Mesoporous bubble-like manganese silicate as a versatile platform for design and synthesis of nanostructured catalysts

Article abstract of DOI:10.1002/chem.201405697

Manganese silicate in bubble-like morphology was used as a versatile platform to prepare a new class of yolk-shell hybrids. The mesoporosity of the shell was generated from the interbubble space and the bubble structure of manganese silica was used to hold and support nanoparticles (e.g., Au, Ag, Pt, Co, Ni, Au-Pd alloy, MoO₂, Fe₃O₄, carbon nanotubes and their combinations). We also used heterogeneous catalysis reactions to demonstrate the workability of these catalysts in both liquid and gas phases.

Full text of DOI:10.1002/chem.201405697

DOI: 10.1002/chem.201405697
Communication
&
Mesoporous Materials
Mesoporous Bubble-like Manganese Silicate as a Versatile
Platform for Design and Synthesis of Nanostructured Catalysts
Guowu Zhan, Christopher C. Yec, and Hua Chun Zeng*[a]
faces of mesoporous MS, including Au, Ag, Pt, Ni, Co, Au–Pd
Abstract: Manganese silicate in bubble-like morphology
alloy, MoO2, Fe3O4, carbon nanotubes (CNTs) and their combi-
was used as a versatile platform to prepare a new class of
nations. We demonstrate their enhanced catalysis for both
yolk–shell hybrids. The mesoporosity of the shell was gen-
liquid- (e.g., 4-nitrophenol reduction) and gas-phase reactions
erated from the interbubble space and the bubble struc-
(e.g. CO2 hydrogenation). MS also allows doping of rare-earth
ture of manganese silica was used to hold and support
ions into its silicate framework;[6] therefore, catalytic properties
nanoparticles (e.g., Au, Ag, Pt, Co, Ni, Au–Pd alloy, MoO2,
brought in by the dopants could be further tuned.
Fe3O4, carbon nanotubes and their combinations). We also
We started with introducing Au NPs into the framework of
used heterogeneous catalysis reactions to demonstrate
MS, in view of their excellent catalytic activity.[7] Shown in Fig-
the workability of these catalysts in both liquid and gas
ure 1a, the Au@SiO2 core–shell structure was first synthesised.
phases.
Hollow-structured mesoporous materials consisting of a cavity
or void within a mesoporous shell have been attracting a great
deal of attention across various fields.[1] Usually, the mesopore
channels are generated by using self-assembled surfactant mi-
celles as structural directors, similar to the preparation of
MCM-41 type mesopores,[2] and the templating surfactants are
removed by thermal treatment or solvent extraction.[3] Howev-
er, these methods are not necessarily very cost-effective at an
industrial scale and it is difficult to remove surfactants com-
pletely. A concept of a “surfactant-free process” may be real-
ised if the mesopore channels can be created by the stacking
of nano-building blocks (e.g., sheets and crystallites).[4] Unfortu-
nately, the uniformity of pores arising from the interparticle
space is usually low due to disordered aggregation and poly-
dispersity of the building blocks. Therefore, the control of mes-
oporous nanostructures without using surfactants is particular-
ly challenging.
Figure 1. TEM images of a) Au@SiO2 core–shell; b) and c) Au@MS yolk–shell;
d) FESEM image of Au@MS yolk–shell; e) XRD patterns of the “bubble cata-
Herein we report a facile surfactant-free approach for fabri-
cating sophisticated yolk–shell nanocatalysts with mesoporous
channels in the shell. Manganese silicate (MS) was used as the
shell framework, since manganese has been explored as an es-
sential component in numerous catalytic materials, especially
important in electron-transfer reactions.[5] In order to pursue
multifunctional catalysis or synergetic effects between the dif-
ferent active species, various catalytic nanoparticles (NPs) were
encapsulated/or immobilised into hollow cavities or on the sur-
lysts”; f) the ideal crystallographic structure of MS: yellow, green, and blue
represent 4-coordinated Si4+ (tetrahedrons), 6-coordinated Mn3+ (octahe-
drons) and 8-coordinated Mn2+ (irregular polyhedron), respectively; and
g) elemental mappings of the Au@MS yolk–shell.
This solid precursor has a highly uniform size (ca. 226 nm) and
contains a single Au NP core (ca. 15 nm) at the centre. Then
the solid precursor (80 mg) was mixed with an aqueous solu-
tion (20 mL) of MnSO4 (6.8 gLꢀ1) and sodium maleate (9 gLꢀ1).
The mixture was stirred vigorously for 10 min before a hydro-
thermal treatment (1808C for 12 h, see the Supporting Infor-
mation section SI-1). Interestingly, the transformations of struc-
ture (core–shell!yolk–shell) and composition (silica!MS)
could be achieved in this one-pot hydrothermal process. The
as-prepared yolk–shell structure with a movable Au NP core
inside the MS shell (denoted as Au@MS) are shown in Fig-
[a] G. Zhan, C. C. Yec, Prof. Dr. H. C. Zeng
NUS Graduate School for Integrative Sciences and Engineering and
Department of Chemical and Biomolecular Engineering
Faculty of Engineering, National University of Singapore
10 Kent Ridge Crescent, Singapore 119260
E-mail: chezhc@nus.edu.sg
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201405697.
Chem. Eur. J. 2015, 21, 1882 – 1887
1882
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
ure 1b,c. The TEM images reveal
that the thickness of MS shell is
about 22 nm, which was com-
posed of numerous smaller
hollow bubbles (ca. 7 nm, see
Supporting Information, section
SI-2), and the overall size of MS
hollow spheres is about 215 nm.
The size of the Au@MS yolk–
shell is very uniform (Figure 1d).
Occasionally, mechanically frac-
tured spheres (<3%) were also
observed under large-area SEM
screening (see Supporting Infor-
mation, section SI-3), proving the
existence of hollow interior. The
bubble-like shell was produced
owing to the presence of CO2
bubbles (from maleate anion de-
composition under hydrothermal
conditions), which serve as soft
templates for the deposition of
manganese silicate (through the
ion-exchange of Mn2+ with
H4SiO4).[6] Our X-ray diffraction
(XRD) results (Figure 1e) indicate
the presence of both metallic
gold
and
blythite
phase
(Mn3+2 Mn2+3[SiO4]3, JCPDS no.
89-5709) in the Au@MS yolk–
shell catalyst, while the Stçber
silica is in its amorphous state.
The crystal structure of MS is
similar to other garnet-group
minerals with a general chemical
formula of A3B2(XO4)3 (Fig-
Figure 2. a) N2 adsorption/desorption isotherms of the “bubble catalysts”; b) pore size distribution of the Au@MS
yolk–shell spheres; c) time-dependent UV/Vis spectra of the reaction mixture of reduction of 4-nitrophenol by
NaBH4 using Au@MS yolk–shell as a catalyst; and d) preparation procedures of “bubble catalysts”: 1) pristine cata-
lytic NPs, 2) NPs@SiO2 core–shell, 3) NPs@mSiO2 core–shell, 4) NPs@MS yolk–shell, 5) silica bead, 6) hollow MS
sphere, 7) MS@NPs, 8) CNT@mSiO2 core–shell, 9) CNT@MS yolk–shell, 10) CNT@NPs@mSiO2 core–shell, and
11) CNT@NPs@MS yolk–shell.
ure 1 f).[8] Furthermore, our energy-dispersive X-ray (EDX) map-
ping study also confirms the formation of the Au@MS yolk–
shell structure (Figure 1g).
of the catalyst. The reduction reaction was completely finished
in 12 min, along with the colour change from bright yellow to
colourless. This reaction can be considered as a pseudo-first-
order reaction, and the rate constant was determined to be
0.18 minꢀ1 (see the Supporting Information, section SI-4), much
higher than that using pristine MS as a catalyst (0.03 minꢀ1).
We also demonstrated a broad scope of applications of this
approach by introducing other catalytic metal species to diver-
sify the functions of the MS-based nanocatalysts. As illustrated
in Figure 2d, we have developed various synthetic strategies
to produce eleven more representative “bubble catalysts”; they
are spherical Ag@MS, Au–Pd@MS, MoO2@MS, Fe3O4@MS,
MS@Pt, MS@Au, MS@Ni and MS@Co, and tubular CNT@MS,
CNT@Au–Pd@MS and Au–Pd@MS. Generally speaking, the con-
trolled fabrication can be divided into three steps, 1) synthesis
of the discrete NPs with a desirable size and shape, 2) deposi-
tion of silica shell on the as-synthesised NP cores to generate
core–shell structures (NPs@SiO2 or NPs@mSiO2, where SiO2 =
nonporous SiO2 and mSiO2 =mesoporous SiO2) and 3) conver-
sion of NPs@SiO2 or NPs@mSiO2 core–shell to NPs@MS yolk–
shell (i.e., nanocatalysts). Likewise, using Fe3O4@SiO2 (Fig-
As expected, N2 physisorption measurements (Figure 2a)
reveal the highly porous nature of Au@MS spheres. The
“bubble catalysts” followed Langmuir type IV isotherm with
a hysteresis loop at high P/P0 (which is characteristic of meso-
pores). The Brunauer–Emmett–Teller (BET) surface areas are
341 m2 gꢀ1 for the pristine MS and 217 m2 gꢀ1 for the Au@MS,
both of which are much higher than that of 15 m2 gꢀ1 for non-
porous Stçber silica beads. Intriguingly, the Au@MS gives pore
volume of 0.20 cm3 gꢀ1 and mesopores of 3.3 nm (Figure 2b),
similar to those obtained by using cetyltrimethylammonium
bromide (CTAB) micelles as a soft-template.[9] Using the reduc-
tion of 4-nitrophenol as a testing example,[10] we confirmed
that the mesoporous MS shell indeed allows a convenient
access for the reactant (namely, 4-nitrophenol with a molecular
size of 0.66 nmꢁ0.43 nm[11]) to travel into the central cavity in
which the Au NP catalyst is located. As shown in Figure 2c, the
intensity of the characteristic absorption peak at 400 nm (as-
signed to 4-nitrophenol) quickly decreased upon the addition
Chem. Eur. J. 2015, 21, 1882 – 1887
www.chemeurj.org
1883
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Figure 4. TEM images of a) CNT@mSiO2 core–shell; b) and c) CNT@MS yolk–
shell; d) CNT@Au–Pd@mSiO2 core–shell; e) and f) CNT@Au–Pd@MS yolk–
shell; and g) and h) tubular Au–Pd@MS. i) XRD patterns of these studied
structures.
Figure 3. TEM images of a) Fe3O4@SiO2 core–shell, b) Fe3O4@MS yolk–shell,
c) Ag@SiO2 core–shell, d) Ag@MS yolk–shell, e) Au–Pd@SiO2 core–shell,
f) Au–Pd@MS yolk–shell, g) MoO2@mSiO2 core–shell, and h) MoO2@MS yolk–
shell. i) XRD patterns of these studied structures.
ure 3a) as a precursor, the yolk–shell structure of Fe3O4@MS
was also successfully obtained (Figure 3b). The Fe3O4@MS ex-
hibits strong magnetic property, as it can be collected easily
with an external magnetic bar (within 7 s, see Supporting Infor-
mation, section SI-5). Other yolk–shell structures with different
compositions and shapes can be readily produced by replacing
the core materials, such as Ag NPs, MoO2 NPs and flower-
shaped Au–Pd alloy NPs. The synthetic conditions were also
optimised (see the Supporting Information section SI-1) in
order to deposit uniform silica coating on the cores (Fig-
ure 3c,e,g). As evidenced in the yolk–shell structures of
Ag@MS (Figure 3d and Supporting Information, section SI-6),
Au–Pd@MS (Figure 3 f and Supporting Information, section SI-
7) and MoO2@MS (Figure 3 h and Supporting Information, sec-
tion SI-8), the silica layer could be converted completely to MS
shells, while the preinstalled cores were kept intact. Moreover,
the XRD patterns of Figure 3i and EDX elemental mapping
(see Supporting Information, section SI-8) also elucidate the
presence of the nanoparticles within the MS shells. These “de-
signed” yolk–shell structures, which could also be viewed as
nanoreactors, endow the trapped core materials with excellent
anti-aggregation properties. In addition to the applications in
solution phase, high thermal stability of MS shells is very suita-
ble for reactions at high temperatures. Such MS-supported cat-
alysts are expected to be workable at harsh conditions,[12]
which will be explored below for CO2 hydrogenation.
ure 4b,c display the typical TEM images of the as-prepared
CNT@MS. As expected, the product preserves the shape of the
original CNT@mSiO2 and the CNTs were encapsulated within
the central cavity. Analogously, we also fabricated CNT@
Au–Pd@MS catalysts employing CNT@Au–Pd@mSiO2 as a pre-
cursor (Figure 2d (10 and 11) and Figure 4e,d,f). TEM images
(Figure 4e,f and Supporting Information, section SI-10) show
that the Au–Pd alloy NPs (size: 1.4 to 3.3 nm) are located on
the CNTs backbones inside the hollow cavity of the CNT@
Au–Pd@MS. The XRD results of Figure 4i affirm the presences
of Au–Pd alloy, CNTs, and manganese silicate in this product.
The CNT@Au–Pd@MS has a surface area of 388 m2 gꢀ1 and
a pore volume of 0.59 cm3 gꢀ1 (Figure 2a). In addition, the
CNTs inside this structure could be removed thermally (e.g., at
5008C), and the thus obtained tubular Au–Pd@MS was also
characterised by TEM (Figure 4 g,h and Supporting Informa-
tion, section SI-10).
The metal NPs not only can be encapsulated inside the MS,
but also be dispersed on the external surface of MS (Figure 2d,
5–7). The MS@Pt catalyst was prepared under in situ hydro-
thermal conditions with potassium tetrachloroplatinate. As re-
vealed by TEM (Figure 5a,b) and STEM (Figure 5c and Support-
ing Information, section SI-11) techniques, Pt NPs with a uni-
form size of 3.3 nm were highly dispersed on the surfaces of
the MS and the bubble feature of MS was well maintained
after metal loading. A high-resolution TEM image in Figure 5d
shows that the lattice fringe of Pt NPs is approximately 2.3 ꢂ,
consistent with the d111 value of platinum. Furthermore, chemi-
cal states of the surface Mn and Pt were investigated by XPS
method (Figure 5e and Supporting Information, section SI-11),
revealing that Pt existed mainly as Pt0 (ca. 68%) and surface-
CNTs are generally recognised as one of the best and most
easily available one-dimensional nanomaterials. In Figure 4a,
we also extended the spherical MS to one-dimensional struc-
ture by using CNT@mSiO2 as a precursor for the same transfor-
mative synthesis (see Supporting Information, section SI-9). Fig-
Chem. Eur. J. 2015, 21, 1882 – 1887
www.chemeurj.org
1884
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
means of a reverse water gas shift reaction (see Supporting In-
formation, section SI-13).[14] A certain amount of methanol and
methane were found as by-products. Using the MS@Pt, the
conversion increased considerably with temperature and pres-
sure due to the endothermic reaction (Figure 6a,b). It should
be mentioned that the hydrogenation of CO2 towards CO was
favourable at a high temperature, but at a low pressure (Fig-
ure 6b). Under the same pressure (0.1 MPa), for example, the
CO2 conversion was achieved at 30% with selectivity to CO at
97% at 673 K, while the selectivity to methanol and methane
was 11 and 8%, respectively, at 548 K. Under the optimal con-
ditions, a high turnover frequency (TOF) of 0.51 sꢀ1 (or
1836 hꢀ1) was achieved, which is significantly higher than
some previous results (see Supporting Information, section SI-
13). In addition, the apparent activation energy (Ea) of the pro-
cess was determined. As shown in Arrhenius plots of Figure 6c,
Ea values of the MS@Pt and CNT@Au–Pd@MS were 47 and
32 kJmolꢀ1, respectively, which are much lower than other lit-
erature data.[15] By contrast, very low activity was observed for
Au@MS catalyst with only 1.7% CO2 conversion at 673 K and
0.1 MPa, largely due to the big size of Au NPs (ca. 15 nm; inac-
Figure 5. a)–d) TEM, STEM and HRTEM images of the MS@Pt catalyst; e) XPS
spectra of Pt 4f regions recorded on the MS@Pt catalyst (top) and pristine
MS sample (bottom); f) XRD patterns of the MS@Pt catalyst; and g) elemen-
tal mappings of the MS@Pt catalyst.
oxidised PtII (ca. 32%). In agree-
ment with the findings from
both TEM and XPS analyses, the
XRD patterns in Figure 5 f show
that the MS@Pt catalyst was
composed of platinum (at a size
of 3.7 nm; Scherrer method), bly-
thite-type manganese silicate
and others (e.g., Mn2SiO4 and
Mn3O4). Again, results from EDX
elemental mapping shown in
Figure 5 g confirm its structural
configuration, and an N2 physi-
sorption study (Figure 2a) indi-
cates that MS@Pt has a specific
surface area of 210 m2 gꢀ1
,
a pore volume of 0.23 cm3 gꢀ1
and an average mesopore size
of 3.9 nm. Moreover, other types
of NPs can also be immobilised
on the external surface of MS,
such as Ni, Co and Au NPs (see
the Supporting Information, sec-
tion SI-12).
We also carried out CO2 hy-
drogenation as a test reaction to
verify the effectiveness of these
hybrid nanocatalysts,[13] namely
the
MS@Pt,
Au@MS
and
CNT@Au–Pd@MS, at 548 to Figure 6. a) Catalytic performance of the MS@Pt as a function of reaction temperature at a feed pressure of
0.1 MPa; b) catalytic performance of the MS@Pt as a function of pressure at 623 K; c) Arrhenius plots of ln(r)
673 K and 0.1, 1, 2 and 3 MPa.
versus 1/T for CO2 hydrogenation with different catalysts; d) catalytic performance of the MS@Pt as a function of
Firstly, we found that the MS@Pt
time-on-stream at 623 K and 0.1 MPa; e) comparison of catalytic performance of Pt catalysts over different sup-
catalyst was very selective to-
ports at 0.1 MPa (feed pressure); and f) comparison of catalytic performance of the MS@Pt catalysts with different
wards the CO formation by rare-earth-metal dopants at 0.1 MPa.
Chem. Eur. J. 2015, 21, 1882 – 1887
www.chemeurj.org
1885
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
polyvinylpyrrolidone (PVP)) and core materials (e.g., Au, Ag, Au–Pd,
MoO2, Fe3O4, CNT and CNT@Au–Pd). To prepare the Au@SiO2, for
example, PVP-stabilised Au NPs (3 mL) was dispersed into a mixture
of isopropanol (25 mL) and H2O (5 mL) with ultrasonication for
0.5 h. Then ammonium (0.6 mL) and TEOS (1.2 mL) were added
successively into the mixture. The mixture was further stirred for
16 h at room temperature before collecting the solid by centrifug-
ing and washing with ethanol twice. Details on the preparation
processes of other core–shell structures Ag@SiO2, Au–Pd@SiO2,
MoO2@mSiO2, Fe3O4@SiO2, CNT@mSiO2 and CNT@Au–Pd@mSiO2
can be found in the Supporting Information, section SI-1.
tive surface sites) and the low metal loading (0.52 wt%). The
operating stability of these catalysts is remarkable; Figure 6d
illustrates the catalyst performance of the MS@Pt over a period
of 24 h, showing that there was no significant decrease in
both the activity and selectivity. Our TEM investigation on the
three types of catalysts after reactions proves that there is no
major change in particle size and morphology (see the Sup-
porting Information, section SI-13).
In order to demonstrate the synergetic effect between the
Pt and MS, we also studied the catalytic performance of Pt cat-
alysts on other support materials (e.g., SBA-15 and Ni-silicate;
see the Supporting Information, section SI-13). Encouragingly,
our MS based Pt catalysts exhibited the highest conversion
(Figure 6e). Concerning the product selectivity, no alcohol
products (only CO and methane) were detected with the SBA-
15@Pt and Ni-silicate@Pt catalysts. Based on these results, it is
believed that the production of methanol is due to the pres-
ence of manganese. Previous studies have shown that Mn can
act as both electronic modifier and structural promoter for cat-
alysts, which favours the hydrogenation of CO2 into value-
added oxygenates or olefin over methane.[16] Furthermore, it
has been reported that the catalytic performance for CO2 hy-
drogenation can be enhanced by adding rare-earth promoters
to the catalyst.[17] Herein, the MS shell of MS@Pt catalyst was
doped with five different rare-earth elements (namely, Y, La,
Ce, Nd and Sm).[6] The effect of the dopant is shown in Fig-
ure 6 f. In general, catalysts with rare-earth doped MS shell
could bring about appreciable enhancement in activity. For ex-
ample, the reaction rate of Y-doped MS@Pt catalyst is about
two times higher than that of the pristine MS@Pt at 548 K. The
apparent activation energy of CO2 hydrogenation over the
doped catalysts decreases in the order of Nd (50 kJmolꢀ1)>
Sm (44 kJmolꢀ1)>La (43 kJmolꢀ1)>Ce (38 kJmolꢀ1)>Y
(35 kJmolꢀ1). The enhanced activities could be ascribed to the
change of surface basicity (or amount of oxygen vacancy) of
the catalysts, which in turn promotes the adsorption and acti-
vation of CO2.[18]
Synthesis of NP@MS yolk–shell: In a typical synthesis, the above-
prepared NPs@SiO2 core–shell (10–50 mg) was dispersed in H2O
(10 mL) with ultrasonication for 0.5 h. During this period, MnSO4
(50–150 mg) and sodium malate (50–200 mg) were dissolved in
H2O (10 mL). Afterwards, the solid suspension was poured into the
MnSO4 and sodium malate solution under vigorous stirring for
10 min. The mixture was then transferred to a Teflon-lined auto-
clave and hydrothermal treated at 1808C for 12 h. After the reac-
tion, the yolk–shell products were collected by centrifugation, fol-
lowed by washing with ethanol (40 mLꢁ2) and drying at 608C
overnight. Information on the optimised parameters during the
synthesis for different structures can be found in the Supporting
Information, section SI-1.
Synthesis of MS@Pt catalysts: Briefly, SiO2 beads (60 mg) were dis-
persed in H2O (10 mL) with ultrasonication for 0.5 h. During this
period, of MnSO4 (136.6 mg), sodium malate (180 mg) and K2PtCl4
(0.6 mL, 75 mm) were dissolved in H2O (10 mL), into which the sus-
pension of SiO2 beads was poured under vigorous stirring for
10 min. The mixture was then transferred to a Teflon-lined stainless
steel autoclave and hydrothermally treated at 1808C for 12 h. The
brown product was harvested by centrifugation, followed by wash-
ing with ethanol (40 mLꢁ2) and drying at 608C overnight. Rare-
earth-metal-doped MS@Pt catalysts were also prepared in a similar
way, except that MnSO4 (136.6 mg) was mixed together with rare-
earth metal salts (0.04 mmol) such as Y(NO3)3, La(NO3)3, Ce(NO3)3,
Nd(NO3)3, and Sm(NO3)3.
Catalytic evaluation: The CO2 hydrogenation was carried out in
a continuous flow fixed bed reactor made of stainless-steel by
using 0.20 g of the catalyst. A feed mixture (H2/CO2 =3) was intro-
duced with the space velocity of 9600 mLgcatalystꢀ1 hꢀ1. The temper-
ature was measured with a thermocouple located at the centre of
the catalyst bed. Concentrations of reactants and products were
measured online by a gas chromatographic system.
In summary, pure or doped nanobubble-like manganese sili-
cate can be used as a versatile platform for the design and
synthesis of a range of spherical and tubular catalysts. In addi-
tion to the flexibility in structural architecture and composi-
tional tailoring, the attained interbubble space can serve as
a new form of mesoporosity for protecting and/or supporting
catalytically active phases. Because of their high chemical and
thermal stability, the “bubble catalysts” can also work effective-
ly in both solution and gas environments.
Acknowledgements
The authors would like to thank the financial support provided
by the Ministry of Education, Singapore, NUS, and GSK Singa-
pore. This project is also partially funded by the National Re-
search Foundation (NRF), Prime Minister’s Office, Singapore
under its Campus for Research Excellence and Technological
Enterprise (CREATE) program. The authors would also like to
acknowledge Mr. Yuan Sheng for helpful discussion in catalytic
experiments.
Experimental Section
Detailed information on the chemicals used, treatments, and char-
acterisation of products can be found in the Supporting Informa-
tion section SI-1. The following are brief experimental highlights.
Synthesis of NP@SiO2 core–shell: The silica beads were prepared
with a modified Stçber process through hydrolysis/condensation
of tetraethyl orthosilicate (TEOS) in aqueous solutions containing
ethanol (or isopropanol), ammonium (or triethanoamine), surfac-
tants (e.g., CTAB, cetyl trimethylammonium chloride (CTAC) and
Keywords: CO2 hydrogenation · heterogeneous catalysis ·
hollow structures · manganese silicate · mesoporous materials
Chem. Eur. J. 2015, 21, 1882 – 1887
www.chemeurj.org
1886
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[1] a) H. C. Zeng, Acc. Chem. Res. 2013, 46, 226–235; b) C. C. Yec, H. C.
Zeng, J. Mater. Chem. A 2014, 2, 4843–4851; c) Y. Li, J. Shi, Adv. Mater.
2014, 26, 3176–3205; d) J. Liu, S. Z. Qiao, S. Budi Hartono, G. Q. Lu,
Angew. Chem. Int. Ed. 2010, 49, 4981–4985; Angew. Chem. 2010, 122,
5101–5105; e) Y. Yang, J. Liu, X. Li, X. Liu, Q. Yang, Chem. Mater. 2011,
23, 3676–3684.
[10] Z. Li, H. C. Zeng, J. Am. Chem. Soc. 2014, 136, 5631–5639.
[11] J. Huang, C. Yan, K. Huang, J. Colloid Interface Sci. 2009, 332, 60–64.
[12] a) S. H. Joo, J. Y. Park, C.-K. Tsung, Y. Yamada, P. Yang, G. A. Somorjai,
Nat. Mater. 2009, 8, 126–131; b) P. M. Arnal, M. Comotti, F. Schꢃth,
Angew. Chem. Int. Ed. 2006, 45, 8224–8227; Angew. Chem. 2006, 118,
8404–8407.
[2] C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli, J. S. Beck, Nature
1992, 359, 710–712.
[3] N. Lang, A. Tuel, Chem. Mater. 2004, 16, 1961–1966.
[13] W. Wang, S. Wang, X. Ma, J. Gong, Chem. Soc. Rev. 2011, 40, 3703–3727.
[14] J. A. Rodriguez, J. Evans, L. Feria, A. B. Vidal, P. Liu, K. Nakamura, F. Illas,
J. Catal. 2013, 307, 162–169.
[4] a) J. Wu, Y.-J. Zhu, S.-W. Cao, F. Chen, Adv. Mater. 2010, 22, 749–753;
b) S. Fujihara, T. Maeda, H. Ohgi, E. Hosono, H. Imai, S.-H. Kim, Langmuir
2004, 20, 6476–6481; c) J. N. Kondo, K. Domen, Chem. Mater. 2008, 20,
835–847; d) Z. Wei Seh, W. Li, J. J. Cha, G. Zheng, Y. Yang, M. T. McDo-
well, P.-C. Hsu, Y. Cui, Nat. Commun. 2013, 4, 1331.
[15] M. C. Romꢄn-Martꢅnez, D. Cazorla-Amorꢆs, A. Linares-Solano, C. Salinas-
Martꢅnez de Lecea, Appl. Catal. A 1996, 134, 159–167.
[16] a) R. W. Dorner, D. R. Hardy, F. W. Williams, H. D. Willauer, Appl. Catal. A
2010, 373, 112–121; b) L. Xu, Q. Wang, D. Liang, X. Wang, L. Lin, W. Cui,
Y. Xu, Appl. Catal. A 1998, 173, 19–25.
[5] a) Y. Meng, W. Song, H. Huang, Z. Ren, S.-Y. Chen, S. L. Suib, J. Am.
[17] a) N. Perkas, G. Amirian, Z. Zhong, J. Teo, Y. Gofer, A. Gedanken, Catal.
Lett. 2009, 130, 455–462; b) Y. S. She, Q. Zheng, L. Li, Y. Y. Zhan, C. Q.
Chen, Y. H. Zheng, X. Y. Lin, Int. J. Hydrogen Energy 2009, 34, 8929–
8936.
[18] a) P. Gao, F. Li, N. Zhao, F. Xiao, W. Wei, L. Zhong, Y. Sun, Appl. Catal. A
2013, 468, 442–452; b) X. Guo, D. Mao, G. Lu, S. Wang, G. Wu, J. Mol.
Catal. A 2011, 345, 60–68.
ˇ
ˇ
Chem. Soc. 2014, 136, 11452–11464; b) N. N. Tusar, D. Maucec, M.
ˇ
ˇ ˇ
Rangus, I. Arcon, M. Mazaj, M. Cotman, A. Pintar, V. Kaucic, Adv. Funct.
Mater. 2012, 22, 820–826; c) F. Jiao, H. Frei, Chem. Commun. 2010, 46,
2920–2922.
[6] C. C. Yec, H. C. Zeng, ACS Nano 2014, 8, 6407–6416.
[7] M. Haruta, Catal. Today 1997, 36, 153–166.
[8] a) T. Arlt, T. Armbruster, R. Miletich, P. Ulmer, T. Peters, Phys. Chem. Miner.
1998, 26, 100–106; b) V. Thangadurai, S. Narayanan, D. Pinzaru, Chem.
Soc. Rev. 2014, 43, 4714–4727.
Received: October 17, 2014
[9] D. P. Wang, H. C. Zeng, Chem. Mater. 2011, 23, 4886–4899.
Published online on December 8, 2014
Chem. Eur. J. 2015, 21, 1882 – 1887
www.chemeurj.org
1887
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim