J. Phys. Chem.

C 2008, 112, 6981-6987 6981

MnO2 Nanorod Supported Gold Nanoparticles with Enhanced Activity for Solvent-free
Aerobic Alcohol Oxidation

Lu-Cun Wang, Yong-Mei Liu, Miao Chen, Yong Cao,* He-Yong He, and Kang-Nian Fan
Department of Chemistry and Shanghai Key Laboratory of Molecular Catalysis and InnoVatiVe Materials,
Fudan UniVersity, Shanghai 200433, People’s Republic of China
ReceiVed: NoVember 30, 2007; In Final Form: February 20, 2008

Gold nanoparticles supported on β-MnO2 with different morphologies, i.e., nanorods and conventional
particulates, were prepared by homogeneous deposition-precipitation using urea as the precipitation agent.
The catalysts were extensively characterized by a combination of different techniques (N2 adsorption, X-ray
diffraction (XRD), transmission electron microscopy (TEM), hydrogen temperature-programmed reduction
(H2-TPR), and X-ray photoelectron spectroscopy (XPS)) in relation to their performance for liquid-phase
aerobic oxidation of benzyl alcohol under solvent-free conditions. TEM analysis showed that the two types
of gold catalysts have similar gold particle size distributions. TPR results indicated that the presence of Au
strongly promotes MnO2 reduction in the Au/MnO2-R system. XPS revealed both reduced and oxidized Au
species on the MnO2 nanorods support before and after the reaction. Significantly enhanced catalytic activity
was observed for gold catalyst supported on MnO2 nanorods, as compared with that on commercial MnO2
powders. The enhanced catalytic activity of the Au/MnO2-R catalyst was attributed to the beneficial presence
of higher amount of oxidized gold species and surface oxygen vacancies resulting from the strong interaction
between Au and the well-defined reactive surface of MnO2 nanorods.

1. Introduction an important role in determining the catalytic activity and

selectivity of gold catalyst.17,18,20 For example, it was found that
Supported gold catalysts have attracted significant attention
compared with activated carbon or graphite, metal oxides such
in the last 2 decades owing to their unique catalytic properties
as CeO2 or TiO2 supported gold catalysts are more active and
under mild conditions.1-4 Studies on Au catalysts have been
selective toward the solvent-free oxidation of alcohols to their
mainly focused on gas-phase oxidation, such as CO oxidation,
corresponding aldehydes or ketones.17,18 Notably, Abad et al.
hydrocarbon combustion, selective oxidation, and water gas shift
have recently reported that Au nanoparticles supported on
(WGS) reaction.1-4 It is generally agreed that the catalytic
nanocrystalline CeO2 were extraordinarily active and selective
activity of gold catalysts depends on the size of the gold
for the solvent-free oxidation of various alcohols to ketones, as
particles, but the nature of the support material, the preparation
compared with the conventional CeO2.17
method, and the activation procedure have also been suggested
to play a key role.1-8 Exceptionally high activities for oxidation Featured with superior ability to activate and supply oxygen,
of CO have been reported for highly dispersed Au on reducible manganese oxide is well-known as a stoichiometric oxidant for
oxides, such as TiO2, Fe2O3, and CeO2, etc. It has been proposed oxidation of hydrocarbons in organic synthesis and has also been
that the reducible metal oxide support supplies oxygen to form extensively investigated as an excellent support or promoter for
active oxidic gold sites.8 Sites at the gold-support interface various metal or metal oxide catalysts.22-27 When combined
have also been claimed to be responsible for the activity in CO with gold nanoparticles, significantly enhanced activity has been
oxidation.9-11 Other explanations focus on the charge transfer achieved over the Au/MnOx system for low-temperature CO
between the support, particularly negatively charged defects (F oxidation in the absence or presence of H2 for fuel cell
centers), and the Au particles.12 In addition, the effects of low- applications. However, to the best of our knowledge, manganese
coordinated sites and surface roughness have also been oxide has rarely been addressed as support of gold nanoparticles
suggested.11,13-15 in the liquid-phase catalytic oxidation reactions.20,27 The present
Over the past decade, much attention has also been paid to work focuses on the exploring of the potential of MnO2 material
the application of gold catalysts in the liquid phase, in particular in the gold-catalyzed oxidation of various structurally different
for the selective oxidation of alcohols, due to its importance in alcohols by molecular oxygen. Special attention is paid to the
both fine chemicals industry and academia.3 The substrates effects of support morphology on their structural properties and
investigated to date include aromatic, aliphatic, and allylic catalytic behavior in the process of aerobic alcohol oxidation.
alcohols, both primary and secondary alcohols, and some The structure-activity relationships of the Au/MnO2 materials
polyols.16-21 In most cases, supported gold catalysts show high are discussed in light of a detailed characterization of the
catalytic activity with much higher selectivity at low temperature physicochemical and surface properties of the catalysts.
and better stability than Pd or Pt catalysts.16 Moreover, it has
been shown that choice of suitable metal oxide as support plays 2. Experimental Section

* Corresponding author. Phone: +86-21-55665287. Fax: +86-21- 2.1. Preparation of Supports and Catalysts. The β-MnO2
65642978. E-mail: yongcao@fudan.edu.cn. nanorods were prepared by hydrothermal method according to
10.1021/jp711333t CCC: $40.75 © 2008 American Chemical Society
Published on Web 04/09/2008
6982 J. Phys. Chem. C, Vol. 112, No. 17, 2008 Wang et al.

the reported procedure.28 Briefly, MnSO4‚H2O (0.008 mol) and

an equal amount of ammonium persulfate ((NH4)2S2O8) were
put into distilled water at room temperature to form a homo-
geneous solution, which was then transferred into a 40 mL
Teflon-lined stainless steel autoclave, sealed and maintained at
140 °C for 24 h. After the reaction was completed, the resulting
solid product was recovered by filtration and washing four times
with distilled water. Gold (5.0 wt %) was deposited on
as-synthesized β-MnO2 nanorods (denoted as Au/MnO2-R) by
homogeneous deposition-precipitation (HDP) using urea as the
precipitating agent.29 In a typical procedure, 3.6 g urea was
dissolved in 200 mL of 1.46 mmol/L HAuCl4 solution (urea/
Au ) 100, molar ratio) at room temperature. An amount of
0.55 g of MnO2 support was then added to this clear solution,
and the temperature of the resulting slurry was increased
gradually to 90 °C. The temperature was maintained for 4 h,
followed by filtering and washing several times with distilled
Figure 1. XRD patterns of (a) MnO2-C, (b) Au/MnO2-C, (c) MnO2-
water. The solid product was dried overnight before calcination R, and (d) Au/MnO2-R. (0) β-MnO2, (b) Au0.
at 300 °C for 4 h in static air. For comparison, a gold catalyst
supported on commercial MnO2 (Sinopharm Chem. Reagent.,
99.9%) (denoted as Au/MnO2-C) was also prepared following to the catalytic reactor charged with 200 mmol of alcohols. The
the same procedure. Elemental analysis results showed that the reactor was sealed and purged five times with oxygen before
gold loadings were 5.0 wt % for both catalysts. reaction. The consumed O2 was replenished by maintaining the
Reference gold catalysts including 1.5 wt % Au/TiO2 (type oxygen pressure constant throughout the experiment. The
A, lot no. Au/TiO2 no. 02-1), 4.5 wt % Au/Fe2O3 (type C, lot magnetic stirring speed was set at 1200 rpm, and the reaction
no. Au/Fe2O3 no. 02-5), and 0.8 wt % Au/C (type D, lot no. mixture was raised to the required temperature. Samples were
Au/C no. 38D) were supplied by the World Gold Council taken periodically during the reaction. The products were
(WGC).30 separated by centrifuging and analyzed by a gas chromatograph
2.2. Catalyst Characterization. The BET specific surface (Trace GC Ultra) fitted with an HP-5 capillary column (25 m
areas of the calcined catalysts were determined by adsorption- × 0.32 mm) and flame ionization detector (FID). The conver-
desorption of nitrogen at liquid nitrogen temperature, using a sions and selectivities were determined using undecane as an
Micromeritics TriStar 3000 equipment. Sample degassing was internal standard in all reactions.
carried out at 300 °C prior to acquiring the adsorption isotherm.
Elemental analysis with respect to Au loading was performed 3. Results and Discussion
using inductively coupled plasma atomic emission spectroscopy
3.1. Structural and Textural Properties of the Catalysts.
(ICP-AES) on a Thermo Electron IRIS Intrepid II XSP
The XRD patterns given in Figure 1 show that the two MnO2
spectrometer. The samples were dissolved in a mixture of
materials have identical diffraction features characteristic of the
concentrated HCl and HNO3 with volumetric ratio of 3:1 prior
β-MnO2 phase (JCPDS 24-0735). However, the relative intensity
to the analysis.
of the peak corresponding to the (110) plane for MnO2 nanorods
The X-ray powder diffraction (XRD) of the samples was is much higher than that for the commercial MnO2 particles.
carried out on a Bruker D8 Advance X-ray diffractometer using According to Xu et al.,31 the preferential plane exposed by a
nickel-filtered Cu KR radiation with a voltage and current of particle depends substantially on the shape of the particle.
40 kV and 20 mA, respectively. Therefore, this observation confirms a disparity in the shape of
Transmission electron microscopy (TEM) images were the MnO2 supporting materials, pointing to a preferential
recorded on a JEOL 2011 electron microscope operating at 200 exposure of the (110) plane in the nanorod samples.28 After the
kV. Before being transferred into the TEM chamber, the samples deposition of gold nanoparticles on the surface of MnO2
dispersed with ethanol were deposited onto a carbon-coated supports, the XRD patterns remain unchanged, although weak
copper grid and then quickly moved into the vacuum evaporator. metallic gold diffraction lines are evidenced.
Temperature-programmed reduction (TPR) profiles were TEM images shown in Figure 2A further reveal that the MnO2
obtained on a homemade apparatus loaded with 20 mg of nanorods have well-defined rodlike morphology with diameters
catalyst. TPR experiments were carried out in 5% H2/Ar flowing of 40-100 nm and lengths ranging between 2.5 and 4.0 µm.
at 40 mL‚min-1, with a ramping rate of 10 °C‚min-1 to a final The high-resolution TEM (HRTEM) image (Figure 2B) shows
temperature of 650 °C. The H2 consumption was monitored that the MnO2 nanorods are characteristic of single-crystalline
using a TCD detector. nature and possess the (110) planes as the main surfaces. The
X-ray photoelectron spectroscopy (XPS) spectra were re- structure feature of MnO2 nanorods remained after depositing
corded with a Perkin-Elmer PHI 5000C system equipped with Au. As shown in Figure 2, parts C and E, Au nanoparticles are
a hemispherical electron energy analyzer. The Mg KR source highly dispersed on the MnO2 surface, and the size distribution
(hν ) 1253.6 eV) was operated at 15 kV and 20 mA. The ranged from 2.0 to 10.0 nm for both catalysts as depicted in
binding energy (BE) scale was referenced to the C 1s peak Figure 2F. Accordingly, on the basis of 200 particles, the mean
(284.6 eV) arising from adventitious carbon in the sample. gold particle sizes are estimated to be 4.4 and 5.0 nm for Au/
2.3. Catalytic Activity Tests. The liquid-phase aerobic MnO2-R and Au/MnO2-C, respectively. In addition, the HRTEM
oxidation of alcohols was carried out in a high-pressure, Teflon- image of the Au/MnO2-R catalyst (Figure 2D) shows that the
lined, stainless steel autoclave reactor (Dalian Tongchan Co. gold nanoparticles on the surface of MnO2 nanorods are mainly
Ltd, China; 200 mL). Typically, 0.2 g of the catalyst was added semispherical in shape.
MnO2 Nanorod Supported Au Nanoparticle Catalysts J. Phys. Chem. C, Vol. 112, No. 17, 2008 6983

Figure 2. (A) TEM image of pure MnO2 nanorods; (B) HRTEM image and electron diffraction pattern (inset) of MnO2 nanorod. In the HRTEM
image, the spacing between the lattice fringes is ca. 0.32 nm, which corresponds well with the (110) crystal plane of β-MnO2 (pyrolusite). (C) TEM
image of Au/MnO2-R catalyst; (D) HRTEM image of Au/MnO2-R catalyst. (E) TEM image of Au/MnO2-C; (F) size distributions of Au nanoparticles
obtained for Au/MnO2-R and Au/MnO2-C.
3.2. Redox Properties and Chemical States. TPR experi- as compared with MnO2-C. With Au introduction, the low-
ments were carried out to investigate the reducibility of the temperature (LT) reduction peak of the Au/MnO2-R catalyst
various MnO2-based materials. Figure 3 presents the reduction shifted dramatically from 325 to 178 °C, whereas much less
profiles of MnO2 nanorods, commercial MnO2, and their variation is achieved for that of Au/MnO2-C, viz., from 365 to
corresponding supported gold catalysts. Two reduction peaks 313 °C. These results clearly demonstrate that the presence of
are observed for all four samples, indicating the stepwise Au nanoparticles strongly promotes MnO2 reduction, with such
reduction of manganese oxides.32 The sample MnO2-R exhibits promotion effect being more pronounced for the Au/MnO2-R
a significantly narrower peak and lower reduction temperature than Au/MnO2-C system. The reduction of MnO2 material was
6984 J. Phys. Chem. C, Vol. 112, No. 17, 2008 Wang et al.

Figure 3. TPR profiles of MnO2 materials before and after the Figure 5. Mn 2p XPS spectra of MnO2 supported gold catalysts: (a)
introduction of gold: (a) MnO2-C, (b) Au/MnO2-C, (c) MnO2-R, and Au/MnO2-C; (b) Au/MnO2-R; (c) used Au/MnO2-R.
(d) Au/MnO2-R.

Figure 6. O 1s XPS spectra of (a) MnO2-C, (b) Au/MnO2-C, (c) MnO2-

R, and (d) Au/MnO2-R.
Figure 4. Au 4f XPS spectra of MnO2 supported gold catalysts: (a)
Au/MnO2-C; (b) Au/MnO2-R; (c) used Au/MnO2-R.

while to note that oxidized gold species (15.6%) were also

also shown to be strongly facilitated by the presence of Ag31 detected on the surface of Au/MnO2-R catalyst after reaction
and Rh,33 which has been attributed to the occurrence of for 8 h in the solvent-free oxidation of benzyl alcohol. Notably,
spillover phenomena involving either hydrogen activated on the this contribution is insignificantly decreased after reaction,
metal phase or lattice MnO2 oxygen induced by intimate metal- suggesting that ionic gold has suffered modest reduction during
support interactions. the reaction. Similar to our findings, the presence of positive
The catalyst surface composition and oxidation state were gold species along with metallic Au have also been reported in
investigated by XPS. Figures 4 and 5 show the Au 4f and Mn Au/CeO2 and Au/Cu5Mg1Al2Ox catalysts for selective oxidation
2p core level XPS spectra of various MnO2 supported gold of alcohols.17,21
catalysts, respectively. To study the surface structural changes The XPS spectra shown in Figure 5 indicate that the Mn 2p3/2
that may occur during the oxidation reaction, the Au 4f and peaks of both Au/MnO2-C and Au/MnO2-R samples have a BE
Mn 2p XPS spectra of Au/MnO2-R catalyst after removal from of about 642.1 eV, which is in good agreement with those
the reactor are also included. The detailed XPS parameters of reported in the literature for MnO2, i.e., 642 ( 0.2 eV.39
all samples are summarized in Table 1. Broad peaks of Au 4f7/2 Moreover, a weak shoulder can be identified at about 646 eV,
and Au 4f5/2 states are observed in all Au/MnO2 samples, the intensity of which is slightly stronger for sample Au/MnO2-C
indicating the presence of both metallic and ionic gold spe- as compared with Au/MnO2-R, as indicated by the two arrows.
cies.34,35 As shown in Figure 4, the variation in the relative The shoulder feature at ca. 646 eV is due to the presence of
intensity at ca. 84 eV indicates the presence of different amounts MnO, whereas the characteristic feature at 663 eV is related to
of Au0 species in the two Au/MnO2 catalysts. Deconvolution Mn species associated with lower valence states, i.e., Mn3O4
analysis results show that, in addition to the main peak or Mn2O3.40 These results point to the presence of a high
characteristic of metallic Au0, the XPS spectra also contain the concentration of oxygen vacancies on the surface of Au/MnO2
4f7/2 signals from Auδ+ ions.36-38 As shown in Table 1, the catalysts. After the oxidation reaction, the 646 eV shoulder
fraction of Auδ+ species, probably at the gold-support interface, feature becomes even more distinct along with the disappearance
in the Au/MnO2-R catalyst (18.5%) is evidently larger than that of the satellite feature near 663 eV. This may imply that slight
of the Au/MnO2-C catalyst (11.8%). Furthermore, it is worth- reduction of the MnO2 support had occurred during reaction.
MnO2 Nanorod Supported Au Nanoparticle Catalysts J. Phys. Chem. C, Vol. 112, No. 17, 2008 6985

TABLE 1: XPS Results of Pure MnO2 and Supported Gold Catalysts

BE of Au 4f7/2 fraction of Au Au/Mn OT/Mn OL/Mn
catalyst (eV) species (%) molar ratio molar ratioa molar ratiob
MnO2-C 2.4 1.8
MnO2-R 2.3 1.7
Au/MnO2-C 84.0 88.2 0.24 2.6 1.7
85.9 11.8
Au/MnO2-R 84.1 81.5 0.26 4.2 1.4
86.2 18.5
Au/MnO2-Rc 83.9 84.4 0.22 7.7 1.1
86.0 15.6
a
OT, total surface oxygen. b OL, surface lattice oxygen. c Used catalyst, after reaction for 8 h in the solvent-free oxidation of benzyl alcohol.

Similar observations have been reported in an XPS investigation TABLE 2: Comparison of the Catalytic Performance of
Various Gold Catalysts in the Solvent-free Oxidation of
of Au/MnOx catalysts prepared by coprecipitation.41 Benzyl Alcohol to Benzaldehydea
The O 1s XPS spectra obtained from MnO2 and the supported Au loading substrate/Au conversion selectivity yield
gold catalysts are depicted in Figure 6. The O 1s main peak at catalyst (wt %) ratio (%) (%) (%)
ca. 528.7 eV is attributed to the lattice oxygen bonded to Mn MnO2-R 1.9 99.2 1.9
atoms.41-43 In addition, distinct shoulders are visible on the high MnO2-C 1.1 99.5 1.1
BE sides of the main peaks, which were assigned to a mixture Au/MnO2-R 5.0 3940 40.7 98.6 40.1
of hydroxyl groups and adsorbed water on the surface of Au/MnO2-C 5.0 3940 13.6 97.8 13.3
Au/Fe2O3b 4.5 3945 9.8 99.3 9.7
catalysts.40,41,43 As shown in Figure 6, with the introduction of Au/TiO2c 1.5 3950 30.5 99.6 30.2
gold, the intensities of the shoulder features at higher binding Au/Cd 0.8 3955 20.6 99.5 20.5
energies significantly increased, pointing to the enrichment of a
Reaction conditions: 200 mg of catalyst, 200 mmol of benzyl
hydroxyl groups and accumulation of adsorbed water on the
alcohol, 0.3 MPa of O2, 120 °C, reaction for 5 h. b Catalyst amount:
surface induced by the gold-support interaction. Moreover, the 222 mg. c Catalyst amount: 665 mg. d Catalyst amount: 1245 mg.
shoulder peak in the O 1s XPS spectrum of Au/MnO2-R catalyst
is apparently stronger than that of Au/MnO2-C, indicating higher
tendency to form hydroxyl groups or adsorb water molecules copy that the ester was directly formed via the hemiacetal
on the surface of Au/MnO2-R catalyst. intermediate.17 However, it cannot be excluded that that the
The surface compositions in terms of Au/Mn and O/Mn molar instant reaction between benzoic acid (the overoxidized product,
ratios are summarized in Table 1. It is clear that the molar ratio which was not detected in our experiments) and the reactant
of the total amount of surface oxygen (OT) to Mn is significantly (benzyl alcohol) may also contribute to some extent toward the
higher than the stoichiometric value in the MnO2 materials. ester formation, as suggested by Choudhary et al.44
Moreover, a pronounced increase in the OT/Mn ratio is identified The catalytic results obtained in the solvent-free liquid-phase
for the Au-containing samples. All these results further cor- oxidation of benzyl alcohol to benzaldehyde by molecular
roborate the presence of a higher concentration of oxygen- oxygen at 120 °C over various catalysts are presented in Table
containing adsorbates, most likely hydroxyl groups or water, 2. Control experiments using the parent MnO2 supports (sub-
on the surface of Au/MnO2 catalysts, especially for sample Au/ strate to MnO2 ∼100) in the absence of gold demonstrate that
MnO2-R. The peak deconvolution of the O 1s lines gives more only less than 2% of the benzyl alcohol can be oxidized to
information about the relative amount of different surface benzaldehyde within 5 h at 120 °C, confirming the stoichio-
oxygen species. As shown in Table 1, the molar ratios of surface metric oxidant nature of the MnO2 materials and the essential
lattice oxygen (OL, 528.7 eV) to Mn for all samples are smaller role of Au for catalytic alcohol oxidation.45-47 Apparently,
than the stoichiometric value in MnO2, inferring the presence results given in Table 2 show that the surface structural
of oxygen vacancies on the surface of the catalysts. Notably, properties of the MnO2 support have a significant influence on
the OL/Mn ratio of MnO2-R decreases from 1.7 to 1.4 after the the activity of Au/MnO2 catalysts. The conversion of benzyl
introduction of gold. In contrast, little discernible change of the alcohol after reaction for 5 h over Au/MnO2-R was 40.5%, about
OL/Mn ratio is observed for Au/MnO2-C catalyst. In addition, 3 times higher than that on Au/MnO2-C, i.e., 13.6%, whereas
the OT/Mn ratio of the spent Au/MnO2-R catalyst is much higher the selectivity to benzaldehyde was always higher than 98%.
than that of the fresh catalyst. This may result from the Moreover, as shown in Table 2, the catalytic activity of Au/
adsorption of reactant or other oxygen-containing organic MnO2-R catalyst is also much higher than the activities of Au/
molecules on the surface of the catalyst, while the lower OL/ TiO2, Au/Fe2O3, and Au/C reference gold catalysts under similar
Mn ratio is consistent with the partial reduction of MnO2 support reaction conditions. This demonstrates that the supports play
as indicated above. an essential role for gold catalysts in the alcohol oxidation
3.3. Solvent-free Aerobic Oxidation of Alcohols. For the reaction due to synergetic effects between Au and carriers.
MnO2 supported gold catalysts, our initial studies focused on The catalytic performance of the Au/MnO2-R catalyst was
the oxidation of benzyl alcohol because this reaction is often also compared with other reported catalysts in the solvent-free
employed as a model reaction for alcohol oxidation.18-20 aerobic oxidation of 1-phenylethanol. It should be noted that
Furthermore, this molecule was chosen because of its relatively the experimental conditions in the literatures concerning the
high reactivity and because the main product is a nonenoliable selective oxidation of alcohols may deviate from each other.
aldehyde thus reducing the number of possible side products.21 Nevertheless, a rough comparison is still feasible in the present
In all cases, benzaldehyde was obtained as the major product, study. As shown in Table 3, the Au/MnO2-R catalyst shows
and only small amounts of other product, viz., benzyl benzoate, higher TOF than the literature results21,57,58 in the solvent-free
were observed. Abad et al. have shown by H1 NMR spectros- oxidation of 1-phenylethanol, except that for the Au-Pd/TiO2
6986 J. Phys. Chem. C, Vol. 112, No. 17, 2008 Wang et al.

TABLE 3: Comparison of the Catalytic Performance of Au/MnO2-R Catalyst with That of Other Catalysts Employed in the
Aerobic Oxidation of 1-Phenylethanol to Acetophenone
1-phenylethanol selectivity PO2 temperature
catalyst TOF (h-1) (%) (atm) (°C) solvent ref
Au/MnO2-R 15 760 a
>99 1 160 none this work
Au/Cu5Mg1Al2Ox 11 748 >99 1 160 none 21
Au-Pd/TiO2 269 000 1 160 none 19
Au/CeO2 12 500 >99 1 160 none 57
Pd/hydroxyapatite 9 800 1 160 none 58
a
Reaction conditions: 1 atm of O2 (60 mL‚min-1), 10 mg of catalyst, 200 mmol of substrate. The turnover frequency (TOF) for the oxidation
of 1-phenylethanol was calculated by dividing the initial reaction rate during the first hour of the reaction to the number of gold atoms.

catalyst,19 demonstrating that the present catalytic system can we infer that the β-MnO2 nanorods bear more reactive surfaces
serve as a promising catalyst for the selective oxidation of as compared with the commercial MnO2 particles.
alcohols. More importantly, it should be noted that introducing Au to
3.4. Discussion. Manganese dioxide is traditionally used as MnO2 produces a strong Au-MnO2 interaction. The different
a versatile stoichiometric oxidation reagent for the direct surface structural features clearly determine the strength of
oxidation of alcohols to aldehydes or ketones.48 However, the metal-support interaction and thus the catalytic activity. TPR
low oxidation efficiency and pollution issues associated with results indicate that the introduction of gold results in much
the byproducts from both economical and ecological points of lower reduction temperature for Au/MnO2-R than that of Au/
view make it less attractive in the oxidation procedures. In the MnO2-C. Stobbe et al. have suggested that the reducibility of
present study, we have shown experimentally that by incorpora- manganese oxide is determined by its crystallinity or defect
tion of gold nanoparticles, the stoichiometric oxidation sites on concentration rather than the oxidation state.32 This is consistent
MnO2 can be converted into catalytic sites for the selective with the findings derived from the analysis of O 1s XPS spectra
oxidation of alcohols. In particular, by using MnO2 nanorods (Table 1) which reveal the presence of more oxygen defects on
as support materials exceptionally high activity in the liquid- the surface of MnO2 nanorods supported gold catalyst. It is likely
phase aerobic oxidation of alcohols has been achieved over the that the introduction of Au to MnO2 may reduce the barrier for
Au/MnO2-R catalyst. Note that the commercial MnO2 has a the creation of an oxygen vacancy that is most likely adjacent
relatively lower BET surface area (5 m2‚g-1) compared with to the Au sites.56 Alternatively, the MnO2 support materials are
MnO2 nanorods (10 m2‚g-1); however, this difference does not activated by the incorporation of gold, to a larger extent for
seem to impose considerable effects on the resulting gold Au/MnO2-R catalyst. The XPS results also reveal that higher
loading, phase structure, and size distribution of gold nanopar- fraction of oxidized Au species are present on the MnO2 nanorod
ticles for the two catalysts, as indicated by the combined supports before and after the reaction, compared with the gold
elemental analysis and TEM results. It should also be mentioned catalyst supported on commercial MnO2. According to the
that it is unlikely the surface contamination of the MnO2 support reaction mechanism for aerobic oxidation of alcohols proposed
due to different preparation history may contribute to some by Abad et al.,57 the activation of oxygen molecules takes place
extent to the superior performance of the present Au/MnO2-R at the oxygen defect sites on the surface of the support, while
sample, since no sulfur or other remnants could be detected by the positive gold ions play an important role in the rate-
XPS measurements on the surface of the MnO2 nanorod determining step involving the hydride shift from the alcohol
materials. Consequently, the key role of the surface structural to gold. Therefore, the presence of more oxygen vacancies and
properties of the support in determining the catalytic perfor- oxidized gold species is considered to be the key point in relation
mances of the gold catalysts could be established. to the excellent activity of the Au/MnO2-R catalyst for the
The performance of catalysts depends strongly on their aerobic oxidation of alcohols, although many details of the
surface structures, which have a direct relationship with the mechanism still remain unraveled.
crystal planes or crystal phases.48-50 In this sense, rational design
of catalysts with desirable properties, i.e., high activity and 4. Conclusions
selectivity, should involve the fabrication of well-defined
reactive crystal planes.51 In this study, both XRD and TEM In this study, gold nanoparticles were deposited on two MnO2
results reveal a predominant exposure of the {110} plane in materials with different morphologies, i.e., particles and well-
the β-MnO2 nanorod material, whereas a significant fraction of defined nanorods, by the HDP method. It is shown that the two
more closely compact facets are expected for the commercial types of gold catalysts have similar gold loading and gold
β-MnO2 particles.52 We propose that this great diversity in the particle size distributions. TPR results revealed that the presence
surface morphology of the support accounts for the different of gold significantly promotes the reduction of the MnO2
redox properties and catalytic behaviors of the two gold nanorods. XPS revealed both reduced and oxidized Au species
catalysts. The β-MnO2 (pyrolusite) has a rutile-type structure on the MnO2 nanorod supports before and after the reaction.
with tetragonal symmetry, with the oxygen ions forming a The catalysts were tested in the liquid-phase aerobic oxidation
distorted hexagonal close packing array.53 Previously, on the of benzyl alcohol under solvent-free conditions. The catalytic
basis of a thorough theoretical calculation of the electrostatic activity tests show that the gold supported on MnO2 nanorods
potential, Woning and van Santen predict that the rutile (110) is much more active than that on the commercial MnO2 particles.
surface can be reduced easier than the anatase (101) surface.54 The far superior catalytic performance of gold supported on
Moreover, a recent theoretical study of the oxygen adsorption MnO2 nanorods is attributed to the collaborative effects resulting
on three planes (001), (110), and (111) of β-MnO2 has shown from beneficial interaction between the electronically modified
that the (110) plane has the highest reactivity toward the Au nanoparticles and the well-defined reactive surface of
activation of oxygen molecules.55 On the basis of these results, nanorods.
MnO2 Nanorod Supported Au Nanoparticle Catalysts J. Phys. Chem. C, Vol. 112, No. 17, 2008 6987

Acknowledgment. This work was supported by the NSF of (30) Bond, G. C.; Louis, C.; Thompson, D. T. Catalysis by Gold;
China (20421303, 20473021, and 20633030), the State Key Imperial College Press: London, 2006; p 113.
(31) Xu, R.; Wang, X.; Wang, D.; Zhou, K.; Li, Y. J. Catal. 2006, 237,
Basic Research Program of PRC (2003CB615807), the National 426.
High Technology Research and Development Program of China (32) Stobbe, E. R.; de Boer, B. A.; Geus, J. W. Catal. Today 1999, 47,
(2006AA03Z336), the Shanghai Science and Technology Com- 161.
(33) Trevino, H.; Lei, G. D.; Sachtler, W. M. H. J. Catal. 1995, 154,
mittee (07QH14003), and the Shanghai Education Committee 245.
(06SG03). (34) Epling, W. S.; Hoflund, G. B.; Weaver, J. F.; Tsubota, S.; Haruta,
M. J. Phys. Chem. 1996, 100, 9929.
References and Notes (35) Minicò, S.; Scire, S.; Crisafulli, C.; Galvagno, S. Appl. Catal., B
2001, 34, 277.
(1) Haruta, M. Catal. Today 1997, 36, 153. (36) Soares, J. M. C.; Morrall, P.; Crossley, A.; Harris, P.; Bowker, M.
(2) Bond, G. C.; Thomson, D. T. Catal. ReV. Sci. Eng. 1999, 41, 319. J. Catal. 2003, 219, 17.
(3) Hashmi, A. S. K.; Hutchings, G. J. Angew. Chem., Int. Ed. 2007, (37) Fan, L.; Ichikuni, N.; Shimazu, S.; Uematsu, T. Appl. Catal., A
45, 7896. 2003, 246, 87.
(4) Min, B. K; Friend, C. M. Chem. ReV. 2007, 107, 2709. (38) Hutchings, G. J.; Hall, M. S.; Carley, A. F.; Landon, P.; Solsona,
(5) Grunwaldt, J. D.; Kiener, C.; Wögerbauer, C.; Baiker, A. J. Catal. B. E.; Kiely, C. J.; Herzing, A.; Makkee, M.; Moulijn, J. A.; Overweg, A.;
1999, 181, 223. Fierro-Gonzalez, J. C.; Guzman, J.; Gates, B. C. J. Catal. 2006, 242, 71.
(6) Grunwaldt, J. D.; Baiker, A. J. Phys. Chem. B 1999, 103, 1002. (39) Kapteijn, F.; van Langeveld, A. D.; Moulijn, J. A.; Andreini, A.;
(7) Sanchez-Castillo, M. A.; Couto, C.; Kim, W. B.; Dumesic, J. A. Vuurman, M. A.; Turek, A. M.; Jehng, J. M.; Wachs, I. E. J. Catal. 1994,
Angew. Chem., Int. Ed. 2004, 43, 1140. 150, 94.
(8) Schubert, M. M.; Hackenberg, S.; van Veen, A. C.; Muhler, M.; (40) Srinivasan, B.; Gardner, S. D. Surf. Interface Anal. 1998, 26, 1035.
Plzak, V.; Behm, R. J. J. Catal. 2001, 113, 197. (41) Gardner, S. D.; Hoflund, G. B.; Davidson, M. R. Langmuir 1991,
(9) Baumweda, G. R.; Tsubota, S.; Nakamura, T.; Haruta, M. Catal. 7, 2140.
Lett. 1997, 44, 83. (42) Lee, S. J.; Gavriilidis, A.; Pankhurst, Q. A.; Kyek, A.; Wagner, F.
(10) Minico, S.; Scire, S.; Crisafulli, C.; Visco, A. M.; Galvagno, S. E.; Wong, P. C. L.; Yeung, K. L J. Catal. 2001, 200, 298.
Catal. Lett. 1997, 47, 273. (43) Peña, D. A.; Uphade, B. S.; Smirniotis, P. G. J. Catal. 2004, 221,
(11) Molina, L. M.; Hammer, B. Phys. ReV. Lett. 2003, 90, 206102. 421.
(12) Sanchez, A.; Abbet, S.; Heiz, U.; Schneider, W. D.; Hakkinen, H.; (44) Choudhary, V. R.; Jha, R.; Jana, P. Green Chem. 2007, 9, 267.
Barnett, R. N.; Landman, U. J. Phys. Chem. A 1999, 103, 9573.
(45) Dimitratos, N.; Villa, A.; Wang, D.; Porta, F.; Su, D.; Prati, L. J.
(13) Lemire, C.; Meyer, R.; Shaikhutdinov, S.; Freund, H. J. Angew.
Catal. 2006, 244, 113.
Chem., Int. Ed. 2003, 43, 118.
(46) Conventional active manganese oxide has been most commonly
(14) Lopez, N.; Nørskov, J. K. J. Am. Chem. Soc. 2002, 124, 11262.
used as the stoichiometric metal oxidants for the direct oxidation of alcohols
(15) Mills, G.; Gordon, M. S.; Metiu, H. Chem. Phys. Lett. 2002, 359,
to aldehydes or ketones. The reactivity of active manganese oxide is
493.
dependent on preparation methods, compositions, and structure. Complicated
(16) Porta, F.; Prati, L. J. Catal. 2004, 224, 397.
preparation methods are often necessary, and the use of freshly made active
(17) Abad, A.; Concepcion, P.; Corma, A.; Garcia, H. Angew. Chem.,
manganese oxide is required.
Int. Ed. 2005, 44, 4066.
(18) Enache, D. I.; Knight, D. W.; Hutchings, G. J. Catal. Lett. 2005, (47) Au and Mn leaching were negligible during the recycling experi-
103, 43. ments. Both the Au and Mn concentration in the filtrate is less than 5 ppb
(19) Enache, D. I.; Edwards, J. K.; Landon, P.; Solsona-Espriu, B.; as found by ICP.
Carley, A. F.; Herzing, A. A.; Watanabe, M.; Kiely, C. J.; Knight, D. W.; (48) Son, Y. C.; Makwana, V. D.; Howell, A. R.; Suib, S. L. Angew.
Hutchings, G. J. Science 2006, 311, 362. Chem., Int. Ed. 2001, 40, 4280.
(20) Choudhary, V. R.; Dhar, A.; Jana, P.; Jha, R.; Uphade, B. S. Green (49) Zhou, K.; Wang, X.; Sun, X.; Peng, Q.; Li, Y. J. Catal. 2005, 229,
Chem. 2005, 7, 768. 206.
(21) Haider, P.; Baiker, A. J. Catal. 2007, 248, 175. (50) Choudary, B. M.; Mulukutla, R. S.; Klabunde, K. J. J. Am. Chem.
(22) Singoredjo, L.; Korver, R.; Kapteijn, F.; Moulijn, J. Appl. Catal., Soc. 2003, 125, 2020.
B 1992, 1, 297. (51) Schlögl, R.; Abd Hamid, S. B. Angew. Chem., Int. Ed. 2004, 43,
(23) Morales, F.; de Smit, E.; de Groot, F. M. F.; Visser, T.; 1628.
Weckhuysen, B. M. J. Catal. 2007, 246, 91. (52) Rotzinger, F. P.; Kesselman-Truttmann, J. M.; Hug, S. J.; Shklover,
(24) Xia, G. G.; Yin, Y. G.; Willis, W. S.; Wang, J. Y.; Suib, S. L. J. V.; Gra1tzel, M. J. Phys. Chem. B 2004, 108, 5004.
Catal. 1999, 185, 91. (53) Thackeray, M. M. Prog. Solid State Chem. 1997, 25, 1.
(25) Radhakrishnan, R.; Oyama, S. T. J. Phys. Chem. B 2001, 105, 4245. (54) Woning, J.; van Santen, R. A. Chem. Phys. Lett. 1983, 101, 541.
(26) Han, Y. F.; Chen, F.; Zhong, Z.; Ramesh, K.; Chen, L.; Widjaja, (55) Li, L.; Wei, Z. D.; Li, L. L.; Sun, C. X. Acta Chim. Sin. 2006, 64,
E. J. Phys. Chem. B 2006, 110, 24450. 287.
(27) Sanchez, R. M. T.; Ueda, A.; Tanaka, K.; Haruta, M. J. Catal. (56) Widmann, D.; Leppelt, R.; Behm, R. J. J. Catal. 2007, 251, 437.
1997, 168, 125. (57) Abad, A.; Almela, C.; Corma, A.; Garcı́a, H. Tetrahedron 2006,
(28) Wang, X.; Li, Y. D. Chem. Eur. J. 2003, 9, 300. 62, 6666.
(29) Zanella, R.; Giorgio, S.; Henry, C. R.; Louis, C. J. Phys. Chem. B (58) Mori, K.; Yamaguchi, K.; Hara, T.; Mizugaki, T.; Ebitani, K.;
2002, 106, 7634. Kaneda, K. J. Am. Chem. Soc. 2002, 124, 11572.