Aerobic oxidation of glucose with gold catalyst: Hydrogen peroxide as intermediate and reagent

Article abstract of DOI:10.1002/adsc.200505389

Careful analytical determinations show that the gold-catalysed aerobic oxidation of glucose occurs through a two-electrons mechanism leading to gluconate and hydrogen peroxide. This latter decomposes before reaching the critical concentration for competing with O₂ in glucose oxidation. A mechanism of glucose oxidation on gold nanoparticles is presented.

Full text of DOI:10.1002/adsc.200505389

COMMUNICATIONS
DOI: 10.1002/adsc.200505389
Aerobic Oxidation of Glucose withGold Catalyst: Hydrogen
Peroxide as Intermediate and Reagent
Massimiliano Comotti, Cristina Della Pina, Ermelinda Falletta, Michele Rossi*
Dipartimento di Chimica Inorganica, Metallorganica e Analitica, CIMAINA and ISTM. University of Milano,
Via Venezian, 21 20133 Milano. Italy
Phone: (þ39)-02-5031-4397, Fax: (þ39)-02-5031-4405, e-mail: michele.rossi@unimi.it
Received: October 10, 2005; Accepted: December 15, 2005
Abstract: Careful analytical determinations show
that the gold-catalysed aerobic oxidation of glucose
occurs through a two-electrons mechanism leading
to gluconate and hydrogen peroxide. This latter de-
composes before reaching the critical concentration
hydrogen peroxide is formed by two-electrons reduction
of dioxygen. The role of H O as a possible competitor of
2
2
O is also discussed.
2
Several oxidation tests have been carried out at differ-
ent pH values, by bubbling dioxygen at atmospheric
pressure and 303 K into the aqueous solution of glucose
in the presence of colloidal gold particles having a mean
diameter of 3.6 nm. The experimental procedure, based
on the automatic titration of gluconic acid with NaOH at
for competing with O in glucose oxidation. A mech-
2
anism of glucose oxidation on gold nanoparticles is
presented.
[
13]
Keywords: catalytic oxidation; glucose oxidation;
gold; hydrogen peroxide; oxygen
a fixed pH value, has been already reported. As de-
rived from Table 1, the initial TON (mol of reacted glu-
cose per mol of total gold per hour), is strongly depen-
À1
dent upon pH, varying from 2700 h at pH 7, through
À1
À1
1
0,500 h at pH 9.5 up to 45,000 h at pH 12. In any
case, the rate declines with time owing to the instability
[13]
Clean syntheses, based on the use of natural renewable of colloidal particles in the presence of gluconate.
reagents, in water solution under mild conditions, are
Analytical determinations allowed us to detect hydro-
[
16]
highly desirable processes and often a catalytic step is gen peroxide, by the titanium sulphate test and quan-
[
1]
[17]
the key factor for successful applications. Inorganic tify it by permanganate titration in the reaction solu-
catalysis, as an alternative to enzymatic catalysis, could tion: at pH 7 it was formed approximately in a 1:1 ratio
represent a powerful tool for many oxidation processes with respect to gluconate and accumulated in the first
using atmospheric oxygen or hydrogen peroxide. In the 900 s. By increasing the alkalinity, we always observed
case of dioxygen, platinum group metals are the most in- H O amounts below the stoichiometric value at
2
2
[
2,3]
vestigated catalysts. In recent years new efficient cat- pH 9.5 and only trace amounts at pH 12. It is known
alytic systems, based on the peculiar properties of nano- that hydrogen peroxide decomposition is favoured by
[
18]
metric gold particles, have been developed for the liquid alkali. However, in addition to H O decomposition,
2
2
[
4–9]
[10]
phase oxidation of polyols,
glucose to carboxylates.
amino alcohols and its reaction with glucose should also be taken into ac-
[11]
In particular, it has been shown that glucose, the abun-
dant and cheap starting material for the production of
Table 1. Analytical data of the reacting solutions. [Au]¼2.5·
À5
À1
1
0
M; [Glucose]¼0.35 M; T¼303 K; O ¼6 NL·h
.
2
[12]
many chemicals, can be efficiently oxidised to gluco-
nate using supported and unsupported gold nanoparti-
cles, gold catalysis being competitive with respect to en-
Test pH t (s) Conversion % M(H O )
M(H O )
2
2
found
2
2
calcd.
À3
À3
À3
À3
À3
À3
À3
À3
À3
À3
À3
1
2
3
4
5
6
7
8
9
0
7
7
7
100 0.45
400 0.92
900 1.68
1.9·10
3.7·10
5.2·10
6.2·10
10.5·10
5.3·10
<0.1·10
0.3·10
0.1·10
1.7·10
3.5·10
6.4·10
[13]
À3
À3
À3
À3
À3
À3
À3
À3
À3
zymatic catalysis.
During these studies, however, information was lack-
ing on the oxidation mechanism and even the stoichiom-
etry of the reaction was never fully ascertained.
9.5 100 2.61
9.5 400 8.12
9.5 900 14.50
9.5 1900 18.34
12
12
12
9.9·10
30.9·10
54.9·10
69.7·10
34.4·10
51.7·10
56.3·10
While the kinetics of the aerobic oxidation of glucose
catalysed by colloidal gold has recently been investigat-
[14]
[15]
110 9.06
170 13.61
230 14.83
ed and compared with the enzymatic catalysis, in
the present study we report new molecular aspects of re-
agent activation on gold particles, demonstrating that
1
<0.1·10
Adv. Synth. Catal. 2006, 348, 313 – 316
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
313

COMMUNICATIONS
Massimiliano Comotti et al.
À5
Table 2. Oxidation of glucose by continuous addition of H O under N . [Au]¼2.5·10 M; [Glucose]¼0.35 M, T¼303 K,
2
2
2
À1
pH¼9.5; N ¼6 NL·h
.
2
Test
t (s)
M(H O )
M(gluconate)
M(H O )
% H O
2
2
added
2
2
found
2
2decomp.
[
[
[
[
[
[
a]
À3
À3
À2
À1
À1
À1
À3
À3
À3
À2
À2
1
1
2
3
4
5
6
100
200
500
950
1800
2300
2.0·10
4.0·10
1.0·10
1.2·10
3.4·10
4.5·10
0
0
0
1.5·10
2.8·10
7.0·10
6.3·10
7.0·10
25
30
30
32
55
40
a]
a]
a, b]
b]
À2
À2
À1
1.9·10
8.2·10
1.1·10
b]
1.2·10·
[
[
a]
b]
À7
À1
[
8
H O ]¼8 10 ·mol·s
.
2
2
À6
À1
·10 mol·s
.
count as a possible event. To verify this point, H O was
2
2
tested as a substitute for the O reagent under similar
2
conditions: it was continuously added by an automatic
syringe to a 0.35 M glucose solution under nitrogen
flow in order to produce, in the first 1000 s, local concen-
À3
À1
À1
trations ranging from 10 to 10 mol·L
.
Following the rate of gluconate formation, reported in
Table 2, we can observe that, for the tests at pH 9.5 no
gluconic acid was titrated in the presence of H O con-
2
2
À3
centrations up to 7·10 M (tests 1–3), whereas aconsis-
tent decomposition of the reagent (25–30%) always
took place. Glucose oxidation started only when the
H O concentration was between 10 and 10 M (tests
–6) meaning that the loss of H O , outlined in Table 1 Figure 2. Initial rate of glucose oxidation at different H O
À2
À1
2
2
4
2 2
2
2
concentrations.
at pH 9.5 and pH 12 (tests 4–10), is mainly due to de-
composition.
More detailed information on the H O reactivity has
been derived considering the rate of glucose oxidation
2
2
[
14]
In ref. an analogous linear plot, related to the glu-
using different initial concentrations of H O , in the cose oxidation by O at different concentrations, was al-
2
2
2
presence of a constant amount of gold sol, as represent- ready determined with a similar experimental proce-
ed in Figure 1. dure using identical values of gold concentration, glu-
From these data, the slope has been calculated in the cose concentration and temperature. Although the an-
2
0–40 s interval owing to experimental uncertainty in gular coefficient calculated from the plot of Figure 2
À1
the first 20 s, and the linear correlation between rate (4 h )cannot representthe true apparentrateconstant,
and H O concentration is represented in Figure 2.
owing to the unknown amount of decomposed H O ,
2
2
2 2
(
evaluated as 25–40% of the total in the experiments
of Table 1), it can be roughly compared with the appa-
À1
rent rate constant (165 h ) calculated when O was
2
[
14]
used as a reagent. According to these data, dioxygen
results as being at least one order of magnitude more ef-
ficient than hydrogen peroxide at the same concentra-
tion. From a practical point of view, however, H O2
2
can be more easily used at higher concentrations, allow-
ing higher initial TOF as shown in Figure 3.
Considering that the experiment with H O , carried
2
2
out in an open reactor under nitrogen flow at atmospher-
ic pressure, allows us to exceed by two times the rate of
glucose oxidation observed using O at atmospheric
2
pressure, we can exclude that O , formed by H O de-
2
2
2
composition, is the effective oxidant during the oxida-
tion with hydrogen peroxide.
In conclusion, it has been demonstrated the tendency
of gold to catalyse the two- electrons reduction of dioxy-
Figure 1. Glucose conversion at different H O concentra-
2
2
À5
À1
tions. [Au]¼2.5·10
M; [Glucose]¼0.35 M, T¼303 K,
pH¼9.5; N 6 NL·h
.
2
314
asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 313 – 316

Aerobic Oxidation of Glucose with Gold Catalyst
COMMUNICATIONS
Figure 3. The reactivity of H O and O in glucose oxidation.
2
2
2
À5
[
~
Au]¼1.4·10 M, [Glucose]¼0.35 M, T¼303 K, pH¼9.5;
À1
À1
[H O ]¼0.35 M, N ¼6 NL·h
;
&
O ¼6 NL·h
at
2
2
2
2
À3
0
.1 MPa (10 M); blanks, without catalyst: lower two lines.
Scheme 1.
gen forming hydrogen peroxide: this occurs not only in
[
19,20]
the case of molecular hydrogen as a reducing agent,
reaction 1, but also using more complex organic mole- behaviour is determined by the electronic properties of
cules, as glucose, according to reaction 2:
the nanometric gold particles (d<10 nm).
þ
À
In the dioxogold intermediate either Au ÀO or
2
2
þ
2À
2
Au ÀO
couples can be formally considered as a
ð1Þ
bridge for the two electron-transfer from glucose to di-
oxygen. Peroxidic-like species have already been sup-
posed as a reaction intermediate during dihydrogen ox-
[
20]
idation to water on a gold catalyst.
ð2Þ
ð3Þ
Experimental Section
A colloidal dispersion of gold, I, was prepared by treating a
À4
1
.25·10
M aqueous solution of Au (as HAuCl ) with
4
NaBH (NaBH :Au¼5) under an N atmosphere, in the pres-
4
4
2
ence of a large excess of glucose (0.35 M). The resultant brown
sol contained metal particles, stable for several hours in the ab-
ð4Þ sence of dioxygen, having a mean diameter of 3.6 nm, deter-
mined by TEM performed on a drop of the dispersion evapo-
rated on a copper grid.
In the presence of alkali, hydrogen peroxide is mainly
decomposed, Equation (3), before reaching a sufficient
concentration for oxidising glucose. It is worth noting
that Equations (2) and (3) are also the basis of the enzy-
The oxidation of glucose with O was carried out in a ther-
mostatted, magnetically stirred reactor (80 mL) with bubbling
dioxygen at atmospheric pressure. The reaction was started by
2
adding the gold sol to the O saturated solution. Gluconic acid
2
matic oxidation catalysed by oxidase and catalase, re- was continuously titrated at fixed pH value (7.0, 9.5 and 12)
[15]
spectively.
with NaOH, using a GPD 751 Titrino apparatus (Metrohm).
Concerning the mechanism of molecular activation, In the reaction product, only gluconate was detected by
HPLC at pH 7 and 9.5 whereas at pH 12 fructose plus other un-
identified products (ca. 10–15%) were formed.
Scheme 1 can be proposed for gold catalysis on the basis
of the promoting effect of alkali and the formation of hy-
drogen peroxide as a reaction product.
The key point is represented by the electron-rich gold
species, formed by the hydrated glucose anion with gold
surface atoms, which is supposed to activate molecular
oxygen by nucleophilic attack. According to previous
The oxidation of glucose with H O , related to the data of
2
2
Tables 1 and 2, was carried out in a similar apparatus where
H O (10% w/w) was continuously added by means of an auto-
2
2
matic syringe to the glucose solution containing the gold sol,
and the kinetics followed by titrating the formed gluconic
acid at the given pH. The data referring to Figure 3 were ob-
experiments correlating catalytic activity with particle tained after one-pot addition of H O (30%) to obtain a 1:1 ra-
2
2
[13]
dimensions, we suppose that an efficient nucleophilic tio to glucose.
Adv. Synth. Catal. 2006, 348, 313 – 316
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
315

COMMUNICATIONS
Massimiliano Comotti et al.
For the analytical determinations, the following procedure
has been set up: a sample (4–10 mL) of the reacting solution
was withdrawn at the indicated times and titrated with a
[8] F. Porta, L. Prati, M. Rossi, S. Coluccia, G. Martra, Catal.
Today 2000, 61, 165–172.
[9] S. Carrettin, P. McMorn, P. Johnston, K. Griffin, G. J.
Hutchings, Chem. Commun 2002, 696–697.
10] S. Biella, G. L. Castiglioni, C. Fumagalli, L. Prati, M.
Rossi, Catal. Today 2002, 72, 43–49.
À2
À3
10
M KMnO , 10 M MnSO and 0.08 M H SO solution
4 4 2 4
at constant pH of 1.8ꢀ0.1 in order to selectively oxidise
[
H O in the presence of a large amount of glucose. Blank ex-
2
2
periments on glucose-H O solutions of similar concentrations
2
2
[11] S. Biella, L. Prati, M. Rossi, J. Catal. 2002, 206, 242–247.
showed an acceptable reproducibility (2%).
th
[
[
[
12] Ullmannꢀs Encyclopedia of Industrial Chemistry, VI
edn., Vol. 15, Wiley-VCH, Weinheim, 2003.
13] M. Comotti, C. Della Pina, R. Matarrese, M. Rossi, An-
gew. Chem. Int. Ed. 2004, 43, 5812–5815.
14] P. Beltrame, M. Comotti, C. Della Pina, M. Rossi, Aero-
bic oxidation of glucose II. Catalysis by colloidal gold.
Appl. Catal.: A, accepted for publication.
Acknowledgements
This work was supported in part by the EC project “AURICAT,
Grant Research Training Network (HPRN-CT 2002–00174)”.
[
[
[
15] P. Beltrame, M. Comotti, C. Della Pina, M. Rossi J. Catal
2
004, 228, 282–287.
References
th
16] Analisi Chimica Qualitativa, VI edn., G. Charlot, Piccin
Editore, Padova, 1993.
17] Vogelꢀs Textbook of quantitative chemical analysis, V
[
1] J. F. Jenck, F. Agterberg, M. J. Droescher, Green Chem.
004, 6, 544–556.
th
2
edn., (Eds.: G. H. Jeffery, J. Bassett, J. Mendham, R. C.
Denney), Longman Scientific & Technical, London,
[
2] R. A. Sheldon, J. K. Kochi, Metal-Catalyzed Oxidations
of Organic Compounds, Academic Press, London, 1981.
3] T. Mallat, A. Baiker, Chem. Rev. 2004, 104, 3037–3058.
4] G. C. Bond, D. Thompson, Cat .Rev. – Sci. Eng. 1999, 41,
1
989.
[
[
th
[
18] The Merck Index, XIII edn., p. 858.
[
19] P. Landon, P. J. Collier, A. J. Papwhorth, C. J.Kiely, G. J.
Hutchings, Chem. Commun. 2002, 2058–2059.
3
19–388.
[
[
5] L. Prati, M. Rossi, J. Catal. 1998, 176, 552–560.
6] C. L. Bianchi, F. Porta, L. Prati, M. Rossi, Top. Catal.
[
20] D. G. Barton, S. G. Podkolzin, J. Phys. Chem. 2005, 109,
2
262–2274.
2
000, 13, 231–236.
7] F. Porta, L. Prati, M. Rossi, G. Scari, J. Catal. 2002, 211,
64–469.
[
4
316
asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 313 – 316