Selective production of lactobionic acid by aerobic oxidation of lactose over gold crystallites supported on mesoporous silica

Article abstract of DOI:10.1016/j.apcata.2011.05.034

Partial oxidation of lactose over Au-based catalyst system using nanostructured silica materials with improved activity, selectivity and stability was investigated as a novel chemo-catalytic approach for selective synthesis of lactobionic acid (LBA) for therapeutic, pharmaceutical and food grad applications. Highly active gold crystallites dispersed on mesoporous silica (SiO₂-meso) using bis-[3-(triethoxysilyl) propyl] tetrasulfide (BTSPT), a silane coupling agent to immobilize gold, were successfully formulated, and their catalytic activity was evaluated in an agitated semi-batch reactor. The catalysts were characterized by N₂ physisorption, XRD, XPS and TEM. The influence of the reaction conditions, i.e., temperature, pH value, metal loading and catalyst/lactose ratio on lactose conversion were investigated. After 100 min of reaction, the catalyst containing 0.7% Au showed the highest catalytic activity (100% lactose conversion) and a 100% selectivity towards LBA, when it was used at a catalyst/lactose ratio of 0.2 under alkaline (pH 9.0) and mild reaction temperature (65 °C).

Full text of DOI:10.1016/j.apcata.2011.05.034

Applied Catalysis A: General 402 (2011) 94–103
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Selective production of lactobionic acid by aerobic oxidation of lactose over gold
crystallites supported on mesoporous silica
Luis-Felipe Gutierreza,b, Safia Hamoudia, Khaled Belkacemia,∗
a Department of Soil Sciences and Food Engineering, Université Laval, Québec, Canada G1V 0A6
b Department of Food Sciences and Nutrition, Université Laval, Québec, Canada G1V 0A6
a r t i c l e i n f o
a b s t r a c t
Article history:
Partial oxidation of lactose over Au-based catalyst system using nanostructured silica materials with
improved activity, selectivity and stability was investigated as a novel chemo-catalytic approach for
selective synthesis of lactobionic acid (LBA) for therapeutic, pharmaceutical and food grad applications.
Highly active gold crystallites dispersed on mesoporous silica (SiO2-meso) using bis-[3-(triethoxysilyl)
propyl] tetrasulfide (BTSPT), a silane coupling agent to immobilize gold, were successfully formulated, and
their catalytic activity was evaluated in an agitated semi-batch reactor. The catalysts were characterized
by N2 physisorption, XRD, XPS and TEM. The influence of the reaction conditions, i.e., temperature, pH
value, metal loading and catalyst/lactose ratio on lactose conversion were investigated. After 100 min of
reaction, the catalyst containing 0.7% Au showed the highest catalytic activity (100% lactose conversion)
and a 100% selectivity towards LBA, when it was used at a catalyst/lactose ratio of 0.2 under alkaline (pH
9.0) and mild reaction temperature (65 ◦C).
Received 8 February 2011
Received in revised form 16 May 2011
Accepted 23 May 2011
Available online 30 May 2011
Keywords:
Lactobionic acid
Lactose
Oxidation
Gold clusters
Mesoporous silica
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
be used as acidulant with a sweet taste; as filler in cheese produc-
tion; as firming agent, and to fortify functional drinks with essential
Lactose is an important by-product of the cheese and casein
manufacture, whose production is estimated to about 1.2 million
tons per year [1]. Because of its low solubility and sweetness, as
well as a certain intolerance of some population segment, the use
of lactose is limited in many applications. However, as a conse-
quence of its worldwide surplus and low cost, there is a great
interest in the research of innovative processes for transforming
lactose, and expanding its applications in the food and pharmaceu-
tical industries as value-added lactose derivatives. Some significant
developments include the production of highly valued pharmaceu-
tical products and functional food ingredients such as lactulose,
lactitol, lactobionic acid, lactosucrose, and galacto-oligosaccharides
[2].
Lactobionic acid (LBA) (4-O-␤-d-galactopyranosyl-d-gluconic
acid) is a high value-added product obtained from lactose oxi-
dation, with excellent properties for food and pharmaceutical
applications. The main commercial use of LBA is as ingredient of the
solutions employed for stabilizing organs prior to transplantation
[3]. LBA is also used in calcium supplementation, and represents
a new ingredient in skin care products featuring potent antioxi-
dant and humectant properties [4,5]. In the food industry, LBA can
minerals such as Fe and Cu [4,6].
LBA production by means of heterogeneous catalytic oxida-
tion of lactose on Pd/C and Pd-Bi/C catalysts was first reported
by Hendriks et al. [7]. Since then, some studies of lactose oxida-
tion over platinum, palladium and bismuth-promoted palladium
supported catalysts have been investigated [8–19]. Although these
reactions are environmentally safer than chemical and biochem-
ical oxidations, the main problem reported for these catalysts is
their deactivation during reaction [20]. Platinum is less prone to
deactivation by over-oxidation, but its selectivity towards aldonic
acids formation is lower than that of palladium. It has been also
demonstrated that the presence of bismuth or lead as promoters,
and the strict control of the oxygen concentration, diminish the
susceptibility of catalyst to poisoning, and it improves its selec-
tivity towards aldonic acids [17,19,21]. In recent published works
of our research group, bimetallic catalyst supported on meso-
porous SBA-15 (1.02%Pd, 0.64%Bi; Bi/Pd molar ratio of 0.3) was
used for lactose oxidation at 65 ◦C under microaerial and alkaline
conditions (pH 9.0). This catalyst was highly active (96% lactose
conversion) and 100% selective towards LBA, when the reaction was
conducted during 3 h with consecutive addition of air and nitro-
gen periodically to avoid the catalyst poisoning by over-oxidation
[17,19]. However, even if bismuth-promoted palladium catalysts
are promising for the oxidation of lactose to LBA, improved cata-
lysts showing high activity and selectivity towards LBA are needed,
∗
Corresponding author. Tel.: +1 418 656 2131x6511; fax: +1 418 656 3723.
E-mail address: khaled.belkacemi@sga.ulaval.ca (K. Belkacemi).
0926-860X/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2011.05.034

L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
95
in order to facilitate the reaction control, to avoid the presence of
toxic substances derived from the possible leaching of Pd and Bi,
and to make the catalytic process more competitive [1].
40 ◦C. Then, a mixture of 20.8 g TEOS and 2.1 g bis-[3-(triethoxysilyl)
propyl] tetrasulfide were added drop wise under continuous and
vigorous agitation. Subsequently, solutions of HAuCl4 in 2 N HCl
were added drop wise, according to the different investigated cata-
lyst gold loadings (0.4, 0.5, 0.7, 0.8 and 1.0%). The obtained mixture
was stirred at 600 rpm for 24 h at 40 ◦C, and then hydrother-
mally aged for 72 h at 100 ◦C. After vacuum filtration, the solid
was thoroughly washed with deionized water and ethanol at room
temperature. Finally, the catalyst was vacuum-dried overnight at
100 ◦C, and the resulting solids were calcined at 500 ◦C for 5 h at a
heating rate of 1 ◦C/min, using a programmable oven.
Supported gold catalysts were shown to outperform palladium
catalysts for the oxidation of carbohydrates, because they are less
prone to overoxidation, and they exhibit high activity and selec-
tivity towards aldonic acids [20]. However, the oxidation of lactose
over gold catalyst has been scarcely investigated. Recent studies
indicate that supporting gold nanoparticles on Al2O3, TiO2, and
CeO2 allowed to achieve moderate to high conversions (>80%) and
selectivities (up to 95%), using Au loadings in the range of 0.4–2.0%
[10,13,22,23]. Even though these relatively high conversions and
selectivities were reached, the oxidation of lactose to LBA still
necessitates cumbersome and costly downstream separation steps.
Since the pioneering work of Haruta [24], intensive research
efforts were initiated to use gold-based catalysts for liquid phase
oxidation. The activity of heterogeneous gold catalysts strongly
depends on the gold cluster size deposited on the catalyst surface.
Then, a weak stability of these clusters against sintering during the
synthesis of the catalyst is problematic and undesirable and could
hamper the catalyst applications [25].
Owing to their high surface area, highly ordered structure,
good thermal stability and excellent mechanical properties, meso-
porous silicas are considered very attractive catalyst supports.
Consequently, synthesis of metal nanoparticles supported on
such materials has received much attention in recent years [26].
However, gold nanocrystallites dispersed on silica materials are
mobile and sinter easily under severe synthesis conditions of
temperature, especially during the calcination step. To avoid this
problem, Hu et al. [27] reported a simple method to confine gold
nanoparticles in the mesoporous silica walls in the presence of
bis-[3-(triethoxysilyl) propyl] tetrasulfide (BTSPT). This catalytic
system exhibited fairly interesting activity for alcohol oxidation.
The aim of this work focuses on the synthesis and character-
ization of gold nanoclusters highly dispersed on mesostructured
silica material using BTSPT and application of this material as a
catalyst for selective oxidation of lactose to LBA. Specifically, the
catalyst characterization by means of nitrogen sorption, powder X-
ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and
transmission electron microscopy (TEM) was investigated, as well
as the influence of the reaction conditions, i.e., temperature, pH
value, gold loading and catalyst-to-lactose ratio (catalyst/lactose),
on lactose conversion. To the best of our knowledge, mesostruc-
tured silica has not been used as support in the oxidation of lactose
over gold catalysts.
2.3. Characterization
Nitrogen adsorption/desorption isotherms of calcined samples
were obtained using a volumetric adsorption analyzer (Model
Autosorb-1, Quantachrome Instruments, Boyton Beach, FL) at 77 K.
Before the adsorption analysis, the samples were degassed for 3 h at
200 ◦C. Total pore volume was estimated from the amount adsorbed
at 0.99 relative pressures. Pore size distributions were calcu-
lated using the desorption branch of the N2 adsorption/desorption
isotherms and the Barrett–Joyner–Halenda (BJH) method [28].
Powder X-ray diffraction (XRD) patterns were obtained using an
Ultima III Rigaku Monochromatic Diffractometer using Cu K␣ radi-
˚
ation (ꢀ = 1.5406 A). Small angle powder diffraction patterns were
acquired at a scanning rate of 1◦/min in the 2ꢁ range of 0.6–3◦, while
wide angle powder XRD were obtained at the same scanning rate
in the 2ꢁ range 25–80◦.
The catalyst surface and the oxidation state of gold in the cat-
alysts were studied by X-ray Photoelectron Spectroscopy (XPS)
analysis using an Axis-Ultra spectrometer from Kratos (UK)
equipped with an electrostatic analyzer of large ray, a source of
double X-rays Al–Mg without monochromator and an Al source
with monochromator. The pressure in the XPS room was main-
tained between 5 × 10−9 and 5 × 10−8 torr during the analysis. All
the spectra were recorded with the Al monochromatic source with
a power of 300 W. The flyover spectrum used to determine the
elementary composition was recorded with pass energy in the ana-
lyzer of 160 eV and an energy step of 1 eV, using lenses in hybrid
mode, which maximizes the sensitivity. The detailed spectra with
high resolution were recorded with pass energy of 40 or 20 eV, and
step energy of 50 or 100 meV. The spectra with high resolution are
used for the chemical analysis [29]. The adjustment of the enve-
lope calculated with the experimental spectrum was carried out
using CasaWPS software from Kratos (UK). The binding energy (BE)
scale was calibrated by measuring C1s peak (BE = 285.0 eV) from
the surface contamination.
2. Experimental
Transmission electron microscopy (TEM) images were taken on
a JEM-3010 electron microscope (JEOL, Japan) with an accelera-
tion voltage of 80–120 kV. The catalysts samples were suspended
in methanol and ultrasonically treated during 10 min. Then, the
suspension (5 ␮L) were disposed uniformly and dried on a nickel
grid.
2.1. Materials
All reagents were of the highest purity and were used without
further purification. Gold (III) chloride trihydrate (HAuCl4·3H2O),
tetreaethyl orthosilicate (TEOS), bis-[3-(triethoxysilyl) propyl]
tetrasulfide (BTSPT), ␣-lactose monohydrate and LBA were
2.4. Lactose oxidation experiments
purchased from Sigma Aldrich. Pluronic P123 (EO20PO70EO20
,
Mav = 5900 g/mol) was graciously offered by BASF. For HPLC anal-
ysis, sodium hydroxide solution 50% and HPLC grade anhydrous
sodium acetate from Fluka, were used for eluents preparation.
Lactose oxidation experiments were carried out at atmospheric
pressure in a thermostated and magnetically stirred 300 mL-glass
reactor. In a typical oxidation reaction, 175 mL of 20 mM lac-
tose solution were heated under continuous agitation (600 rpm)
until the desired reaction temperature was reached, whereas
500 mL/min of nitrogen flow were simultaneously bubbled for
stripping off the dissolved oxygen. Once the reaction temperature
was reached, the stirring was increased to 900 rpm, the catalyst
was added and the pH of the solution was adjusted at the desired
value, by the addition of few drops of 1 M NaOH aqueous solution.
2.2. Catalyst preparation
Gold catalysts were synthesized as reported recently by Hu et al.
[27]. In a typical synthesis, 10 g of P123 were dissolved in 375 mL
of a 2 N HCl solution at room temperature. After complete disso-
lution of the template agent, the temperature was increased to

96
L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
Table 1
The textural parameters of the mesoporous silica support as well as Au catalyst
samples prepared at different Au loadings.
Materials
BET surface
area (m2 g−1
Pore diameter
(nm)
Total pore volume
)
(cm3 g−1
)
SiO2-meso
995.7
813.9
827.3
790.1
3.70 and 4.73
3.50
3.70
1.087
0.716
0.779
0.651
0.4% Au/SiO2-meso
0.7% Au/SiO2-meso
1.0% Au/SiO2-meso
3.70
Then, air was bubbled through the reaction mixture at a constant
flow rate of 40 mL/min. During the reaction, the pH value was kept
constant by adding 1 M aqueous NaOH solution, by means of a pH
STAT Titrator (Metrohm® 842 Titrando®, Switzerland), and the dis-
solved oxygen concentration was monitored using an oxygen probe
(Ocean Optics, Inc., USA) connected to an optic fiber sensor system
(TauTheta Instruments LLC, USA).
The time ‘zero’ of the reaction corresponded to the moment
when air was bubbled through the reaction mixture. Reactions
were carried out during 180 min. In order to optimize the reac-
tion conditions, the effects of reaction temperature (50–70 ◦C),
pH (7.0–10.0), catalysts metal loading (0.4–1.0%), as well as cat-
alyst/lactose ratio (0.1–0.3) were evaluated. Liquid samples were
taken at regular intervals for the subsequent HPLC analyses.
Fig. 1. Nitrogen adsorption/desorption of mesoporous silica material and Au on
mesoporous silica at different Au loadings.
porous silica support indicating that some structural defects were
formed in the mesoporous framework [31,32]. Similarly when they
incorporated gold nanoparticles in mesoporous silica, Hu et al. [27]
reported the formation of void defects in their catalysts and con-
cluded that the formation of these structural defects that could be
attributed to the incorporation of gold in the mesoporous silica
walls.
The nitrogen physisorption of the support and Au catalysts
reveals also that they have high BET surface areas. The support
exhibited dual pores size distribution which reduces to one pore
distribution of nearly 4 nm of average pore size after gold incorpo-
ration as depicted in Fig. 2 and Table 1.
It is noticed in Table 1, that the BET surface area of mesoporous
silica support approaches 1000 m2 g−1 with a total pore volume of
≈1.1 cm3 g−1 and dual-pore size distributions centered at around
3.7 and 4.7 nm. By incorporation of gold crystallites, the BET sur-
face area decreased notably to ∼820 m2 g−1 for the catalysts with
2.5. Reaction sample analysis
Samples of reaction mixture were analyzed using a chro-
matographic system DX-ICS 2500 (Dionex, USA) equipped with a
gradient pump GP50 and an electrochemical detector ED50 with
a thin-layer type amperometric cell outfitted with gold electrodes
and an Ag/AgCl reference electrode. The chromatographic separa-
tion of lactose and LBA was performed on a CarboPac PA1 analytical
column (4 mm × 250 mm i.d., Dionex) and a CarboPac PA1 guard
column (4 mm × 50 mm i.d., Dionex) at a flow-rate of 1 mL/min.
Sodium hydroxide solution 50% and anhydrous sodium acetate
(NaOAc) from Fluka diluted in deionized water were used as elu-
ents. Chromeleon software (Dionex, USA) was used to perform all
the calibration and integration of chromatographic data. All sup-
plementary details can be found in Belkacemi et al. [17].
3. Results and discussion
3.1. Material characterization
3.1.1. Nitrogen physisorption
Textural properties of the mesoporous silica (SiO2-meso) and
Au/SiO2-meso catalysts were obtained from low-temperature
(77 K) nitrogen adsorption/desorption isotherm measurements,
which allow calculation of the specific surface areas, specific
pore volumes, and mesopore size distributions, as summarized in
Table 1. These isotherms and the corresponding pore size distri-
butions using the BJH theory are depicted in Figs. 1 and 2 for the
initial mesoporous silica material and Au catalysts with Au load-
ings of 0.4, 0.7 and 1%. As shown in Fig. 1, the shape of the obtained
isotherms corresponds to type IV according to IUPAC classifica-
tion and displayed a broad H1 type hysteresis loop characteristic of
large pore mesoporous solids [30]. The initial increase in adsorption
capacity at low relative pressure is due to monolayer adsorption on
mesopores. The upward deviation in the range of P/P0 = 0.4–0.8 is
associated with progressive mesopores filling. As the relative pres-
sure increases, the adsorption isotherm for the mesoporous support
displays a sharp increase characteristic of capillary condensation
inside uniform mesopores. There were uncommon hysteresis loops
at P/P0 values between 0.4 and 0.7 for the Au/SiO2-meso catalysts.
These loops look more flat in comparison to that of initial meso-
Fig. 2. Pore size distributions of mesoporous silica material and Au on messoporous
silica at different Au loadings.

L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
97
Fig. 4. Wide angle XRD profiles of Au on mesoporous silica at different Au loadings.
Fig. 3. Low angle XRD profiles of mesoporous silica material and Au on mesoporous
silica at different Au loadings.
the present mesoporous material presents a wormhole-like frame-
work structure containing interconnected 3D-mesopores that can
be ideal for the minimization of diffusion limitation phenomena
often encountered in adsorption and catalytic reactions.
Au loadings of 0.4 and 0.7%. Further increment of Au loading to
1% decreases further the BET surface area. Similar tendency can be
observed for the total pore volume where about 41% of the initial
pore volume of the support is lost when 1% of Au is impregnated by
filling the support pore size of 4.7 nm. Therefore, incorporation of
metallic gold has a consequential impact on the BET specific surface
area, specific pore volume and pore size of the silica material. The
pore and structural morphology of the synthesized mesoporous sil-
ica material and those of the Au catalysts can be further verified by
transmission electron microscopy (TEM).
The TEM image of 0.4% Au/SiO2-meso shows similar wormhole-
like framework mesostructure (Fig. 5b). The average pore size is
generally up to 3–4 nm. Some of gold crystallites are distinguish-
able and their dimension can be located between 11 and 18 nm,
with an average value of ∼14 nm then excluded from the pores
and indicating that they are on the external catalyst surface. The
TEM image of 0.7 and 0.8% Au/SiO2-meso shows identical pattern
where Au crystallites seemed to be better dispersed having dimen-
sions between 3 and 14 nm, with an average values of ∼7 and 9 nm,
respectively (Fig. 5d and e). Au crystallites in 0.5 and 1.0% Au/SiO2-
meso have larger size up to 21 nm with average values of circa 11
for both catalysts (Fig. 5c and f).
3.1.2. X-ray diffraction
The small angle XRD patterns of the mesoporous silica support
and the synthesized catalysts with different gold loadings (0.4, 0.7
and 1.0%) are shown in Fig. 3. It can be noticed that there is a peak at
about 2ꢁ = 1.0 degree, which is broader than the narrow peak nor-
mally reported for the well-ordered hexagonal mesoporous silica
such as SBA-15 [33]. Then the structure obtained by co-condensing
silica source and BTSPT as a silane coupling agent to immobilize
gold in the presence of P123 is not that of SBA-15 for all tested
gold loadings. Similar low angle XRD patterns where previously
reported by Besson et al. [34] using bis-[3-(trietoxysilyl) propyl]
disulfide and TEOS in the presence of P123. These patterns signify
the presence of a correlated distribution of framework pores but
the lack of a regular cylindrical pore structure. This is in agreement
with the TEM findings presented below.
Table
2 compares the average Au crystallite dimensions
obtained by TEM and those computed using wide angle XRD analy-
sis data and Scherrer’s equation. The average Au crystallite sizes
obtained with the two methods are somewhat different. As the
estimated crystallite size using Sherrer’s equation gives a rough
estimate of the crystallite size and is known to lake precision, the
more trustful crystallite size data could be those estimated using
TEM analysis.
3.1.4. X-ray photoelectron spectroscopy (XPS) analysis
The oxidation state of the loaded gold on the catalysts was
investigated by XPS. The presence of Au 4f, Si 2p, Si 2s, C1s and
O photoelectron peaks in all catalysts was detected by XPS mea-
surements. For the sake of clarity only Au 4f photoelectron peaks
are presented and discussed for the Au/SiO2-meso catalysts with
Au loadings of 0.4, 0.7 and 1.0%. The spectra of the Au 4f peaks
for the 3 catalysts are depicted in Fig. 6a–c for Au loadings of 0.4,
0.7 and 1.0%, respectively. The experimental curves of Au 4f core
Fig. 4 shows the wide angle XRD patterns of the catalysts loaded
with gold at 0.4, 0.7 and 1.0%. The well resolved characteristic
diffraction peaks at around 38◦, 45◦ and 65◦, assigned to the (1 1 1),
(2 0 0), and (2 2 0) lattice planes of face-centered cubic structure of
gold, indicate that gold is in crystalline form after the catalysts syn-
thesis, and reveal the presence of metallic gold species [35,36]. It
can be noticed that for a 0.4 and 0.7% gold loadings, the diffraction
peaks are weak. This can be attributed to the low Au loading and the
better Au crystallite dispersion on the mesoporous silica support.
Table 2
3.1.3. Transmission electron microscopy
Average crystallite size of Au catalyst samples obtained with TEM and XRD analysis
of catalyst samples prepared at different Au loadings.
Fig. 5a–f shows the transmission electron microscopy (TEM)
micrographs of the mesoporous silica material, and Au catalysts
with Au loading of 0.4, 0.5, 0.7, 0.8 and 1.0%, respectively. The TEM
image of the mesoporous silica material (Fig. 5a) clearly shows the
wormhole pore arrangement structure confirming the small angle
XRD analysis stipulating the existence of wormhole structure. It
can be understood from these results that unlike SBA-15 which
have a hexagonal framework structure with cylindrical mesopores,
Catalyst
TEM average metal
crystallite size (nm)
XRD average metal
crystallite size (nm)
0.4% Au/silica-meso
0.5% Au/silica-meso
0.7% Au/silica-meso
0.8% Au/silica-meso
1.0% Au/silica-meso
14
11
7
9
11
1
1
1
1
1
20
–
14–15
–
1
1
1
24–25

98
L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
Fig. 5. (a)TEM micrographs of mesoporous silica material; (b) 0.4% Au/SiO2-meso; (c) 0.5% Au/SiO2-meso; (d) 0.7% Au/SiO2-meso; (e) 0.8% Au/SiO2-meso; (f) 1.0% Au/SiO2-
meso.
levels spectra present well-defined 2 spin orbit split Au 4f5/2 and
2Au 4f7/2 components, after the peak deconvolution process using
Kratos software. Table 3 summarizes the bending energies (BE) of
these spin orbit split components for all Au loadings. The compo-
nents Au 4f7/2 (1) and Au 4f7/2 (2) binding energies are important
and can be used to determine the oxidation state of gold species
[37–39]. The Au 4f7/2 (1) component with BE values between 84.1
and 84.4 eV can typically be assigned to pure metallic Au0 species
for all Au loadings [29,37,39]. The 4f7/2 (2) component at 85.2–85.5
ascertains the presence of Au+ in catalyst loadings of 0.4 and 0.7%,
while the 4f7/2 (2) component at ∼86.5 eV confirmed the presence
of Au3+ species in catalyst loading of 1.0% [39]. Owing to the peak
surfaces of Au+1 and Au+3 species which are significantly smaller
than that of Au0 species, most of the gold crystallites are present
in the metallic state even though Au+ and Au3+ presence can be
detectable.
Table 3
Binding energies of Au 4f7/2 and 4f5/2 deconvoluted peaks of catalyst samples pre-
pared at different Au loadings.
Au loading (%)
4f7/2 (1) (eV)
4f5/2 (1) (ev)
4f7/2 (2) (eV)
4f5/2 (2) (eV)
0.4
0.7
1.0
84.20
84.10
84.37
87.85
87.60
88.07
85.45
85.20
86.47
89.05
88.80
90.17

L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
99
Fig. 6. XPS-spectra of Au 4f photoelectron peak for: (a) 0.4% Au/SiO2-meso; (b) 0.7% Au/SiO2-meso; (c) 1.0% Au/SiO2-meso.
3.2. Catalyst activity
such as the reaction temperature, pH, catalyst to lactose ratio and
catalyst gold loading are investigated.
As mentioned in the introduction section, gold nonoclusters
dispersed on ordered mesoporous silica has never been tested to
evaluate their capability to partially oxidize lactose to LBA. The next
sections will treat on the optimization of lactose conversion to LBA
over Au/SiO2-meso as an oxidation catalyst where main conditions
3.2.1. Effect of Au catalyst loading on lactose oxidation kinetics
Fig. 7 shows the kinetic profiles of lactose oxidation to LBA
over Au/SiO2-meso at 0.7%. The Au/SiO2-meso silica at 0.7% Au
loading exhibited a pretty good activity to oxidize lactose to LBA.

100
L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
Fig. 8. Trends of the initial rate and the Au crystallite size versus Au loading. Bares
stand for the standard deviation for two repetitions for each catalyst synthesis.
Fig. 7. Kinetic profiles of lactose uptake and LBA production during lactose oxidation
over 0.7% Au/SiO2-meso catalyst. Conditions: T = 65 ◦C; pH 9; Catalyst to lactose ratio
R = 0.2. Bares stand for the standard deviation for two repetitions.
this figure, the trend in the particle size with Au loading variation
shows a minimum value of ∼7 nm at Au loading of 0.7% where the
catalyst activity is at its maximum value. It is possible to state that
the Au catalyst, as prepared by the reported procedure, is very sen-
sitive to the catalyst loading which may control the catalyst size.
Furthermore, in the liquid phase reactions, which is the case for
the present work, the activity of supported gold catalysts could
depend on the gold cluster size and the reaction kinetics could be
structure-sensitive depending on the support chemical nature [42],
particularly if the support is a metal oxide which is the case in the
present study. This could be ascribed to the significant interactions
between the noble metal and the support [43]. Actually, the reason
why this behavior occurs is not known for the moment and needs
deep investigations.
The lactose oxidation over this catalyst exhibited favorable and
outstanding lactose conversion to LBA where the rates of lactose
consumption as well as LBA formation were markedly high and the
lactose conversion essentially completed (it reached about 100%)
after 80 min of reaction. It is worth mentioning that lactose was
almost depleted after only 80 min of reaction where LBA concen-
tration reached a value of 7.8 g/L starting from exactly 7.5 g/L of
lactose. This means that 7.5/342.29 ∼ 0.022 moles of lactose were
oxidized to give 7.8/358.3 ∼ 0.022 moles of LBA consequently, 100%
of selectivity towards LBA formation was reached using this cata-
lyst.
The oxidation of lactose over other investigated catalysts (0.4,
0.5, 0.8 and 1% Au/SiO2-meso) showed clearly much lower kinetic
patterns. Even if the conversion of lactose was not complete for
these catalysts, the selectivity towards lactose was also 100%.
At Au loading of 0.4, 0.5 and 1%, the formation of larger metal
crystallites (see Table 2) built up and lowered the metal availability
at the reaction surface area. This diminution of the catalyst dis-
persion is one possible reason which could presumably affect the
catalyst activity. However, the small metal cluster size is not the
only sufficient condition to ensure a good catalyst activity. It is
well known that the oxidation state of metals is a very important
parameter in the catalytic activity. In the case of the carbohydrate
oxidation over gold catalysts, the oxidation state of gold species
could play an important role in the catalyst activity and this has
been hardly investigated. As shown by XPS analysis, our results
indicate that the presence of oxidized gold species may affect the
activity of gold catalysts in carbohydrate oxidation.
The initial rate r0 is plotted versus the gold crystallite size as
depicted in Fig. 9, shows a volcano-type tendency if the trivial point
(0, 0), corresponding to the reaction activity of the pure support, i.e.,
SiO2-meso, which was verified to not exhibit any oxidation activ-
ity) is included in the data. From this figure, the crystallite size
where maximum of activity is obtained is in the vicinity of 6 nm.
This finding fairly corroborates those of other works reported in the
literature for reactions catalyzed by gold [44–46].
Although some works found in the literature for the oxida-
tion of alcohols over Au supported catalysts reported the presence
of mainly the metal Au0 with the presence of Au+ and rarely
Au3+ species [40,41], in the similar manner as reported in the
present work, however, it cannot be stated clearly which species
are responsible for the Au oxidation activity in the liquid phase and
the role of the oxidation state of gold is still matter of discussion.
Exploiting the determination of the initial reaction rate of
lactose conversion to LBA for each investigated Au loading, as eval-
uated by Eq. (1):
ꢀ
d[LBA]t
dt
ꢀ
r0
=
(1)
ꢀ
Fig. 9. Trend of the initial rate versus the Au crystallite size. The vertical bares stand
for the standard deviation in the initial rate while horizontal bares stand for the stan-
dard deviation in the crystallite size for two repetitions for each catalyst synthesis.
Dashed line represents only a general tendency.
t=0
allowed to show that maximum catalyst activity was obtained
at Au loading of 0.7% as depicted in Fig. 8. As we can observe in

L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
101
it is decided to select the 0.2 catalyst/lactose ratio as the optimal
value for the evaluation of the effects of temperature and pH on
the lactose conversion. This ratio is comparable to that obtained
in our recent work [17] for the catalytic oxidation of lactose over
bimetallic 1.02%Pd–0.64%Bi/SBA-15 catalyst, with the advantage
that in the case of gold catalysts the bubbling of nitrogen is
not necessary as in Bi–Pd catalysts, making easier the reaction
control.
3.2.3. Effect of pH on the oxidation kinetics using 0.7%
Au/SiO2-meso
The influence of pH, in the range of 7–10, on the lactose
conversion was investigated using the 0.7%Au/SiO2-meso at 0.2
catalyst/lactose ratio. During these experiments, temperature, agi-
tation and airflow were kept constant at 65 ◦C, 900 rpm and
40 mL/min, respectively. As shown in Fig. 11, the oxidation of
lactose over gold catalyst is pH-dependent as reported for the
oxidation of other carbohydrates over metal catalysts [17,20,47].
The initial reaction rate was clearly enhanced by increasing the
pH value up to 9.0, where a maximum of activity was reached
which led to nearly 100% lactose conversion with 100% selectivity
towards LBA. Similar to our recent findings for lactose oxida-
tion over bimetallic 1.02%Pd–0.64%Bi/SBA-15 catalyst [47], it has
been recently reported that gold catalysts could be deactivated
by the strong adsorption of the reaction products, which occupy
the active sites of the catalyst surface, when pH is neutral (pH 7)
[10]. However, the catalyst activity declined readily at pH 10.
This strongly alkaline condition induced a partial decomposition
of the catalyst due to the dissolution of silica material at very
high alkaline conditions. Similar results were reported by Mirescu
and Prü␤e [23], who found a pronounced decrease in the activ-
ity and selectivity in the oxidation of lactose with Au/TiO2 and
Au/Al2O3 catalysts, when the pH values were higher than 9.0. This
behavior was also reported in our investigations on the partial
oxidation of lactose to LBA over bimetallic Pd–Bi/SBA-15 catalyst
[17,47].
Fig. 10. Effect of the catalyst/lactose ratio, R, on the initial reaction rate, r0 during
the partial oxidation of lactose to LBA over 0.7% Au/SiO2-meso catalyst. Conditions:
T = 65 ◦C; pH 9. Bares stand for the standard deviation for two repetitions of the
catalyst synthesis.
3.2.2. Effect of catalyst/lactose ratio on the oxidation kinetics
using 0.7% Au/SiO2-meso
In
our
recent
work
[17],
using
0.21 g
catalyst
(1.02%Pd–0.64%Bi/SBA-15)/g lactose for the microaerial oxi-
dation of lactose, we found the maximal conversion of lactose
to LBA at 65 ◦C in alkaline conditions (pH 9.0). Based on these
results, in this study we have evaluated the effect of different
catalyst/lactose ratio on the oxidation of lactose over 0.7% Au/SiO2-
meso at 65 ◦C and pH 9.0 under agitation of 900 rpm, using air
(40 mL/min) as oxidizing agent. Fig. 10 shows the effect of this
ratio on the reaction kinetics. The evaluation effect of this ratio
on the initial reaction rate r0 as depicted in Fig. 10, allows clearly
to observe that the kinetic behavior is well favored at R = 0.2 and
levels off. Irrefutably, the more concise observation of the reaction
kinetics with r0 determination at the beginning of the reaction
asserted that the reaction kinetics can be considered identical
at R ratios of 0.2 and 0.3. Decreasing this ratio to 0.1 led to a
significant decrease in the reaction kinetics. From these results,
Consequently, the pH of 9 was kept constant to the further
experiments evaluating the effect of the reaction temperature on
the catalytic activity of the 0.7%Au/SiO2-meso.
Fig. 11. Effect of the pH on the initial reaction rate, r0 during the partial oxidation of lactose to LBA over 0.7% Au/SiO2-meso catalyst. Conditions: T = 65 ◦C; R = 0.2. Bares stand
for the standard deviation for two repetitions of the catalyst synthesis.

102
L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
4. Conclusions
For the first time, gold catalysts supported on mesoporous silica
material functionalized with silane coupling agent (BTSPT) were
successfully investigated for lactose oxidation using air as oxidizing
agent.
XPS analysis suggested that gold nanoparticles were present in
the catalysts as both metallic and oxidized species. Furthermore,
these nanoparticles seemed to be within the mesoporous frame-
work and on the surface of silica material, as indicated by TEM
measurements.
After catalyst evaluation in a certain range of gold loadings,
catalyst/lactose ratio, temperature and pH, the 0.7%Au/SiO2-meso
catalyst was found highly active (100% of conversion, after only
80–100 min of reaction) and 100% selective towards LBA when
it was added at a ratio of 0.2 catalyst/lactose under the reaction
conditions of 65 ◦C, pH of 9, 40 mL/min, 900 rpm. Comparing with
the palladium or bismuth-promoted palladium supported cata-
lysts, the reaction control using 0.7%Au/SiO2-meso catalyst is easier
because the control of oxygen flow, required to avoid the over oxi-
dation of Pd and Bi–Pd catalysts, is not necessary, and additionally,
no side products are detected. Thus, due to the ease of synthe-
sis, as well as because of its excellent catalytic performance, the
0.7%Au/SiO2-meso catalyst can be regarded as major breakthrough
on the way to an industrial process of catalytic oxidation of lactose.
Fig. 12. Effect of the reaction temperature on the initial reaction rate, r0 during
the partial oxidation of lactose to LBA over 0.7% Au/SiO2-meso catalyst. Conditions:
pH 9; R = 0.2. Bares stand for the standard deviation for two repetitions of the catalyst
synthesis.
3.2.4. Effect of temperature on the oxidation kinetics using 0.7%
Au/SiO2-meso
Acknowledgments
The effect of temperature on lactose oxidation kinetics over
fresh 0.7% Au/SiO2-meso catalyst was investigated in the range
[50–70 ◦C].
The authors gratefully acknowledge the FQRNT who provided
funds for this research. They also acknowledge the Canadian Foun-
dation for Innovation (CFI) for funding K. Belkacemi and S. Hamoudi
allowing them to purchase the Autosorb-1 and 1-C as well as XRD-
DSC equipment.
The pH has been kept constant at 9 and the ratio catalyst/lactose
used was 0.2 g catalyst/g lactose. The reaction kinetics in terms of
production rate of LBA versus the temperature variation is depicted
in Fig. 12. At 50 ◦C the reaction kinetics is slow. The main reason
could be the poor activation state of the reaction at this tempera-
ture. The reaction kinetics carried out at 60 ◦C exhibited clearly a
more pronounced rate of LBA production. The greater activation of
the reaction at this temperature could constitute the main reason
of the kinetic improvement. Even if the obtained rate is relatively
high, it is a room to further improvement at higher temperature.
Indeed, the kinetics of lactose oxidation to LBA performed at 65 ◦C
showed during the first 100 min of reaction a very high rate of lac-
tose depletion, reaching a conversion of almost 100%. After only
80 min, the lactose conversion was 96%. The selectivity of the cata-
lyst during the reaction of lactose oxidation into LBA was found
always 100% (see Fig. 7 of this work). The rate of LBA produc-
tion increased significantly and reached ∼0.12 g LBA L−1 min−1, as
exhibited in Fig. 12. The kinetics of lactose oxidation performed
at 70 ◦C exhibited an acceptable but slower consumption rate of
lactose, reaching a conversion 94% after 100 min and residual non
consumed lactose representing about 6% of the initial concentra-
tion even after prolonged time of reaction (results not shown). This
indicates that the reaction halted after 100 min due to a possible
depletion of oxygen in the reaction medium. Indeed, at relatively
higher temperature (80 ◦C), the oxygen solubility decreased read-
ily and its concentration is almost zero at atmospheric pressure
as measured using the optic fiber oxygen sensors. Consequently,
the available O2 concentration was too low to sustain and main-
tain the oxidation activity, which could explain the decrease in
the lactose conversion. This is clearly shown in terms of LBA pro-
duction rate (Fig. 12), where further increment in the temperature
higher than 65 ◦C affected markedly the kinetics due to the lack
of oxygen at the surface of the catalyst. However, it is worth
to mention that the temperature has not affected the catalyst
selectivity towards LBA production, which remained 100% in all
cases.
References
[1] J. Kowalczyk, U. Prusse, H. Berndt, I. Pitsch, US Patent 2007/0112186 A1 (2007).
[2] M.G. Gänzle, G. Haase, P. Jelen, Int. Dairy J. 18 (2008) 685–694.
[3] M. Nordkvist, P.M. Nielsen, J. Villadsen, Biotechnol. Bioeng. 97 (2007) 694–707.
[4] M.J. Playne, R.G. Crittenden, in: P.L.H. McSweeney, P.F. Fox (Eds.), Advanced
Dairy Chemistry, vol. 3, 3rd Edition, Springer-Verlag, New York, 2009, pp.
121–201.
[5] E. Berardesca, F. Distante, G.P. Vignoli, C. Oresajo, B. Green, Br. J. Dermatol. 137
(1997) 934–938.
[6] S. Toshiaki, Y. Shuichi, A. Seiichiro, S. Sakanori, Japanese Patent 07277991
(1995).
[7] H.E.J. Hendriks, B.F.M. Kuster, G.B. Marin, Carbohydr. Res. 204 (1990) 121–129.
[8] H.E.J. Hendriks, Selective Oxidation of Lactose and Related Carbohydrates, Ph.D.
Thesis, Technische Universiteit Eindhoven, The Netherlands, 1991.
[9] A. Abbadi, K.F. Gotlieb, J.B.M. Meiberg, H. van Bekkum, Appl. Catal. A: Gen. 156
(1997) 105–111.
[10] J. Kuusisto, A.V. Tokarev, E.V. Murzina, M.U. Roslund, J.-P. Mikkola, D.Y. Murzin,
T. Salmi, Catal. Today 121 (2007) 92–99.
[11] A.V. Tokarev, E.V. Murzina, K. Eränen, H. Markus, A.J. Plomp, J.H. Bitter, P. Mäki-
Arvela, D.Y. Murzin, Res. Chem. Intermed. 35 (2009) 155–174.
[12] A.V. Tokarev, E.V. Murzina, P.K. Seelam, N. Kumar, D.Y. Murzin, Microporous
Mesoporous Mater. 113 (2008) 122–131.
[13] A.V. Tokarev, E.V. Murzina, J.-P. Mikkola, J. Kuusisto, L.M. Kustov, D.Y. Murzin,
Chem. Eng. J. 134 (2007) 153–161.
[14] Y.N. Chia, M.P. Latusek, J.H. Holles, Ind. Eng. Chem. Res. 47 (2008) 4049–4055.
[15] S. Karski, J. Mol. Catal. A: Chem. 253 (2006) 147–154.
[16] S. Karski, I. Witónska, J. Gołuchowska, J. Mol. Catal. A: Chem. 245 (2006)
225–230.
[17] K. Belkacemi, M.C. Vlad, S. Hamoudi, J. Arul, Int. J. Chem. React. Eng. 5 (2007)
A64.
[18] M.S.C. Vlad, Production of Bioactive Lactobionic Acid Using a Novel Catalytic
Method, MSc Thesis, Université Laval, 2007, 116p.
[19] H. Bouasker, Étude expérimentale et optimisation des conditions opératoires
de l’oxydation catalytique du lactose en acide lactobionique, MSc Thesis, Uni-
versité Laval, 2009, 99p.
[20] M. Besson, M.P. Gallezot, Catal. Today 57 (2000) 127–141.
[21] T. Mallat, A. Baiker, Catal. Today 24 (1995) 143–150.
[22] E.V. Murzina, A.V. Tokarev, K. Kordás, H. Karhu, J.-P. Mikkola, D.Y. Murzin, Catal.
Today 131 (2008) 385–392.
[23] A. Mirescu, U. Prü␤e, Appl. Catal. B: Environ. 70 (2007) 644–652.

L.-F. Gutierrez et al. / Applied Catalysis A: General 402 (2011) 94–103
103
[24] M. Haruta, Catal. Today 36 (1997) 153–166.
[35] L. Li, C. Jin, X.C. Wang, W.J. Ji, Y. Pan, T. van der Knaap, R. van der Stoel, C.T. Au,
Catal. Lett. 129 (2009) 303–311.
[36] J. Lee, J.C. Park, H. Song, Adv. Mater. 20 (2008) 1523–1528.
[37] E.D. Park, J.S. Lee, J. Catal. 186 (1999) 1–11.
[25] C.W. Corti, R.J. Holliday, D.T. Thompson, Appl. Catal. A 291 (2005) 253–261.
[26] B. Lee, Z. Maa, Z. Zhang, C. Park, S. Dai, Microporous Mesoporous Mater. 122
(2009) 160–167.
[27] J. Hu, L. Chen, K. Zhu, A. Suchopar, R. Richards, Catal. Today 122 (2007) 277–283.
[28] E.P. Barrett, L.G. Joyner, P.P. Halenda, J. Am. Chem. Soc. 73 (1951) 373–380.
[29] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, in: J. Chastain, R.C. King Jr.
(Eds.), Handbook of X-Ray Photoelectron Spectroscopy, Physical Electronics,
Eden Prairie, MN, 1995.
[30] J. Lynch, Analyse physico-chimique des catalyseurs industriels, éditions TECH-
NIP, Paris, 2001.
[31] V.R. Choudhary, A. Dhar, P. Jana, R. Jha, B.S. Uphade, Green Chem. 7 (2005)
768–770.
[38] A.M. Venezia, G. Pantaleo, A. Longo, G. Di Carlo, M.P. Casaletto, L. Liotta, G.
Deganello, J. Phys. Chem. B 109 (2005) 2821–2827.
[39] M.P. Casaletto, A. Longo, A. Martorana, A. Prestianni, A.M. Venezia, Surf. Inter-
face Anal. 38 (2006) 215–218.
[40] A. Abad, P. Concepción, A. Corma, H. García, Angew. Chem. Int. Ed. 44 (2005)
4066–4069.
[41] N. Pestryakov, V.V. Lunin, J. Mol. Catal. A 158 (2000) 325–329.
[42] T. Ishida, N. Kinoshita, H. Okatsu, T. Akita, T. Takei, M. Haruta, Angew. Chem.
Int. Ed. 47 (2008) 9265–9268.
[32] J. Liu, J. Yang, Q. Yang, G. Wang, Y. Li, Adv. Funct. Mater. 15 (2005)
1297–1302.
[33] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky,
Science 279 (1998) 548–552.
[34] E. Besson, A. Mehdi, C. Reyé, R.J.P. Corriu, J. Mater. Chem. 19 (2009)
4746–4752.
[43] O.A. Simakova, B.T. Kusema, B.C. Campo, A.-R. Leino, K. Kordás, V. Pitchon, P.
Mäki-Arvela, D.Y. Murzin, J. Phys. Chem. C 115 (2011) 1036–1043.
[44] G.J. Hutchings, Catal. Today 100 (2005) 55–61.
[45] G.J. Hutchings, M. Haruta, Appl. Catal. A: Gen. 291 (2005) 2–5.
[46] R. Zanella, C. Louis, S. Giorgio, R. Touroude, J. Catal. 223 (2004) 328–339.
[47] K. Belkacemi, S. Hamoudi, Ind. Eng. Chem. Res. 49 (2010) 6878–6889.