Highly chemoselective hydrogenation of crotonaldehyde over Ag-In/SBA-15 fabricated by a modified "two solvents" strategy

Article abstract of DOI:10.1039/c1cc11013f

In the challenging crotonaldehyde hydrogenation to crotyl alcohol, an Ag-In/SBA-15 catalyst fabricated by a modified "two solvents" strategy shows an unprecedentedly high yield of 86% at a selectivity of 87%, which exceeds the best results on Pt-, Au- and

Full text of DOI:10.1039/c1cc11013f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 6168–6170
www.rsc.org/chemcomm
COMMUNICATION
Highly chemoselective hydrogenation of crotonaldehyde over
Ag–In/SBA-15 fabricated by a modified ‘‘two solvents’’ strategy
Li Tian,a Qiuyun Yang,a Zheng Jiang,b Yuan Zhu,a Yan Pei,*a Minghua Qiao*a and
Kangnian Fana
Received 21st February 2011, Accepted 2nd April 2011
DOI: 10.1039/c1cc11013f
In the challenging crotonaldehyde hydrogenation to crotyl alcohol,
an Ag–In/SBA-15 catalyst fabricated by a modified ‘‘two solvents’’
strategy shows an unprecedentedly high yield of 86% at a
selectivity of 87%, which exceeds the best results on Pt-,
Au- and other Ag-based heterogeneous catalysts reported so far.
Table 1 Physicochemical properties of the Ag/SBA-15 and Ag–In/
SBA-15 catalysts fabricated by the modified ‘‘two solvents’’ strategy
Catalyst
Loading (wt%)
Ag/Sia SBET
/
Vpore
/
dpore/
Ag
In
(%)
m2 gÀ1 cm3 gÀ1 nm
SBA-15
Ag/SBA-15
Ag–In/SBA-15 6.1
—
6.9
—
—
1.9
—
4.2
4.6
750
582
496
1.02
1.09
0.91
6.9
6.7
6.6
Among a,b-unsaturated aldehydes, crotonaldehyde (CRAL)
is one of the most difficult molecules to be selectively
hydrogenated to a,b-unsaturated alcohols over heterogeneous
catalysts. However, Ag, a well-documented oxidation catalyst,
unexpectedly exhibits notably high selectivity in CRAL
hydrogenation to crotyl alcohol (CROL).1,2 Unlike Pt and Au
that always rely on promoter and reducible support to evoke
their high selectivity, Ag is intrinsically selective in saturating the
CQO bond in conjugation with the CQC bond.2–5 Moreover,
Claus and co-workers experimentally,6,7 and later on Zhang and
co-workers theoretically and experimentally5 demonstrated, that
the formation of a,b-unsaturated alcohols is preferred on small
Ag nanoparticles (AgNPs).
a
Surface atomic ratio determined by XPS.
a poor catalytic performance of the Ag/SBA-15 catalyst prepared
following the usual ‘‘two solvents’’ strategy.
Siliceous mesoporous SBA-15 was synthesized according to the
published procedures.8 Then, 1.0 g of the calcined SBA-15 was
homogeneously dispersed in 50 ml of cyclohexane under vigorous
stirring at room temperature. Following this, 5.0 ml of the AgNO3
aqueous solution was added dropwise into the dispersion. The
nominal Ag loading relative to SBA-15 was 9.0 wt%. The slurry
was evaporated to dryness at 343 K and heated at 373 K
overnight. The resulting solids were calcined at 623 K for 4 h
and reduced in a 5 vol% H2/Ar flow at 573 K for 4 h. The
as-fabricated catalyst is designated Ag/SBA-15. The Ag–In/
SBA-15 catalyst was similarly fabricated, except that the aqueous
solution contained both AgNO3 and In(NO3)3. The nominal
Ag and In loadings relative to SBA-15 in this catalyst were 9.0
and 3.0 wt%, respectively. The practical loadings and textural
properties of these catalysts are summarized in Table 1.
In the family of mesoporous materials, SBA-15 is one of the
most ideal catalyst supports due to its high surface area, high
porosity, and large and uniform channel size.8 These features
provide a consistent and well-isolated environment for the
growth of nanoparticles, which is of particular interest for size-
dependent reactions. Herein, we report the fabrication and
characterization of SBA-15-supported Ag and In-promoted Ag
catalysts by a modified ‘‘two solvents’’ strategy for CRAL
hydrogenation to CROL. The usual ‘‘two solvents’’ strategy is
based on a volume of the metal salt aqueous solution set equal to
the pore volume of SBA-15. The introduction of the metal salt
into the hydrophilic channels of SBA-15 is driven by the capillary
force in the presence of an organic solvent.9–11 In this approach,
several regular metal oxide nanostructures have been fabricated
successfully. We modified this strategy by using an excess volume
of aqueous solution, since our preliminary experiments revealed
Fig. 1a presents transmission electron micrographs (TEM;
JEOL JEM2011) of the Ag/SBA-15 catalyst fabricated by the
modified ‘‘two solvents’’ strategy. It shows that the AgNPs are
dispersed evenly inside the entire SBA-15 host instead of on the
exterior, as most AgNPs are located between two immediate
neighboring channel walls. The mean particle size of the AgNPs
is 6.1 Æ 1.3 nm. The decreased BET surface area and pore
volume of Ag/SBA-15 in comparison to SBA-15 (Table 1)
substantiate the encapsulation of the AgNPs in the channels of
SBA-15. Early methods for the fabrication of SBA-15-supported
AgNPs can be mainly classified into three categories: (1) wetness
impregnation of SBA-15 with silver salt,12 (2) synthesis of
SBA-15 in the coexistence of silver salt and chemical reductant,13
and (3) synthesis of SBA-15 in the presence of pre-formed
AgNPs.14 However, the former two methods inevitably produce
large Ag particles or aggregates on the external surface, while the
a Department of Chemistry and Shanghai Key Laboratory of
Molecular Catalysis and Innovative Materials, Fudan University,
Shanghai, 200433, P. R. China. E-mail: mhqiao@fudan.edu.cn,
peiyan@fudan.edu.cn; Fax: +86 21 6564 1740;
Tel: +86 21 55664679
b Shanghai Synchrotron Radiation Facility, Shanghai Institute of
Applied Physics, Chinese Academy of Sciences, Shanghai 201204,
P.R. China
c
6168 Chem. Commun., 2011, 47, 6168–6170
This journal is The Royal Society of Chemistry 2011

View Article Online
Fig. 1 TEM images and particle size distribution histograms (insets)
of (a) Ag/SBA-15 and (b) Ag–In/SBA-15 fabricated by the modified
‘‘two solvents’’ strategy recorded along the [110] zone axis of SBA-15.
Fig. 3 Ag K-edge XAS spectra of (a) Ag foil, (b) Ag/SBA-15 and
(c) Ag–In/SBA-15 fabricated by the modified ‘‘two solvents’’ strategy.
Table 2 Hydrogenation results over the Ag/SBA-15 and Ag–In/
SBA-15 catalysts fabricated by the modified ‘‘two solvents’’ strategya
Selectivityb (%)
Conv.b
CROL BUALd BUOLe
Yb (%) r0
c
Catalyst
tb/min (%)
Ag/SBA-15 245
Ag–In/SBA-15 740
95
99
54
87
21
3
25
10
51
86
391
283
a
Reaction conditions: 0.2 g catalyst, 0.1 ml CRAL, 29 ml n-hexane,
reaction temperature 413 K, H2 pressure 2.0 MPa, and stirring rate
b
1000 rpm. Values corresponding to the maximum yield of CROL.
c
Initial hydrogenation rate in mmol minÀ1 gAg
BUOL: butanol.
À1 d
. BUAL: butanal.
e
respectively, falling well in the ranges of the particle sizes
measured by TEM.
Fig. 2 Wide- and small-angle (inset) XRD patterns of (a) Ag/SBA-15
and (b) Ag–In/SBA-15 fabricated by the modified ‘‘two solvents’’
strategy.
In Fig. 2, there is an additional small peak at 2y of 30.61 for
Ag–In/SBA-15. This feature is assigned to the (222) reflection
of In2O3 (JCPDS 06-0416), since when we further increased
the loading of In, this feature was intensified, along with the
appearance of other peaks relating to In2O3. The inset in Fig. 2
confirms again that the mesoscopic ordering of SBA-15 was
retained in both catalysts fabricated by the modified ‘‘two
solvents’’ strategy, with the clear observation of all three
small-angle peaks characteristic of the hexagonal mesostructure
of SBA-15.
last method involves complicated pre-synthesis of size-controlled
AgNPs and rigorous encapsulation steps. The modified ‘‘two
solvents’’ strategy is thus advantageous relative to those methods
in its convenience and effectiveness.
Surprisingly, when the same strategy was used to fabricate the
Ag–In/SBA-15 catalyst, the AgNPs could hardly be discerned in
the low-magnification image as Fig. 1a, although more metal was
loaded in this case (see Table 1). Only in the TEM image with
higher magnification the AgNPs of 3.4 Æ 0.6 nm adhering to the
channel walls of SBA-15 were identified (Fig. 1b), suggesting that
In is efficient in further dispersing the AgNPs. This observation is
corroborated by the distinct difference between their wide-angle
X-ray diffraction patterns (XRD; Bruker AXS D8 Advance).
For Ag/SBA-15, besides the broad peak at 2y E 221 due
to amorphous silica, there are pronounced diffraction peaks
attributable to fcc Ag (JCPDS 04-0783) (Fig. 2). For Ag–In/
SBA-15, the diffractions of Ag are attenuated markedly, leaving
only the once strongest Ag(111) peak that now becomes weak
and broad. The assignment of this peak to fcc Ag is validated by
the identical Ag K-edge X-ray absorption spectra (XAS; SSRF)
of Ag–In/SBA-15 and Ag/SBA-15 (Fig. 3). It is known
that for Ag metal, the intensity of the whiteline (first continuum
resonance) in the Ag K-edge spectrum is nearly the same as
that of the adjacent resonances. Based on the Scherrer equation,
the Ag crystallite sizes perpendicular to the {111} planes were
estimated as 6.4 and 2.7 nm for Ag/SBA-15 and Ag–In/SBA-15,
Table 2 compiles the catalytic results over the Ag/SBA-15
and Ag–In/SBA-15 catalysts fabricated by the modified ‘‘two
solvents’’ strategy in liquid-phase hydrogenation of CRAL.
Consistent with previous work on selective hydrogenation of
acrolein15 and citral on Ag catalysts,16 In is remarkable in
improving the selectivity to a,b-unsaturated alcohol. On the
Ag–In/SBA-15 catalyst prepared by the modified ‘‘two
solvents’’ strategy, the maximum yield of CROL was drastically
increased to 86% with the corresponding selectivity as high as
87%, as compared to the yield of 51% and the selectivity of
54% on the Ag/SBA-15 catalyst (Table 2). It is worth
emphasizing here that the CROL yield on the present
Ag–In/SBA-15 catalyst excels the best values on literature
Pt-, Au- and Ag-based catalysts via catalytic hydrogenation.
For example, the highest CROL yield reported so far was 71%
obtained on an Au5.0In0.75/APTMS-SBA-15 catalyst.17
The most selective Ag-based catalyst ever reported is the
Ag–MnO2/Al2O3Á5AlPO4 catalyst developed by Nagase
et al.,1 on which a CROL yield of 69.5% was obtained under
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 6168–6170 6169

View Article Online
more stringent reaction conditions (reaction temperature
423 K, H2 pressure 5 MPa) than ours.
Lewis acid sites, which aid in the adsorption and activation of
the CQO bond via the electron lone pair on oxygen.7 This may
be another important reason underlying the unprecedentedly
high selectivity to CROL on the Ag–In/SBA-15 catalyst.
In summary, the Ag–In/SBA-15 catalyst prepared by the
modified ‘‘two solvents’’ strategy is highly selective in the
challenging CRAL hydrogenation to CROL. Its excellent
catalytic performance is attributed to the combined effects
of the reduced AgNPs bearing a high population of the coordi-
natively unsaturated or positively charged Ag sites and the
presence of In2O3 functioning as Lewis acid sites, both favoring
the adsorption and activation of the CQO bond in CRAL. The
present work, together with our previous work on SBA-15-
supported Ru-based catalysts prepared by the same strategy,20,21
shows the prospect of the modified ‘‘two solvents’’ strategy as a
routine and effective method for the fabrication of highly
dispersed metal nanocrystals accommodated in mesoporous
materials with improved catalytic performances.
It is acknowledged that the chemisorption mode of the
substrate is decisive to the selectivity in the hydrogenation of
a,b-unsaturated aldehydes.18 The preferential binding through
the CQO bond, e.g., the Z1-on-top mode via the terminal
oxygen atom, should favor a high selectivity to a,b-unsaturated
alcohols. Based on density functional theory, it is found that
while the adsorption of CRAL on the Ag(111) surface is
negligible, the energetically most favored adsorption mode
on an Ag19 cluster with fcc structure is the Z1-on-top mode
bound on coordinatively unsaturated Ag atoms.5 Likewise,
Fujii et al. found that acrolein can coordinate through
the CQO bond to the positively charged sites (probably
ad-adatom and/or kink sites) on an evaporated Ag film.19
Alternatively, the positively charged Ag sites may arise from
electronic modification by subsurface oxygen in Ag formed
during the process of catalyst preparation or hydrogenation.
These sites are more abundant on supported Ag than on
unsupported polycrystalline Ag.7 Both interpretations point
to a higher selectivity to a,b-unsaturated alcohols on smaller
AgNPs, since the abundance of either the coordinatively
unsaturated or the positively charged Ag sites is inversely
proportional to the Ag particle size. This idea was validated
by Zhang and co-workers on Ag/SiO2 catalysts with Ag
particle sizes ranging from 4.8 to 411.3 nm for CRAL
hydrogenation5 and by Claus and co-workers on Ag(111)
and (100) single crystals, sputtered Ag, and Ag/SiO2 prepared
by three different methods for acrolein hydrogenation.6 Such a
size effect may operate on the Ag/SBA-15 and Ag–In/SBA-15
catalysts, which is reflected in the much higher CROL selectivity
on the Ag–In/SBA-15 catalyst with much smaller AgNPs.
However, the selectivity enhancement is not exclusively
limited to the size effect of Ag. Claus and Hofmeister reported
a CROL selectivity of only 57% on an Ag/SiO2 catalyst with
an average particle size of 3.7 nm.2 Although the reaction
conditions are in variance with ours, such a distinctive
selectivity difference between their catalyst and our Ag–In/
SBA-15 catalyst with similar Ag particle size clearly indicates
that In2O3 plays an important role in improving CROL
selectivity. The involvement of In2O3 in CRAL hydrogenation
is evidenced by the lower hydrogenation rate on Ag–In/SBA-
15 than on Ag/SBA-15 (Table 2), or smaller AgNPs would
have led to a higher rate.7 This activity loss infers that the
AgNPs were decorated by In2O3, thus blocking some Ag sites.
Because of the inclusion of the AgNPs in the channels and the
amorphous nature of SBA-15, it is impossible to directly
observe the decoration of the AgNPs by In2O3 by high-
resolution TEM. However, X-ray photoelectron spectroscopy
(XPS; Perkin Elmer PHI5000C) revealed an Ag/Si surface
atomic ratio of 4.6% for Ag–In/SBA-15, which is only slightly
higher than the ratio of 4.2% for Ag/SBA-15 (Table 1).
Assuming that the presence of In2O3 and the improved
dispersion of Ag do not alter the surface exposure of
SBA-15, a simple calculation shows that the Ag/Si surface
ratio on Ag–In/SBA-15 should be at least 1.6 times of that on
Ag/SBA-15, if the surface of the AgNPs was In2O3-free. In
light of these facts, we propose that a portion of In2O3 covered
the AgNPs. Metal oxides, including In2O3, can function as
Financial support from the NSF of China (20803011,
21073043), the Shanghai Science and Technology Committee
(10JC1401800, 08DZ2270500), and the Program of New
Century Excellent Talents in Universities (NCET-08-0126) is
gratefully acknowledged.
Notes and references
1 Y. Nagase, H. Muramatu and T. Sato, Chem. Lett., 1988, 1695.
¯
2 P. Claus and H. Hofmeister, J. Phys. Chem. B, 1999, 103, 2766.
3 P. G. N. Mertens, F. Guypers, P. Vandezande, X. P. Ye,
F. Verpoort, I. F. J. Vankelecom and D. E. De Vos, Appl. Catal.,
A, 2007, 325, 130.
4 C. E. Volckmar, M. Bron, U. Bentrup, A. Martin and P. Claus,
J. Catal., 2009, 261, 1.
5 X. F. Yang, A. Q. Wang, X. D. Wang, T. Zhang, K. L. Han and
J. Li, J. Phys. Chem. C, 2009, 113, 20918.
6 M. Bron, D. Teschner, A. Knop-Gericke, B. Steinhauer,
A. Scheybal, M. Havecker, D. Wang, R. Fodisch, D. Honicke,
¨
A. Wootsch, R. Schlogl and P. Claus, J. Catal., 2005, 234, 37.
¨
¨
¨
7 M. Bron, D. Teschner, A. Knop-Gericke, F. C. Jentoft,
J. Krohnert, J. Hohmeyer, C. Volckmar, B. Steinhauer,
¨
¨
R. Schlogl and P. Claus, Phys. Chem. Chem. Phys., 2007, 9, 3559.
8 D. Y. Zhao, J. L. Feng, Q. S. Huo, N. Melosh, G. H. Fredrickson,
B. F. Chmelka and G. D. Stucky, Science, 1998, 279, 548.
9 M. I. Clerc, D. Bazin, M. D. Appay, P. Beaunier and A. Davidson,
Chem. Mater., 2004, 16, 1813.
10 I. Lopes, N. El Hassan, H. Guerba, G. Wallez and A. Davidson,
Chem. Mater., 2006, 18, 5826.
11 Q. S. Lu, Z. Y. Wang, J. G. Li, P. Y. Wang and X. L. Ye,
Nanoscale Res. Lett., 2009, 4, 646.
12 M. Boutros, J. M. Trichard and P. Da Costa, Appl. Catal., B, 2009,
91, 640.
13 D. Tian, G. P. Yong, Y. Dai, X. Y. Yan and S. M. Liu, Catal.
Lett., 2009, 130, 211.
´
14 J. Zhu, Z. Konya, V. F. Puntes, I. Kiricsi, C. X. Miao, J. W. Ager,
A. P. Alivisatos and G. A. Somorjai, Langmuir, 2003, 19,
4396.
15 M. Lucas and P. Claus, Chem. Eng. Technol., 2005, 28, 867.
16 M. Steffan, M. Lucas, A. Brandner, M. Wollny, M. Oldenburg and
P. Claus, Chem. Eng. Technol., 2007, 30, 481.
17 Q. Y. Yang, Y. Zhu, L. Tian, S. H. Xie, Y. Pei, H. Li, H. X. Li,
M. H. Qiao and K. N. Fan, Appl. Catal., A, 2009, 369, 67.
18 V. Ponec, Appl. Catal., A, 1997, 149, 27.
19 S. Fujii, N. Osaka, M. Akita and K. Itoh, J. Phys. Chem., 1995, 99,
6994.
20 J. Bu, J. L. Liu, X. Y. Chen, J. H. Zhuang, S. R. Yan, M. H. Qiao,
H. Y. He and K. N. Fan, Catal. Commun., 2008, 9, 2612.
21 J. L. Liu, L. J. Zhu, Y. Pei, J. H. Zhuang, H. Li, H. X. Li,
M. H. Qiao and K. N. Fan, Appl. Catal., A, 2009, 353, 282.
c
6170 Chem. Commun., 2011, 47, 6168–6170
This journal is The Royal Society of Chemistry 2011