Supported silver catalysts prepared via melt infiltration: Synthesis, characterization and performance in selective hydrogenation

Article abstract of DOI:10.1016/j.cattod.2020.03.017

Heterogeneous supported catalysts are often synthesized by impregnation or precipitation methods. Recently, melt infiltration has emerged as an alternative method that allows high metal loadings and eliminates the need for a solvent, but challenges arise regarding control over the particle size and distribution. In this work, melt infiltration for the synthesis of supported silver catalysts is explored. The narrow pore size distribution of the chosen ordered mesoporous silica support, SBA-15, allowed in depth in-situ and ex-situ characterization of the infiltration of the precursor, molten silver nitrate, into the support and its subsequent decomposition to form metallic silver nanowires or nanoparticles. The heat treatment parameters during decomposition played a key role in determining whether nanowires or nanoparticles were formed. The supported silver catalysts containing high silver weight loadings were investigated in the selective hydrogenation of cinnamaldehyde, where the silver nanowires showed superior activity and selectivity over the nanoparticles. Hence, melt infiltration shows great promise for the synthesis of supported silver catalysts containing high silver weight loadings, which are applicable in, e.g., selective oxidation or hydrogenation reactions.

Full text of DOI:10.1016/j.cattod.2020.03.017

Catalysis Today xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Supported silver catalysts prepared via melt infiltration: Synthesis,
characterization and performance in selective hydrogenation
P.H. Keijzer, B. Donoeva, K.P. de Jong, P.E. de Jongh*
Inorganic Chemistry and Catalysis, Debye Institute for Nanomaterials Science, Utrecht University, Utrecht, the Netherlands
A R T I C L E I N F O
A B S T R A C T
Keywords:
Heterogeneous supported catalysts are often synthesized by impregnation or precipitation methods. Recently,
melt infiltration has emerged as an alternative method that allows high metal loadings and eliminates the need
for a solvent, but challenges arise regarding control over the particle size and distribution. In this work, melt
infiltration for the synthesis of supported silver catalysts is explored. The narrow pore size distribution of the
chosen ordered mesoporous silica support, SBA-15, allowed in depth in-situ and ex-situ characterization of the
infiltration of the precursor, molten silver nitrate, into the support and its subsequent decomposition to form
metallic silver nanowires or nanoparticles. The heat treatment parameters during decomposition played a key
role in determining whether nanowires or nanoparticles were formed. The supported silver catalysts containing
high silver weight loadings were investigated in the selective hydrogenation of cinnamaldehyde, where the silver
nanowires showed superior activity and selectivity over the nanoparticles. Hence, melt infiltration shows great
promise for the synthesis of supported silver catalysts containing high silver weight loadings, which are ap-
plicable in, e.g., selective oxidation or hydrogenation reactions.
Melt infiltration
Silver
SBA-15
Nanowires
Cinnamaldehyde hydrogenation
1
. Introduction
Silver catalysts are industrially used on a large scale to produce for-
alternative method [19]. Melt infiltration makes use of capillary forces
to let a liquid precursor enter a porous material spontaneously. Whether
or not melt infiltration takes place depends on the degree of wetting
and the related contact angle between the liquid and solid. Melt in-
filtration can be used for the synthesis of materials with a high weight
loading. Moreover, since it is a solvent free technique, no drying step or
waste water treatment is needed. However, controlling the distribution
of the precursor over the support is less straightforward than when
using the synthesis methods mentioned before. For example, below
maximum pore filling, some pores might be completely filled, where
others are empty. To prevent the uneven distribution of the precursor,
the pore volume of the support can be reduced by partially filling the
pores with surfactants or polymers [20,21].
maldehyde from methanol, and to selectively oxidize ethylene to ethylene
oxide, but are also relevant for other selective oxidation and hydrogena-
tion reactions [1–5]. If the active phase of a catalyst reside with metal
nanoparticles, these are most often stabilized on a high specific surface
area support material, usually a metal oxide. Porous metal oxides with an
ordered pore structure and narrow pore size distribution are preferred
over disordered systems for fundamental research on both the synthesis
and the performance of the catalyst. One of these ordered systems is SBA-
1
5, a mesoporous silica first synthesized by Zhao et al. [6] which has been
used extensively as a model support since its discovery [7–11].
In industry mostly impregnation and precipitation are used for the
synthesis of heterogenous catalysts [12–15]. For pore volume impreg-
nation, on the one hand, the maximum weight loading is constrained by
the limited solubility of precursors, and moreover, the inevitable drying
step may lead to large intra- and interparticle heterogeneities in metal
distribution [16–18]. Precipitation, on the other hand, allows high
loadings but is accompanied by the production of large amounts of
waste water [13].
There are restrictions to which precursors can be used for melt in-
filtration. For example, the precursor should not react with the support
and only precursors that have a relative low melting temperature, at
least below their decomposition temperature, can be used. Among in-
organic salts, metal nitrates are particularly interesting because of their
wide availability and easy decomposition into metal oxides. In litera-
ture, mainly metal nitrate hydrates have been reported, as of their low
melting temperature resulting from dissolution of the salt in its own
crystal water [19,22,23]. For several catalysts synthesized via melt in-
filtration using metal nitrates, the decomposition pathways are well
For the synthesis of nanostructured materials for applications such
as catalysis and batteries, melt infiltration has been proposed as an
⁎
Corresponding author.
E-mail address: P.E.deJongh@uu.nl (P.E. de Jongh).
https://doi.org/10.1016/j.cattod.2020.03.017
Received 20 November 2019; Received in revised form 4 February 2020; Accepted 12 March 2020
0
920-5861/ © 2020 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(
http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Please cite this article as: P.H. Keijzer, et al., Catalysis Today, https://doi.org/10.1016/j.cattod.2020.03.017

P.H. Keijzer, et al.
Catalysis Today xxx (xxxx) xxx–xxx
understood, and catalyst morphology and effectiveness were reported
2.3. Characterization
[
21,23–25]. To the best of our knowledge, this is not the case for silver
nitrate, whereas the low melting temperature of silver nitrate (212 °C)
in combination with its high decomposition temperature (440 °C)
makes this non-hydrated metal nitrate very interesting for the melt
infiltrated synthesis of high weight loading silver catalysts. Moreover,
the work of Worboys et al. already shows the promise of this synthesis
method, as it was reported that the pores of SBA-15 can be infiltrated
with silver nitrate [26].
The pore size, pore volume and surface area of the synthesized SBA-
15 were analyzed using N -physisorption. Isotherms were measured at
2
−196 °C on a Micromeritics TriStar 3000 apparatus. The specific surface
area of the support was calculated using the BET equation (0.05 < p/
p0 < 0.25). Differential scanning calorimetry (DSC) measurements were
performed on a METTLER TOLEDO HP DSC 1 system by heating the
sample in a 40 μL sealed aluminum pan to 250 °C with a heating ramp of
−1
−1
In this work, the use of melt infiltration for the synthesis of highly
loaded silver catalysts supported on silica is described. SBA-15 was used
as model support, as its narrow pore size distribution provides the op-
portunity to follow the infiltration process with silver nitrate using dif-
ferential scanning calorimetry. After the infiltration process is complete,
the decomposition of the silver nitrate into metallic silver was followed
both in-situ and ex-situ using x-ray diffraction and electron microscopy,
showing how the morphology of the final catalyst is influenced by the
decomposition parameters. Finally, we show the application of these
catalysts in the selective hydrogenation of cinnamaldehyde, where the
performance is coupled to the structural properties of the catalysts.
5 °C min , at 2 bar, under an argon flow (10 mL min ). The SBA-15,
the AgNO /SBA-15 composites before and after melt infiltration and the
3
obtained Ag/SBA-15 composites were analyzed by transmission electron
microscopy (TEM) using a Tecnai20FEG and Talos F200X microscope
both operated at 200 kV. For details on the surface and volume averaged
particle size analysis, see Supplementary Information, Section A. TEM
samples were prepared by dropwise addition of a dispersion of the ma-
terial in ethanol (SBA-15 and Ag/SBA-15) or hexane (AgNO /SBA-15)
3
obtained by sonication to a TEM grid. To prevent decomposition of the
silver nitrate by the electron beam, the TEM grid with the AgNO /SBA-
15 composite and grid holder were cooled with liquid nitrogen to circa
180 °C. Crystal phase analysis was performed with X-Ray Diffraction
3
−
2
. Experimental
(XRD), in-situ on a Bruker D8 Phaser and ex-situ on a Bruker D2 Phaser
diffractometer, both equipped with a Co Kα source (λ = 0.1789 nm).
The composite structures were analyzed by comparing the XRD dif-
fractograms with crystal structures from the PDF-4 + 2016 database. For
the catalysts with a bimodal distribution of silver particles, TOPAS V5
software was used to deconvolute silver peaks into two silver phases and
to calculate their crystallite sizes. Diffuse-Reflectance UV/Vis spectra of
the catalysts were obtained by measuring circa 100–200 mg material in
the range of 800-200 nm with a 4 nm interval and a 4 nm slit size using a
Perkin Elmer Lambda 950S UV/Vis-NIR spectrophotometer with an in-
tegrating sphere detector.
2.1. Synthesis of the support
SBA-15 was obtained via sol-gel synthesis using Pluronic P123
(
EO20PO70EO20, average Mw = 5800, Aldrich) as template and tetra-
ethyl orthosilicate (TEOS, > 99%, Aldrich) as silica precursor. Following
the procedure of Lee et al. [27], 23.4 g Pluronic P123 was dissolved in
6
06.8 g deionized water and 146.4 g hydrochloric acid (HCl, 37 wt%,
fuming, Merck, analysis grade) in a 1 L polypropylene bottle (cylindrical,
height 19.7 cm, diameter 10.5 cm). After the mixture was vigorously
stirred for at least 3 h in an oil bath at 55 °C, the stirring rate was set to
6
00 rpm (PFTE stirring bar, 50 x 7 mm) and 50 g TEOS was added at
2.4. Catalytic testing
once. After 2 min, the stirring bar was removed and the lid of the bottle
was closed tightly. The mixture was first kept for 24 h at 55 °C, then for
The catalysts (56 wt% Ag/SiO
2
) were investigated for the liquid
2
4 h at 90 °C. Next, the material was filtrated and washed with deionized
phase hydrogenation of cinnamaldehyde. The reaction was carried out
water using a Büchner funnel until the pH of the filtrate was around 5–6
and hence no HCl was left in the solution). The filtrate was dried at 60
C for 2–3 days and afterwards crushed into a fine powder and calcined
at 70 °C and under 40 bar H
2
pressure in an autoclave reactor con-
(
taining 125 μL t-cinnamaldehyde, 75 mg catalyst, 100 μL tetradecane
(internal standard), 6 mL isopropanol and 1 mL milliQ water. The re-
action mixture was stirred at 900 rpm and samples were taken at reg-
ular time intervals of 1 or 2 h and analyzed using a Varian 430 GC.
°
−1
at 550 °C for 6 h (heating ramp 1 °C min ) in static air.
2.2. Synthesis of the catalysts
3
. Results and discussion
Silver was deposited on the SBA-15 via melt infiltration with silver
nitrate and subsequent decomposition of the silver nitrate. In a typical
3.1. Melt infiltration process
synthesis, silver nitrate (AgNO ≥ 99%, Sigma Aldrich) and SBA-15 were
3
2
−1
physically mixed in a 2:1 wt ratio inside a glovebox using a pestle and
mortar for 5 min. This ratio corresponded to half of the pore volume of
silica being filled with silver nitrate, resulting in a theoretical loading of
The SBA-15 particles had a specific surface area of 800 m g and a
well-defined pore structure with mesopores of 6.8 nm (see
Supplementary Information Fig. B1). The monodisperse and ordered
pore system of SBA-15 makes it possible to follow the melt infiltration
process with silver nitrate by differential scanning calorimetry (DSC).
With this technique, information about phase transitions of materials is
obtained, as during these phase transitions, heat is released or consumed.
5
6 wt% Ag on SiO . The physical mixture was heated for 20 h at 250 °C
2
−1
under a top-down nitrogen flow (100 mL min ). The silver nitrate was
decomposed by thermal treatment at 425 °C for 2 h (heating ramp of 2 °C
−1
−1
min ) in nitrogen flow (100 mL min ) or by reduction for 2 h in 10%
−1
hydrogen in nitrogen flow (100 mL min ). This reduction either started
at room temperature, whereupon the temperature was increased to 130
In Fig. 1, DSC thermograms of AgNO /SBA-15 (left frame) and
3
macrocrystalline AgNO (right frame) are shown. Curve A shows the
3
−1
°
C (heating ramp of 0.1 or 1 °C min ) or started at 250 °C, after prior
measured heat flow while heating a physical mixture of silver nitrate
and SBA-15 to 250 °C. Curves B and C show the same sample as in curve
A, but after in-situ and ex-situ melt infiltration, respectively. The amount
of silver nitrate corresponded to 50% of the pore volume of the SBA-15.
heating in nitrogen flow. As reference, a 15 wt% Ag/SBA-15 catalyst was
prepared via impregnation and drying. SBA-15 was dried under vacuum
for 2 h at 250 °C prior to impregnation with an aqueous AgNO solution
3
(
1.64 M, 90 % of pore volume). After impregnation, the composite was
In all heating curves, including curves D–F of macrocrystalline AgNO ,
3
dried overnight under vacuum at room temperature. Next, the silver
nitrate was decomposed by thermal treatment at 500 °C for 2 h (heating
an endothermic peak is visible around 212 °C. In curve A, an additional
endothermic peak is present at 200 °C, while in curves B and C, a second
peak is observed at 148 °C.
−1
ramp of 1 °C min ) in static air. After cooling down, the material was
−1
reduced for 2 h at 250 °C (heating ramp of 5 °C min ) in 10 % hydrogen
All peaks in Fig. 1 are ascribed to phase transitions of AgNO , as in
3
−1
in nitrogen flow (100 mL min ).
the measurement of pristine SBA-15, no peaks were observed (see
2

P.H. Keijzer, et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 1. DSC thermograms of AgNO
3
/SBA-15
composites (left) and macrocrystalline AgNO
3
.
−
1
(
right) recorded during heating at 5 °C min
Curve A: physical mixture of AgNO
3
and SBA-
1
5, kept at 250 °C for 4 h; curve B: second
heating of composite shown in curve A; curve
C: ex-situ melt infiltrated material (20 h, 250
°
C, N flow), curves D, E and F: first, second
2
and third heating of macrocrystalline AgNO ,
3
respectively. Only thermograms during heating
are displayed and they are offset for clarity.
Supplementary Information, Fig. B2). The endothermic peak at 200 °C
in curve A is ascribed to a solid-solid phase transition of silver nitrate
from the orthorhombic phase to the rhombohedral phase. X-ray dif-
fraction (XRD) on the cooled down material confirms the transition to
this rhombohedral phase and shows that the transition is irreversible
of a peak corresponds directly to the amount of molten silver nitrate, a
quantitative analysis on the amount of silver nitrate that infiltrated the
pores is possible. From the peaks at 212 °C in curves A and C, it was
calculated that 93% of the silver nitrate infiltrated the pores of the SBA-
15 during the ex-situ melt infiltration. The higher efficiency of this ex-
situ melt infiltration is attributed to the improved contact between the
silver nitrate and the SBA-15 as result of the longer heating time.
Another method to analyze the outcome of the infiltration process is
by transmission electron microscopy (TEM). With TEM, the pore
structure and filling of the SBA-15 particles were imaged. In the left
frame of Fig. 2, a TEM image of pristine SBA-15 is shown. With the pore
structure of a SBA-15 particle aligned with the electron beam, the or-
dered structure became visible. The inset shows the hexagonal ordering
of the pores at the end of a SBA-15 particle. The right frame of Fig. 2
(
see Supplementary Information, Fig. C1). Therefore, this transition
only occurred during the first heating ramp, and no peaks are visible at
2
00 °C in curves B and C. The solid-solid phase transition is not ob-
served when silver nitrate was heated in the absence of silica (see
Supplementary Information, Fig. C2) but has been reported before upon
heating a composite of silver nitrate and silica [26].
The melting of macrocrystalline silver nitrate gave an endothermic
peak at 212 °C (curves D–F), which is in agreement with literature [28].
This peak is at the same position in every thermogram of macro-
crystalline AgNO
3
(curves D–F), while this peak slightly shifted to lower
shows the AgNO /SBA-15 composite (56 wt% Ag/SBA-15) after melt
3
temperatures for the AgNO
3
/SBA-15 composites in curve B and C
infiltration. Silver nitrate in an electron beam results in immediate
decomposition, however, by cooling the TEM grid in liquid nitrogen, we
were able to stabilize and image the silver nitrate. Here, approximately
half of the pores are filled with silver nitrate and therefore, instead of
bright pores, dark lines are visible within the silica structure, which
confirms the successful infiltration.
compared to curve A. Presumably, during prior heating of materials in
curves B and C in the presence of the SBA-15, the macrocrystalline
silver nitrate spread out over the external surface of the SBA-15 parti-
cles, which slightly lowered the melting temperature.
By comparing the second thermogram on the AgNO /SBA-15 com-
3
posite (curve B) with the second thermogram on macrocrystalline AgNO
curve E), it becomes evident that infiltration took place. In curve E, only
the peak corresponding to melting of macrocrystalline silver nitrate, at
12 °C is present, whereas in curve B, an extra peak at 148 °C is visible,
3
(
3.2. Decomposition of the AgNO3 precursor
2
After the distribution of the silver precursor over the support ma-
terial, the precursor was decomposed to obtain metallic nanoparticles.
For this, several methods are available, such as thermal decomposition
(Eq. 1), or reduction of silver nitrate (Eq. 2). The particle size and
morphology are influenced by the gas atmosphere, heating ramp and
final temperature [12,30]. Using in-situ XRD, we followed the decom-
which corresponds to the melting of infiltrated silver nitrate. At the start
of heating curve A, the pores are empty, and therefore, only the melting
of macrocrystalline silver nitrate is observed. Upon melting, part of the
silver nitrate entered the pores of the SBA-15. The melting temperature
of material inside nanopores is significantly lower, as a result of pore
confinement [29]. Since the pores of the SBA-15 are uniform in size, the
melting point depression is the same for all silver nitrate inside the pores.
DSC thermograms of infiltrated material therefore show a single extra
peak at lower temperature. The melting point depression of 64 °C, is
close to the expected maximum decrease in melting temperature of circa
position of silver nitrate by heating the melt infiltrated AgNO /SBA-15
3
composite under inert, oxidizing or reducing atmosphere up to 500 °C.
2
AgNO3(l)
2 Ag(s) + O2(g) + 2 NO2(g)
(1)
AgNO + H2(g)
Ag+ H2 O+ NO2(g)
(2)
3
7
0 °C, calculated using the Gibbs-Thomson equation and assuming
complete wetting, [29], see Supplementary Information, Section D.
However, the low intensity of this peak and the high intensity of the peak
corresponding to the melting of macrocrystalline silver nitrate indicate
that most silver nitrate remained outside the pores and the in-situ melt
infiltration is not very effective.
Fig. 3 shows the diffractograms at different temperatures, where the
left frame shows the thermal decomposition in He flow and the right
frame shows its reduction in a 5% H in He flow. In both graphs, peaks
2
corresponding to silver nitrate in the rhombohedral structure are visible
up to 100 °C, while large differences are observed at higher tempera-
tures. In inert atmosphere, the peaks corresponding to silver nitrate
decreased in intensity when the temperature was increased to 150 °C,
and at 200 °C, crystalline phases were no longer observed. This is in line
with the melting of the silver nitrate in the pores of the SBA-15, which
takes place at the depressed melting temperature around 150 °C (see
Fig. 1). When the composite reached 450 °C, new peaks were observed,
Curve C in Fig. 1 shows the heating of the composite after ex-situ
melt infiltration, where the mixture was heated for 20 h at 250 °C,
under nitrogen flow. The intensity of the peak of melting silver nitrate
inside the pores at 148 °C is significantly higher than for curve B, while
the intensity of the peak corresponding to the melting of macro-
crystalline silver nitrate at 212 °C is much lower. Since the surface area
3

P.H. Keijzer, et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 2. TEM images of pristine SBA-15 (A) and ex-situ melt infiltrated (20 h, 250 °C, N
2
flow) AgNO /SBA-15 (B).
3
which increased in intensity upon further heating to 500 °C. These
peaks correspond to silver, indicating the decomposition of the silver
nitrate into metallic silver, as is expected for macrocrystalline silver
nitrate which is reported to decompose at 450 °C [28]. More in-depth
analysis on the composite showed that next to the formation of silver,
reduced at 125 °C, even before melting of the silver nitrate occurred.
Furthermore, no formation of silver silicate was observed, probably
because of the formation of metallic silver at much lower temperature.
Fig. 4 shows TEM images of Ag/SBA-15 catalysts prepared by ex-situ
decomposition of the infiltrated material. For more TEM images giving a
larger scale overview, see Supplementary Information, Fig. E2. In frame A,
silver nitrate was thermally decomposed by heating to 425 °C in nitrogen
flow. Just as with the material heated during the in-situ XRD measure-
ments, SBA-15 particles no longer contained ordered pores, due to the
formation of silver silicate. In order to prevent the formation of silver si-
licate, the decomposition of the silver nitrate was in all other cases con-
ducted under reductive atmosphere, resulting in the materials shown in
Fig. 4, frames B, C and D. Silver nanowires which filled some of the pores
of the SBA-15 completely, are visible in frames B and C, while small silver
nanoparticles distributed over the pores are visible in frame D.
also some silver silicate, Ag10Si O13, had formed (see Supplementary
4
Information, Fig. E1). Literature on the formation of silver silicate by
reaction of silver nitrate and silica has, to the best of our knowledge,
not been reported before. The liquid silver nitrate presumably reacted
with the silica support at high temperature to form silver silicate and
NO . As is visible in the TEM image of this material (Fig. 3, left frame),
x
the pore system of the SBA-15 was completely destroyed, indicating
that this silver silicate formed at the expense of the silica pore walls.
Analogous to the composite heated in inert atmosphere, the peaks in
the diffractogram after heating at 30–50 °C in reducing atmosphere
correspond to the rhombohedral structure of silver nitrate. However, in
this case, already at 100 °C a broad peak of metallic silver appeared.
The diffractograms taken at 125 °C and higher contain only peaks of
metallic silver. This shows that the silver nitrate was completely
The silver nanowires were obtained by starting the reduction at
room temperature by applying a flow of 10 % H
2
in N , and by slowly
2
heating the composite to 130 °C in this atmosphere. Both when a
heating ramp of 0.1 °C min−1 or 1 °C min
−1
was used, SBA-15 with
Fig. 3. In-situ XRD of melt infiltrated AgNO
3
/SBA-15 treated in inert atmosphere (helium, left) or reducing atmosphere (5 % hydrogen in helium, right), with the
inset of a TEM image of the material after heating.
4

P.H. Keijzer, et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 4. Ag/SBA-15 composites after decomposition of the silver nitrate in nitrogen (A) and in 10 % hydrogen (B–D). Silver nanowires inside the pores of the SBA-15
−1
−1
were obtained when reduction was started at room temperature with a heating ramp of 1 °C min (B) or 0.1 °C min (C), while starting the reduction at 250 °C led
to the formation of silver nanoparticles inside the pores (D).
silver nanowires inside the pores were obtained. Although the heating
ramp was different, the silver nanowires were very similar in both
catalysts, and the pores of the SBA-15 were either completely filled, or
completely empty. The similarities of these samples is further supported
by the similar plasmon absorption peaks in the UV/Vis spectra, with a
maximum absorption at 388 nm for both materials (see Supplementary
Information, Fig. E3). Considering that volume of silver nitrate is 3.8
times larger than for metallic silver, it is striking that completely filled
pores were observed. However, the formation of silver nanowires in
SBA-15 has been reported before and is ascribed to the mobility of in-
termediate phases during the decomposition [26,30].
some particles of 5–10 nm also present outside of the pores that shift
the average to a higher value. Furthermore, silver nanoparticles below
2 nm give a very broad peak with a low intensity, barely visible in the
diffractogram, and therefore, they are at least partly unaccounted for.
The amount of silver inside versus outside of the pores was esti-
mated by assuming that the small crystallites were inside the pores of
the SBA-15, while the larger crystallites were outside. These values are
given in Table 1 and are clearly different for the three composites.
When considering the two materials decomposed in H and containing
2
silver nanowires, the heating ramp during the decomposition had a
crucial influence on how much silver resided in the pores upon the
decomposition of the silver nitrate. Much more silver remained in the
By altering the heat treatment parameters, the morphology of the
silver was changed considerably. When the infiltrated material was first
heated to 250 °C in inert atmosphere, and the reduction was started at
this temperature, the pores of the SBA-15 contained silver nanoparticles
of 2.3 ± 0.7 nm (Fig. 4, frame D). The silver nanoparticles were uni-
formly distributed over the pores. The nanoparticles show a maximum
plasmonic absorption at 368 nm (see Supplementary Information, Fig.
E3). The intensity of this peak is lower than for the nanowires and the
peak location is at a shorter wavelength, confirming the formation of
smaller silver nanoparticles. By increasing the temperature before
starting the reduction, probably more nuclei were created, just as with
the “hot injection” synthesis method of colloidal particles [31]. This
prevented the formation of nanowires and facilitated the formation of
many small nanocrystallites.
−1
pores when a very low heating ramp of 0.1 °C min was used (73 %),
−
1
compared to when a relative high heating ramp of 1 °C min was used
(39 %). Likely, the gaseous decomposition products formed during the
reduction pushed some of the silver nitrate out of the pores. By in-
creasing the heating ramp, not only the amount of gas produced in a
short time, but also the amount of mobile species increased, leading to a
higher expulsion of silver nitrate. When the reduction was started at
250 °C after heating in nitrogen (Ag/SBA-15_N ), even less silver re-
2
mained in the pores (only 19%). Here, probably complete reduction of
all silver nitrate took place in a very short time, allowing the formation
of silver nanoparticles, but also producing large volumes of decom-
position products, pushing most of the silver nitrate out of the pores.
Even though part of the silver formed large particles outside the
SBA-15 pores, for all three catalysts, more than 90 % of the of silver
surface atoms was located on the intraporous silver nanowires and
nanoparticles. These percentages were calculated using the obtained
crystallite sizes and percentages of silver present inside the pores, and
the high numbers are a result of the high surface area to volume ratio of
the silver nanowires and nanoparticles in comparison to the large silver
particles outside of the pores. Therefore, we show the synthesis of
supported silver catalysts, in which, by carefully tuning the decom-
position parameters, we are able to tailor the shape of the silver and the
amount of silver inside the pores of the SBA-15.
Next to TEM, XRD was used to determine the average particle size,
since the width and intensity of silver peaks gives information on the
crystallite size of the material [32,33]. In Fig. 5, the XRD diffractograms
of the Ag/SBA-15 composites are shown, in which all peaks correspond
to metallic silver (black curves). The shape of these peaks indicate the
presence of a bimodal size distribution, since the broad base of the peaks
suggests the presence of small crystallites, whereas the sharp tips imply
larger crystallites [33]. Using TOPAS V5 software, each peak in the dif-
fractograms was deconvoluted into two separate peaks, one sharp peak
for large crystallites (red curves) and one broad peak for smaller crys-
tallites (green curves), where the sum of these two peaks, the total fit, is
in good agreement with the experimental data (see Supplementary In-
formation, Fig. E4). In doing so, both the average crystallite sizes and the
percentages of silver in each phase were estimated [34].
3.3. Catalysis
The supported silver catalysts (56 wt% Ag/SBA-15) were in-
vestigated for the hydrogenation of cinnamaldehyde. The reaction
scheme of this reaction is given in Fig. 6. Both the C]C and the C]O
bond of the α,β-unsaturated aldehyde can be hydrogenated, to form
hydrocinnamaldehyde and cinnamyl alcohol, respectively. Cinnamyl
alcohol (CALC) is used in the pharmaceutical, flavor and perfume in-
dustry, and is therefore the preferred product [5,35]. However, the
hydrogenation of the C]C bond is thermodynamically favored [36,37],
showing the need for a selective catalyst. Furthermore, hydrogenation
of both the C]C and the C]O bonds, yielding hydrocinnamyl alcohol,
should be avoided. Although limited in activity, silver has proven to be
more selective towards cinnamyl alcohol than gold, palladium, pla-
tinum and ruthenium catalysts [35].
In Table 1, an overview of the different materials with the measured
particle sizes using TEM and the calculated crystallite sizes from the
deconvoluted peaks and percentages of silver present in the small
crystallites are given. The calculated size of the large crystallites
overlap with the size of the large silver particles on the external surface
area of the SBA-15, present in all three catalysts. The small crystallite
sizes obtained from the XRD data were very similar to the measured
nanowire thicknesses obtained via TEM. The small crystallite size cal-
culated in Ag/SBA-15_N however, was larger than the measured na-
2
noparticle size on TEM images. This is expected, since in the TEM image
analysis, only nanoparticles in the pores were measured, while for XRD,
a volume average of all smaller crystallites was taken, which includes
5

P.H. Keijzer, et al.
Catalysis Today xxx (xxxx) xxx–xxx
Fig. 5. XRD diffractogram of Ag/SBA-15_1 (left), Ag/SBA-15_0.1 (middle) and Ag/SBA-15_N
2
(right) in black and the deconvoluted peaks for small crystallites in
green and large crystallites in red. (For interpretation of the references to colour in the Figure, the reader is referred to the web version of this article).
Table 1
Overview of particle/crystallite sizes of the Ag/SBA-15 composites.
Material
Heating ramp (°C
H
2
(%)
during heating Ag nanowire / nanoparticle
Small / large crystallite
Small crystallitesb (%) Ag surface atoms inside
−
1
diametera (nm)
b
c
min
1
)
sizes (nm)
pores (%)
Ag/SBA-15_1
10
10
0
10.0 ± 1.4
9.1 ± 0.9
2.3 ± 0.7
9.9 / 252
9.7 / 150
5.1 / 227
39
73
19
92
97
91
Ag/SBA-15_0.1 0.1
Ag/SBA-15_N
2
2
a
b
c
Volume averaged sizes determined by TEM.
Calculated from deconvoluted XRD peaks using TOPAS V5 software.
Based on the volume averaged crystallite sizes and fractions calculated using TOPAS V5 software.
Fig. 6. Reaction scheme of cinnamaldehyde hydrogenation.
Fig. 7. Amount of cinnamaldehyde (black circles) and products, cinnamyl alcohol (green squares), hydrocinnamaldehyde (orange triangles) and hydrocinnamyl
alcohol (red diamonds), during the hydrogenation of cinnamaldehyde over supported silver catalysts, Ag/SBA-15_1 (A), Ag/SBA-15_0.1 (B) and Ag/SBA-15_N (C).
O (1 mL), tetradecane (100
2
Conditions: Cinnamaldehyde (1 mmol), catalyst (75 mg 56 wt% Ag/SiO
2
, cinnamaldehyde to Ag mole ratio is 2.6), isopropanol (1 mL), H
2
μL), H
2
(40 bar), 70 °C, 900 rpm stirring. (For interpretation of the references to colour in the Figure, the reader is referred to the web version of this article).
6

P.H. Keijzer, et al.
Catalysis Today xxx (xxxx) xxx–xxx
Table 2
Catalytic properties of Ag/SBA-15 composites.
Catalyst
Shape Ag in pores
Ag in pores [%]
TOFa Ag in pores [molCALC molAg, surf−1 h-1
]
Selectivity to CALC at X50% [%]
Ag/SBA-15_1
Nanowires
39
12
82
70
50
80
Ag/SBA-15_0.1
Nanowires
73
12
Ag/SBA-15_N
2
Nanoparticles
Nanoparticles
19
1.4
1.0
1
5Ag/SBA-15_ref
n.a.
a
TOF for first hour, based on the initial surface averaged particle size determined by TEM.
In Fig. 7, the conversion of cinnamaldehyde over three different Ag/
4. Conclusion
SBA-15 catalysts, and the formation of products versus the reaction
time is shown. The decreasing amount of cinnamaldehyde (black cir-
cles) shows that all three catalysts were active. At low conversions,
mainly cinnamyl alcohol (green squares) and hydrocinnamaldehyde
The synthesis of supported silver catalysts via melt infiltration and
their application in the selective hydrogenation of cinnamaldehyde has
been explored. After melt infiltration, over 90 % of the precursor, silver
nitrate, was inside the pores of the SBA-15. By subsequent reduction,
supported silver catalysts with a weight loading of 56 wt% were ob-
tained. Depending on the temperature of the reduction, either silver
nanowires or silver nanoparticles had formed inside the pores.
However, after reduction, also some silver was located outside the
pores. The external silver particles were significantly larger than those
inside the pores, and could be neglected when evaluating the catalytic
activity. The silver nanowires, although filling the pores of the SBA-15,
were accessible for the reactants and provided more active and selective
catalysts for the hydrogenation of cinnamaldehyde to cinnamyl alcohol
than the silver nanoparticles.
(
orange triangles) formed, whereas hydrocinnamyl alcohol (red dia-
monds) started to form only after high conversion levels (> 90 %) were
reached. Control experiments confirmed that without silver and upon
catalyst removal, no (more) cinnamaldehyde was converted (see Sup-
plementary Information, Fig. F1). As a reference, the reaction was run
with a catalyst containing 15 wt% Ag/SBA-15 prepared via impregna-
tion and drying (see Supplementary Information, Fig. F2).
The three catalysts clearly showed different activities and selectiv-
ities. Full conversion of cinnamaldehyde within 24 or even 7 h was
reached for both catalysts containing silver nanowires (frame A and B),
but not for the catalyst with the silver nanoparticles (frame C). This is
partly explained by the presence of large silver crystallites in the latter,
which do not contribute significantly to the catalysis. To better compare
the activity of the catalysts, the turn over frequencies (TOF) for all three
catalysts and the reference catalyst prepared via impregnation and
drying are listed in Table 2. These TOF values were calculated using the
estimated values for the silver present inside the pores and using initial
surface averaged particle diameters from TEM analysis. For the nano-
wires, only the surface area on the sides of the wires were taken into
account. The obtained TOF values are in line with activities of silver
catalysts in liquid phase hydrogenation of α,β-unsaturated aldehydes
reported in literature [4,38,39].
CRediT authorship contribution statement
P.H. Keijzer: Methodology, Validation, Investigation, Writing - ori-
ginal draft, Visualization. B. Donoeva: Methodology, Writing - review &
editing. K.P. de Jong: Conceptualization, Writing - review & editing,
Supervision, Funding acquisition. P.E. de Jongh: Conceptualization,
Writing - review & editing, Supervision, Funding acquisition.
Declaration of Competing Interest
For both the catalysts with silver nanowires, cinnamyl alcohol is the
main product, whereas similar amounts of hydrocinnamaldehyde and
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
cinnamyl alcohol are produced using Ag/SBA-15_N (see Supplementary
2
Information, Fig. G1). TEM analysis on the used catalysts (see
Supplementary Information, Fig. H1) show that both catalysts containing
silver nanowires were stable and retained their morphology during the
hydrogenation reaction. However, this was not the case for both catalysts
containing silver nanoparticles, which sintered to form significantly
larger particles and nanowires. Since it is unknown when the sintering
took place, no reliable conclusions for activity and selectivity based on
particle size can be drawn. Still, the considerable difference in activity
and selectivity between the catalyst containing nanoparticles or nano-
wires, suggests that large silver facets are favorable for the hydrogena-
tion reaction, which is in agreement with literature on the hydrogenation
of various unsaturated aldehydes [3,37,40].
Acknowledgements
This research was funded by the Netherlands Center for Multiscale
Catalytic Energy Conversion (MCEC), an NWO Gravitation programme
funded by the Ministry of Education, Culture and Science of the gov-
ernment of the Netherlands and by a NWO-Vici grant (16.130.344). The
authors would like to thank Lee Durndell and Jovana Zečević for sci-
entific discussions and Hans Meeldijk (TEM), Dennie Wezendonk
(
XRD), Marjan Versluijs-Helder (XRD), Sander Lambregts (DSC, N -
2
physisorption), Jan Willem de Rijk (DSC) and Pascal Wijten (catalytic
setup) for technical support.
Even though the nanoparticles were not stable, the low activity of
this catalyst compared to the catalysts containing nanowires show that
the silver inside the pores of the SBA-15 is active in the hydrogenation
reaction, and that the activity of the silver outside of the pores can
indeed be neglected. This is even further evidenced by the similar TOF
values for the two catalysts containing different amount of silver na-
nowires. Moreover, this shows that even though the nanowires fill the
pores of the SBA-15, the surfaces of the silver nanowires are accessible
to the reactants. Most likely, the micropores present in the SBA-15
provide accessibility for products and reactants. This means however,
that part of the surface of the silver nanowires is covered by the silica
and therefore, the real active surface area is lower, and hence the real
TOF values are even higher for the nanowires than reported here.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.cattod.2020.03.017.
References
[
[
1] I.E. Wachs, R.J. Madix, The oxidation of methanol on a silver (110) catalyst, Surf.
Sci. 76 (1978) 531–558, https://doi.org/10.1016/0039-6028(78)90113-9.
2] M.O. Ozbek, I. Onal, R.A. Van Santen, Why silver is the unique catalyst for ethylene
epoxidation, J. Catal. 284 (2011) 230–235, https://doi.org/10.1016/j.jcat.2011.08.
0
04.
[3] P. Claus, H. Hofmeister, Electron microscopy and catalytic study of silver catalysts:
structure sensitivity of the hydrogenation of crotonaldehyde, J. Phys. Chem. B 103
7

P.H. Keijzer, et al.
Catalysis Today xxx (xxxx) xxx–xxx
(
1999) 2766–2775, https://doi.org/10.1021/jp983857f.
Adv. Mater. 15 (2005) 323–327, https://doi.org/10.1002/adma.200400860.
[22] T.M. Eggenhuisen, H. Friedrich, F. Nudelman, J. Zec, N.A.J.M. Sommerdijk, P.E. de
Jongh, K.P. de Jong, Controlling the distribution of supported nanoparticles by
aqueous synthesis, Chem. Mater. 25 (2013) 890–896.
[
4] X. Yang, A. Wang, X. Wang, T. Zhang, K. Han, J. Li, Combined experimental and
theoretical investigation on the selectivities of Ag, Au, and Pt catalysts for hydro-
genation of crotonaldehyde, J. Phys. Chem. C 113 (2009) 20918–20926, https://
doi.org/10.1021/jp905687g.
[23] T.M. Eggenhuisen, J.P. den Breejen, D. Verdoes, P.E. de Jongh, K.P. de Jong,
Fundamentals of melt infiltration for the preparation of supported metal catalysts.
[
[
[
5] P. Gallezot, D. Richard, Selective hydrogenation of α,β-unsaturated aldehydes,
Catal. Rev. – Sci. Eng. 40 (1998) 81–126, https://doi.org/10.1080/
2
The case of Co/SiO for Fischer-Tropsch Synthesis, J. Am. Chem. Soc. 132 (2010)
0
1614949808007106.
18318–18325, https://doi.org/10.1016/j.cattod.2010.02.052.(41).
[24] M.Y. Wang, Z.Y. Wu, H.J. Wang, J.H. Zhu, Fabrication of metal oxides occluded in
ordered mesoporous hosts via a solid-state grinding route: the influence of host –
guest interactions, Adv. Funct. Mater. 16 (2006) 2374–2386, https://doi.org/10.
1002/adfm.200500613.
6] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky,
Triblock copolymer synthesis of mesoporous silica with periodic 50 to 300 ang-
strom pores, Science 279 (1998) 548–552.
7] J.R. a Sietsma, J.D. Meeldijk, M. Versluijs-Helder, A. Broersma, A. Jos Van Dillen,
P.E. de Jongh, K.P. de Jong, Ordered mesoporous silica to study the preparation of
[25] S. Yang, W. Su, S.D. Lin, J. Rick, J. Cheng, J. Liu, C. Pan, D. Liu, J. Lee, T. Chan,
Ni/SiO
2
ex nitrate catalysts: impregnation, drying, and thermal treatments, Chem.
2 4
H. Sheu, B. Hwang, Preparation of nano-sized Cu from a rod-like CuFe O : suitable
Mater. 20 (2008) 2921–2931, https://doi.org/10.1021/cm702610h.
for high performance catalytic applications, Appl. Catal. B Environ. 106 (2011)
650–656, https://doi.org/10.1016/j.apcatb.2011.06.030.
[
[
8] J.R.A. Sietsma, H. Friedrich, A. Broersma, M. Versluijs-Helder, A. Jos van Dillen,
P.E. de Jongh, K.P. de Jong, How nitric oxide affects the decomposition of sup-
ported nickel nitrate to arrive at highly dispersed catalysts, J. Catal. 260 (2008)
[26] L.M. Worboys, P.P. Edwards, P.A. Anderson, Silver nanowires: inclusion in and
extrusion from a mesoporous template, Chem. Commun. 2 (2002) 2894–2895,
https://doi.org/10.1039/b209126g.
2
27–235, https://doi.org/10.1016/j.jcat.2008.10.007.
9] R.M. Rioux, H. Song, J.D. Hoefelmeyer, P. Yang, G.A. Somorjai, High-surface-area
catalyst design: synthesis, characterization, and reaction studies of platinum na-
noparticles in mesoporous SBA-15 silica, J. Phys. Chem. B 109 (2005) 2192–2202,
https://doi.org/10.1021/jp048867x.
[27] H.I. Lee, J.H. Kim, G.D. Stucky, Y. Shi, C. Pak, J.M. Kim, Morphology-selective
synthesis of mesoporous SBA-15 particles over micrometer, submicrometer and
nanometer scales, J. Mater. Chem. 20 (2010) 8483–8487, https://doi.org/10.1039/
c0jm00820f.
[
[
[
10] A. Martínez, C. López, F. Márquez, I. Díaz, Fischer-Tropsch synthesis of hydro-
carbons over mesoporous Co/SBA-15 catalysts: the influence of metal loading, co-
balt precursor, and promoters, J. Catal. 220 (2003) 486–499, https://doi.org/10.
[28] K.H. Stern, High temperature properties and decomposition of inorganic salts part
3, nitrates and nitrites, J. Phys. Chem. Ref. Data 1 (1972) 747–772, https://doi.org/
10.1063/1.3253104.
1
016/S0021-9517(03)00289-6.
[29] G.H. Findenegg, S. Jähnert, D. Akcakayiran, A. Schreiber, Freezing and melting of
water confined in silica nanopores, Chem. Phys. Chem. 9 (2008) 2651–2659,
https://doi.org/10.1002/cphc.200800616.
11] S. Singh, R. Kumar, H.D. Setiabudi, S. Nanda, D.V.N. Vo, Advanced synthesis
strategies of mesoporous SBA-15 supported catalysts for catalytic reforming appli-
cations: a state-of-the-art review, Appl. Catal. A Gen. 559 (2018) 57–74, https://
doi.org/10.1016/j.apcata.2018.04.015.
[30] R.L. Oliveira, M. Shakeri, J.D. Meeldijk, K.P. de Jong, P.E. de Jongh, Mapping
nanocavities in plugged SBA-15 with confined silver nanostructures, Microporous
Mesoporous Mater. 201 (2015) 234–239, https://doi.org/10.1016/j.micromeso.
2014.09.026.
12] J.E. van den Reijen, S. Kanungo, T.A.J. Welling, M. Versluijs-Helder, T.A. Nijhuis,
2 3
K.P. de Jong, P.E. de Jongh, Preparation and particle size effects of Ag/α-Al O
catalysts for ethylene epoxidation, J. Catal. 356 (2017) 65–74, https://doi.org/10.
[31] C. de Mello Donegá, P. Liljeroth, D. Vanmaekelbergh, Physicochemical evaluation
of the hot-injection method, a synthesis route for monodisperse nanocrystals, Small
1 (2005) 1152–1162, https://doi.org/10.1002/smll.200500239.
[32] P. Scherrer, Bestimmung der inneren Struktur und der Größe von Kolloidteilchen
mittels Röntgenstrahlen, Kolloidchem. Ein Lehrb. 277 (1912) 387–409, https://doi.
org/10.1007/978-3-662-33915-2_7.
1
016/j.jcat.2017.10.001.
[
[
13] P. Munnik, P.E. de Jongh, K.P. de Jong, Recent developments in the synthesis of
supported catalysts, Chem. Rev. 115 (2015) 6687–6718, https://doi.org/10.1021/
cr500486u.
14] M. Wolters, L.J.W. Van Grotel, T.M. Eggenhuisen, J.R.A. Sietsma, K.P. de Jong,
P.E. de Jongh, Combining confinement and NO calcination to arrive at highly
dispersed supported nickel and cobalt oxide catalysts with a tunable particle size,
Catal. Today 163 (2011) 27–32, https://doi.org/10.1016/j.cattod.2010.02.052.
15] C.E. Pompe, M. Slagter, P.E. de Jongh, K.P. de Jong, Impact of heterogeneities in
silica-supported copper catalysts on their stability for methanol synthesis, J. Catal.
[33] J.I. Langford, A.J.C. Wilson, Scherrer after sixty years: a survey and some new re-
sults in the determination of crystallite size, J. Appl. Crystallogr. 11 (1978)
102–113, https://doi.org/10.1107/S0021889878012844.
[
[
[34] V. Uvarov, I. Popov, An estimation of the correctness of XRD results obtained from
the analysis of materials with bimodal crystallite size distribution, CrystEngComm
17 (2015) 8300–8306, https://doi.org/10.1039/c5ce01799h.
3
65 (2018) 1–9, https://doi.org/10.1016/j.jcat.2018.06.014.
16] E. Plessers, J.E. van den Reijen, P.E. de Jongh, K.P. de Jong, M.B.J. Roeffaers,
Origin and abatement of heterogeneity at the support granule scale of silver on
silica catalysts, ChemCatChem 9 (2017) 4562–4569, https://doi.org/10.1002/cctc.
[35] Y. Yuan, S. Yao, M. Wang, S. Lou, N. Yan, Recent progress in chemoselective hy-
drogenation of α,β-unsaturated aldehyde to unsaturated alcohol over nanomater-
ials, Curr. Org. Chem. 17 (2013) 400–413.
2
01700753.
[36] M.A. Vannice, B. Sen, Metal-support effects on the intramolecular selectivity of
crotonaldehyde hydrogenation over platinum, J. Catal. 115 (1989) 65–78, https://
doi.org/10.1016/0021-9517(89)90007-9.
[
17] E. Plessers, I. Stassen, S.P. Sree, K.P.F. Janssen, H. Yuan, J. Martens, J. Hofkens,
D. De Vos, M.B.J. Roeffaers, Resolving interparticle heterogeneities in composition
and hydrogenation performance between individual supported silver on silica cat-
alysts, ACS Catal. 5 (2015) 6690–6695, https://doi.org/10.1021/acscatal.5b02119.
18] P. Munnik, N.A. Krans, P.E. De Jongh, K.P. De Jong, Effects of Drying Conditions on
[37] P. Claus, Selective hydrogenation of α,β-unsaturated aldehydes and other C=O and
C=C bonds containing compounds, Top. Catal. 5 (1998) 51–62, https://doi.org/10.
1023/A:1019177330810.
[
[
[
the Synthesis of Co/SiO
2014) 3219–3226.
2
and Co/Al
O
2 3
Fischer − Tropsch Catalysts, ACS Catal. 4
[38] E. Plessers, D.E. De Vos, M.B.J. Roeffaers, Chemoselective reduction of α,β-un-
saturated carbonyl compounds with UiO-66 materials, J. Catal. 340 (2016)
136–143, https://doi.org/10.1016/j.jcat.2016.05.013.
(
19] P.E. de Jongh, T.M. Eggenhuisen, Melt infiltration: an emerging technique for the
preparation of novel functional nanostructured materials, Adv. Mater. 25 (2013)
[39] P.G.N. Mertens, P. Vandezande, X. Ye, H. Poelman, I.F.J. Vankelecom, D.E. De Vos,
Recyclable Au0, Ag0 and Au0-Ag0 nanocolloids for the chemoselective hydro-
genation of α,β-unsaturated aldehydes and ketones to allylic alcohols, Appl. Catal.
A Gen. 355 (2009) 176–183, https://doi.org/10.1016/j.apcata.2008.12.013.
[40] H. Wei, C. Gomez, J. Liu, N. Guo, T. Wu, R. Lobo-Lapidus, C.L. Marshall, J.T. Miller,
R.J. Meyer, Selective hydrogenation of acrolein on supported silver catalysts: A
kinetics study of particle size effects, J. Catal. 298 (2013) 18–26, https://doi.org/
10.1016/j.jcat.2012.10.027.
6
672–6690, https://doi.org/10.1002/adma.201301912.
20] C. Ciotonea, B. Dragoi, A. Ungureanu, C. Catrinescu, S. Petit, H. Alamdari,
E. Marceau, E. Dumitriu, S. Royer, Improved dispersion of transition metals in
mesoporous materials through a polymer-assisted melt infiltration method, Catal.
Sci. Technol. 7 (2017) 5448–5456, https://doi.org/10.1039/c7cy00963a.
21] Y.M. Wang, Z.Y. Wu, L.Y. Shi, Rapid functionalization of mesoporous materials:
directly dispersing metal oxides into as-prepared SBA-15 occluded with template,
[
8