of the
13C NMR spectra of the oxidized products
(Fig. 1). Sodium polyglucuronates should
show one signal in the carboxylic acid region
(170–180 ppm) and one signal around 98
ppm corresponding to the acetal C-1 carbon
atom, since all the internal (1“4)-linked
D
-
glucose units are similar. On examination,
the observed 13C NMR spectra appeared to
be more complex, especially for the product
oxidized at pH 11.4. This could be due to
the presence of smaller chains, in which both
the carboxylate and acetal carbon atoms
have different chemical shifts from those of
the higher terms. Deviation of the terminal
glucose unit can be excluded since it would
be independent of the oxidation pH.
Both conversion ratio and selectivity show
that oxidation is highly selective for the pri-
mary hydroxyl functions, which is in good
agreement with the corresponding good mo-
lar yield. This latter value is directly related
to the very low amounts of recovered
sodium oxalate (13C NMR signal at 173.4
Table 2
Effect of pH on the TEMPO-mediated oxidation of polysaccharides at 5 °C
Entry Carbohydrate (trade
name)
pH
Converted CH2OH Carboxylates (mmol/g) Molar yield
Sodium oxalate
(wt%)
(%)
(%)
Titration
Theory
5.12
1
2
3
4
Glucidex 2
Glucidex 2
Pregeflo P250
Pregeflo P250
9.5
\98
5.3
5.2
5.1
5.1
\95
\95
\95
\95
B1.0
B1.0
B1.0
B1.0
11.4 \98
9.5 \98
11.4 \98
5.06
Fig. 1. 13C NMR spectra of oxidized Glucidex 2 according to pH.

J.-F. Thaburet et al. / Carbohydrate Research 330 (2001) 21–29
25
Table 3
GPC analysis of oxidized polysaccharides
Entry Polysaccharide (trade
name)
pH
Converted
CH2OH (%)
Carboxylates
(mmol/g)
Mw (×10−3
(g/mol)
)
Mn (×10−3
(g/mol)
)
1
2
3
4
5
6
Glucidex 2
Glucidex 2
Glucidex 2
Pregeflo 250P
Pregeflo 250P
Pregeflo 250P
starting product
60096%
27095%
3940%
2500920%
25096%
6.4923%
68950%
38934%
2.3940%
9.5 \98
11.4 \98
starting product
9.5 \98
11.4 \98
5.3
5.2
5.1
5.1
67911%
2.6948%
Table 4
Effect of pH on TEMPO-mediated oxidation of oligosaccharides at 5 °C
Entry Oligosaccharide (trade
name)
pH
Converted
CH2OH (%)
Carboxylates (mmol/g) Molar yield
Sodium oxalate
(wt%)
(%)
Titration
Theory
6.64
1
2
3
4
5
6
7
8
9
Maltitol
Maltitol
10.2
11.4 \98
8.6
9.5
10.5
11.4 \98
9.5
10.2
11.4 \98
11.4 \98
11.4 \98
11.4 \98
91
6.8
6.6
6.2
6.2
6.1
6.1
6.5
6.4
6.3
6.5
6.2
5.9
87
96
84
87
92
94
88
92
95
94
95
95
7.5
1.5
11.0
8.0
3.0
1.5
6.0
2.0
B1.0
B1.0
B1.0
B1.0
Maltotriitol
Maltotriitol
Maltotriitol
Maltotriitol
Glucidex 39
Glucidex 39
Glucidex 39
Glucidex 47
Glucidex 32
Glucidex 17
85
89
97
6.15
88
92
6.33
10
11
12
6.51
6.11
5.64
GPC analysis was achieved on the starting
etal for Glucidex or primary alcohols for poly-
ols) compared to the C-6 primary alcohol. For
these oligosaccharides, it has been demon-
strated that the oxidation selectivity is highly
dependent on the pH (Entries 3–6 and 7–9
for example). The best selectivity is obtained
for pH higher than 11 (Table 4, Entries 2, 6,
and 9–12): the rate of converted CH2OH in-
creases with pH, and the formation of by-
products such as oxalates decreases.
In order to understand the oxidative degra-
dation of oligosaccharides, the oxidation
products of a low DP sugar (Glucidex 47)
were analyzed by GLC–MS, after trimethyl-
silylation of the analytical sample.
The resulting GLC–MS spectrogram (Fig.
2) presents two groups of signals, correspond-
ing to the oxidation products of the two main
families of the starting sugar, i.e., DP 2 and
DP 3. Compounds a–d, corresponding to the
DP 2 family, were identified by their MS
polysaccharides and on the resulting oxidation
products in order to confirm the hypothesis of
a competitive depolymerization. Comparison
of both number and mass average molecular
weights (Mn and Mw) of starting sugars and
oxidized products (Table 3, Entries 2, 3 and 5,
6, respectively, compared to Entries 1 and 4),
clearly proves that hydrolysis of these polysac-
charides occurs during the oxidation process.
So, oxidation of maltodextrins by the
TEMPO–NaClO–NaBr system must be
achieved at a pH lower than 10 (Entries 2 and
5) in order to minimize hydrolysis of the
polymeric structure. A similar behaviour has
been observed in the oxidation of pullulan.15
Contrary to polysaccharides, oligosaccha-
rides and reduced oligosaccharides of low DP,
such as maltitol (DP 2), maltotriitol (DP 3) or
Glucidex 39 (average DP 2.6), present a high
ratio of terminal oxidizable functions (hemiac-

26
J.-F. Thaburet et al. / Carbohydrate Research 330 (2001) 21–29
signals (Fig. 2). They have been oxidized at
C-6, and only differ from their terminal unit,
this one being more or less decarboxylated
(compounds a and b), or unchanged (com-
pounds c and d). Compounds e–h are similar
products, corresponding to the DP 3 series
with m/z values moved forward 392 units.
Electrospray analysis (ESIMS) of the same
sample — sodium salt — shows a similar
distribution for the DP 3, DP 2 and DP 1
terms, and confirms that oligosaccharides
were converted to polyglucuronic acids with
dicarboxylic hydroxylated terminal units being
more or less decarboxylated. No degradation
was detected on the internal glucosidic units.
These analytical results clearly show that
when degradation takes place, it mainly af-
fects the terminal units by stepwise decarboxy-
lation or by diol cleavage, but very few of the
diols of other units. From a theoretical point
of view, oxidation of a polymeric sugar could
be simply schematized by three basic oxidative
pathways as described in Scheme 2. Oxidation
of primary alcohol and hemiacetal units by
the TEMPO–NaOCl–NaBr system would
only compete with the oxidative path 2, but
not with path 3 (the rate of reaction 3 seems
to be very low, since no corresponding dicar-
boxylic derivatives were observed). The sec-
ondary reaction 2 could be due to the direct
reaction of the hypobromite and/or hypochlo-
rite ions on the diol functions.
In summary, oxidation of low DP carbohy-
drates, or reduced carbohydrates is selective at
the primary alcohol and hemiacetal functions
for pH’s higher than 11, thus indicating a
kinetic constant k1 higher than k2 and k3
(Scheme 2). On the other hand at pH lower
than 10.5, the oxidation selectivity decreases
due to greater influence of side reaction 2, and
the higher ‘concentration’ of terminal oxidiz-
able units of the oligosaccharides.
The catalytic effect of the bromide anions
was clearly established with an optimum ratio
of 0.15 equiv against primary alcohol and
hemiacetal functions. This effect is especially
important for pH values higher than 10.5. It
has been demonstrated that oxidation of
maltodextrins and starch by sodium hypochlo-
rite only leads to the C-2ꢀC-3 bonds cleavage
without oxidation of the primary alcohol
groups.16 The catalytic effect of the bromide
anions has been reported for the oxidation of
inulin by alkaline sodium hypochlorite.17
More recently, it has been claimed that sono-
catalysis allows TEMPO-mediated oxidation
Fig. 2. GLC–MS characterization of oxidized Glucidex 47 (R=SiMe3).

J.-F. Thaburet et al. / Carbohydrate Research 330 (2001) 21–29
27
Scheme 2. Hypohalite ion behaviour in the TEMPO-mediated oxidation of maltodextrins. Path a: selective oxidation at primary
hydroxyl and hemiacetal units by TEMPO. Path b: cleavage of the C-2ꢀC-3 bonds of the terminal hemiacetal units. Path c:
cleavage of the C-2ꢀC-3 bonds of the glucose units.
Table 5
Effect of pH on the TEMPO-mediated oxidation of monosaccharides at 5 °C
Entry
Sugar
pH
Sodium glucarate (% isolated yield)
By-products (% molar yield)a
1
2
3
4
5
6
D
D
D
D
D
D
-glucose
9.5
B10
25
64
90
27
ST: 35; SG: 45; SO: 20
ST: 28; SG: 40. SO: 15
ST: 10; SG: 20; SO: 7
ST: 5; SG: 1; SO: 2
ST: 18; SG: 24; SO: 14
ST: 5; SG: 1; SO: 3
-glucose
-glucose
-glucose
-glucitol
-glucitol
10.5
11.2
11.7
10.5
11.7
89
a ST, sodium tartrate; SG, sodium gluconate; SO, sodium oxalate.
Pure glucaric acid was therefore isolated in
good yield (90%) as the mono potassium salt
(precipitation at pH 3.4), after oxidation of
of glucosides using basic sodium hypochlorite
without bromide.18
Our experience of the control of carbohy-
D-glucose at pH 11.7 by the TEMPO–
drates oxidation was then successfully ex-
tended to glucose and its reduction product,
glucitol. Since oxidation of oligosaccharides
proved to be strongly pH dependent due to
the competitive degrading oxidation of the
terminal glucose unit, oxidation of monosac-
charides to glucaric acid should be more
difficult to control.
As expected, oxidation of glucose and gluci-
tol (Table 5) with the TEMPO–NaOCl–NaBr
system proved to be selective only for the
highest pH values, 11.5–12 (Entries 4 and 6).
At pH 9.5, less than 10% glucaric acid was
obtained (Entry 1). Mainly degradation prod-
ucts such as sodium tartrate, oxalate and car-
bonate were formed by oxidative diol cleavage
(Scheme 3).
Scheme 3. Oxidation reactions of
D-glucose.

28
J.-F. Thaburet et al. / Carbohydrate Research 330 (2001) 21–29
Table 6
Sequestering capacity of TEMPO-mediated oxidized carbohydrates
Entry
Oxidized sugar
Oxidation pH
Oxidation degree (%)
‘SC’ (mg of Ca/g)
1
2
3
4
5
6
7
8
glucose
maltitol
11.7
11.4
11.4
11.4
11.4
11.4
9.5
\98
\98
\98
\98
\98
\98
\98
\98
6
8
8
10
14
20
32
32
Glucidex 47
Glucidex 32
Glucidex 17
Glucidex 2
Glucidex 2
Pregeflo P250
9.5
NaOCl–NaBr system. Oxidation of protected
resulting from partial oxidation of glucose or
glucitol due to the inadequate amount of oxi-
dant (Scheme 3).
The previously prepared sodium polyglu-
curonates obtained by oxidation of various
maltodextrins were compared with sodium
tripolyphosphate (TPPNa) as water softeners.
The sequestering ability was measured by an
addition method with determination of their
sequestering capacities (SC) (Table 6).
Compared to sodium tripolyphosphate (SC
120 mg of Ca/g), we observed lower sequester-
ing properties for the sodium salts of polyglu-
curonic acids obtained by oxidation of
maltodextrins. The sequestering power is
clearly related to the molar mass of the start-
ing carbohydrates (Table 6), most likely due
to the double-helical structure of polysaccha-
rides which could favour the Ca2+ cations
complexation. So, oxidation of Glucidex 2 at
pH 11.4 led to a sodium salt with a SC of 20
mg Ca/g (Entry 6), whereas oxidation at pH
9.5 gave a less depolymerized product with a
higher SC value of 32 (Entry 7).
In conclusion, TEMPO-mediated oxidation
of carbohydrates is a reliable method to
chemoselectively oxidize primary alcohol
groups of starch hydrolysis products
(maltodextrins). The applied pH is quite im-
portant and depends on the degree of poly-
merization of the starting carbohydrates.
Oxidation of polysaccharides has been best
achieved at pH 9.5 to limit depolymerization,
whereas more alkaline conditions were re-
quired for lower DP oligosaccharides (i.e., pH
sugar such as a-methyl glucoside to sodium
methyl a- -glucuronate, with a selectivity of
D
95% at pH 10.0, has been reported,9 but no
selective oxidations to glucaric acid are cur-
rently available.
When the oxidation was achieved at a pH
lower than 11, other carboxylates observed in
13C NMR were formed such as sodium tar-
trate, oxalate or carbonate (Entries 1, 2 and
5). For information, signals of the different
13
carbons in C NMR in D2O are as follows: l
(sodium glucarate) 179.2, 179.0, 74.0(2), 73.9,
71.9, l (sodium gluconate) 179.5, 74.3, 72.8,
71.4, 71.2, 62.9, l (sodium tartrate) 178.9 and
177.9, 75.1 and 74.1, l (sodium oxalate) 73.4,
l (sodium carbonate) 168.5 (depending on
pH).
Careful identification of the 13C signals of
tartaric acid in the crude oxidation product
was done by addition of meso or racemic
sodium tartrate. This allows us to identify the
sodium tartrate as a mixture of the two
diastereoisomers, with only 20% of the meso
compound. The ratio of the two diastereoiso-
1
mers was calculated from the H NMR spec-
trum of the crude oxidation product (l 3.98
and 4.05). This would be in favour of two
degradative oxidations of
D
-glucose by cleav-
age of both the C-4ꢀC-5 and the C-2ꢀC-3
bonds, the former being major (characteriza-
tion of the major stereoisomer, presumably
the
L
form, and mechanistic explanation of its
formation are currently in progress).
Formation of these by-products by degrad-
ing oxidation of diols requires greater
amounts of oxidant; this could explain the
simultaneous presence of sodium gluconate,
11.0–11.5), as well as for
D
-glucose or -gluci-
D
tol (i.e., pH\11.5).

J.-F. Thaburet et al. / Carbohydrate Research 330 (2001) 21–29
29
3. Dirkx, J. M. H.; van der Baan, H. S.; van der Broek, J.
M. A. Carbohydr. Res. 1977, 59, 63.
4. Leupold, E. I.; Schoenwaelder, K. H.; Fritsche-Lang, W.
DE Patent 3 900 677 (1989); Chem. Abstr. 1990, 113,
214316h.
5. Painter, T. J.; Cesaro, A.; Delben, F.; Paoletti, S. Carbo-
hydr. Res. 1985, 140, 61–68.
Glucaric acid has been synthesized with
high selectivity and good yield (90%), which is
better than the known methods for oxidation
of unprotected glucose (highest yield of
65%).3,19
The sequestering capacities of the oxidation
products increase with DP, and can reach a
value of 32 mg Ca/g of substrate. This value
could be increased by creating more carboxy-
late functions, for example by partial oxida-
tive cleavage of internal vicinal C-2ꢀC-3 diols
of maltodextrins without degrading the struc-
ture. Such investigations have been realized
and will be reported in a following paper.
6. de Nooy, A. E. J.; Pagliaro, M.; van Bekkum, H. Carbo-
hydr. Res. 1997, 304, 117–123.
7. Floor, M.; Schenck, K. M.; Kieboom, A. P. G.; van
Bekkum, H. Starch 1989, 41, 303–309.
8. Semmelhack, M. F.; Chou, C. S.; Cortes, D. A. J. Am.
Chem. Soc. 1983, 105, 4492–4494; Anelli, P. L.; Biffi, C.;
Montanari, F.; Quici, S. J. Org. Chem. 1987, 52, 2559–
2562.
9. Davis, N. J.; Flitsch, S. L. Tetrahedron Lett. 1993, 34,
1181–1184.
10. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H. Recl.
Tra6. Chim. Pays-Bas 1994, 113, 165–166.
11. de Nooy, A. E. J.; Besemer, A. C. WO Patent 07 303
(1995); Chem. Abstr. 1995, 123, 202726k.
12. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H.
Carbohydr. Res. 1995, 269, 89–98.
Acknowledgements
13. Fleche, G. CA Patent 2 193 034 (1997); Chem. Abstr.
1997, 127, 307619p; Chang, P. S.; Robyt, J. F. J. Carbo-
hydr. Chem. 1996, 15, 819–830; Jiang, B.; Drouet, E.;
Milas, M.; Rinaudo, M. Carbohydr. Res. 2000, 327,
455–461.
14. Hollingsworth, M. W. J. Am. Oil Chem. Soc. 1978, 55,
49–51.
15. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H.; van
Dijk, J. A. P. P.; Smit, J. A. M. Macromolecules 1996, 29,
6541–6547.
We acknowledge the Roquette Fre`res Co.,
the Association Nationale de Recherche Tech-
nique (ANRT), and the Centre National de la
Recherche Scientifique (CNRS) for their
financial support.
16. Floor, M.; Kieboom, A. P. G.; van Bekkum, H. Starch
1989, 41, 348–354.
17. Besemer, A. C.; van Bekkum, H. Recl. Tra6. Chim.
Pays-Bas 1994, 113, 398–402.
18. Brochette-Lemoine, S.; Joannard, D.; Descotes, G.;
Bouchu, A.; Queneau, Y. J. Mol. Catal. 1999, 150,
31–36.
19. Mehltretter, C. L.; Rist, C. E.; Alexander, B. J. US
Patent 2 472 168 (1948); Chem. Abstr. 1949, 48, 7506c;
Besson, M.; Fleche, G.; Fuertes, P.; Gallezot, P.; Lahmer,
F. Recl. Tra6. Chim. Pays-Bas 1996, 115, 217–221.
References
1. van Bekkum, H. In Carbohydrates as Organic Raw Mate-
rials; Lichtenthaler, F. W., Ed.; VCH: Weinheim, 1991;
pp. 289–310; Besemer, A. C.; van Bekkum, H. In Carbo-
hydrates as Organic Raw Materials III; van Bekkum, H.;
Ro¨per, H.; Voragen, F., Eds.; VCH: Weinheim, 1996; pp.
273–293.
2. Heyns, K.; Paulsen, H. Ad6. Carbohydr. Chem. Biochem.
1962, 17, 169.
.