Facile nitroxide-mediated oxidations of D-glucose to D-glucaric acid

Article abstract of DOI:10.1016/S0008-6215(01)00231-2

The oxidation of D-(+)-glucose to D-glucaric acid using the TEMPO-like nitroxide oxidation catalyst, 4-acetamido-2,2,6,6-tetramethyl-1-piperidinyloxy (4-acetamido-TEMPO) was carried out using several oxidizing agents and co-catalyst. The pH and temperature of the reactions were closely monitored to decrease degradations during the oxidation, and several isolation methods were explored.

Full text of DOI:10.1016/S0008-6215(01)00231-2

www.elsevier.com/locate/carres
Carbohydrate Research 336 (2001) 75–78
Note
Facile nitroxide-mediated oxidations of
D-glucose to
D-glucaric acid
Nabyl Merbouh,b Jean Francois Thaburet,a Mathias Ibert,a Francis Marsais,a
James M. Bobbittb,*
aIRCOF-INSA de Rouen, BP08, 76131 Mont Saint Aignan, France
bDepartment of Chemistry, Uni6ersity of Connecticut, Storrs, CT 06269-3060, USA
Received 7 June 2001; accepted 10 August 2001
Abstract
The oxidation of
D
-(+)-glucose to D-glucaric acid using the TEMPO-like nitroxide oxidation catalyst, 4-aceta-
mido-2,2,6,6-tetramethyl-1-piperidinyloxy (4-acetamido-TEMPO) was carried out using several oxidizing agents and
co-catalyst. The pH and temperature of the reactions were closely monitored to decrease degradations during the
oxidation, and several isolation methods were explored. © 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Primary alcohol oxidation; Glucaric acid; Nitroxide; Bleach
1. Introduction
present detailed experimental procedures for
the oxidation of
D
-glucose using 4-acetamido-
A number of preparations of
D
-glucaric
TEMPO,13,14 rather than TEMPO. The acet-
amido derivative is more stable, much less
volatile, and is commercially available. The
catalyst was used in combination with two
acid from
D
-glucose or starch have been re-
ported in the past.1–4 In general, the methods
involved such strong oxidants as nitric acid or
NO2, were expensive, and gave poor yields
(B55%). In recent years, several nitroxide-
mediated oxidations of alcohol groups have
been carried out.5,6 The reaction is particularly
suitable for carbohydrates.7–11 The various
catalytic cycles are shown below (Fig. 1),
based on our work and that of others.5,8
We recently described the nitroxide-cata-
lyzed oxidation of
D
-glucose to -glucaric acid
D
in connection with more detailed work on the
oxidation of polysaccharides.12 Here, we
* Corresponding author. Fax: +1 860 486 2981.
E-mail address: bobbitt@nucleus.chem.uconn.edu (J.M.
Bobbitt).
Fig. 1. Redox system for TEMPO-like nitroxide mediated
oxidations.
0008-6215/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(01)00231-2

76
N. Merbouh et al. / Carbohydrate Research 336 (2001) 75–78
1
Fig. 2. H Spectrum of the crude oxidation mixture.
different oxidants, sodium hypochlorite
(bleach) and potassium hypochlorite. Sodium
or potassium bromide was present as a co-cat-
alyst. Other oxidants such as chlorine and
bromine are under exploration.
D
-glucaric acid. For the sodium experiments,
the disodium salt was precipitated with etha-
nol, (twice) to give a crude yield of 95%.
Using H NMR and gravimetric analysis for
halides, the crude product was shown to con-
tain about 90% disodium glucarate, 5% dis-
odium tartrate, and less than 5% halide salts.
Thus, the corrected yield of disodium glu-
carate was 85%.
Since the disodium salt is amorphous and
does not seem to have been carefully charac-
terized, this crude material was converted into
the less soluble and crystalline monopotas-
sium salt, which was isolated by precipitation.
This was done by passing the aqueous dis-
odium salt solution through an acid ion-ex-
change resin (Amberlite IR-120) to convert it
into the free acid and then by half neutraliza-
tion with potassium hydroxide to afford the
monopotassium salt in about 70% yield. When
the filtrates from the monopotassium salt were
concentrated, the yields were raised to ꢀ90%.
The monopotassium -glucaric acid was char-
acterized by mp, C NMR spectroscopy, and
optical rotation.
For the potassium experiments, the salt was
half neutralized with hydrochloric acid, and
the slightly soluble monopotassium salt was
collected by filtration. Although this reaction
was very simple, the potassium hypochlorite
had to be prepared from calcium hypochlorite
and potassium carbonate.
1
2. Results and discussion
D
-Glucose was oxidized with either sodium
or potassium hypochlorite in the presence of
catalytic amounts of 4-acetamido-TEMPO
(0.078 molar%) and sodium or potassium bro-
mide. Two solutions of sodium hypochlorite
were used, commercial bleach (5.25% NaOCl)
and Aldrich ‘available chlorine, 10–13%’
(11% NaOCl, by iodometric titration). Con-
centrated potassium hypochlorite, 1.25 M,
was prepared by the method of Newman and
Holmes.15 We showed in previous work12 that
carbohydrate oxidations with the system
ClO−/Br−/TEMPO were dependent on the
pH and temperature, with optimum conver-
sions (for monosaccharides) at pH 11.4–11.6
and at temperatures between 0 °C and 5 °C.
Under these conditions, little chain degrada-
D
13
1
tion was detected by 13C or H NMR (about
90% of the product was
D
-glucaric acid salts,
Fig. 2). The rest was a mixture of tartrate
salts, gluconate salts, oxalate salts, carbonates,
and halides.
The primary products in both reactions
were the dipotassium or disodium salts of

N. Merbouh et al. / Carbohydrate Research 336 (2001) 75–78
77
odium
D
-glucarate and some inorganic salts.
3. Experimental
The supernatant liquid was decanted and the
gummy precipitate was dissolved again in 30
mL of water and precipitated with 100 mL of
EtOH. The resulting gum, after decantation,
was washed with 4:1 EtOH–water to give 3.9 g
General methods.—Melting points were de-
termined with a Kofler-block apparatus and
are uncorrected. Optical rotations were mea-
sured with a Jasco DIP-1000 digital polarime-
ter. The 13C and 1H NMR spectra were
recorded in D2O with a Bruker DRX-400
instrument. All chemicals were commercially
available (Acros or Aldrich) and used as
received.
of crude disodium
D
-glucarate (95% yield) af-
ter drying at 50 °C under reduced pressure.
The very small amount of nitroxide re-
mained in the water was ignored. Gravimetric
halide analysis and NMR studies showed that
the crude material contained less than 4%
halide (measured as sodium chloride), about
4–5% tartrate salts and very little gluconate
Oxidation apparatus.—The reactions were
carried out in an open beaker fitted with a
pH-meter electrode, a thermometer and a 50
mL syringe pump controlled by the pH meter
in such a manner as to keep the pH above any
given value by the addition of sodium or
potassium hydroxide solution. The beaker was
cooled in ice. The hypochlorite solution was
added through a burette.
salts. The crude disodium
D
-glucarate was a
fine white, non-crystalline powder, and no fur-
ther decolorization was required.1
Oxidation
with
concentrated
sodium
hypochlorite.—The procedure just described
was carried out with concentrated sodium
hypochlorite (11%, 1.37 M). The reaction mix-
ture (about 105 mL) was processed in the
same manner to give 3.9 g of crude sodium
Oxidation with commercial bleach.— -(+)-
D
Glucose (3.00 g, 16.6 mmol), 4-acetamido-
TEMPO (0.04 g, 0.013 mmol), and NaBr (0.40
g, 3.36 mmol) in water (50 mL) were cooled in
ice to 0–5 °C, and the pH was adjusted to
11.5 with 2 M NaOH solution. Bleach solu-
tion (a total of 77 mL of 5.25% NaOCl, 3.3
eq/mole of sugar) was then added slowly (2
mL every 2 min for the first 25 mL, then 5 mL
every 20 min for the rest). Using the auto-
matic syringe system, the pH was kept be-
tween 11.4 and 11.6 with 2 M NaOH solution.
The end of the reaction was detected by a
negative KI–starch paper test and occurred
about 1 h after the last bleach had been
D
-glucarate (the only advantage of this reac-
tion was an appreciable reduction in the
amount of water to be removed at the end).
Isolation of potassium
D
-glucarate from
-glucarate.—A solution of the crude
-glucarate (1.5 g in 10 mL) was
sodium
D
sodium
D
passed slowly over a column (2 cm×15 cm)
of Amberlite IR-120 that had previously been
washed with water. The eluant was treated
with a KOH (2 M) solution to a final pH of
3.8. The solution was concentrated to ꢀ20
mL and, after several min, the monopotas-
1
added. H NMR spectra were measured in
sium
D
-glucarate crystallized. The solid was
D2O after removal of H2O by freeze drying;
13C NMR spectra were measured on freeze-
dried samples or on the actual reaction mix-
tures after the addition of a small amount of
D2O for locking purposes. The data showed
collected by filtration and dried to yield 1.2 g
(80%) of potassium
D
-glucarate, m.p. 188–
190 °C, [h]20+5° (c1, H2O). Commercial sam-
ples (Aldrich) melted at 188 °C, with [h]21+5°
(c1, H2O), and 13C NMR (D2O): l=71.9,
72.8, 73.0, 73.8 (C-2,3,4,5), 177.4, 177.5 (C-
1,6). The NMR spectrum was identical to that
of a commercial sample.
1
that 90% (quantitative H and 13C NMR) of
the mixture was sodium
D
-glucarate, about
4% was sodium gluconate, about 4% were
sodium tartrates (meso: l=4.30 ppm and dl:
l=4.37 ppm), and 1% was sodium oxalate
(not always observed).
Attempts to isolate the potassium salt di-
rectly from the sodium salt by partial neutral-
ization and the addition of excess KCl gave
poor yields of the potassium salt.
Oxidation with potassium hypochlorite.—
The oxidation was carried out in the same
manner as just described except that the pH
The reaction mixture (about 150 mL) was
concentrated under good vacuum (0.1 mm
Hg) to ꢀ30 mL, and 100 mL of 95% EtOH
was added to precipitate a mixture of dis-

78
N. Merbouh et al. / Carbohydrate Research 336 (2001) 75–78
was controlled with KOH, and potassium
hypochlorite15 (1.075 M) was the oxidant,
with KBr as co-oxidant. At the end of the
reaction, the mixture (ꢀ130 mL) was par-
tially acidified with one equivalent of concd
HCl to give a final pH of 3.8. The precipitated
product, after drying under vacuum at 40 °C
was identical to the one obtained above and
amounted to 2.4 g (55%). The filtrate was
concentrated; the pH was adjusted to 3.8 and
a small amount of KCl was added. An addi-
tional 1.00 g of product precipitated (total
yield was 85%).
Acknowledgements
The authors thank Professors Steven Suib
and Christian Bru¨ckner of this Department
for help and support and Mr. Charles Gallo
of the University of Connecticut Technical
Services for designing and building the pH
control device.
References
1. Mehltretter, C.L. Methods Carbohydr. Chem. 1963, II,
46–48.
2. Mehltretter, C. L. Ad6. Carbohydr. Chem. 1953, 8, 231–
249.
3. Mehltretter, C. L.; Rist, C. E. J. Agr. Food Chem. 1953,
1, 779–783.
The resulting material was a white crys-
talline compound with properties identical to
those already cited.
4. Kiliani, H. Chem. Ber. 1925, 58, 2344–2362.
5. Bobbitt, J. M.; Flores, M. C. L. Heterocycles 1988, 27,
509–533.
6. de Nooy, A.E.J.; Besemer, A.C.; van Bekkum H. Synthe-
sis 1996, 1153–1174.
4. Conclusion
7. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H. Rec.
Tra6. Chim. Pays-Bas 1994, 113, 165–166.
8. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H.
Carbohydr. Res. 1995, 269, 89–98.
9. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H.
Tetrahedron 1995, 29, 8023–8032.
10. de Nooy, A.E.J.; Besemer A.C., Patent, WO 95 07,303
(16 March 1995).
11. Davis, N. J.; Flitsh, S. L. Tetrahedron Lett. 1993, 34,
1181–1184.
12. Thaburet, J. F.; Merbouh, N.; Ibert, M.; Marsais, F.;
Queguiner, G. Carbohydr. Res. 2001, 330, 21–29.
13. Bobbitt, J. M. J. Org. Chem. 1998, 25, 9367–9374.
14. Bragd, P. L.; Besemer, A. C.; van Bekkum, H. J. Mol.
Catal. A. 2001, 170, 3542.
15. Newman, M.S.; Holmes, H.L. Org. Syn. Coll. 1966, II,
428–430.
At pH values between 11.4 and 11.6, and at
temperatures below 5 °C, the nitroxide medi-
ated oxidation of
D
-glucose yields glucaric
acid salts with high selectivity and good yields
(more than 85%, isolated as the monopotas-
sium salt). This would appear to be an excel-
lent method for oxidation of unprotected
glucose.
The use of 4-acetamido-TEMPO instead of
TEMPO as catalyst reduces the price and
allows more freedom for oxidation conditions
due to its greater stability and much lower
volatility.