Highly efficient and stable Au/Mn2O3 catalyst for oxidative cyclization of 1,4-butanediol to γ-butyrolactone

Article abstract of DOI:10.1016/j.apcata.2013.03.020

In this work, gold nanoparticles deposited on manganese sesquioxide (Mn₂O₃) showed high activity in the oxidative lactonization of 1,4-butanediol to γ-butyrolactone with air as the oxidant. On the contrary, nano-sized gold supported on manganese dioxide (MnO₂) and manganese oxide (MnO) presented much lower activities. XRD, TEM and XPS characterizations revealed that the superior catalytic performance of Au/Mn₂O₃ could be ascribed to the highest dispersion of gold particles and unique intrinsic properties of Mn ₂O₃, which possesses the appropriate point of zero charge for deposition-precipitation method, the most quantity of oxygen vacancies and the appropriate gold-support interaction. Au/Mn₂O₃ exhibited selectivity of 100%, which could facilitate the purification of target product from the reaction mixture. Furthermore, the as-prepared Au/Mn ₂O₃ catalyst could be recovered by simple filtration, and used repetitively or preserved for a long period without apparent loss of activity.

Full text of DOI:10.1016/j.apcata.2013.03.020

Applied Catalysis A: General 458 (2013) 63–70
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Highly efficient and stable Au/Mn2O3 catalyst for oxidative cyclization
of 1,4-butanediol to ␥-butyrolactone
Xian Lia, Yuanyuan Cuia, Xinli Yangb, Wei-Lin Daia,∗, Kangnian Fana
a Department of Chemistry and Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials, Fudan University, Shanghai 200433, PR China
b College of Chemistry & Chemical Engineering, Henan University of Technology, Zhengzhou 450001, Henan, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
In this work, gold nanoparticles deposited on manganese sesquioxide (Mn2O3) showed high activity in
the oxidative lactonization of 1,4-butanediol to ␥-butyrolactone with air as the oxidant. On the con-
trary, nano-sized gold supported on manganese dioxide (MnO2) and manganese oxide (MnO) presented
much lower activities. XRD, TEM and XPS characterizations revealed that the superior catalytic perfor-
mance of Au/Mn2O3 could be ascribed to the highest dispersion of gold particles and unique intrinsic
properties of Mn2O3, which possesses the appropriate point of zero charge for deposition–precipitation
method, the most quantity of oxygen vacancies and the appropriate gold–support interaction. Au/Mn2O3
exhibited selectivity of 100%, which could facilitate the purification of target product from the reaction
mixture. Furthermore, the as-prepared Au/Mn2O3 catalyst could be recovered by simple filtration, and
used repetitively or preserved for a long period without apparent loss of activity.
Received 2 February 2013
Received in revised form 11 March 2013
Accepted 16 March 2013
Available online xxx
Keywords:
Au/Mn2O3 catalyst
Manganese oxides supports
1,4-Butanediol
Oxidative lactionzation
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
activities. Conventionally, support plays a crucial role in the per-
formance of active gold catalysts apart from gold particle size
The selective synthesis of lactones has attracted much inter-
est to both academic and industrial chemists. Among the lactones,
␥-butyrolactone is widely used as a significant solvent, extraction
agent and intermediate in agriculture, petroleum industry, phar-
maceutics and fibers [1–4]. Moreover, the aerobic oxidation of
1,4-butanediol to ␥-butyrolactone is considered as one of the most
promising pathway for industrially acceptable process, owing to
the absence of waste or pollutant.
[15,16], and reducible support is supposed to provide oxygen
adsorption and activation sites, which can facilitate the oxidation
process [16,17], thus it is believed that gold supported on reducible
oxides are more active and stable than on inert oxides [17,18]. As
expected, the outstanding catalytic performance for the oxidation
of 1,4-butanediol to ␥-butyrolactone has been obtained from finely
dispersed Au on reducible oxides, such as TiO2 [19], Fe2O3 [20,21],
and Fe–Al–O composite [22].
Traditionally, the lactonization of 1,4-butanediol was carried
out in metal reagents [5,6]. Then several homogeneous [7,8] or het-
erogeneous catalysts [9,10] were used in the presence of organic
co-oxidants such as ketones and alkenes, but these chemicals are
expensive and/or toxic. Zhao and Hartwig reported that ruthe-
nium complexes catalyzed this reaction with high activity and
high turnover numbers even without solvents or hydrogen accep-
tors [11]. In recent years, supported gold catalysts, including gold
deposited on hydrotalcite [12], gold particles supported on ␥-
AlOOH and ␥-Al2O3 [13], and gold supported on anion-exchange
resin Dowex Marathon MSA [14], have attracted tremendous atten-
tion to the synthesis of lactones from diols with admirable catalytic
Manganese oxides (MnOx), as reducible oxides, have long been
used as highly active, durable and low cost catalysts for aero-
bic oxidation of organic compounds [23–26]. For example, MnO2
exhibited remarkably high catalytic activity and selectivity for
the ammoxidation of alcohols [27]. When cooperated with gold
nanoparticles, the catalytic activities were enhanced markedly for
the oxidation of CO [28–31] or alcohols [27,32,33]. However, to the
best of our knowledge, there has been no reports on the lactoniza-
tion of 1,4-butanediol catalyzed by gold supported on manganese
oxides. Herein, for the first time, we report the synthesis and
catalytic application of Au nanoparticle catalysts supported on
different manganese oxides in the oxidation of 1,4-butanediol to
␥-butyrolactone with air as the sole oxidant. Additionally, the gold-
supported catalysts are generally thought to be unstable and easy
to deactivate, especially after a long period preservation or reused
for many cycles because of the high activity of the nano-sized gold
particles [34,35]. In this work, we also investigated the stability of
the as-prepared Au/Mn2O3 catalyst.
∗
Corresponding author. Tel.: +86 21 55664678; fax: +86 21 55665701.
E-mail address: wldai@fudan.edu.cn (W.-L. Dai).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.03.020

64
X. Li et al. / Applied Catalysis A: General 458 (2013) 63–70
2. Experimental
2.1. Catalyst preparation
stirring was set at 800 rpm simultaneously. The reaction mixture
was sampled at regular time intervals and analyzed by gas chro-
matography (GC9560 (HuaAi Co. Ltd.) with a FID detector and HP5
capillary column) to determine the conversion and selectivity. Gas
chromatography–mass spectrometry (GC 8000 TOF-MS voyager)
was utilized to detect the product distribution.
Recycling experiments over Au/Mn2O3 catalyst were conducted
under identical reaction conditions as the above. The used catalyst
was collected by filtration, and recovered by being washed with
ethanol 3 times and drying overnight at 373 K in air.
Three kinds of different manganese oxides (MnO2, Mn2O3
and MnO) were prepared according to the following prepa-
ration procedures as described elsewhere [25,26]. MnO2 was
prepared via hydrothermal method. Equal amount of MnSO4·H2O
and (NH4)2S2O8 (both 50 ml, ca. 0.04 M) was mixed in a 250 ml
round bottomed flask. The as-obtained mixture was hydrother-
mally processed at 393 K for 12 h, followed by filtration, washed
with deionized water for 4 times and dried at 373 K for 12 h. Mn2O3
was obtained by calcination of MnCO3 powder at 773 K for 4 h in
static air and then cooled to room temperature. MnO was obtained
by decomposition of MnCO3 sample under pure N2 atmosphere at
673 K for 4 h, and then cooled to room temperature in flowing N2
stream.
3. Results and discussion
3.1. Catalyst characterization
3.1.1. BET and ICP-AES
In Table 1, the data show that MnO2 has the highest BET sur-
face area (52 m2/g), while Mn2O3 is 35 m2/g and MnO is the lowest
(10 m2/g). Because Mn2O3 and MnO were prepared from the same
precursor, the difference in BET values is attributed to the effect of
different treating atmosphere. Support effect is always significant
in gold catalysts, so we tried to synthesize manganese oxides with
similar morphology for comparing the effect of the support crystal
phase reasonably. After the deposition of gold, the BET surface area
decreases moderately, and the values turn to 8.0, 39 and 31 m2/g for
Au/MnO, Au/MnO2 and Au/Mn2O3 with no sharp changes for the
used one. In addition, the real gold loadings of the three catalysts
are all near the nominal amount (5%) as determined by ICP-AES
method with the error range of 0.5%.
Gold nanoparticles deposited on the as-synthesized manganese
oxides with a nominal Au loading of 5 wt.% were carried out by
deposition–precipitation (DP) method using urea as precipitation
agent [36–38]. In a typical procedure, 15 ml of HAuCl4 aqueous
solution (2.43 × 10−3 M) was mixed with 60 ml of deionized water
with continuous stirring, followed by the addition of 2.5 g of urea
and 1.4 g of MnOx. The mixture was heated to 363 K gradually and
maintained for 4 h. After this step, the pH values of solutions were
about 8.0, 8.0 and 8.5 for Au/MnO, Au/MnO2 and Au/Mn2O3, respec-
tively. The solid mass was filtered, washed with deionized water for
4 times, and dried in air overnight at 373 K. Finally, the Au/MnO2
and Au/Mn2O3 samples were calcined at 573 K for 4 h in air, while
the Au/MnO sample was calcined at 573 K for 4 h in flowing N2 to
avoid the oxidation of MnO by air.
3.1.2. XRD patterns
2.2. Catalyst characterization
Fig. 1 shows XRD patterns of Au/MnOx. It can be found that all
of the supports are in pure crystal phases, and the addition of gold
causes no structural changes for the support. In fresh (Fig. 1(a)) and
used Au/Mn2O3 (Fig. 1(b)), the XRD patterns exhibit well-defined
diffraction characteristics of Mn2O3 (JCPDS 41-1442), and there is
no distinct gold reflection owing to the high dispersion of Au parti-
cles. For Au/MnO (Fig. 1(c)), the reflections are mainly attributed to
MnO (JCPDS 06-0592) with a small amount of Mn3O4 (marked with
*). However, the presence of small amount of Mn3O4 has little effect
on the properties of MnO according to the results in Ref. [25], and
the following activity tests can also confirm this deducing. Although
the broad and weak diffraction lines in Fig. 1(d) demonstrate the
bad crystallization, Au/MnO2 can still be indexed to MnO2 (JCPDS
012-0713). Moreover, the sharp diffraction peaks of Au, indicat-
ing the presence of large gold particles, are discerned on Au/MnO
and Au/MnO2 samples. Thus, according to the Scherrer equation,
the average size of gold particles calculated from XRD patterns are
23.8 and 8.5 nm for Au/MnO and Au/MnO2.
The specific surface area of samples was determined by nitro-
gen adsorption at 77 K (Micromeritics Tristar ASAP 3000) using
the Brunauer–Emmett–Teller (BET) method. The gold loadings
were determined by the inductively coupled plasma method (ICP,
thermo E. IRIS). The XRD patterns were recorded on a Bruker D8
Advance diffractometer with Cu K␣ radiation (ꢀ = 0.154 nm), oper-
ated at 40 mA and 40 kV. The TEM micrographs were obtained on
a JOEL JEM 2010 transmission electron microscope. The average
size of the Au particles and relative distributions were estimated
by counting more than 300 Au particles. The turnover frequency
(TOF) was calculated by the amount of surface gold atoms and the
mole diol reacted in the initial 2 h on per mole surface gold. The
gold particles were assumed as sphere shape with a single gold
atom diameter of 0.1442 nm. According to the gold atom diam-
eter and the average gold particle size from TEM, the amount of
surface gold atoms could be estimated, and the TOF value could be
obtained [19–21]. The turnover number (TON) can also be obtained
by the mole diol reacted per mole gold during the reaction. The
XPS spectra were acquired under ultra high vacuum (<10−6 Pa) at
a pass energy of 93.90 eV on a Perkin-Elmer PHI 5000C ESCA sys-
tem equipped with a dual X-ray source using an Mg K␣ (1253.6 eV)
anode and a hemispheric energy analyzer. All binding energies
were calibrated by using contaminant carbon (C 1s = 284.6 eV) as
a reference.
Table 1
The physico-chemical properties of Au/MnOx catalysts.
b
c
Sample
SBET (m2/g)
Aua (wt.%)
dAu (nm)
DAu (nm)
MnO
MnO2
Mn2O3
Au/MnO
Au/MnO2
Au/Mn2O3
Used Au/Mn2O3
10
52
35
8
39
31
33
–
–
–
5.5
5.3
5.0
5.3
–
–
–
–
–
–
23.8
8.5
–
3.1/26.2d
3.0/11.5d
2.6
2.3. Activity test
e
2.7
–
Catalyst tests were performed using a stainless steel autoclave
equipped with a magnetic stirrer at 393 K. 1.4 g of 1,4-butanediol
was dissolved in 20 ml of tributyl phosphate (TBP), followed by
the addition of 0.3 g of Au/MnOx catalyst. Then the autoclave
was sealed and filled with 1.25 MPa air. The rate of magnetic
a
Determined by ICP-AES analysis with the error range of 0.5%.
Determined by TEM.
Calculated by Scherrer equation from the XRD results.
The average size of the small and the large gold particles.
After being used for 4 times.
b
c
d
e

X. Li et al. / Applied Catalysis A: General 458 (2013) 63–70
65
After the deconvolution of the Au 4f peaks, the binding energy
difference in the two chemical states is 1.3 eV in Au/MnO and
Au/Mn2O3 and an important component of Au is Auı−, which
means the formation of lots of anionic Au species with more than
one charge (ı > 1) if the charge effect was corrected by 5.4 eV.
This finding is quite different from those reported in Au/MnOx
previously [30,31,45]. We think that the strange chemical shift
of Au 4f may be an overcorrection of Au peaks by contaminant
carbon in Au/MnO and Au/Mn2O3. Similar result has been also
observed in the same XPS system on Ag/SiO2 catalyst [43]. The
two series of catalysts both have metals on the metal oxide sup-
port while the metals show different particle size or chemical
states. Because the conductivity of metals is much better than the
metal oxide supports, the charge effect many be different on the
metals and the support. Moreover, there are always metallic and
cationic Au species in the literatures [19–21,30–33] whose prepa-
ration methods or catalysts are similar to the present work. Thus,
it is reasonable to believe that there are metallic and positively
charged gold in our Au/MnO and Au/Mn2O3 catalysts. Detailed
binding energies after the re-correction are listed in Table 2 in
the parentheses. Under the same condition, due to the semicon-
ductor character of MnO2, the charge effect is very small and can
be neglected. As a result, no anionic gold species was found on
Au/MnO2.
Combined with the above discussion, the support phase influ-
ences the electronic nature of gold particles and the proper content
of Auı+ suggests an appropriate gold–support interaction [20],
and the amount of Auı+ are 20% and 67% on the surface of
Au/Mn2O3 and Au/MnO (after the reaction, the amount of Auı+
increased from 20% to 57% in the used Au/Mn2O3 (Fig. 3(d))),
indicating the proper metal–support interaction in Au/Mn2O3.
Meanwhile, the previous investigations have found that Mn2O3
can promote the catalytic activity of gold catalyst better than
other manganese oxides due to the intimate metal–support inter-
action, thus we believe the cationic gold is beneficial to the high
activity, which is in well agreement with those reported earlier
[28,44,45].
Mn 2p XPS spectra for various catalysts are shown in Fig. 4(a).
The Au/MnO2 catalyst exhibits an asymmetric Mn 2p3/2 peak
located at 642.2 eV, which is in agreement with those reported in
literature for MnO2, i.e., 642.0 0.2 eV [46]. It is found that the bind-
ing energies of Mn 2p3/2 shift to lower values in following order:
Au/MnO2 > Au/Mn2O3 > Au/MnO, and they are all close to the val-
ues of pure supports. Nevertheless, the variation of XPS binding
energies of Mn 2p alone is too small to precisely evaluate the Mn
valence of MnOx [47]. Therefore, the corresponding Mn 3s spec-
tra were measured simultaneously (Mn 3s ꢁE) to estimate the Mn
chemical states. Yet the Au 4f doublet overlaps with the Mn 3s mul-
tiplet splitting peaks, so we constrained the relative position and
intensity of Mn 3s components by assuming for the gold-containing
samples the same Mn 3s/Mn 2p intensity ratio as for the bare sup-
ports [48] and the curve fitting results are listed in Table 2. It is
shown that the values of Mn 3s ꢁE before and after Au dispersion
are nearly invariable, suggesting the Mn chemical states are lit-
tle effected by the deposition of gold particles. Comparing with the
literatures, the values of the Mn 3s ꢁE does not decrease monotoni-
cally with the increase of nominal valence of MnOx, this is attributed
to the difference of preparation methods and test conditions. It is
interesting to find that the Mn 3s ꢁE values of MnO and Mn2O3
both are lower than that in Ref. [47], but they present the same
decreasing trend as that in the reference. Because of the very dif-
ferent synthesis process, the Mn 3s ꢁE value in MnO2 is higher
than that in Ref. [47]. However, the Mn 3s ꢁE value of MnO2 is in
agreement with the results observed by Longo et al. [48], whose
preparation method and crystal phase of MnO2 are similar to the
present work.
(a)
(b)
(c)
(d)
*
*
20
30
40
50
60
70
80
2-theta (degree)
Fig. 1. XRD patterns of (a) Au/Mn2O3, (b) Au/Mn2O3 used for 4 times, (c) Au/MnO
and (d) Au/MnO2.
3.1.3. TEM results
Fig. 2 shows the Au particle-size distributions of the three kinds
of catalysts. In Au/Mn2O3 (Fig. 2(e) and (f)), the images demonstrate
that gold particles are in a narrow distribution and well dispersed
with the average particle size of 2.6 nm. Clearly, the morphology
of Au particles keeps intact after being used for 4 times (Fig. 2(g)
and (h)). In contrast, the Au size distribution is relatively broad in
Au/MnO (Fig. 2(a) and (b)) and Au/MnO2 (Fig. 2(c) and (d)), sug-
gesting a large heterogeneity of gold particles. Meanwhile, the two
catalysts both display bimodal particle size distribution compris-
ing two populations. For Au/MnO2, the average gold sizes are 3.0
and 11.5 nm, while they are 3.1 and 26.2 nm in Au/MnO. The diver-
sity of Au particle-size distribution may be related to the various
phase structures and intrinsic properties of different supports (as
discussed below). In the present work, the preparation method for
the gold-based catalysts was the same (DP) and the results (the
gold particle size and distribution) were the best under identical
conditions.
The previous work has revealed that the applicability of urea
DP method strongly depends on the nature of substrates, and it
is beneficial to the deposition of gold onto oxide supports with
point of zero charge (PZC) close to 6.0 [30,37]. According to the Zeta
potential measurements, the PZC values for MnO, MnO2 and Mn2O3
were estimated to be 7.3, 4.0 and 6.7, which are in agreement
with the findings in literatures [39,40]. Therefore, this result well
explains the presence of small and high dispersed gold nanoparti-
cles obtained on the surface of Mn2O3.
3.1.4. XPS
Surface gold content and electronic state of the catalysts were
characterized by XPS and the results are given in Table 2 and
Figs. 3 and 4. In Au/MnO2 (Fig. 3(a)), the Au 4f peak is quite sym-
metric and narrow, representing that there is only one kind of Au
species (elemental state, 83.8 eV), or the amount of ionic gold is too
small to be distinguished from the background. Broad peaks of Au
4f were observed in Au/MnO (Fig. 3(b)) and Au/Mn2O3 (Fig. 3(c)),
indicating the possible presence of both metallic and ionic gold
species [41,42]. Thus, the original peaks (broken line) of Au/MnO
(Fig. 3(b)) and Au/Mn2O3 (Fig. 3(c)) were deconvolutioned into dif-
ferent peaks (solid line), and the binding energies of Au 4f7/2 peaks
are 82.5 eV and 83.8 eV after the correction of charge effect. It is
strange to find that the charge effect of binding energy in Au/MnO
(5.4 eV, C 1s = 290.0 eV) and Au/Mn2O3 (5.4 eV, C 1s = 290.0 eV) is
much bigger than that in Au/MnO2 (0.6 eV, C 1s = 285.2 eV).

66
X. Li et al. / Applied Catalysis A: General 458 (2013) 63–70
Fig. 2. TEM images and particle size distributions of Au/MnO (a, b), Au/MnO2 (c, d), Au/Mn2O3 (e, f), and Au/Mn2O3 used for 4 times (g, h).
The O 1s spectra of various Au/MnOx are also presented in
Au/MnO catalyst has the highest surface molar ratio of gold to Mn
(Au/Mn, 0.50), which may be a result of the lowest BET of MnO sup-
port. The Au/Mn ratio is 0.48 in Au/Mn2O3, which is much higher
than that in Au/MnO2 (0.27). The relatively lower Au/Mn value in
Au/MnO2 sample is due to the highly acidic nature of MnO2 [30],
and Mn2O3 can facilitate the dispersion of gold on the surface of
Fig. 4(b). The main peak at about 529.0 eV is attributed to the lattice
oxygen bonded to Mn atoms [49]. Simultaneously, distinct shoul-
ders, which can be assigned to the mixture of hydroxyl groups and
adsorbed water on the surface of catalysts [50], are visible on the
higher binding energy side of the main peaks. It is found that the

X. Li et al. / Applied Catalysis A: General 458 (2013) 63–70
67
Table 2
XPS results of Au/MnOx catalysts.
Sample
MnO
BE of Au 4f 7/2 (eV)c
Au species (%)
–
BE of Mn 2p 3/2 (eV)
641.8
Mn 3s ꢁE
Au/Mn
–
OT/Mna
2.08
OL/Mnb
–
5.5
1.01
82.5 (83.8)
83.8 (85.1)
Au0 (33)
Auı+ (67)
Au/MnO
641.7
5.3
0.50
1.99
0.98
MnO2
Au/MnO2
Mn2O3
–
83.8
–
–
642.3
642.2
642.1
5.4
5.4
4.6
–
0.27
–
2.52
2.56
2.43
1.33
1.35
0.95
Au0 (100)
–
82.5 (83.8)
83.8 (85.1)
Au0 (80)
Auı+ (20)
Au/Mn2O3
641.9
641.9
4.6
4.6
0.48
0.48
2.37
2.67
0.93
0.99
82.5 (83.8)
83.8 (85.1)
Au0 (43)
Auı+ (57)
Used Au/Mn2O3
a
OT, total surface oxygen.
OL, surface lattice oxygen.
Data in the parentheses, charge effect was corrected by 4.1 eV.
b
c
support [30]. After being used for 4 times, no variation of Au/Mn
ratio is found in Au/Mn2O3, suggesting this catalyst has superior
stability.
in Au/Mn2O3, Au/MnO, and Au/MnO2, respectively. They are all
less than the stoichiometric values (1.5, 1.0 and 2.0 in Au/Mn2O3,
Au/MnO, and Au/MnO2), inferring the emerging of surface oxy-
gen vacancies which are filled with water or hydroxyl groups.
It is clear that the Au/Mn2O3 catalyst owns the maximum oxy-
gen vacancies because of the minimum OL/Mn ratio (0.93) that
is less than the stoichiometric value (1.5). The surface oxygen
vacancies are considered as a key role in promoting the catalytic
activity by enhancing the electron transfer between the support
and gold particles and can be regenerated under reaction condi-
tions [48,51–54]. After four times reaction over Au/Mn2O3, the
Furthermore, the molar ratios of the total amount of surface
oxygen (OT) to Mn were calculated and these values are much
higher than the stoichiometric ones, indicating the presence of a
high concentration of oxygen-containing adsorbates on the sur-
face of catalysts [30,50]. The peak deconvolution of the O 1s lines
gives more information about the relative amount of different
surface oxygen species. As shown in Table 2, the molar ratios of
surface lattice oxygen (OL, 529.0 eV) to Mn are 0.93, 0.98 and 1.35
(a)
(b)
Au0 4f
Au +4f
Au 4f
Mn 3s
Mn 3s
78
80
82
84
86
88
90
92
80
82
84
86
88
90
92
Binding Energy (eV)
Binding Energy (eV)
(c)
(d)
Au 4f
Au +4f
Au 4f
Au +4f
Mn 3s
Mn 3s
74
94
76
78
80
82
84
86
88
90
92
94
76
78
80
82
84
86
88
90
92
Binding Energy (eV)
Binding Energy (eV)
Fig. 3. Mn 3s and Au 4f XPS spectra of (a) Au/MnO2, (b) Au/MnO, (c) Au/Mn2O3, and (d) Au/Mn2O3 used for 4 times.

68
X. Li et al. / Applied Catalysis A: General 458 (2013) 63–70
(a) Mn 2p
(b) O 1s
Main peak
642.2 eV
Shoulder peak
Au/MnO2
Au/MnO2
Au/Mn2O3
Used Au/Mn2O3
Au/MnO
Au/MnO
Au/Mn2O3
Used Au/Mn2O3
534 537
525
665
528
531
Binding Energy (eV)
635
640
645
650
655
660
Binding Energy (eV)
Fig. 4. (a) Mn 2p XPS spectra of Au/MnO2, Au/Mn2O3, Au/Mn2O3 used for 4 times and Au/MnO. (b) O 1s XPS spectra of Au/MnO2, Au/MnO, Au/Mn2O3 and Au/Mn2O3 used
for 4 times.
OT/Mn and OL/Mn ratios increased slightly, revealing the oxygen
vacancies relatively decreased, and it also implies that the oxy-
gen vacancies participated in the oxidation reaction process and
the produced oxygen-containing group would occupy the oxygen
vacancies during the reaction. Thus, the oxygen vacancies are sup-
posed to facilitate the reaction. In addition, the OT/Mn and OL/Mn
ratios could be recovered after the activation of the used catalyst
with air at 573 K.
particles estimated from XRD patterns and TEM images, both of
them expressed admirable activities.
For evaluating the catalytic ability, we estimated the TON val-
ues for all the gold-supported catalysts. The TON value of Au/Mn2O3
100
(a)
80
60
3.2. Catalytic oxidative lactonization of 1,4-butanediol
The catalytic results of Au/MnOx in the oxidative lactonazation
of 1,4-butanediol to ␥-butyrolactone are given in Table 3 and
Fig. 5. For comparison, activities of bare carriers are also included.
Obviously, the pure supports hardly catalyzed the reaction, but the
catalytic activities were distinctly improved after the deposition
of gold. All the three catalysts exhibit excellent selectivity of 100%,
which can greatly facilitate the process of product separation after
the reaction, and the Au/Mn2O3 catalyst displays a prominent
activity with 4-h-conversion of 91%. After 8 h reaction, the con-
versions are 98%, 75% and 50% for Au/Mn2O3, Au/MnO2, Au/MnO.
Besides, we tested a catalyst with gold deposited on pure Mn3O4,
whose catalytic activity was similar to that of Au/MnO2. Thus, we
suppose that the small amount of Mn3O4 would not lower the
activity of Au/MnO at least. However, there is a little surprising
for Au/MnO2 and Au/MnO. Considering the relatively large gold
Mn2O3
40
MnO2
MnO
20
0
2
4
6
8
Reaction time (h)
(b)
Selectivity
100
80
100
80
60
40
20
Table 3
Catalytic activity of Au/MnOx catalysts in the oxidative lactonization of 1,4-
butanediol.a
Au/Mn2O3
Au/MnO2
Sample
Conversion (%)
Yield (%)
TOFb (h−1
)
TONc
60
40
20
MnO
MnO2
1
2
1
2
–
–
–
–
–
142
193
261
253
Mn2O3
5
5
–
Au/MnO
Au/MnO2
Au/Mn2O3
Au/Mn2O3
50
75
98
95
50
75
98
95
473
728
1352
1341
d
Au/MnO
2
a
Reaction conditions: 0.3 g catalyst, 1.4 g 1,4-butanediol, 20 ml TBP, 1.25 MPa air,
4
6
8
393 K, 8 h.
b
TOF value was calculated as mol 1,4-butanediol reacted in the initial 2 h on per
Reaction time (h)
mol surface gold calculated from the gold dispersion by TEM.
c
TON value was calculated as the amount of 1,4-butanediol reacted per mole
Fig. 5. Time course of the oxidative lactonization of 1,4-butanediol catalyzed by
parent MnOx (a) and Au/MnOx (b). Reaction conditions: 0.3 g catalyst, 1.4 g 1,4-
butanediol, 20 ml TBP, 1.25 MPa air, 393 K.
gold.
d
The catalyst was preserved for one year.

X. Li et al. / Applied Catalysis A: General 458 (2013) 63–70
69
the conversions are 50%, 48% and 10%, respectively. These values are
much lower than that in 1,4-butanediol, due to the shorter carbon
chain and smaller steric hindrance of cyclization in 1,4-butanediol
molecule. Meanwhile, ␥-butyrolactone is a very stable lactone with
five-membered ring, which is beneficial to the reaction.
Selectivity
Conversion
100
80
60
40
20
0
4. Conclusions
We have demonstrated that Au/MnOx catalysts prepared by
urea DP method were very active for the aerobic lactonization of
1,4-butanediol to ␥-butyrolactone with air as the oxidant. These
catalysts showed selectivity of 100% that could simplify the sep-
aration of the products after the reaction. The highest conversion
of 98% was achieved by using Au/Mn2O3 after 8 h reaction. Com-
pared with other Au/MnOx catalysts, we believed that the superior
activity of Au/Mn2O3 could be attributed to the presence of highly
dispersed gold particles and the intrinsic properties of Mn2O3,
including the appropriate point of zero charge for DP method, the
most quantity oxygen vacancies and the formation of the appropri-
ate metal–support interaction. Moreover, the Au/Mn2O3 catalyst
could be recycled by simple treatment and kept for a long time
without losing any catalytic activity. However, studies on the intrin-
sic nature of Au/MnOx by using Mn2O3 with different gold particle
size and different manganese oxides with the same sized gold par-
ticles are being under way.
0
1
2
3
Catalyst recycles
Fig. 6. Recyclability of the oxidative lactonization of 1,4-butanediol catalyzed by
Au/Mn2O3 catalyst. Reaction conditions: 0.3 g catalyst, 1.4 g 1,4-butanediol, 20 ml
TBP, 1.25 MPa air, 393 K.
is 261, which is much inferior to the ruthenium complexes (TON:
3170-17000) [11] and Au/hydrotalcite (TON: 1400) [12]. It is known
that the ruthenium complexes are homogeneous which can make a
good contact with the reactant and there is strong anion-exchange
ability in hydrotalcite, which can both greatly facilitate the reac-
tion. However, compared with other reducible oxide supported
gold catalysts: Au/TiO2 (TON: 101) [19], and Au/FeOx (TON: 300)
[20], the TON value of Au/Mn2O3 is comparable to those reported
earlier. Additionally, the TOF values were also calculated, and they
are 1352, 728 and 473 h−1 for Au/Mn2O3, Au/MnO2 and Au/MnO,
respectively. To our knowledge, the first two values are much
Acknowledgements
This work was financially supported by the Major State
Basic Resource Development Program (Grant No. 2012CB224804),
NNSFC (Project 20903035, 20973042, 21173052), and the Natural
Science Foundation of Shanghai Science and Technology Commit-
tee (08DZ2270500). The authors sincerely thank the reviewers for
their useful suggestions on the chemical state analysis of Au/MnOx.
greater than that in Au/TiO2 (64 h−1) [19] and Au/FeOx (624 h−1
)
[20]. Especially, the Au/Mn2O3 showed at least 2 times higher activ-
ity than Au/FeOx, which can be considered as a promising candidate
in the lactonization of 1,4-butanediol.
References
[1] Y. Sasaki, Electrochemistry 76 (2008) 161–164.
[2] A.C. Albertsson, I.K. Varma, Biomacromolecules 4 (2003) 1466–1486.
[3] S. Kobayashi, Macromol. Symp. 240 (2006) 178–185.
Recycling experiments were also conducted over Au/Mn2O3.
The sample was used for 4 successive reactions without apparent
loss of catalytic activity as shown in Fig. 6. Moreover, we have tested
a Au/Mn2O3 catalyst that was preserved in glass desiccators with-
out any special protection for more than one year. To our great
surprise, this catalyst also showed analogous great catalytic activ-
ity (95%), which alters the traditional impression that gold catalyst
is always hard to be stored and to keep its initial catalytic perform-
ances for a long period of time [55,56]. Additionally, we operated
the leaching test by ICP-AES analysis, there was no leaching of Au
or Mn being detected in the reaction mixture, indicating the special
stability of Au/Mn2O3.
Furthermore, the Au/Mn2O3 catalyst was also used in the
lactonizations of a series of ␣,␻-diols to afford corresponding
lactones, including 1,2-benzenedimethanol, 1,5-pentanediol and
1,6-hexanediol, and the corresponding catalytic activities are listed
in Table 4. It is found that their selectivities are all above 90%, but
[4] H. Lee, F.Q. Zeng, M. Dunne, C. Allen, Biomacromolecules 6 (2005) 3119–3128.
[5] M. Fetizon, M. Golfier, J.M. Louis, Chem. Commun. (1969) 1118–1119.
[6] P. Johnton, R.C. Sheppard, C.E. Stehr, S. Turner, J. Chem. Soc. C (1966) 1847–1856.
[7] T. Suzuki, K. Morita, M. Tsuchida, K. Hiroi, Org. Lett. 4 (2002) 2361–2363.
[8] M. Ito, A. Osaku, A. Shiibashi, T. Ikariya, Org. Lett. 9 (2007) 1821–1824.
[9] K. Hari Prasad Reddy, N. Anand, P.S. Sai Prasad, K.S. Rama Rao, B. Dacid Raju,
Catal. Commun. 12 (2011) 866–869.
[10] W.B. Hou, N.A. Dehm, R.W.J. Scott, J. Catal. 253 (2008) 22–27.
[11] J. Zhao, J.F. Hartwig, Organometallics 24 (2005) 2441–2446.
[12] T. Mitsudome, A. Noujima, T. Mizugaki, K. Jitsukawa, K. Kaneda, Green Chem.
11 (2009) 793–797.
[13] J. Huang, Y. Wang, J.M. Zheng, W.L. Dai, K.N. Fan, Appl. Catal. B: Environ. 103
(2011) 343–350.
[14] W. Oberhauser, A. Lavacchi, F. Vizza, L. Capozzoli, Appl. Catal. A: Gen. 451 (2013)
58–64.
[15] A.S.K. Hashmi, G.J. Hutchings, Angew. Chem. Int. Ed. 45 (2006) 7896–7936.
[16] B.K. Min, C.M. Friend, Chem. Rev. 107 (2007) 2709–2724.
[17] M.M. Schubert, S. Hackenberg, A.C. van Veen, M. Muhler, V. Plzak, R.J. Behm, J.
Catal. 197 (2001) 113–122.
[18] S. Carrettin, P. Concepción, A. Corma, J.M. Lopez Nieto, V.F. Puntes, Angew.
Chem. Int. Ed. 43 (2004) 2538–2540.
[19] J. Huang, W.L. Dai, H.X. Li, K.N. Fan, J. Catal. 252 (2007) 69–76.
[20] J. Huang, W.L. Dai, K.N. Fan, J. Phys. Chem. C 112 (2008) 16110–16117.
[21] J. Huang, W.L. Dai, K.N. Fan, J. Catal. 266 (2009) 228–235.
[22] J.M. Zheng, J. Huang, X. Li, W.L. Dai, K.N. Fan, RSC Adv. 2 (2012) 3801–3809.
[23] F. Wang, G. Yang, W. Zhang, W. Wu, J. Xu, Adv. Synth. Catal. 346 (2004) 633–638.
[24] V.R. Choudhary, J.R. Indurkar, V.S. Narkhede, R. Jha, J. Catal. 227 (2004) 257–261.
[25] X.Q. Li, J. Xu, F. Wang, J. Gao, L.P. Zhou, G.Y. Yang, Catal. Lett. 108 (2006) 137–140.
[26] Y. Su, L.C. Wang, Y.M. Liu, Y. Cao, H.Y. He, K.N. Fan, Catal. Commun. 8 (2007)
2181–2185.
[27] T. Ishida, H. Watanabe, T. Takei, A. Hamasaki, M. Tokunaga, M. Haruta, Appl.
Catal. A: Gen. 425–426 (2012) 85–89.
[28] L.M. Martinez.T, F. Romero-Sarria, W.Y. Hernandez, M.A. Centeno, J.A. Odrio-
zola, Appl. Catal. A: Gen. 423–424 (2012) 137–145.
Table 4
Catalytic activity of Au/Mn2O3 catalyst in the oxidative lactonization of ␣,␻-diols.a
Diol
Conversion (%)
Selectivity (%)
Yield (%)
1,4-Butanediol
1,5-Pentanediol
1,6-Hexanediol
1,2-Benzenedimethanol
98
48
10
50
100
90
90
98
43
9
96
48
a
Reaction conditions: 16 mmol ␣,␻-diols, Au/diol = 1/200, 20 ml TBP, 1.25 MPa
air, 393 K, 8 h.

70
X. Li et al. / Applied Catalysis A: General 458 (2013) 63–70
[29] Q. Ye, J.S. Zhao, F.F. Huo, D. Wang, S.Y. Cheng, T.F. Kang, H.X. Dai, Microporous
Mesoporous Mater. 172 (2013) 20–29.
[30] L.C. Wang, Q. Liu, X.S. Huang, Y.M. Liu, Y. Cao, K.N. Fan, Appl. Catal. B: Environ.
88 (2009) 204–212.
[43] Y. Cao, W.L. Dai, J.F. Deng, Appl. Catal. A: Gen. 158 (1997) 27–34.
[44] H. Trevino, G.D. Lei, W.M.H. Sachtler, J. Catal. 154 (1995) 245–252.
[45] L.C. Wang, X.S. Huang, Q. Liu, Y.M. Liu, Y. Cao, H.Y. He, K.N. Fan, J.H. Zhuang, J.
Catal. 259 (2008) 66–74.
[31] L.C. Wang, L. He, Y.M. Liu, Y. Cao, H.Y. He, K.N. Fan, J.H. Zhuang, J. Catal. 264
(2009) 145–153.
[32] L.C. Wang, Y.M. Liu, M. Chen, Y. Cao, H.Y. He, K.N. Fan, J. Phys. Chem. C 112
(2008) 6981–6987.
[46] F. Kapteijn, A.D. van Langeveld, J.A. Mpulijin, A. Andreini, M.A. Vuurman, A.M.
Turek, J.M. Jehng, I.E. Wachs, J. Catal. 150 (1994) 94–104.
[47] Y.F. Han, F.X. Chen, Z.Y. Zhong, K. Ramesh, L.W. Chen, E. Widjaja, J. Phys. Chem.
B 110 (2006) 24450–24456.
[33] L.C. Wang, L. He, Q. Liu, Y.M. Liu, M. Chen, Y. Cao, H.Y. He, K.N. Fan, Appl. Catal.
A: Gen. 344 (2008) 150–157.
[48] A. Longo, L.F. Liotta, G.D. Carlo, F. Giannici, A.M. Venezia, A. Martorana, Chem.
Mater. 22 (2010) 3952–3960.
[34] G.C. Bond, C. Louis, D.T. Thompson, in: G.J. Hutchings (Ed.), Catalysis by Gold,
Catalytic Science Series, vol. 6, Imperial College Press, London, 2006.
[35] X.Y. Deng, B.K. Min, A. Guloy, C.M. Friend, J. Am. Chem. Soc. 127 (2005)
9267–9270.
[49] S.D. Gardner, G.B. Hoflund, M.R. Davidson, D.R. Schryer, J. Schryer, B.T.
Upchurch, E.J. Kielin, Langmuir 7 (1991) 2135–2140.
[50] S.J. Lee, A. Gavriilidis, Q.A. Pankhurst, A. Kyek, F.E. Wagner, P.C.L. Wong, K.L.
Yeung, J. Catal. 200 (2001) 298–308.
[36] R. Zanella, S. Giorgio, C.R. Henry, C. Louis, J. Phys. Chem.
7634–7642.
B
106 (2002)
[51] J.A. Rodriguez, G. Liu, T. Jirsak, J. Hrek, Z. Chang, J. Dvorak, A. Maiti, J. Am. Chem.
Soc. 124 (2002) 5242–5250.
[37] R. Zanella, S. Giorgio, C.H. Shin, C.R. Henry, C. Louis, J. Catal. 222 (2004) 357–367.
[38] J. Radnik, L. Wilde, M. Schneider, M.M. Pohl, D. Herein, J. Phys. Chem. B 110
(2006) 23688–23693.
[39] M. Kosmulski, J. Colloid Interface Sci. 253 (2002) 77–87.
[40] P.S. Nico, R.J. Zasoski, Environ. Sci. Technol. 34 (2006) 3363–3367.
[41] S.D. Gardner, G.B. Hoflund, M.R. Davidson, H.A. Laitinen, D.R. Schryer, B.T.
Upchurch, Langmuir 7 (1991) 2140–2145.
[52] A.K. Sinha, K. Suzuki, M. Takanara, H. Azuma, T. Nonaka, K. Fukumoto, Angew.
Chem. Int. Ed. 46 (2007) 2891–2894.
[53] A.K. Sinha, K. Suzuki, M. Takahara, H. Azuma, T. Nonaka, N. Suzuki, N. Takahashi,
J. Phys. Chem. C 112 (2008) 16028–16035.
[54] D. Andreeva, I. Ivanov, L. Ilieva, J.W. Sobczak, G. Avdeev, K. Petrov, Top. Catal.
44 (2007) 173–182.
[55] R. Zanella, C. Louis, Catal. Today 768 (2005) 107–108.
[56] G.F. Zhao, M. Ling, H.Y. Hu, M.M. Deng, Q.S. Xue, Y. Lu, Green Chem. 13 (2001)
3088–3092.
[42] R. Leppelt, B. Schumacher, V. Plzak, M. Kinne, R.J. Behm, J. Catal. 244 (2006)
137–152.