Efficient and recyclable catalysts for selective oxidation of polyols in H2O with molecular oxygen

Article abstract of DOI:10.1039/c1gc15811b

Advanced carbon materials with improved physical properties and tailored surface morphology and chemistry were developed and applied as the support for platinum nanoparticles. The carbon materials are in regular spherical form (around 100 μm) and have a scale-like surface morphology. Platinum nanoparticles of high dispersion was achieved and detailedly characterized by EPMA, TEM, and XPS. The resulting catalysts show high activity for the selective oxidation of glycol, propylene glycol, and glycerol with molecular oxygen as the oxidant. The correlation between catalytic behaviour and the materials' surface properties (active phase dispersion, support surface morphology and chemistry, and bimetallic effect) were further investigated. The recyclability and stability of the catalysts were tested over a five-run recycling experiment. No marked deactivation was observed and the transformation of the dispersion state of platinum nanoparticles during this process was detected by XPS and TEM to address the stabilizing mechanism of the carbon support for platinum nanoparticles.

Full text of DOI:10.1039/c1gc15811b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 3414
www.rsc.org/greenchem
PAPER
Efficient and recyclable catalysts for selective oxidation of polyols in H2O
with molecular oxygen
Zhijun Huang,a,b Fengbo Li,*a Bingfeng Chen,a,b Fei Xue,a,b Yin Yuan,a,b Guochang Chena,b and
Guoqing Yuan*a
Received 7th July 2011, Accepted 30th August 2011
DOI: 10.1039/c1gc15811b
Advanced carbon materials with improved physical properties and tailored surface morphology
and chemistry were developed and applied as the support for platinum nanoparticles. The carbon
materials are in regular spherical form (around 100 mm) and have a scale-like surface morphology.
Platinum nanoparticles of high dispersion was achieved and detailedly characterized by EPMA,
TEM, and XPS. The resulting catalysts show high activity for the selective oxidation of glycol,
propylene glycol, and glycerol with molecular oxygen as the oxidant. The correlation between
catalytic behaviour and the materials’ surface properties (active phase dispersion, support surface
morphology and chemistry, and bimetallic effect) were further investigated. The recyclability and
stability of the catalysts were tested over a five-run recycling experiment. No marked deactivation
was observed and the transformation of the dispersion state of platinum nanoparticles during this
process was detected by XPS and TEM to address the stabilizing mechanism of the carbon
support for platinum nanoparticles.
Selective oxidation of primary and secondary alcohols to the
Introduction
corresponding carbonyl compounds is a very important process
Molecular oxygen is a steadily available and environmentally
in organic synthesis.3,4 Classical oxidation methods use stoichio-
clean reagent for chemical oxidation. There is 20% of oxygen in
metric quantities of inorganic oxidants, chromium(VI) reagents
air and water is the only byproduct of oxidations with oxygen.
or ruthenium or manganese salts,3 which are highly toxic and
In every aerobic organism in nature, various selective oxidation
environmentally polluting. Hypervalent iodine reagents5 and
reactions with molecular oxygen occur every moment. However,
Swern oxidation (the stoichiometric use of DMSO)4 are other
it remains a difficult problem for molecular oxygen to be
two classical non-green oxidation methods. However, modern
selectively activated for the synthesis of complicated molecules
oxidation methods prefer using either O2 or H2O2 as green
under mild conditions. Oxygen can be easily activated under
oxidants. Homogeneous catalysis provides powerful solutions
harsh conditions and organic compound burns out quickly
for selective green oxidation, but the use of organic solvent
with the product of carbon dioxide. This quick and complete
and the problems related to handling, recovery, and reuse of
oxidation is not applicable to selective synthesis of target organic
the catalyst limit the application of these processes. Many
molecules. Activation of oxygen is thermodynamically favorable,
applications of the heterogeneous oxidation of simple alcohols
but kinetically inert.1 This is caused by the mismatch in the spin
using O2 have been well-established.6 However, the selective
state of triplet dioxygen with singlet organic molecules. This
oxidation of polyols to target compounds remains problematic
limitation can be lifted by introducing transition metal catalysts
in developing practical and green methods.
for triplet molecular oxygen. Oxygen activation with transition
metals received extensive attention, and it shows great promise
a or b-hydroxyl carboxylic acids are important intermediate
chemicals for degradable or biocompatiable polymers and
from the standpoint of developing practical, environmentally
medicine. General synthesis methods involve highly toxic HCN.
clean oxidations.2
Selective oxidation of polyols is a powerful green method to
synthesize a or b-hydroxyl carboxylic acids. Heyns et al. first
reported that 2-keto-L-gulonic acid can be obtained in high yield
by bubbling oxygen through an aqueous solution of L-sorbose
and suspended finely divided platinum.7,4 Heyns oxidation shows
highly efficient oxidations of primary alcohols to acids under
mild conditions. After the initial studies by Heyns et al. on
the oxidation of L-sorbose, the platinum-catalyzed oxidation of
aBeijing National Laboratory of Molecular Science, Laboratory of New
Materials, Institute of Chemistry, Chinese Academy of Sciences, Beijing,
100190, P. R. China. E-mail: lifb@iccas.ac.cn, yuangq@iccas.ac.cn;
Fax: (86)-10-62559373; Tel: (86)-10-62634920
bGraduate University of Chinese Academy of Sciences, Beijing, 100049,
P. R. China
3414 | Green Chem., 2011, 13, 3414–3422
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
primary alcohols with oxygen was expanded to numerous sub-
strates by many researchers. All subsequent studies confirmed
the early observations and the original protocol had not been
essentially improved.4 From a practical angle, Heyns oxidation
is very costly for large-scale synthesis because of the expensive
platinum. For best results, a generous amount of catalyst must
be employed. Fine black platinum (0.6–2.1 equivalent) or 5–
10% Pt/C (95–210 g mol-1 alcohol) is required for catalyzing the
oxidation of one equivalent of alcohol.7,4
In this work, well-dispersed platinum nanoparticles over
advanced carbon materials were prepared by incipient wet
impregnation and performed as efficient and recyclable catalysts
for the selective oxidation of glycol, propylene glycol, and
glycerol with molecular oxygen. In our catalytic systems, the
total platinum amount is only 0.05–0.1% of Heyns oxidation
method for catalyzing oxidation of one equivalent alcohol,
however, the activity is comparable to gold nanoparticles over
activated carbon. TOF is around 1000 h-1. More appealingly,
this catalyst can been separated by simple filtration and recycled
batch by batch for over ten runs without marked deactivation.
All these properties are attributed to the unique carbon support,
which is a well-prepared polymer-derived carbon microsphere
with tailored surface chemistry and morphology. In our previous
works, analogous carbon microspheres have been proven to
be suitable support for monodispersed metal nanoparticles8,9
and the resulting materials are excellent catalysts for methanol
carbonylation.9,10 Herein, green and recyclable catalysts for the
oxidation of polyol are developed based on such advanced
carbon supports.
Fig. 2 XPS survey spectrum of the pre-oxidized carbon microspheres.
The insert is the XPS peak of Si2p and fitting of the Si2p envelope
individualizes three silicon states.
The presence of silicon may be caused by application of a silica
sol–gel as a dispersion agent in the preparation process. Si
surface atomic mole concentration is about 5%. The insert of
Fig. 2 is XPS peak of Si2p and fitting of the Si2p envelope
individualizes three silicon states: silicon oxycarbide phase
(SiOxCy) (101.2 eV, 47.88%), silicon carbide (100.5 eV, 41.07%),
and silicon (97.8 eV, 11.05%).
The occurrence and transforming of silicon is evidenced by
the X-ray powder diffraction patterns of the samples treated
under different temperatures. Fig. 3a shows the XRD patterns of
Results and discussion
Surface properties of the carbon support and the resultant
catalyst
The as-synthesized carbon materials have a regular spheri-
cal shape and uniform size distribution (100–125 mm) (Fig.
1a). Fig. 1b shows the topography of the local surface of
a carbon spherule. Scale-like layers are observed and such
surface morphology can offer steric restriction to confine metal
nanoparticles and prevent their interaction and agglomeration.11
Fig. 1 a) A SEM image of as-synthesized carbon microspheres; b) a
SEM image of the surface morphology of a selected carbon microsphere.
X-ray photoelectron spectroscopy (XPS) was applied to detect
the surface chemical composition of the carbon spherule. Fig. 2
is the XPS survey spectrum of the sample that has been treated
with diluted nitric acid. Three elements can be discriminated:
carbon (C1s), oxygen (O1s), and silicon (Si2s, 2p). Oxygen is
introduced by partially oxidative functionalization of surface
and this treatment can improve anchoring and dispersing of
metal nanoparticles’ precursors over carbon support surface.12,13
Fig. 3 a) XRD patterns of carbon microspheres treated at different
temperatures; b) the N2 adsorption isotherm and BET calculation (the
insert).
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 3414–3422 | 3415
©

View Article Online
the samples treated at different temperatures. With the increase
of treatment temperature, three typical XRD peaks of b-SiC:
[111], [202], and [113], are detected, and the carbon materials are
graphitized gradually to a high degree. Another marked change
is the disappearing of the SiO2 peaks, which clearly shows the
transforming of SiO2 to SiOxCy and SiC. These transformations
caused by thermal treatment improve physical and mechanical
properties. These improvements overcome the main drawbacks
of carbon materials, such as graphite, carbon black, activated
carbon, pyrolytic carbon, fullerenes, and carbon nanotubes. The
increase of graphitization degree ameliorates surface resistant
to both acidic and basic media and structure stability at high
temperatures.
Thermal treatment improves mechanical properties of the
carbon spherules. The resulting materials show high resistance to
mechanical crushing and abrasion in catalytic process operation.
The surface chemical properties are modified to control polarity
and hydrophobicity by introducing heteroatom species during
thermal treatment.
Nitrogen sorption analyses were carried out to characterize
the porosity properties. The isotherm of the resultant carbon ma-
terials is best described as a type I isotherm (Fig. 3b). There is es-
pecially the steep slope at low relative partial pressure, with a hys-
teresis loop. The isotherm and adsorption–desorption curve are
typical for porous carbon materials. A BET surface area of 700
m2 g-1 was calculated for the carbon materials. The total pore vol-
ume is 0.394 cc g-1 (P/P0 = 0.997) for mesopores and micropores.
The metal nanoparticle precursors are introduced by incipient
wet impregnation. The metal nanoparticles are obtained by
reducing the impregnated sample in diluted hydrogen flow under
thermal treatment. The dispersion state of metal nanoparticles
directly determines their catalytic performance. Electron Probe
Microanalysis (EPMA) images (Fig. 4) give the global view of
platinum nanoparticle dispersion over a carbon microspherule
(metal load: 0.5 wt%). Fig. 4a shows the image of the selected
carbon spherule, and it is clearly demonstrated that it is in a
regular spherical shape and surface morphology, as depicted by
Fig. 1a. Fig. 4b shows the corresponding dispersion of platinum
nanoparticles. There is some background noise in Fig. 4b, but it
is negligible. Bright points are platinum enriched spots.
Transmission electron microscopy (TEM) is employed to
analyze the particle size and morphology of the supported
platinum nanoparticles. As shown by Fig. 5a and b, platinum
nanoparticles are in a monodispersed state and the particle
size is under 10 nm. On closer inspection, it is found that
platinum nanoparticles are located over the scale-like surface
of the carbon spherule, which has been characterized by SEM
(Fig. 1b). Such surface morphology provides steric restriction to
stabilize the metal nanoparticles. Chemical states of supported
platinum nanoparticles were measured by X-ray photoelectron
spectroscopy (XPS). The binding energy of Pt 4f7/2 is 71.3 eV
(Fig. 5c), which indicates that supported platinum species are
kept in a zero-valent chemical state. Herein, monodispersed
platinum nanoparticles over a carbon spherule were obtained.
Fig. 4 Electron Probe Microanalysis images of the nitrogen-doped
carbon microsphere supporting platinum nanoparticles. a) The se-
lected microsphere and b) the corresponding dispersion of platinum
nanoparticles.
For a given support of noble metal nanoparticles, there is a
suitable metal load range to achieve high catalytic activity with
the least metal consuming. The best metal loads for Pt over
the present carbon spherules are from 0.5 wt% to 0.75 wt%.
In the follwing experiments the Pt metal load is 0.5 wt%. The
size and dispersion of supported nanoparticles are affected by
many factors, including support surface properties, preparation
method, and nanoparticle precursors. It is difficult to rationalize
the relation between particle size and the metal load. In our
experimental observations, low metal load leads to relatively
low catalytic activity.
Fig. 7 shows the results of reaction condition screening
experiments. The oxidation TOF is sensitive to the oxygen
pressure. The oxidation reaction can proceed smoothly over
50 ◦C, but the fluctuating of reaction temperature does not affect
the activity as markedly as the oxygen pressure. Acidification
in the oxidation of alcohol to carboxylic acid can cause very
substantial decrease in oxidation speed. Base must be added
to avoid acidification. In the present catalytic system, NaOH
is used and its amount is 0.8–1.0 equivalents of the substrate.
Excessive base can cause the decrease of oxidation selectivity,
due to over oxidation and base-catalyzed decomposition of the
target compound. Different carbon materials were selected as
the supports for platinum nanoparticles. Pt/carbon spherule
(Pt/CS) was observed as the most active catalyst with relatively
Catalytic oxidation of polyols with molecular oxygen
The catalysts with different metal loads were tested. The relation
between metal load and catalytic activity is shown in Fig. 6.
3416 | Green Chem., 2011, 13, 3414–3422
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 7 Screening experiment results of reaction temperature and oxygen
pressure. Reaction conditions: 10 mmol glycol, 8 mmol NaOH, 8 ml
deionized water were added to a 50 ml autoclave. 0.2 g catalyst (metal
load: 0.5 wt%) was used. Reaction time is 2 h. Glycol/Pt is 1950.
Table 1 Comparison of supporting behavior of different carbon
materials
Catalysta
TOF (h-1) Conversion (%)b Selectivity (%)
Pt/Carbon Spherule
Pt/Activated Carbon 242
Pt/Carbon Black
Pt/Graphite
Pt/Carbon Nanotube 357
852
87.3
24.8
21.1
22.1
36.6
93
81
77
78
86
206
216
Reaction conditions: 10 mmol glycol, 8 mmol NaOH, 8 ml deionized
water were added to a 50 ml autoclave. 0.2 g catalyst (metal load: 0.5 wt%)
was used. Oxygen pressure is 1.0 MPa. Reaction temperature is 70 ◦C.
Reaction time is 2 h. Glycol/Pt is 1950.a All catalysts are prepared by
the same incipient wet impregantion and reducing process and the metal
load is 0.5 wt%. b Data are determined at reaction time of 2 h.
Fig. 5 a) and b) TEM images of the dispersion state of supported
platinum nanoparticles; c) XPS peaks of platinum nanoparticles (Pt
4f7/2: 71.3 eV).
supported platinum nanoparticles exhibit excellent performance
in catalyzing this conversion. The reaction TOF is 891 h-1 and
the selectivity to lactic acid is 90%. Glycerol, a by-product
of biodiesel, has stimulated much research in transforming
this inexpensive compound into valuable chemicals, due to its
large availability.16,17 Selective oxidation of glycerol under mild
conditions is a very useful tool to convert glycerol to value-
added products.18 In our research, carbon spherule supported
platinum nanoparticles can oxidize glycerol to glyceric acid
with a selectivity of 82% under a TOF of 645 h-1. Compared
with many reported results,19 our catalyst is very efficient in
converting glycerol to glyceric acid, whose D-enantiomer is an
anticirrhosis agent.
The catalyst can be separated from the reaction mixture
by simple filtration. This allows quick recovery for reuse in
the next run. The recyclability of platinum nanoparticles over
carbon spherule is tested by a five-run recycling experiment
(Fig. 8). The activity of the catalyst is kept constant after
five-run recycling. In our practical application, the catalyst
remains considerably activity after over ten batches of operation.
Platinum nanoparticles over other carbon supports, such as
activated carbon, carbon black, graphite, and carbon nanotubes,
are not very mechanically stable and suffer from serious attrition
in use. Recycling tests of the catalysts over these carbon materials
show marked catalytic activity loss after three runs.
Fig. 6 Screening experiment results of Pt metal load over catalysts.
Reaction conditions: 10 mmol glycol, 8 mmol NaOH, 8 ml deionized
water were added to a 50 ml autoclave. 0.2 g catalyst with different metal
load was used. Oxygen pressure is 1.0 MPa. Reaction temperature is
70 ◦C. Reaction time is 2 h. Glycol/Pt is 1950.
high selectivity for glycolic acid in catalyzing the oxidation of
glycol (Table 1).
Based on the above well-established method, selective oxida-
tion of propylene glycol and glycerol are attempted over carbon
spherule supported platinum nanoparticles with molecular
oxygen (Table 2). Lactic acid is an important platform chemical
for biodegradable polymers and valuable compounds.14 It can
be obtained from glucose through fermentation or catalytic
conversion.15 Direct oxidation of propylene glycol to lactic
acid is another powerful synthesis method. Carbon spherule
The introducing of the second metal may modify the sur-
face chemistry and morphology of metal nanoparticles and
the support. This may lead to improving or deactivation in
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 3414–3422 | 3417
©

View Article Online
Table 2 Selective oxidation of typical polyols
Polyol
Catalyst
Pt/CS
Conversion (%)b
TOF (h-1)
852
Product Selectivity (%)a
87.3
Pt/CS
Pt/CS
91.4
66.2
891
645
Reaction conditions: 10 mmol polyol, 8 mmol NaOH, 8 ml dei◦onized water were added to a 50 ml autoclave. 0.2 g catalyst (metal load: 0.5 wt%) was
used. Oxygen pressure is 1.0 MPa. Reaction temperature is 70 C. Reaction time is 2 h. Polyol/Pt is 1950.a Only target products are listed. b Data are
determined at reaction time of 2 h.
treatment in diluted hydrogen flow. In presence of La and Bi, the
reaction TOF is increased to near 1100 h-1. Ce and Nd deactivate
the platinum nanoparticles. It shows no marked affecti on the
catalyst. The mechanistic aspect of such a bimetallic effect will
be discussed in the following section.
The correlation between surface properties and catalytic
performance
In heterogeneous catalysis, carbon materials can satisfy most of
the properties desired for a suitable support.20,21 They have some
advantages over other traditional catalyst supports:22 surface
resistant to both acidic and basic media, structure stability at
Fig. 8 The five-run recycling test. Every batch was carried out under
the same reaction conditions.
high temperatures, the tailored pore structure and pore size
distribution needed for a given application, porous carbons with
a variety of macroscopic shapes (e.g., granules, powder, fibers,
cloths, pellets, monoliths, disks), the surface chemical properties
modified to control polarity and hydrophobicity, lower cost than
other conventional catalyst supports, and easy recovery of active
phases by burning off carbon.
catalytic activity and selectivity. The bimetallic effect of platinum
nanoparticles on polyol oxidation was further investigated
(Fig. 9). The second metal precursors were introduced onto
the support by incipient wet co-impregnation with the platinum
precursor and the final catalysts were obtained by thermal
High surface area and well-developed porosity are generally
considered to be very important for achieving a high dispersion
of the active phase in the catalyst. However, for carbon materials,
most of their surface area may be contained in narrow micro-
pores. Such pore distribution is not suitable for mass transfer
in the catalytic process involving large molecules. Some studies
report the positive effect of porosity and surface area on the
active-phase dispersion and the catalytic activity.23 There are also
reports concluding that the surface area and porosity of carbons
do not affect either the active-phase dispersion or the catalytic
activity.24,25 Actually, the relation between porosity and surface
area on the active-phase dispersion and the catalytic activity
has no fixed rule and mostly depends on the special catalyst
compositions and reaction conditions. In this study, a high
surface area of the carbon support may be detrimental. Polyols
and the corresponding hydroxyl carboxylic acids are very viscous
molecules due to hydrogen bonds and the diffusion of reactants
Fig. 9 The bimetallic effect on the catalytic behavior of carbon
microsphere supported platinum nanoparticles.
3418 | Green Chem., 2011, 13, 3414–3422
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
and products may be hindered by the narrow porosity. This has
been confirmed by experimental observation (Table 1). Carbon
spherules have a relatively lower surface area (nearly 700 m2 g-1)
than activated carbon (3000 m2 g-1), but platinum nanoparticles
over carbon spherules show higher catalytic activity. This is
mainly attributed to the unique surface morphology of the
carbon spherule. The layer-structured surface offers enough
surface area to disperse and stabilize metal nanoparticles (Fig.
1b and Fig. 5). Most metal nanoparticles are located on the
support surface and this exposes them directly to reactant
molecules (Fig. 4) and makes them more active than metal
nanoparticles embedded in micropore systems. This is the main
reason that carbon spherule supported platinum catalysts show
high catalytic activity even with very low metal loading.
Surface chemical properties are extremely important carbon
characteristics for interpreting the catalytic behavior of carbon-
supported catalysts. The common heteroatoms on carbon sur-
face are hydrogen, oxygen, nitrogen, sulphur and phosphorus,
which are in the forms of functional groups, such as carboxylic,
lactonic, phenoclic, pyranes, chromene, and so on. These groups
directly determine the carbon surface acid–base and hydrophilic
properties. More importantly, they affect the dispersion of the
active phase across the carbon support and the interaction
between active phases and the support. There are many reports
attempting to clarify the influence of surface functional groups
on the catalytic behavior of carbon-supported catalysts.26 In
this work, four types of carbon spherules, which have different
surface chemical compositions, are selected to address the
role of surface functional groups in affecting the dispersion
of platinum nanoparticles and the catalytic activity. S1 is the
carbon spherules that have been treated at 1000 ◦C under diluted
hydrogen flow and their surface oxygen atomic concentration
determined by XPS is less than 0.1%. S2 is the as-synthesized
carbon spherules and their XPS oxygen atomic concentration
is 5.3%. S3 is the carbon spherules that have been treated
with 30% nitric acid aqueous solution and their XPS oxygen
atomic concentration is 12.1%. S4 is the nitrogen-doped carbon
spherules that are obtained by added acrylonitrite (20%) to the
precursors’ polymerization process. Their XPS oxygen atomic
concentration is 4.1% and nitrogen atomic concentration is
5.7%. Chemical modification of the support surfaces show little
influence on the porosity properties of the carbon spherules.
Surface areas of support S1–S4 have no marked difference.
Supported platinum nanoparticles are prepared with the same
method and metal load (0.5 wt%). Fig. 10a shows Pt/C XPS
intensity ratio of the above-mentioned four types of carbon
spherules (determined by XPS). The XPS intensity ratio of
signals from metal particles and the support (Ip/Is) reflects the
dispersion of metal nanoparticles over the carbon support.27
The well-dispersed state leads to a high intensity ratio. It is
clearly demonstrated that platinum dispersion increases with
the increase in concentration of surface functional groups.
This tendency is very similar to the results obtained from
platinum dispersed over carbon black or activated carbon.26
The platinum nanoparticles dispersion state directly affects the
catalytic activity. Better dispersion leads to higher catalytic
activity (Fig. 10b).
Fig. 10 a) The XPS intensity ratio IPt/IC of platinum nanoparticles
over four types of carbon microsphere surface and b) the corresponding
catalytic activity for selective oxidation of glycol.
mechanical hardness, and good resistance to attrition. The
chemical stability of platinum nanoparticles and the support
during the oxidation reaction is another important factor
affecting the recyclability. Fig. 11a shows the Pt/C XPS intensity
ratio of the catalyst after every test run in the five-run recycling
experiment (Fig. 8). It is clearly revealed that the dispersed
state of platinum nanoparticles is kept at a relatively high level.
The chemical state of platinum nanoparticles is kept constant
during the oxidation reaction, as determined by XPS (Fig. 11b).
Fig. 11 a) The XPS intensity ratio IPt/IC of the catalysts in five runs
of recycling test and b) comparison of XPS Pt peaks’ binding energy of
the fresh catalyst (A) and the used catalyst (B).
The good recyclability of platinum nanoparticles over carbon
spherules is partly attributed to regular physical form, high
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 3414–3422 | 3419
©

View Article Online
TEM images of Run-1 catalyst, Run-3 catalyst, and Run-5
catalyst give a direct and close view of the dispersion state of
platinum nanoparticles during the oxidation reaction (Fig. 12).
The characterized local surfaces are selected randomly, but the
comparison indicates that there is no marked agglomeration and
leaching out of platinum nanoparticles during the oxidation
reaction. This confirms the result obtained from Pt/C XPS
intensity ratio analysis and the five-run recycling experiment.
Fig. 13 a) XPS Pt peaks of the fresh catalyst (A) and the used catalyst
(B); b) XPS La peaks of the fresh catalyst (A) and the used catalyst (B).
adsorption of polyols over the catalyst surface and then promote
the catalytic conversion. Fig. 14a is the comparison of Pt XPS
peaks between the fresh Pt–Bi (1 : 1) catalyst and the used Pt–
Bi (1 : 1) catalyst and Fig. 14b is the comparison of Bi XPS
peaks. The chemical states of Pt and Bi show marked changes.
This indicates that Bi is involved in the oxidation reaction and
meanwhile modifies the platinum nanoparticles. The bimetallic
effect of Pt–Bi can be interpreted from the following three points:
1) Bi is involved in the catalytic oxidation process. Bi has oxygen
storage ability and shows catalytic activity in many oxidation
reactions. 2) The presence of Bi can facilitate the chemical
adsorption of polyols over the catalyst surface. 3) The presence
of Bi can improve the dispersion of platinum nanoparticles. An
increase of the content of Bi can directly increase the adsorbed
activated oxygen over the catalytic active sites and promote the
catalytic process (Fig. 9). However, there is a limit to Bi content,
over which the promoting effect has little fluctuation. The Bi pro-
moting effect shows a different mechanism from the La involved
promoting effect. A more elaborate model catalyst is needed to
address this rising problem. Further investigation is under way.
Fig. 12 TEM images of the dispersion state of the catalysts in the
recycling test. a) Run-1 (the largest nanoparticle: 25 nm, the smallest:
5 nm, average size: 10 nm), b) Run-3 (the largest nanoparticle: 25 nm, the
smallest: 4 nm, average size: 9 nm), c) Run-5 (the largest nanoparticle:
22 nm, the smallest: 5 nm, average size: 12 nm).
The bimetallic effect also affects the catalytic activity of the
supported platinum nanoparticles (Fig. 9). La and Bi show a
patent promoting effect on the catalytic activity. Fig. 13a is the
comparison of Pt XPS peaks between the fresh Pt–La catalyst
and the used Pt–La catalyst and Fig. 13b is the comparison
of La XPS peaks. Platinum nanoparticles show no chemical
state changing during the oxidation. The chemical state of La is
also kept constant and the XPS peaks are attributed to La2O3.
This reveals that La is not involved in the catalytic oxidation
reaction. Its promoting effect may result from two aspects. First,
the presence of La2O3 can improve the dispersion of platinum
nanoparticles. Pt and La precursors were coimpregnated. The La
precursors precipitate as La(OH)3 and restrain agglomeration
of Pt salts. In the thermal treatment, the presence of La2O3
can lower the sintering of Pt nanoparticles. Second, La is the
oxyphilic element and has a strong chelating ability to hydroxyl
groups. The presence of La2O3 can facilitate the chemical
Conclusion
In summary, we have developed green and recyclable catalysts
for selective oxidation of glycol, propylene glycol, and glycerol
to the corresponding hydroxyl carboxylic acids. Advanced
carbon materials with a regular spherical shape and improved
mechanical properties are applied as the support for platinum
3420 | Green Chem., 2011, 13, 3414–3422
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
added to the solution under stirring. The mixture was kept
stirring at 45 ◦C for 24 h and then cooled to 5 ◦C. Purified
vinylidene chloride, with the initiator azobisisobutylvaleronitrile
(1% mole ratio) dissolved, was added rapidly. A copolymer
of bis(2-aminoethyl) ethylene diamine and epichlorohydrin was
added to make the resultant polymer materials a good spherical
shape. Such a mixture was kept at 40 ◦C in a water bath for 24
h under agitation of 250–300 RPM. The final solid materials
were filtered and washed by a great amount of deionized water
and then dried in vacuum at room temperature. The precursors
were first treated in a quartz tube under argon stream at 180 ◦C.
After complete removal of HCl, the treatment temperature was
◦
◦
increased to 800 C under the stream of argon at rate of 1 C
min-1 and then kept at 1000 ◦C for 3 h. By this method, carbon
microspherules were obtained.
Before introducing the metal precursors, the carbon support
are treated with 30% HNO3 aqueous solution at 40 ◦C for
12 h. The metal precursors were introduced into the support
by incipient wet impregnation. The carbon spherules were
immersed in metal salt methanol or aqueous solution for 40 min,
followed by drying in vacuum. The solution volume is equal to
total pore volume of the given support. The fresh impregnated
sample was reduced at 200 ◦C in the stream of argon and
hydrogen (4 : 1) for 8 h.
Characterization of the resultant materials
X-ray photoelectron spectroscopy (XPS) was used to detect the
change of chemical states of carbon and silicon species during
evolution of the composite materials. All peak positions in
XPS experiments were calibrated by the binding energy of C1s
as reference of 284.6 eV. Eclipse V2.1 data analysis software
supplied by the VG ESCA-Lab200I-XL instrument manufac-
turer was applied in manipulation of the acquired spectra.
The shape and surface morphology of the as-synthesized com-
posite microspherules were characterized by scanning electron
microscope (SEM) conducted over Hitachi S530. Transmission
electron microscopy (TEM) was obtained by a JEOL 2010 TEM
with an accelerating voltage of 200 kV. Nitrogen adsorption
isotherms and textural properties of the as-synthesized materials
were determined at 77.3 K using nitrogen by a Quantachrome
Autosorb Automated Gas Sorption System (Quantachrome
Corporation). Before analysis, the samples were dried at 300 ◦C
under vacuum. Electron Probe Microanalysis (EPMA) (carried
out on EPMA1600, SHIMADZU) was used to characterize the
morphological structure of the nanocomposite spherules and
the dispersion state of nanoparticles. Powder X-ray diffraction
(XRD) data were recorded using a Rigaku D/max-2500.
Fig. 14 a) XPS Pt peaks of the fresh catalyst (A) and the used catalyst
(B); b) XPS Bi peaks of the fresh catalyst (A) and the used catalyst (B).
nanoparticles. This makes the catalyst separation and recycling
easy to operate. The tailored surface morphology and chemistry
of the carbon support leads to high dispersion of platinum
nanoparticles. High catalytic activity is achieved even under low
metal load (0.5 wt%). All oxidation processes are carried out
in water under mild reaction conditions with molecular oxygen
as the oxidant. No organic solvents are used and no additives
except NaOH are added. The catalyst shows high selectivity for
partial oxidation products and can been recycled for five runs
without any deactivation. Herein, a green route for the catalytic
selective oxidation of polyols is well established.
Experimental
Chemicals and Materials
Vinylidene chloride, azobisisobutylvaleronitrile, cetyltrimethy-
lammonium chloride (CTAC), glycol, propylene glycol, and
glycerol were purchased from Sinopharm Chemical Reagent Co.
Ltd. Vinylidene chloride was distilled before using. H2PtCl4 was
purchased from Strem. All the reagents were used as received.
Catalytic oxidation of glycol, propylene glycol, and glycerol
The catalytic oxidation was carried out in a 50 ml autoclave
with magnetic stirring. To 10 mmol polyol was added 5–8 ml
of deionized water containing 8 mmol NaOH. 0.2 g of catalyst
(metal load 0.5 wt%) was used and the ratio of substrate to
metal amount is 1950. The autoclave was flushed by oxygen
three times and the reaction pressure of oxygen is 1.0 MPa. The
Preparation of the catalysts
Carbon microspherules were derived from a polymer precursor.
The surfactant, cetyltrimethylammonium chloride (CTAC) and
ammonium hydroxide were dissolved in deionized water at
35 ◦C. The required amount of tetraethyl orthosilicate was then
◦
reaction temperature is 70 C. The substrate can be converted
totally in 3 h. The catalyst was separated by simple filtration and
washed by deionized water, then dried for next use. The acidified
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 3414–3422 | 3421
©

View Article Online
reaction mixture was analyzed by GC over SHIMADZU GC-
2014 with the use of capillary column MXT-1 (30 m, 0.25 mm
ID, 0.25 mm df). The conversion of polyol was inter-calibrated by
1H-NMR and 13C-NMR over a Varian-Unity 400 MHz NMR
Spectrophotometer.
15 M. S. Holm, S. Saravanamurugan and E. Taarning, Science, 2010,
328, 602.
16 M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi and C. D. Pina,
Angew. Chem., Int. Ed., 2007, 46, 4434.
17 M. Pagliaro and M. Rossi, The Future of Glycerol: New Uses of a
Versatile Raw Material, RSC Publishing, Cambridge, 2008.
18 B. Katryniok, H. Kimura, E. Skrzynska, J. S. Girdon, P. Fongarland,
M. Capron, R. Ducoulombier, N. Mimura, S. Paul and F. Dumeignil,
Green Chem., 2011, 13, 1960.
Notes and references
19 T. Imanaka, H. Terasaki, A. Fujio and Y. Yokota, (Kao Corp.)
JP1992133500, 1992; T. Imanaka, H. Terasaki and A. Fujio, (Kao
Corp.) JP1992149195, 1993; S. Carretin, P. McMorn, P. Johnston,
K. Griffin, C. Kiely and G. J. Hutchings, Phys. Chem. Chem. Phys.,
2003, 5, 1329; S. Carretin, P. McMorn, P. Johnston, K. Griffin, C.
Kiely, A. Attard and G. J. Hutchings, Top. Catal., 2004, 27, 131; S.
Carrettin, P. McMorn, P. Johnston, K. Griffin and G. J. Hutchings,
Chem. Commun., 2002, 696; C. L. Bianchi, P. Canton, N. Dimitratos,
F. Porta and L. Prati, Catal. Today, 2005, 203, 102; N. Dimitratos,
A. Villa, C. L. Bianchi, L. Prati and M. Makkee, Appl. Catal., A,
2006, 311, 185; F. Porta and L. Prati, J. Catal., 2004, 224, 397; N.
Dimitratos, J. A. Lopez-Sanchez, D. Lennon, F. Porta, L. Prati and
A. Villa, Catal. Lett., 2006, 108, 147; N. Dimitratos, C. Messi, F.
Porta, L. Prati and A. Villa, J. Mol. Catal. A: Chem., 2006, 256, 21;
N. Dimitratos, F. Porta and L. Prati, Appl. Catal., A, 2005, 291, 210;
L. Prati, A. Villa, F. Porta, D. Wang and D. Su, Catal. Today, 2007,
122, 386.
1 A. Bakac, Physical Inorganic Chemistry: Reactions, Processes, and
Applications, John Wiley & Sons, New Jersey, 2010, p. 109.
2 D. T. Sawyer, Oxygen Chemistry, Oxford University Press: New
York, 1991; C. S. Foote, J. S. Valentine, A. Greenberg and J.
F. Liebman, Active Oxygen in Chemistry, Blackie Academic &
Professional, Chapman & Hall, Glasgow, 1995; J. S. Valentine,
C. S. Foote, A. Greenberg and J. F. Liebman, Active Oxygen in
Biochemistry, Blackie Academic & Professional, Chapman & Hall,
Glasgow, 1995; B. Meunier, Biomimetic Oxidations Catalyzed by
Transition Metal Complexes, Imperial College Press, London, 2000;
L. I. Simandi, Advances in Catalytic Activation of Dioxygen by Metal
Complexes, Kluwer, Dordrecht, 2003; W. B. Tolman, Activation of
Small Molecules, Wiley-VCH, Weinheim, 2006, p187; J. E. Backvall,
Modern Oxidation Methods, Wiley-VCH, Weinheim, 2nd edn, 2010,
p. 336; N. Mizuno, Modern Heterogeneous Oxidation Catalysis,
Wiley-VCH, Weinheim, 2009; W. Nam, Acc. Chem. Res., 2007, 40,
522; R. van Eldik, Chem. Rev., 2005, 105, 1917; R. H. Holm and E. I.
Solomon, Chem. Rev., 2004, 104, 347; A. Bakac, Prog. Inorg. Chem.,
1995, 43, 267; A. Bakac, Adv. Inorg. Chem., 2004, 55, 1; A. Bakac,
Coord. Chem. Rev., 2006, 250, 2046.
3 G. Tojo, M. Fernandez, Oxidation of Alcohols to Aldehydes and
Ketones, Springer, New York, 2006.
4 G. Tojo, M. Fernandez, Oxidation of Primary Alcohols to Carboxylic
Acids, Springer, New York, 2007.
5 A. Varvoglis, Hypervalent Iodine in Organic Synthesis, Academic
Press, San Diego, 1997.
6 T. Mallat and A. Baiker, Chem. Rev., 2004, 104, 3037; Y. Wang, J.
Zheng, W. Dai and K. Fan, Green Chem., 2011, 13, 1644.
7 O. Dalmer, K. Heyns, U.S. Pat. 2,190,377, 1940.
20 P. Serp and J. L. Figueiredo, Carbon Materials for Catalysis, John
Wiley & Sons, Hoboken, New Jersey, 2009.
21 V. Calvino-Casilda, A. J. Lopez-Peinado, C. J. Duran-Valle and R.
M. Martin-Aranda, Catal. Rev. Sci. Eng., 2010, 52, 325.
22 F. Rodr’ıguez-Reinoso, in Porosity in Carbons: Characterization and
Applications, ed. J. W. Patrick, Edward Arnold, London, 1995.
23 A. Linares-Solano, F. Rodriguez Reinoso, C. Salinas-Martinez de
Lecea, O. P. Mahajan and P. L. Walker Jr., Carbon, 1982, 20, 177; A.
Martingullon, C. Pradoburguete and F. Rodriguezreinoso, Carbon,
1993, 31, 1099; J. Laine, F. Severino and M. Labady, J. Catal., 1994,
147, 355; J. Laine, M. Labady, F. Severino and S. Yunes, J. Catal.,
1997, 166, 384.
24 F. Rodriguez-Reinoso, C. Salinas-Martinez de Lecea, A. Sepulveda-
Escribano and J. D. Lopez-Gonzalez, Catal. Today, 1990, 7,
287.
8 F. Li, J. Huang, J. Zou, P. Pan and G. Yuan, Carbon, 2002, 40, 2871.
9 F. Li, Q. Qian, F. Yan and G. Yuan, Carbon, 2006, 44, 128.
10 F. Li, J. Huang, J. Zou, P. Pan and G. Yuan, Appl. Catal., A, 2003,
251, 295; F. Li, Q. Qian, F. Yan and G. Yuan, Mater. Chem. Phys.,
2008, 107, 310.
25 L. Li, Z. Zhu, Z. Yan, G. Lu and L. Rintoul, Appl. Catal., A, 2007,
320, 166.
26 F. Derbyshire, V. H. J. De Beer, G. M. K. Abotsi, A. W. Scaroni, J.
M. Solar and D. J. Skrovanek, Appl. Catal., 1986, 27, 117; C. Prado-
Burguete, A. Linares-Solano, F. Rodriguez-Reinoso and C. Salinas-
Martinez De Lecea, J. Catal., 1989, 115, 98; C. Prado-Burguete, A.
Linares-Solano, F. Rodriguez-Reinoso and C. Salinas-Martinez De
Lecea, J. Catal., 1991, 128, 397; W. Han, H. Liu and H. Zhou, Catal.
Commun., 2007, 8, 351; A. Sepulveda-Escribano, F. Coloma and F.
Rodriguez-Reinoso, Appl. Catal., A, 1998, 173, 247.
11 C. H. Bartholomew, Appl. Catal., A, 2001, 212, 17.
12 L. R. Radovic and F. Rodriguez-Reinoso, Carbon materials in
catalysis, Chemistry and Physics of Carbon, New York, Marcel
Dekker, 1997, vol. 15, p. 243.
13 F. Rodriguez-Reinoso, Carbon, 1998, 36, 159.
14 C. H. Holten, Lactic acid, Verlag Chemie: Copenhagen, Denmark,
1971; K. Datta, Hydroxycarboxylic acids, in Kirk-Othmer Encyclope-
dia of Chemical Technology, ed. J. Kroschwitz and M. Howe-Grant,
Wiley, New York, 1997, 4th edn, Vol. 13, p. 1042; R. Datta, S. P. Tsai,
P. Bonsignore, S. H. Moon and J. R. Frank, FEMS Microbiol. Rev.,
1995, 16, 221; Y. Fan, C. Zhou and X. Zhu, Catal. Rev. Sci. Eng.,
2009, 51, 293; J. C. Serrano-Ruiz and J. A. Dumesic, ChemSusChem,
2009, 2, 581.
27 J. W. Neimantsverdriet, Spectroscopy in catalysis, WILEY-VCH,
Weinheim, 3rd edn, 2007, p. 54; A. Frydman, D. G. Castner, M.
Schmal and C. T. Campbell, J. Catal., 1995, 152, 164; F. P. J. M.
Kerkhof and J. A. Moulijn, J. Phys. Chem., 1979, 83, 1612; H. P. C.
E. Kuipers, H. C. E. van Leuven and W. M. Visser, Surf. Interface
Anal., 1986, 8, 235.
3422 | Green Chem., 2011, 13, 3414–3422
This journal is
The Royal Society of Chemistry 2011
©