Mono- and bimetallic catalysts for glucose oxidation

Article abstract of DOI:10.1016/j.molcata.2006.02.014

Mono- and bimetallic catalysts (Au, Pt, Pd and Rh) in form of supported particles or colloidal dispersions were tested in the aerobic oxidation of d-glucose to d-gluconic acid in water solution under mild conditions. Whereas the activity of single metals under acidic conditions was weak in the case of Au and Pt (TOF = 51-60 h^-1) and very low in the case of Rh and Pd (TOF < 2), the activity of the bimetallic particles resulted enhanced by combining Au with Pd and Pt. In the latter case high TOFs (max 924 h^-1) allow a new route to the synthesis of gluconic acid. In the presence of alkali at pH 9.5, the oxidation of glucose to gluconate resulted remarkably improved either in the case of monometallic gold or bimetallic catalysts, but no evident synergetic effects among metals have been detected.

Full text of DOI:10.1016/j.molcata.2006.02.014

Journal of Molecular Catalysis A: Chemical 251 (2006) 89–92
Mono- and bimetallic catalysts for glucose oxidation
Massimiliano Comotti, Cristina Della Pina, Michele Rossi∗
Dipartimento di Chimica Inorganica, Metallorganica e Analitica, CIMAINA and ISTM, University of Milano, Via Venezian, 21-20133 Milano, Italy
Available online 9 March 2006
Abstract
Mono- and bimetallic catalysts (Au, Pt, Pd and Rh) in form of supported particles or colloidal dispersions were tested in the aerobic oxidation
of d-glucose to d-gluconic acid in water solution under mild conditions. Whereas the activity of single metals under acidic conditions was weak in
the case of Au and Pt (TOF = 51–60 h−1) and very low in the case of Rh and Pd (TOF < 2), the activity of the bimetallic particles resulted enhanced
by combining Au with Pd and Pt. In the latter case high TOFs (max 924 h−1) allow a new route to the synthesis of gluconic acid.
In the presence of alkali at pH 9.5, the oxidation of glucose to gluconate resulted remarkably improved either in the case of monometallic gold
or bimetallic catalysts, but no evident synergetic effects among metals have been detected.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Glucose oxidation; Gold; Bimetallic catalysts; Gluconic acid; Oxygen
1. Introduction
1,2-diols [11–13]. Moreover, the extension of gold catalysis to
the selective oxidation of glucose led to promising results, thus
suggesting a new route to gluconates in competition with enzy-
matic catalysis [14–18]. An important feature of gold is to be
active either in form of supported metal or in form of nanometric
colloidal particles (sol) and an advantage of gold with respect
to biological systems is the possibility to work in a wide range
of pH, from alkaline to acidic conditions. Since glucose oxi-
dation in the absence of alkali resulted a slow process, in this
research we investigated the possibility of improving the perfor-
mance using bimetallic catalysts. In the present paper we report
the preparation of bimetallic Au–M (M = Pt, Pd, Rh) catalysts
and their use in the oxidation of glucose using dioxygen as the
oxidant under mild conditions.
Industrial production of gluconic acid and derivatives is man-
ufactured by glucose or glucose containing raw materials bio-
chemical transformation [1]. The great commercial interest for
aldonic acids, in particular gluconic acid and relative salts, is
due to their wide use as water-soluble cleansing agents or addi-
tives for food and beverages. The difficulties in moulds and free
enzymes separation, waste-water removal and low space–time
yield of biochemical plants are spurring the research of alter-
native catalytic routes for producing gluconates since the 1970s
[2–5]. These methods, however, can be industrially applied only
if the catalyst shows good activity, selectivity, productivity and
durability.
Beside platinum metal catalysts, sophisticated bi- and
trimetallic Bi–Pd and Bi–Pd–Pt on carbon catalysts were
proposed [6–10]. However, side reactions (i.e. carbohydrates
isomerization) and catalyst deactivation reduce gluconate pro-
ductivity avoiding their practical use.
2. Experimental
2.1. Reagents and apparatus
Recently, gold has received a growing interest as a catalyst
for the selective oxidation of organic molecules using oxygen
under mild conditions and our research group has emphasized
its unusual selectivity in oxidizing the primary alcoholic func-
tion to the corresponding carboxylate during the oxidation of
Gold sponge of 99.9999% purity from Fluka, d-glucose
monohydrate (Fluka) was used without further purification,
PdCl2 solution (21% w/w Pd and density 1.53 g/ml), RhCl3
(40% w/w Rh), K2PtCl4 (46% w/w Pt, Alfa Aesar-Johnson
Matthey GmbH), NaBH4 purity > 96% from Fluka, carbon
powder Vulcan XC72R from Cabot (average superficial area
254 m2/g, pH 8.0, pore volume 0.192 ml/g, density 96 g/l,
sulphur content 0.5%), NaOH (Fluka, of 98% purity), HCl
37%(Fluka), H2O milliQ, gaseous oxygen (SIAD, 99.9% pure).
∗
Corresponding author. Tel.: +39 02 503 14397; fax: +39 02 503 14405.
E-mail address: michele.rossi@unimi.it (M. Rossi).
1381-1169/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2006.02.014

90
M. Comotti et al. / Journal of Molecular Catalysis A: Chemical 251 (2006) 89–92
2.2. Preparation of the catalysts
2.4. Analysis of products
2.2.1. Monometallic catalysts on carbon
The products were identified and quantified by a compari-
son with original samples. Quantitative and qualitative analyses
were performed by HPLC; C NMR was also used for qualitative
analysis. In the case of free pH experiments, the conversion to
gluconic acid was determined by NaOH–HCl inverse titration.
Monometallic gold, platinum, rhodium and palladium cat-
alysts on carbon were prepared by immobilizing the corre-
sponding metal sol on carbon. Aqueous solutions of HAuCl4,
K2PtCl4, RhCl3 or PdCl2 (100 mg/l) were prepared and glucose
was added as a protecting agent (metal:glucose = 1:50 w/w).
Under stirring, a freshly prepared solution of NaBH4 (0.1 M,
NaBH4/M = 1 w/w) was added to obtain a dark yellow (Au)
and dark brown (Pt, Rh and Pd) colloidal dispersion (sol).
The metal particles were immobilized by adding carbon to the
metal dispersion, under stirring, until the solution became clear.
The amount of support was calculated in order to produce a
metal loading of 1%. After 1 h the slurry was filtered, the solid
washed many times with distilled water and dried in the air for
12 h. XRD analysis was performed, using a Rigaku D III-MAX
horizontal-scan powder diffractometer with Cu K␣ radiation, to
determine the size of the metal crystallites using the Scherrer
equation.
2.4.1. HPLC analysis
Analyses were performed on an ionic exchange Hamilton
HC-75H form (305 mm × 7.75 mm) column, using a succinic
acid 0.55 mM solution as the eluent and a Shimazu LC 10 AD
instrument provided with a RID-10 A detector.
2.4.2. C NMR analysis
13C NMR spectra were recorded in water on a Bruker
300 MHz without adjusting the pH. The assignment of peaks
in the oxidation of glucose was made by comparison with stan-
dard spectra.
2.4.3. Titration procedure
2.2.2. Bimetallic catalysts on carbon
Samples of the reacting mixtures, carried out at free pH,
were treated with 1 M NaOH until pH 11–12 and then back-
titrated with HCl 0.25 M. The titration curve (pH versus ml HCl
added) highlighted two equivalent points, one corresponding to
the excess of NaOH, the second to the organic acid at about pH
3.
Bimetallic catalysts were prepared as above by immobiliz-
ing the preformed metal sols. Mixtures of HAuCl4 + K2PtCl4,
HAuCl4 + PdCl2, HAuCl4 + RhCl3 were used as metal precur-
sors of gold–platinum, gold–palladium and gold–rhodium cata-
lysts, respectively.
2.2.3. Mono- and bimetallic catalysts as sols
3. Results and discussion
According to the procedure described in Sections 2.2.1 and
2.2.2 monometallic and bimetallic sols were prepared and used
without carbon support. The desired amount of sol was diluted
in the reacting solution for catalytic tests.
The reduction of the noble metal compounds HAuCl4,
K2PtCl4, PdCl2 and RhCl3 with NaBH4 in the presence of glu-
cose produces colloidal dispersions stable for some hours. The
dimension of the particles was determined by XRD technique
after immobilization on carbon: it ranged from 2.0 to 5.0 nm. In
the case of the Au + Pt colloids, the peak position lied between
the extreme values of the corresponding monometallic particles,
suggesting that these bimetallic systems were in form of alloys
(Table 1).
With the above catalysts, two series of experiments have been
planned. In the first one the oxidation of glucose to free glu-
conic acid was investigated under moderate conditions (pressure
300 kPa, temperature 343–363 K), while in a second series of
experimentstheoxidationofglucosetogluconateinthepresence
2.3. Oxidation procedure
2.3.1. Oxidation at uncontrolled pH
Catalyst evaluation at uncontrolled pH (from initial pH 6.0
to ca. pH 2.6) was carried out varying the type of metal and, in
the case of Au + Pt catalysts, the Au/Pt ratio. Reactions were
performed in a thermostatted glass reactor (30 ml), provided
with an electronically controlled magnetic stirrer connected to
a large reservoir (5000 ml) containing dioxygen at 300 kPa.
Glucose and the catalyst (reactant:metal = 3000 mol/mol) were
mixed in distilled water (10 ml total volume). The reactor was
pressurised at 300 kPa of O2 and thermostatted at the appro-
priate temperature (343–363 K). After an equilibration time
of 15 min, the mixture was stirred and analysis was carried
out.
Table 1
XRD parameters of mono- and bimetallic catalysts
Metals
2θ◦
D (nm)
Au
Pt
Pd
Rh
38.1
39.7
40.2
43.2
38.4
38.6
38.9
39.0
39.1
3.4
5.0
2.9
2.0
4.8
3.8
3.3
3.2
3.0
2.3.2. Oxidation at controlled pH
Glucose and catalyst (glucose:total metal = 20,000) in dis-
tilled water (total volume 15 ml) were stirred in a thermostatted
glass reactor and dioxygen was bubbled (1000 ml min−1) at
atmospheric pressure. Glucose conversion versus time plot was
registered by an automatic titrator (Titrino®, Metrohm) at the
controlled pH value of 9.5 and temperature 323 K.
Au:Pt = 4
Au:Pt = 2
Au:Pt = 1
Au:Pt = 0.5
Au:Pt = 0.25

M. Comotti et al. / Journal of Molecular Catalysis A: Chemical 251 (2006) 89–92
91
Table 2
Catalytic activity of mono- and bimetallic catalysts. 1%M on carbon; T = 343 K;
P(O2) = 300 kPa; glucose/Au = 3000; reaction time = 6.5 h
Metals
Conversion %
TOF (h−1
)
Au
Pt
Pd
Rh
Au–Pt, 1:1
Au–Pd, 1:1
Au–Rh, 1:1
11
13
<2
<2
64
20
<2
51
60
<2
<2
295
92
<2
of alkali was evaluated working at the fixed pH value of 9.5,
atmospheric pressure and 323 K.
Kinetic tests have been carried out mainly with carbon sup-
ported metal particles owing to the high stability of these cat-
alytic systems. In some cases catalysts were used in form of
colloidal particles.
Fig. 1. Comparison between supported and unsupported catalysts. Au:Pt =
2:1 w/w; P(O2) = 300 kPa; T = 363 K; glucose/Au = 3000.
3.2. Controlled pH experiments
3.1. Uncontrolled pH experiments
Since a promoting effect of alkali in gold catalysed oxida-
tion of glucose was established during previous studies [12,18],
it seemed of interest to investigate the performance of Au–Pt
bimetallic systems under basic conditions. For this purpose,
the most effective catalytic combinations used under acidic
conditions, Au:Pt = 2 and Au:Pd = 1, have been compared with
monometallic gold, platinum and palladium catalysts, working
with a Titrino apparatus at pH 9.5, atmospheric pressure and
323 K.
Table 2 compares the catalytic performance of monometal-
lic Au, Pd, Pt and Rh on carbon catalysts and bimetallic Au–M
on carbon catalysts, having a common Au:M = 1 ratio(w/w) in
the synthesis of free gluconic acid. Whereas monometallic sys-
tems are scarcely active, the conversion of glucose is enhanced
by using Au–Pd and, in particular, Au–Pt catalysts. Mean TOF
values up to 295 h−1, referred to the total metal, have been cal-
culated by titrating the produced gluconic acid after a reaction
time of 6.5 h.
Under these conditions, the kinetic results reported in Table 4
showed faster reactions leading to gluconate: a stronger pro-
moting effect of alkali resulted mainly in the case of su−pp1orted
catalysts, monometallic gold (TOF from 51 to 17200 h ) and
bimetallic gold–platinum catalyst (TOF from 924 to 17600 h−1).
The similar activity obtained at pH 9.5 indicates the absence
of synergetic effects between Au and Pt under these condi-
tions. Correlations among unsupported catalysts are difficult to
discuss owing to the instability of colloidal dispersions in the
presence of alkali. However, after the 2.5 h test, gold–platinum
and gold–palladium particles behave better than monometallic
No other products, beside gluconic acid, have been detected
by HPLC and NMR techniques.
Owing to the highest efficiency of the Au–Pt system, we
made an optimization by changing the relative amount of the
two metals from 4:1 to 1:4 (w/w). According to the results,
reported in Table 3, the best performance was reached with sup-
ported catalysts having the Au:Pt = 2 ratio. The strong synergetic
effect between gold and platinum metals is represented in Fig. 1
where unsupported bimetallic particles are also compared with
the monometallic Au and Pt on carbon catalysts. We can out-
line that supported and unsupported bimetallic particles show a
similar catalytic activity, which is ca. five times higher than the
monometallic systems. The reported data were always achieved
under catalyst-controlled kinetics, using appropriate stirring rate
and catalyst:substrate ratio.
gold particles. In this latter case, a lower TOF value (4600 h−1
)
has been found with respect to the value of 18043 h−1 previ-
ously calculated as initial rate [18]. A strong destabilizing effect
of gluconate, leading in a short time to inactive large metal par-
ticles [18], could explain the lower TOF value mediated in the
long time test. High TOF values found in the case of bimetallic
Table 3
Table 4
Influence of Au:Pt ratio on supported catalysts. 1% (Au + Pt)/C; T = 343 K;
P(O2) = 300 kPa; glucose/Au = 3000; reaction time 2.5 h
Oxidation with mono- and bimetallic catalysts at pH 9.5. Glucose/Au = 20,000;
T = 323 K; reaction time = 2.5 h
Au:Pt
Conversion %
TOF (h−1
)
Metals
Type of catalyst
Conversion %
TOF (h−1
)
4
2.5
2
1.5
1
0.5
0.25
29
68
77
62
20
23
26
348
816
924
744
240
276
312
Au
Au–Pt (2:1)
Au
Pt
Pd
Supported
Supported
43
44
12
<2
5
17200
17600
4600
<500
2000
Unsupported
Unsupported
Unsupported
Unsupported
Unsupported
Au–Pt (2:1)
Au–Pd (1:1)
26
28
10500
11600

92
M. Comotti et al. / Journal of Molecular Catalysis A: Chemical 251 (2006) 89–92
unsupported catalysts could be due to a good stability of these
colloids.
References
[1] R.H. Blom, V.F. Pfeifer, A.J. Moyer, et al., Ind. Eng. Chem. 44 (1952)
435.
4. Conclusions
[2] Johnson, Matthey and Co., Ltd., GB Patent 1,208,101 (1970).
[3] Kawaken Fine Chemicals Co. Ltd., J Patent 8,007,230 (1980).
[4] Mitsui Toatsu Chemicals Inc., J Patent 7,652,121 (1976).
[5] Kawaken Fine Chemicals Co. Ltd./Kao Corp., US Patent 4,843,173
(1989).
[6] T. Mallat, A. Baiker, Catal. Today 19 (1994) 247.
[7] M. Wenkin, C. Renard, P. Ruiz, B. Delmon, M. Devillers, Stud. Surf.
Sci. Catal. 110 (1997) 517.
[8] M. Besson, F. Lahmer, P. Gallezot, P. Fuertes, G. Fleche, J. Catal. 152
(1995) 116.
[9] A. Abbadi, H. van Bekkum, Appl. Catal. A: Gen. 124 (1995)
409.
[10] C. Bronnimann, Z. Bodnar, P. Hug, T. Mallat, A. Baiker, J. Catal. 150
(1994) 199.
[11] L. Prati, M. Rossi, Stud. Surf. Sci. Catal. 110 (1997) 509.
[12] L. Prati, M. Rossi, J. Catal. 176 (1998) 552.
[13] C. Bianchi, F. Porta, L. Prati, M. Rossi, Topics Catal. 13 (2000)
231.
[14] S. Biella, L. Prati, M. Rossi, J. Catal. 206 (2002) 242.
[15] S. Biella, G.L. Castiglioni, C. Fumagalli, L. Prati, M. Rossi, Catal. Today
72 (2002) 43.
[16] M. Comotti, C. Della Pina, R. Matarrese, M. Rossi, New perspectives
in gold catalysed oxidation, in: XIV Congresso Nazionale di Catalisi
GIC 2004, 6–10 June, Lerici (SP), 2004.
Comparing experiments in the absence and in the presence of
alkali it is evident the great synergetic effect of platinum, along
with a lower effect of palladium, working at low pH whereas
almost no effect is present at pH 9.5. In the latter case bimetallic
colloidal particles appeared more stable towards agglomeration
than monometallic gold particles, allowing higher conversions.
The different promoting effect of Pt and Pd under acidic and
alkaline solution could be related to different reaction mecha-
nisms. If we consider a reaction mechanism where alkali plays
a strategic role, which favours the slow dehydrogenation step of
glucose producing gluconate and H2O2 [18,19], no important
effects of Pt and Pd metals on Au should be expected. However,
no mechanistic studies are presently available in the case of glu-
cose oxidation in acidic conditions for understanding the strong
synergetic effect of platinum here detected.
From a synthetic point of view, the catalytic oxidation of
glucose represents a target of practical interest in industrial
chemistry [20] and in the present paper we have developed a
catalytic system tailored for the synthesis of free gluconic acid
having the preferred composition Au:Pt = 2 (w/w).
[17] P. Beltrame, M. Comotti, C. Della Pina, M. Rossi, J. Catal. 228 (2004)
282.
[18] M. Comotti, C. Della Pina, R. Matarrese, M. Rossi, Angew. Chem. Int.
Ed. 43 (2004) 5812.
[19] P. Beltrame, M. Comotti, C. Della Pina, M. Rossi, Appl. Cat. A 297
(2006) 1–7.
Acknowledgment
This work was supported in part by the EC project “AURI-
CAT, grant Research Training Network (HPRN-CT 2002-
00174)”.
[20] S. Biella, M. Comotti, M. Rossi PCT/WO2005/003072 A1.