Influence of Adding Molybdenum on Structure and Performance of FexOy/SBA-15 Catalysts in Selective Oxidation of Propene

Article abstract of DOI:10.1002/open.201900219

Mixed iron and molybdenum oxide catalysts supported on nanostructured silica, SBA-15, were synthesized with various Mo/Fe atomic ratios ranging from 0.07/1.0 to 0.57/1.0. Structural characterization of as-prepared Mo_xO_y_Fe_xO_y/SBA-15 samples was performed by nitrogen physisorption, X-ray diffraction, and DR-UV-Vis spectroscopy. Adding molybdenum resulted in a pronounced dispersion effect on supported iron oxidic species. Increasing atomic ratio up to 0.21Mo/1.0Fe was accompanied by decreasing species sizes. Strong interactions between iron and molybdenum during the synthesis resulted in the formation of Fe?O?Mo structure units, possibly Fe₂(MoO₄)₃-like species. Reducibility of Mo_xO_y_Fe_xO_y/SBA-15 catalysts was investigated by temperature-programmed reduction experiments with hydrogen as reducing agent. The lower reducibility obtained when adding molybdenum was ascribed to both dispersion and electronic effect of molybdenum. Catalytic performance of Mo_xO_y_Fe_xO_y/SBA-15 samples was studied in selective gas-phase oxidation of propene with O₂ as oxidant. Adding molybdenum resulted in an increased acrolein selectivity and a decreased selectivity towards total oxidation products.

Full text of DOI:10.1002/open.201900219

DOI: 10.1002/open.201900219
Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Influence of Adding Molybdenum on Structure and
Performance of Fe O /SBA-15 Catalysts in Selective
x
y
Oxidation of Propene
[a]
Nina Sharmen Genz and Thorsten Ressler*
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Mixed iron and molybdenum oxide catalysts supported on
nanostructured silica, SBA-15, were synthesized with various
Mo/Fe atomic ratios ranging from 0.07/1.0 to 0.57/1.0. Structural
characterization of as-prepared Mo O _Fe O /SBA-15 samples
was performed by nitrogen physisorption, X-ray diffraction, and
DR-UV-Vis spectroscopy. Adding molybdenum resulted in a
pronounced dispersion effect on supported iron oxidic species.
Increasing atomic ratio up to 0.21Mo/1.0Fe was accompanied
by decreasing species sizes. Strong interactions between iron
and molybdenum during the synthesis resulted in the forma-
tion of FeÀ OÀ Mo structure units, possibly Fe (MoO ) -like
2
4 3
species. Reducibility of Mo O _Fe O /SBA-15 catalysts was
x
y
x
y
investigated by temperature-programmed reduction experi-
ments with hydrogen as reducing agent. The lower reducibility
obtained when adding molybdenum was ascribed to both
dispersion and electronic effect of molybdenum. Catalytic
performance of Mo O _Fe O /SBA-15 samples was studied in
x
y
x
y
x
y
x
y
selective gas-phase oxidation of propene with O as oxidant.
Adding molybdenum resulted in an increased acrolein selectiv-
ity and a decreased selectivity towards total oxidation products.
2
1
. Introduction
optimized or newly designed catalysts. However, catalysts
applied in selective oxidation reactions are highly complex and
seem to become more complex with increasing efficiency.
Rational catalyst design constitutes a promising alternative to
conventional trial-and-error approach in developing optimized
The awareness for the depleting of our fossil feedstocks
resources has been growing in the last years. The search for
alternatives to fossil feedstocks, like, for example, bio-based
[1–3]
[6,7]
feedstocks, is ongoing.
Green chemistry aims to achieve the
catalysts. Rational catalyst design requires an in-depth under-
overall goal of sustainability for chemical processes. Chemical
processes conform with green chemistry if they are less toxic
standing of reliable structure-activity correlations.
Revealing reliable structure-activity correlations requires
reducing chemical and structural complexity of industrial
applied catalysts to model catalysts. In heterogeneous catalysis,
catalytic reactions occur on the surface of the catalysts, while
the surface structure differs significantly from that of the bulk.
Therefore, dispersing metal oxides on well-defined support
materials may result in suitable model catalysts. For elucidating
structure-activity correlations, a detailed knowledge of struc-
ture, composition, and certain chemical functions of the model
catalysts is indispensable. Thus, determining the important
variables for structure and catalytic performance of model
catalysts is a starting point for a rational design of improved
catalysts.
[4]
for human health and environment. This affects also the
selective oxidation of propene. Propene is the second most
[5]
demanded product in petrochemical industry. Glycerol as
possible new bio-based feedstock for synthesis of acrolein
recently attracted attention in catalysis research. Glycerol is
abundantly available and relatively inexpensive as a by-product
from biodiesel production. It can be transformed to acrolein by
[2,3]
catalytic dehydration.
However, this bio-based synthesis
route is still in its early stages and considerable research effort
is required for generating an industrially applicable process.
Therefore, interest in further improvement of already well-
investigated selective oxidation of propene to acrolein reap-
peared. An improvement in both selectivity towards acrolein
and propene conversion would lead to a considerably increased
cost efficiency. Moreover, selective oxidation of propene would
become greener. Such an improvement might be realized by
In terms of green chemistry, the search for alternative
greener catalysts becomes more relevant. Iron is the second
most abundant metal and constitutes 4.7 wt% of the earth
[8]
crust. Therefore, iron-based catalysts constitute a promising
alternative to more expensive metal catalysts. Moreover, various
iron compounds are commercially available or easy to synthe-
size. The probably most important advantage in view of
principles of green chemistry, is that iron is less toxic for human
[
a] N. S. Genz, Prof. Dr. T. Ressler
Institut für Chemie
Technische Universität Berlin
Straße des 17. Juni 135
[4,9]
2+
3+
health and environment.
Furthermore, Fe /Fe are often
1
0623 Berlin, Germany
E-mail: thorsten.ressler@tu-berlin.de
2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA.
used as redox promotors to improve redox properties of
[7,10,11]
©
selective oxidation catalysts.
Replacing less environmen-
This is an open access article under the terms of the Creative Commons
Attribution Non-Commercial NoDerivs License, which permits use and dis-
tribution in any medium, provided the original work is properly cited, the
use is non-commercial and no modifications or adaptations are made.
tally friendly and more expensive metals by iron without
diminishing catalytic performance would be an advancement in
catalysis research.
ChemistryOpen 2019, 8, 1133–1142
1133
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Figure 1. Left: Wide-angle X-ray diffraction patterns of Mo
SBA-15. Right: Small-angle X-ray diffraction patterns of Mo
x
O
y
_Fe
_Fe
x
O
y
/SBA-15 samples at various Mo/Fe atomic ratios, 1.3 wt% Mo_SBA-15, and 10.7 wt% Fe_
x
O
y
x y
O /SBA-15 samples, 1.3 wt% Mo_SBA-15, and 10.7 wt% Fe_SBA-15.
Currently, catalysts employed for the industrial synthesis of
acrolein are complex multicomponent metal oxides. Examples
for applied metals are bismuth, molybdenum, iron, vanadium,
2.1.1.2. Surface and Porosity Characteristics
N physisorption measurements were conducted for determin-
2
[12]
cobalt, nickel, and tungsten. In this work, a binary model
catalyst system was synthesized by changing the chemical
composition of the green Fe O /SBA-15 catalysts by adding
ing specific surface area and investigating pore structure of the
Mo O _Fe O /SBA-15 samples, 1.3 wt% Mo_SBA-15, and 10.7 wt
x
y
x
y
% Fe_SBA-15. All samples exhibited type IV nitrogen adsorp-
tion/desorption isotherms with H1 hysteresis loops indicating
x
y
molybdenum. Therefore, iron and molybdenum mixed oxides
supported on SBA-15 were synthesized and characterized
regarding structure, reducibility, and catalytic performance in
selective oxidation of propene. An interpretation of molybde-
num induced differences in structure and reducibility of Fe O /
[16]
mesoporous materials (Figure 2). Adsorption and desorption
branches were nearly parallel at the hysteresis loop, as expected
for regularly shaped pores. However, a knee in the hysteresis
loop was observed for all Mo O _Fe O /SBA-15 samples except
x
y
x
y
x
y
SBA-15 and their influence on catalytic performance was
eventually presented.
for sample 1.3 wt% Mo_SBA-15. A knee in the hysteresis loop
might have three possible reasons. First, synthesis procedure
could have varied the structure of SBA-15. A changed SBA-15
structure was excluded based on XRD results confirming
structure preservation. Furthermore, despite the observed knee
2. Results and Discussion
in the hysteresis loop, N adsorption/desorption isotherms of all
2
samples were identified to be type IV with H1 hysteresis loop,
being characteristic for SBA-15 support material. Second,
partially blocked mesopores of SBA-15 could have yielded a
knee in the hysteresis loop. This possible reason was also
excluded since desorption and adsorption branches converged
2
.1. Structural Characterization of Mo O _Fe O /SBA-15
x y x y
Catalysts
2
.1.1. Mesoporous Structure of Mo O _Fe O /SBA-15
x y x y
2
.1.1.1. Long-Range Ordered Structure
Long-range ordered structure of the Mo O _Fe O /SBA-15
x
y
x
y
samples was investigated by wide-angle and small-angle X-ray
diffraction. Figure 1, left depicts wide-angle X-ray diffraction
patterns of all Mo O _Fe O /SBA-15 samples, 1.3 wt% Mo_SBA-
x
y
x
y
1
5, and 10.7 wt% Fe_SBA-15. The absence of sharp diffraction
peaks was indicative of no long-range ordered phases and high
dispersion of small iron and molybdenum oxides within the
Mo O _Fe O /SBA-15 samples. Additionally, neither Mo O nor
x
y
x
y
x
y
Fe O species of the reference samples showed a long-range
x
y
ordering. The broad diffraction peak at 23° 2θ corresponded to
[13]
the amorphous SBA-15 support material. Small-angle diffrac-
tion patterns of all samples exhibited characteristic diffraction
[14,15]
peaks (10 l), (11 l), and (20 l) of SBA-15 (Figure 1, right).
Hence, two-dimensional hexagonal symmetry of the support
material SBA-15 was retained after supporting iron and
molybdenum oxides.
Figure 2. N adsorption/desorption isotherms of Mo O _Fe O /SBA-15 sam-
2
x
y
x
y
ples, 1.3 wt% Mo_SBA-15, and 10.7 wt% Fe_SBA-15.
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1134
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
iron during synthesis. At low Mo/Fe atomic ratios, significantly
higher amounts of iron compared to molybdenum were present
in the Mo O _Fe O /SBA-15 samples. Hence, only a small
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
x
y
x
y
fraction of iron atoms directly interacted with molybdenum
atoms during synthesis, resulting in a bimodal particle size
distribution. Fe O species which were directly influenced by
x
y
molybdenum atoms differed in species size compared to Fe Oy
x
without direct influence of molybdenum atoms. Accordingly,
differences in chemical environment during synthesis were
probably responsible for bimodal particle size distributions, and
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
hence, for the knee in the hysteresis loop of N adsorption/
2
desorption isotherms of the Mo O _Fe O /SBA-15 samples. For
x
y
x
y
sample 0.57Mo/1.0Fe/SBA-15, possessing a more balanced Mo/
Fe atomic ratio, such differences in chemical environment were
less pronounced. The higher Mo/Fe atomic ratio effected more
balanced interactions between iron and molybdenum atoms
during synthesis, inducing a broadened particle size distribu-
tion. However, the broadened particle size distribution agreed
with the knee in the hysteresis loop because of the excess of
iron compared to molybdenum.
x y x y
Figure 3. Pore radius distribution of Mo O _Fe O /SBA-15 samples, 1.3 wt%
Mo_SBA-15, and 10.7 wt% Fe_SBA-15.
after the hysteresis loop at higher p/p , and all samples
The specific surface area, as,BET, of Mo O _Fe O /SBA-15
0
x
y
x
y
possessed similar pore radius distributions (Figure 3). Besides,
partial blocking of mesopores, as reported for CuNi/SBA-15
samples systematically decreased with increasing Mo/Fe atomic
ratio. Whereas SBA-15 possessed a specific surface area
[
17]
[18]
2
(
5 wt% Cu, 5 wt% Ni), NiO/SBA-15 (24 wt% Ni), or CoMoW/
between 706.8 and 756.8 m /g, surface areas determined after
[19]
SBA-15 (9 wt% MoO , 14 wt% WO , 4 wt% CoO), was always
supporting iron and molybdenum oxides were between 447.6
3
3
2
associated with pronounced two-step desorption branches,
whereas adsorption branches were unaffected. Hence, the third
possible reason, a bimodal particle size distribution, seemed to
be reasonable. The knee in the hysteresis loop of the Mo O _
and 675.7 m /g. This decrease in a
together with a decrease
s,BET
3
in pore volume, V , from 1.036–1.203 cm /g of SBA-15 to
0.705–1.012 cm /g of as-prepared samples, indicated the iron
pore
3
and molybdenum species being successfully located in the
mesopores of SBA-15. Details on surface and porosity character-
x
y
Fe O /SBA-15 samples, was probably induced by bimodal
x
y
particle size distributions, complying with pore radius distribu-
tions determined by BJH method. Samples 0.07Mo/1.0Fe/SBA-
istics evaluated by N physisorption and XRD measurements are
2
summarized in Table 1.
Mesopore surface area, apore, calculated by BJH method
and specific surface area, as,BET, calculated by BET method can
[
20]
1
1
5, 0.10Mo/1.0Fe/SBA-15, 0.15Mo/1.0Fe/SBA-15, and 0.21Mo/
.0Fe/SBA-15 showed a bimodal pore radius distribution with a
[
21]
main maximum at 4.0 nm and a second maximum at ~3–
be used for estimating the contribution of micropores to the
[22,23]
3.5 nm. In contrast, pore radius distribution of 1.3 wt% Mo_SBA-
15 was narrow with a sharp maximum at 4.0 nm and that of
0.57Mo/1.0Fe/SBA-15 was broad and ranged from ~2.6 through
entire surface area of SBA-15.
Therefore, the ratio of
mesopore surface area and specific surface area, apore/as,BET, was
calculated (Table 1). For Mo O _Fe O /SBA-15 samples with
x
y
x
y
~
4.6 nm (Figure 3). Pore radius distribution of 10.7 wt% Fe_
atomic ratios between 0.07Mo/1.0Fe and 0.21Mo/1.0Fe, ratio of
apore/as,BET was nearly invariant at 0.91 and 0.90, respectively.
Ratio of apore/as,BET of 0.91 was also obtained for sample 10.7 wt
% Fe_SBA-15, the Fe O /SBA-15 sample without molybdenum
SBA-15 showed a main maximum at 4.0 nm. Differences in pore
radius contributions dependent on Mo/Fe atomic ratio probably
originated from various interactions between molybdenum and
x
y
[
a]
Table 1. Surface and porosity characteristics of Mo O _Fe O /SBA-15 samples, sample 1.3 wt% Mo_SBA-15, and sample 10.7 wt% Fe_SBA-15.
x
y
x
y
2
[b]
3
[c]
[d]
pore/as,BET
[e]
[f]
[g]
a
s,BET/m /g
V
pore/cm /g
a
a
0
/nm
d
w
/nm
ΔD
f
1
0
.3 wt% Mo_SBA-15
.07Mo/1.0Fe/SBA-15
675.7�0.7
581.1�0.6
579.1�0.6
573.2�0.6
554.7�0.7
447.6�0.4
605.7�0.6
1.012�0.001
0.865�0.001
0.869�0.001
0.851�0.001
0.832�0.001
0.705�0.001
0.898�0.001
0.88
0.91
0.91
0.90
0.91
0.96
0.91
11.14�0.01
11.09�0.01
11.01�0.01
11.17�0.01
11.18�0.01
11.05�0.01
10.94�0.01
3.08�0.01
3.03�0.01
2.95�0.01
3.11�0.01
3.12�0.01
2.99�0.01
2.88�0.01
0.06�0.02
0.11�0.02
0.13�0.01
0.11�0.02
0.14�0.02
0.27�0.03
0.15�0.02
0.10Mo/1.0Fe/SBA-15
0.15Mo/1.0Fe/SBA-15
0.21Mo/1.0Fe/SBA-15
0.57Mo/1.0Fe/SBA-15
10.7 wt% Fe_SBA-15
[
a] Measurement errors according to standard deviation determined from multiple measurements. [b] Specific surface area, as,BET, was calculated in the
relative pressure range, p/p , of 0.05–0.25. [c] Pore volume, V , is based on relative pressure, p/p , of 0.99. [d] Ratio of mesopore surface area, apore, and
0
pore
0
specific surface area, as,BET, as measure of micropore contribution to the entire surface of SBA-15. [e] Lattice constant, a
0
, of hexagonal unit cell. [f] Wall
thickness between the mesopores of SBA-15, d
w
f
. [g] Differences in fractal dimension, ΔD , between SBA-15 and corresponding Mo _Fe /SBA-15 samples
x
O
y
x y
O
as measure of the roughness of the surface.
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1135
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
addition. Hence, adding molybdenum at atomic ratios between
.07Mo/1.0Fe and 0.21Mo/1.0Fe did not affect the contribution
of micropores to the entire surface area of SBA-15. A higher
atomic ratio of 0.57Mo/1.0Fe induced an increased ratio of apore
as,BET up to 0.96 indicating a decreased contribution of micro-
pores. Conversely, an increased contribution of micropores to
the entire surface area of SBA-15 was observed for sample
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
0
/
1
.3 wt% Mo_SBA-15 possessing a ratio of apore/as,BET of 0.88. The
low molybdenum loading of this sample correlated with small
and highly dispersed molybdenum species. Therefore, a lower
degree of micropore filling was observed.
In addition to BET and BJH method, modified FHH method
was used to analyze the nitrogen physisorption data. Herein,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
the fractal dimension D was determined as a measure of the
f
[22,24]
roughness of the surface.
.3 wt% Mo_SBA-15, 10.7 wt% Fe_SBA-15, and SBA-15, the
fractal dimension ranged between 2 and 3, indicating a rough
For Mo O _Fe O /SBA-15 samples,
x y x y
Figure 4. DR-UV-Vis spectra of Mo
ratios between 0.07/1.0 and 0.57/1.0, 1.3 wt% Mo_SBA-15, and 10.7 wt% Fe_
SBA-15.
x
O
y
_Fe
x
O
y
/SBA-15 samples at Mo/Fe atomic
1
[25]
surface. In order to elucidate the effect of supported iron and
molybdenum species on surface roughness of the support
material, ΔD values were calculated as difference between D
position of the low-energy rise in DR-UV-Vis spectra. Analogous
to the model of the particle in the box, decreasing DR-UV-Vis
edge energy is reported to correlate with an increasing size of
f
f
values of SBA-15 and those of corresponding Mo O _Fe O /SBA-
x
y
x
y
1
5 samples. For Mo O _Fe O /SBA-15 samples with atomic ratios
x y x y
[13,26,27]
between 0.07Mo/1.0Fe and 0.21Mo/1.0Fe, ΔDf values were
invariant within the error limits. Consequently, adding molybde-
num at low Mo/Fe atomic ratios did not influence the surface
roughness of SBA-15. An invariant surface roughness of support
material at atomic ratios between 0.07Mo/1.0Fe and 0.21Mo/Fe
coincided with a nearly invariant contribution of micropores to
transition metal oxide domain.
DR-UV-Vis edge energies
of Mo O _Fe O /SBA-15 samples, 1.3 wt% Mo_SBA-15, and iron
x
y
x
y
[
27]
oxide references were determined according to Weber, and
depicted as function of Mo/Fe atomic ratio (Figure 5). Sample
1.3 wt% Mo_SBA-15 possessed the highest value of DR-UV-Vis
edge energy, and hence, the smallest Mo O species size.
Determined edge energy of 4.24 eV was close to that reported
for mononuclear molybdate species, [MoO ] , indicating that
x
y
the entire surface area of SBA-15, apore/as,BET
.
2
À
Only for samples 1.3 wt% Mo_SBA-15 and 0.57Mo/1.0Fe/
4
SBA-15, ΔD was significantly lower or higher, respectively. For
this sample mainly consisted of isolated mononuclear Mo O
y
f
x
[27,28]
sample 1.3 wt% Mo_SBA-15, consisting of small and highly
species.
The smallest species size together with the
dispersed Mo O species, lowest value of ΔD agreed with the
presence of mainly isolated, mononuclear species was in
accordance with the low metal oxide loading and the results
from N physisorption measurements. All Mo O _Fe O /SBA-15
x
y
f
highest contribution of micropores to the entire surface area of
SBA-15. At this low metal oxide loading, most of the pores of
SBA-15 remained unfilled, and hence, surface roughness of SBA-
2
x
y
x
y
1
0
5 was only slightly affected. The higher ΔD value for sample
.57Mo/1.0Fe/SBA-15 indicated a smoother surface of SBA-15.
f
The decreased contribution of micropores to the entire surface
area of SBA-15 for this sample was in accordance with the
smoother surface. Depositing iron and molybdenum oxides in
the pores of SBA-15 at higher Mo/Fe atomic ratio resulted in an
enhanced micropore filling, accompanied by an enhanced
smoothing of the support surface.
2
.1.2. Dispersion Effect of Molybdenum on the Size of the Fe O
x y
Species
Figure 4 depicts DR-UV-Vis spectra of all Mo O _Fe O /SBA-15
x
y
x
y
samples, 1.3wt%Mo_SBA-15, and 10.7 wt% Fe_SBA-15. DR-UV-
Vis spectra of Mo O _Fe O /SBA-15 samples and 10.7 wt% Fe_
x
y
x
y
SBA-15 showed a broad absorption, resulting from overlapping
absorption bands, while absorption of sample 1.3 wt% Mo_SBA-
Figure 5. DR-UV-Vis edge energy of Mo O _Fe O /SBA-15 samples (dots),
x
y
x
y
1
5 was significantly narrower.
1
.3 wt% Mo_SBA-15 (square), 10.7 wt% Fe_SBA-15 (triangle), and α-Fe
(star) as function of Mo/Fe atomic ratio. Inset depicts the enlarged trend for
the Mo _Fe /SBA-15 samples.
2 3
O
The average metal oxide particle size was estimated by
analyzing the DR-UV-Vis edge energy, determined from the
x
O
y
x y
O
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1136
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
samples possessed edge energy values higher than those of
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
crystalline α-Fe O3 and 10.7 wt% Fe_SBA-15, corresponding
2
Fe O /SBA-15 catalyst without molybdenum addition. Accord-
x
y
ingly, adding molybdenum induced an increased iron oxide
dispersion, and furthermore, a decreased average species size.
A dispersion effect of molybdenum on iron oxidic species was
also reported for MoÀ Fe catalysts supported on activated
[
29]
[30]
carbon and FeÀ Mo catalysts supported on Al O . DR-UV-Vis
2
3
edge energy of the Mo O _Fe O /SBA-15 samples increased
x
y
x
y
with increasing Mo/Fe atomic ratio between 0.07/1.0 and 0.21/
.0, indicating a decreasing average species size. Apparently,
dispersion effect of molybdenum on Fe O species was
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
1
x
y
enhanced by increasing Mo/Fe atomic ratio up to 0.21/1.0. A
further increased Mo/Fe atomic ratio to 0.57/1.0 yielded a
slightly decreased DR-UV-Vis edge energy, indicating a slightly
increased average species size. It seemed reasonable that Mo/
Fe atomic ratio of 0.57/1.0 induced a slightly increased Mo O
Figure 6. TPR traces of Mo
atomic ratios, 1.3 wt% Mo_SBA-15, and 10.7 wt% Fe_SBA-15, measured in
% H in argon at a heating rate of 10 K/min.
x y x y
O _Fe O /SBA-15 samples at various Mo/Fe
5
2
x
y
species size due to the increased molybdenum loading.
However, Fe Oy species remained still smaller and higher
x
dispersed on the support compared to those without molybde-
num addition.
2
.2. In Situ Characterization
2
.2.1. Reducibility of Mo O _Fe O /SBA-15 Catalysts
x
y
x
y
Gaining insight into reducibility of Mo O _Fe O /SBA-15 samples
x
y
x
y
was an important starting point for elucidating the influence of
adding molybdenum on catalytic performance of Fe O /SBA-15
x
y
catalysts for selective oxidation of propene. Selective oxidation
reactions proceed according to the so-called Mars-van-Krevelen
mechanism where catalysts are partial reduced and re-oxidized
[11,31,32]
during catalytic cycle.
Suitable selective oxidation cata-
lysts must possess an intermediate metalÀ oxygen bond
strength. A weak metalÀ oxygen bond strength will lead to
mainly total oxidation products. Conversely, if the
metalÀ oxygen bond strength is too strong, catalytic reaction
Figure 7. Temperature of the first TPR maxima, Tmax1(TPR), as function of Mo/
Fe atomic ratio. Samples 1.3 wt% Mo_SBA-15 and 10.7 wt% Fe_SBA-15 are
depicted at Mo/Fe atomic ratio of 0. TPR experiments were performed in 5%
2
H in argon at a heating rate of 10 K/min.
[32]
will not proceed.
Figure 6 depicts TPR traces of all Mo O _Fe O /SBA-15
x
y
x
y
samples, 10.7 wt% Fe_SBA-15, and 1.3 wt% Mo_SBA-15. Signifi-
cant differences in TPR traces were discernible. 10.7 wt% Fe_
SBA-15 showed a two-step reduction mechanism where the first
molybdenum addition. These smaller iron oxidic species with
strong interactions to the surface of SBA-15 possessed a lower
reducibility. Decreased reducibility with decreased iron oxidic
species size was already observed for Fe O /SBA-15 samples
III
and second reduction step were assigned to reduction of Fe
x
y
II
0
[25]
oxidic species to Fe oxidic species and further reduction to Fe ,
obtained from nitrate precursor. Moreover, adding molybde-
num and thereby formed FeÀ OÀ Mo structure units may further
inhibit reducibility of the Fe O system because of a charge
[25]
respectively. Adding molybdenum at atomic ratios between
.07Mo/1.0Fe and 0.21Mo/1.0Fe induced a rising shoulder at
0
x
y
the first TPR maxima at higher temperatures while the TPR
maxima were shifted to higher temperatures. Conversely, at an
atomic ratio of 0.57Mo/1.0Fe, the first TPR maximum was sharp
but further shifted to higher temperatures. The shift of the first
TPR maxima towards higher temperatures (Figure 7) indicated a
lower reducibility of the Fe O species. Apparently, adding
transfer between iron and molybdenum species. A postulated
electron transfer from iron to oxygen, and further to
[33]
molybdenum yielded a strengthened FeÀ O bond, and thus,
hindered the reducibility. The rising shoulder at the first TPR
maxima at atomic ratios between 0.07Mo/1.0Fe and 0.21Mo/
1.0Fe was assigned to the bimodal particle size distribution of
x
y
III
molybdenum induced both an electronic and a dispersion
Mo O _Fe O /SBA-15 (Figure 3). Smaller Fe oxidic species
x
y
x
y
III
effect on the Fe O /SBA-15 system. Interactions between Fe
resulting from the dispersion effect of molybdenum were
correlated with the shoulder at the first TPR maxima at higher
temperatures. However, presence of FeÀ OÀ Mo units in small
x
y
VI
and Mo ions during synthesis resulted in higher dispersed and
smaller iron oxidic species compared to those without
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1137
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
oligomeric species might also influence the first TPR maxima.
Because of an excess of iron compared to molybdenum,
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
III
II
reduction of Fe oxidic species to Fe oxidic species was still
ascribed to the first TPR maxima. Besides bimodal particle size
distribution and correlated rising shoulder at the first TPR
maxima, the shift of these maxima towards higher temperatures
was also induced by the dispersion effect of molybdenum on
the Fe O /SBA-15 system.
x
y
The second TPR maxima were also affected by adding
molybdenum. Increasing Mo/Fe atomic ratio induced a decreas-
ing intensity and an increasing shift of the second TPR maxima
to higher temperatures, while a third TPR maximum arose for
atomic ratios up from 0.10Mo/1.0Fe. This new TPR maximum at
higher temperatures was attributed to the formation of small,
hardly reducible FeÀ OÀ Mo structure units, possibly Fe (MoO ) -
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
2
4 3
[34]
like species. Boulaoued et al. reported for FeÀ Mo/KIT-6 strong
interactions between iron and molybdenum in mixed oxide
catalysts, resulting in a changed reduction mechanism and a
more difficult reduction of the iron oxidic species. Interestingly,
they ascribed a more important role to molybdenum compared
to iron. An analogous influence of adding molybdenum was
Figure 8. Acrolein selectivity of Mo O _Fe O /SBA-15 catalysts at various Mo/
Fe atomic ratios and 10.7 wt% Fe_SBA-15 as function of time on stream at
653 K in 5% propene and 5% oxygen in helium.
x
y
x
y
[
30]
reported by Kharaji et al.
for FeÀ Mo/Al O catalysts. The
2
3
authors correlated the new third reduction peak with the
presence of hardly reducible FeÀ Mo composite oxides, such as
Fe (MoO ) . These findings corroborate the assumed strong
desired selective oxidation product, carbon oxides, CO and CO2,
were the main reaction products in selective oxidation of
propene using Mo O _Fe O /SBA-15 catalysts. Selectivity to-
2
4 3
interactions between iron and molybdenum, and the involved
x
y
x
y
formation of FeÀ OÀ Mo structure units, in the Mo O _Fe O /SBA-
wards total oxidation products, CO , was significantly decreased
x
y
x
y
x
1
5 catalysts.
after molybdenum addition (Figure 9, right). Lowest acrolein
TPR traces of sample 1.3 wt% Mo_SBA-15 differed signifi-
cantly from those of Mo O _Fe O /SBA-15 catalysts and also
selectivity and highest CO selectivity for 10.7 wt% Fe_SBA-15
x
was in accordance with the largest Fe O species size and the
x
y
x
y
x
y
[34,35]
from those reported for bulk MoO3.
This sample possessed
highest reducibility compared to Mo O _Fe O /SBA-15 catalysts.
x y x y
only one single TPR peak. The broad TPR peak was indicative of
a high dispersion of small and mainly mononuclear Mo O
Obviously, FeÀ O bond strength without molybdenum addition
was weaker leading to a favored total oxidation of propene. The
lower reducibility of the Mo O _Fe O /SBA-15 catalysts, and the
x
y
species, as observed by XRD and DR-UV-Vis spectroscopy.
Furthermore, high reduction temperature of 1.3 wt% Mo_SBA-
x
y
x
y
increased FeÀ O bond strength, correlated with an increased
acrolein selectivity and a lower amount of total oxidation
products.
1
5 was ascribed to the small Mo O species possessing strong
x y
interactions to the support material.
2
.2.2. Catalytic Performance in Selective Oxidation of Propene
2.3. Correlation Between Molybdenum Induced Effects and
Catalytic Performance
Catalytic performance of Mo O _Fe O /SBA-15 catalysts in
x
y
x
y
selective oxidation of propene was investigated at 653 K and at
comparable propene conversions between 5 and 8%. Figure 8
depicts the evolution of acrolein selectivity as function of time
on stream at 653 K for Mo O _Fe O /SBA-15 catalysts and
As starting point for deducing reliable correlations between
adding molybdenum and catalytic performance, Figure 10
depicts acrolein selectivity, DR-UV-Vis edge energy, and temper-
ature of first TPR maxima as function of Mo/Fe atomic ratio.
Adding molybdenum induced a dispersion effect on supported
iron oxidic species as observed from DR-UV-Vis spectra. This
dispersion effect, and furthermore, resulting smaller, higher
dispersed Fe O species could be a possible explanation for the
x
y
x
y
1
0.7 wt% Fe_SBA-15, respectively. Acrolein selectivity and
further product distribution for sample 1.3 wt% Mo_SBA-15
were omitted from the figure. For this low loaded sample,
propene conversion was too low to yield reliable results.
x
y
Acrolein selectivity showed a significant increase due to
adding molybdenum. At an atomic ratio of 0.07Mo/1.0Fe,
acrolein selectivity was increased by a factor of 1.7. A further
increase in Mo/Fe atomic ratio induced a decrease in acrolein
selectivity. However, all Mo O _Fe O /SBA-15 catalysts pos-
observed catalytic performance of the Mo O _Fe O /SBA-15
x y x y
catalysts. DR-UV-Vis edge energy was used as a measure for
particle size of Mo O _Fe O /SBA-15 catalysts. The significantly
x
y
x
y
increased DR-UV-Vis edge energy due to adding molybdenum
from 2.67 eV for 10.7 wt% Fe_SBA-15 to 2.80 eV for 0.07Mo/
1.0Fe_SBA-15 correlated with a significantly increased acrolein
selectivity (Figure 10, left). However, the subsequent decrease
x
y
x
y
sessed higher acrolein selectivity compared to corresponding
Fe O /SBA-15 catalyst (Figures 8 and 9, left). Besides acrolein as
x
y
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1138
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
in acrolein selectivity with increasing Mo/Fe atomic ratio could
not be explained merely based on a dispersion effect. With
increasing Mo/Fe atomic ratio, the influence of the electronic
effect of molybdenum addition seemed to be crucial for the
observed trend in acrolein selectivity. With increasing molybde-
num addition, increasing FeÀ O bond strength and decreasing
reducibility was revealed for Mo O _Fe O /SBA-15 catalysts.
2.4. Interpretation of Molybdenum Induced Differences in
Structure and Reducibility of Fe /SBA-15 Catalysts and their
Impact on Catalytic Performance
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
O
y
x
Figure 11 depicts a schematic representation of the main
structural motifs of Mo O _Fe O /SBA-15 catalysts dependent
x
y
x
y
on Mo/Fe atomic ratio. Based on this schematic representation,
the influence of adding molybdenum on Fe O /SBA-15 catalysts,
x
y
x
y
Decreasing reducibility, expressed as increasing temperature of
the first TPR maxima (Figure 10, right), correlated with a
decreasing acrolein selectivity. These observations agreed with
the postulated intermediate metalÀ oxygen bond strength
x
y
i.e. on their structure, reducibility, and catalytic performance,
might be interpreted.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Increasing Mo/Fe atomic ratio correlated with a decreasing
amount of FeÀ OÀ Fe structure units (Figure 11, (A)). This was
ascribed to smaller, higher dispersed Fe O species of Mo O _
[32]
being required for selective oxidation catalysts.
x
y
x
y
x
Figure 9. Selectivity towards acrolein (left) and CO (right) as function of Mo/Fe atomic ratio. Selectivity was measured at 653 K in 5% propene and 5%
oxygen in helium.
Figure 10. Correlation between acrolein selectivity and DR-UV-Vis edge energy (left) and temperature of first TPR maxima (right) as function of Mo/Fe atomic
ratio. Acrolein selectivity was measured at 653 K in 5% propene and 5% oxygen in helium, DR-UV-Vis edge energy was determined at ambient temperature in
air, and TPR measurements were conducted in 5% H in argon at 10 K/min up to 1223 K.
2
Figure 11. Schematic representation of structural differences of iron oxidic species supported on SBA-15 dependent on Mo/Fe atomic ratio. Only the main
structural motifs of Mo
x
O
y
_Fe
x
y
O /SBA-15 catalysts are depicted.
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1139
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
Fe O /SBA-15 catalysts compared to those of corresponding
to 0.21Mo/1.0Fe was accompanied by decreasing species sizes.
Strong interactions between iron and molybdenum during
synthesis resulted in formation of FeÀ OÀ Mo structure units. At
higher Mo/Fe atomic ratio, these strong interactions probably
induced small Fe (MoO ) -like species. Moreover, adding molyb-
x
y
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Fe O /SBA-15 catalyst without molybdenum. The dispersion
x
y
effect of molybdenum on Fe O species induced this decreasing
x
y
species size. Bridging oxygen atoms in FeÀ OÀ Fe structure units
were presumably required for reducibility of Fe O /SBA-15
x
y
2
4 3
catalysts. Accordingly, the decreased Fe O species size corre-
denum significantly influenced the reducibility of Fe O /SBA-15.
x y
x
y
lated with a decreased reducibility of Mo O _Fe O /SBA-15
Lower reducibility due to adding molybdenum was ascribed to
both dispersion and electronic effect of molybdenum. Smaller
and higher dispersed iron oxidic species possessed a lower
reducibility compared to larger and less dispersed species.
Additionally, a charge transfer from iron to oxygen, and further
to molybdenum in FeÀ OÀ Mo structure units yielded a strength-
ened FeÀ O bond and, hence, hindered reducibility. The change
of the two-step reduction mechanism for 10.7 wt% Fe_SBA-15
towards a three-step reduction mechanism for Mo O _Fe O /
x
y
x
y
catalysts compared to corresponding Fe O /SBA-15 catalyst.
x
y
Furthermore, with increasing Mo/Fe atomic ratio, FeÀ OÀ Fe
structure units were partially replaced by FeÀ OÀ Mo structure
units (Figure 11, (B)). Accordingly, the electronic effect of
molybdenum on Fe O species became more important for their
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
x
y
reducibility. This electronic effect was associated with an
electron transfer from iron to oxygen, and further to molybde-
num in FeÀ OÀ Mo structure units. As a result, FeÀ O bonds
between iron and bridging oxygen atoms were strengthened,
yielding a lower reducibility (Figure 11, (B)). Only for highest
Mo/Fe atomic ratio, an influence of adding molybdenum on
micropore filling was observed. This high Mo/Fe atomic ratio
induced a higher degree of micropore filling being ascribed to
a higher amount of FeÀ OÀ Si (Figure 11, (C)) and MoÀ OÀ Si
bonds (Figure 11, (D)), as well as hydrogen bonds (Figure 11,
x
y
x
y
SBA-15 catalysts corroborated the presence of small, hardly
reducible FeÀ OÀ Mo structure units.
Catalytic performance of Mo O _Fe O /SBA-15 was studied
x
y
x
y
under selective propene oxidation conditions. Adding molybde-
num resulted in an increased acrolein selectivity and a
decreased selectivity towards total oxidation products. Increas-
ing Mo/Fe atomic ratio induced a decreasing acrolein selectivity
while still being higher than that of Fe O /SBA-15 catalyst.
(
E)). These hardly reducible FeÀ OÀ Si and MoÀ OÀ Si bonds
x
y
additionally hindered the reducibility of Mo O _Fe O /SBA-15
Influence of adding molybdenum on catalytic performance was
correlated with both dispersion and electronic effect of
molybdenum. The strengthened FeÀ O bonds, and the lower
reducibility, with increasing Mo/Fe atomic ratio, led to an
inferior total oxidation. This coincided with an intermediate
FeÀ O bond strength being required for selective oxidation of
propene.
x
y
x
y
catalysts. Therefore, selectivity towards acrolein decreased with
increasing Mo/Fe atomic ratio. Highest acrolein selectivity for an
atomic ratio of 0.07Mo/1.0Fe might be explained by an optimal
combination of intermediate reducibility and high electron
transfer efficiency in Fe O oligomers. This might facilitate the
x
y
formation of nucleophilic surface oxygen species, yielding a
higher acrolein selectivity. Despite decreasing reducibility with
increasing Mo/Fe atomic ratio, formation of nucleophilic oxygen
species might still be favored in Mo O _Fe O /SBA-15 catalysts. Experimental Section
x
y
x
y
This might explain the higher acrolein selectivity of all Mo O _
x
y
Fe O /SBA-15 catalysts compared to that of corresponding
Sample Preparation
x
y
Fe O /SBA-15 catalyst. However, acrolein selectivity as function
x
y
Mesoporous silica SBA-15 was prepared according to Zhao
of Mo/Fe atomic ratio was presumably determined by a
combination of various effects, such as reducibility, species size,
and electron transfer efficiency. Characterization of Mo O _
[14,36]
et al..
Mixed iron and molybdenum oxide catalysts supported
on SBA-15 were prepared by incipient wetness technique and
denoted Mo _Fe /SBA-15. Therefore, an aqueous solution of
O
O
x y
x
y
x
y
ammonium iron(III) citrate (~18% Fe, Roth) and ammonium
heptamolybdate tetrahydrate (ꢀ99%, Fluka) was used. The pH
value of the aqueous oxide precursor solution was adjusted to 7.5–
Fe O /SBA-15 catalysts corroborated the assumed importance of
x
y
FeÀ OÀ Fe structure units (Figure 11, (A)) in selective oxidation of
propene.
8
. After drying in air for 24 h, calcination was carried out at 723 K
for 5 h. To investigate the influence of adding molybdenum on
Fe O /SBA-15 catalysts, iron loading was kept invariant, while Mo/Fe
x
y
3. Conclusion
atomic ratio was varied between 0.07/1.0 and 0.57/1.0. According
to the Mo/Fe atomic ratio, samples were denoted 0.07Mo/1.0Fe/
SBA-15, 0.10Mo/1.0Fe/SBA-15, 0.15Mo/1.0Fe/SBA-15, 0.21Mo/1.0Fe/
SBA-15, and 0.57Mo/1.0Fe/SBA-15. Furthermore, samples with
The synthesis of iron and molybdenum mixed oxides on
mesoporous SBA-15 yielded suitable binary model catalysts for
investigating structure-activity correlations. Invariant iron load-
ing permitted investigating the influence of adding molybde-
num on Fe O /SBA-15 catalysts. Mo O and Fe O species of the
1
1
.3 wt% Mo supported on SBA-15, 1.3 wt% Mo_SBA-15, and with
0.7 wt% Fe supported on SBA-15, 10.7 wt% Fe_SBA-15, were
prepared as references. Metal oxide loadings of the samples were
quantified by X-ray fluorescence spectroscopy. Moreover, CHN
analysis was performed to confirm a complete decomposition of
the precursors.
x
y
x
y
x
y
Mo O _Fe O /SBA-15 catalysts were highly dispersed on SBA-15
x
y
x
y
without changing the structure of the support. Formation of
long-range ordered structures was excluded. Adding molybde-
num yielded a pronounced dispersion effect on supported iron
oxidic species. Accordingly, average species size was decreased
compared to Fe O /SBA-15 catalyst. Increasing atomic ratio up
x
y
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1140
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
Sample Characterization
placed vertically in a tube furnace. A P3 frit was centered in the
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
SiO tube in the isothermal zone, where the sample was placed.
2
The samples were diluted with boron nitride (99.5%, Alfa Aesar) to
achieve a constant volume in the reactor and to minimize thermal
effects. Overall sample masses were 0.25 g. The reactor was
operated at low propene conversion levels (5–8%) to ensure
differential reaction conditions. For catalytic testing in selective
oxidation of propene, a gas mixture of 5% propene and 5% oxygen
in helium was used in a temperature range of 298–653 K with a
heating rate of 5 K/min. Reactant gas flow rates of propene, oxygen,
and helium were adjusted through separate mass flow controllers
(Bronkhorst) to a total flow of 40 ml/min. All gas lines and valves
were preheated to 473 K. Hydrocarbons and oxygenated reaction
products were analyzed using a Carbowax 52CB capillary column,
connected to an Al O /MAPD capillary column (25 m×0.32 mm) or
X-Ray Fluorescence Analysis
Quantitative analysis of the metal oxide loadings on SBA-15 was
conducted by X-ray fluorescence spectroscopy on an AXIOS X-ray
spectrometer (2.4 kW model, PANalytical), equipped with a Rh K X-
ray source, a gas flow detector, and a scintillation detector. Prior to
measurements, samples were mixed with wax (Hoechst, Merck),
ratio 1:1, and pressed into pellets of 13 mm diameter. Quantifica-
tion was performed by standardless analysis using the software
package SuperQ5 (PANalytical).
α
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Elemental Analysis
2
3
a fused silica restriction (25 m×0.32 mm), each connected to a
flame ionization detector (FID). Permanent gases (CO, CO , N , O )
Elemental contents of C, H, and N were determined using a Thermo
FlashEA 1112 Organic Elemental Analyzer (ThermoFisher Scientific)
with CHNSÀ O configuration.
2
2
2
were separated and analyzed using a “Permanent Gas Analyzer”
CP-3800, Varian) with a Hayesep Q (2 m×1/8’’) and a Hayesep T
(
packed column (0.5 m×1/8’’) as precolumns combined with a back
flush. For separation, a Hayesep Q packed column (0.5 m×1/8’’)
was connected via a molecular sieve (1.5 m×1/8’’) to a thermal
conductivity detector (TCD). Additionally, product and reactant gas
flow were continuously monitored by a connected mass spectrom-
eter (Omnistar, Pfeiffer) in a multiple ion detection mode.
Powder X-Ray Diffraction
Powder X-ray diffraction patterns were obtained using an X’Pert
PRO diffractometer (PANalytical, 40 kV, 40 mA) in theta/theta
geometry equipped with a solid-state multi-channel detector
(
PIXel). Cu K radiation was used. Wide-angle diffraction scans were
α
conducted in reflection mode. Small-angle diffraction patterns were
measured in transmission mode from 0.4° through 6° 2θ in steps of
0
.013° 2θ with a sampling time of 90 s/step.
Acknowledgements
The authors are grateful to A. Knoch and S. Schwarz for assistance
Nitrogen Physisorption
during synthesis and N physisorption measurements, respectively.
2
Nitrogen adsorption/desorption isotherms were measured at 77 K
using a BELSORP Mini II (BEL Inc. Japan). Prior to measurements, the
samples were pre-treated under reduced pressure (10 kPa) at
J. Krone is acknowledged for CHN elemental analysis.
À 2
3
1
68 K for 35 min and kept under the same pressure at 448 K for
5 h (BELPREP II vac).
Conflict of Interest
The authors declare no conflict of interest.
DR-UV-Vis Spectroscopy
Diffuse reflectance UV-Vis (DR-UV-Vis) spectroscopy was conducted
on a two-beam spectrometer (V-670, Jasco) using a barium sulfate
coated integration sphere. (Scan speed 100 nm/min, slit width
5.0 nm (UV-Vis) and 20.0 nm (NIR), and spectral region 220–
Keywords: dispersion effect
· electronic effect · green
chemistry · heterogeneous catalysis · iron oxides
2000 nm). SBA-15 was used as white standard for all samples.
[
[
[
1] N. R. Shiju, V. V. Guliants, Appl. Catal. A 2009, 356, 1–17.
2] L. Liu, X. P. Ye, J. J. Bozell, ChemSusChem 2012, 5, 1162–1180.
3] B. Katryniok, S. Paul, V. Bellière-Baca, P. Rey, F. Dumeignil, Green Chem.
2010, 12, 2079–2098.
Temperature-Programmed Reduction
Temperature-programmed reduction (TPR) was performed using a
BELCAT_B (BEL Inc. Japan). Samples were placed on silica wool in a
silica glass tube reactor. Evolving water was trapped using a
molecular sieve (4 Å). Gas mixture consisted of 5% H in 95% Ar
with a total gas flow of 40 ml/min. Heating rate used was 10 K/min
to 1223 K. A constant initial sample weight of 0.03 g was used and
[
4] P. Anastas, N. Eghbali, Chem. Soc. Rev. 2010, 39, 301–312.
[5] “The Propylene Gap: How Can It Be Filled? – American Chemical
Society”, can be found under https://www.acs.org/content/acs/en/
pressroom/cutting-edge-chemistry/the-propylene-gap-how-can-it-be-fil-
led.html, 2018.
2
[
6] a) R. K. Grasselli, Catal. Today 1999, 49, 141–153; b) B. L. Kniep, T. Ressler,
A. Rabis, F. Girgsdies, M. Baenitz, F. Steglich, R. Schlögl, Angew. Chem.
Int. Ed. Engl. 2004, 43, 112–115.
H2 consumption was continuously monitored by
conductivity detector.
a thermal
[
[
7] U. S. Ozkan, R. B. Watson, Catal. Today 2005, 100, 101–114.
8] A. F. Holleman, E. Wiberg, N. Wiberg, Lehrbuch der anorganischen
Chemie, de Gruyter, Berlin, New York, 2007.
Catalytic Measurements
[9] a) S. Enthaler, K. Junge, M. Beller, Angew. Chem. Int. Ed. Engl. 2008, 47,
317–3321; b) A. S. Al-Fatesh, A. H. Fakeeha, A. A. Ibrahim, W. U. Khan,
Quantification of catalytic activity in selective oxidation of propene
was performed using a laboratory fixed-bed reactor connected to
an online gas chromatography system (CP-3800, Varian) and a mass
spectrometer (Omnistar, Pfeiffer Vacuum). The fixed-bed reactor
3
H. Atia, R. Eckelt, B. Chowdhury, J. Chin. Chem. Soc. 2016, 63, 205–212.
[10] J. C. Védrine, G. Coudurier, J.-M. M. Millet, Catal. Today 1997, 33, 3–13.
[11] B. Grzybowska-Świerkosz, Top. Catal. 2000, 11–12, 23–42.
[
12] a) M. M. Lin, Appl. Catal. A 2001, 207, 1–16; b) C. Zhao, I. Wachs, J. Catal.
008, 257, 181–189.
consisted of a SiO tube (30 cm length, 9 mm inner diameter)
2
2
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1141
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA

Full Papers
[
13] J. He, Y. Li, D. An, Q. Zhang, Y. Wang, J. Nat. Gas Chem. 2009, 18, 288–
94.
[14] D. Zhao, Science 1998, 279, 548–552.
15] a) M. Kruk, M. Jaroniec, C. H. Ko, R. Ryoo, Chem. Mater. 2000, 12, 1961–
[27] R. S. Weber, J. Catal. 1995, 151, 470–474.
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
2
[28] Y. Lou, H. Wang, Q. Zhang, Y. Wang, J. Catal. 2007, 247, 245–255.
[29] W. Ma, E. L. Kugler, J. Wright, D. B. Dadyburjor, Energy Fuels 2006, 20,
2299–2307.
[
1
2
968; b) C.-M. Yang, B. Zibrowius, W. Schmidt, F. Schüth, Chem. Mater.
004, 16, 2918–2925.
[30] A. G. Kharaji, A. Shariati, M. A. Takassi, Chin. J. Chem. Eng. 2013, 21,
1007–1014.
[16] K. S. W. Sing, Pure Appl. Chem. 1985, 57, 603–619.
[31] G. Centi, F. Cavani, F. Trifirò, Selective oxidation by heterogeneous
catalysis, Kluwer Acad./Plenum Publ, New York, 2001.
[32] R. Grasselli, Top. Catal. 2002, 21, 79–88.
[33] S. Qin, C. Zhang, J. Xu, Y. Yang, H. Xiang, Y. Li, Appl. Catal. A 2011, 392,
118–126.
[34] A. Boulaoued, I. Fechete, B. Donnio, M. Bernard, P. Turek, F. Garin,
Microporous Mesoporous Mater. 2012, 155, 131–142.
[35] H. Zhang, J. Shen, X. Ge, J. Solid State Chem. 1995, 117, 127–135.
[36] D. Zhao, Q. Huo, J. Feng, B. F. Chmelka, G. D. Stucky, J. Am. Chem. Soc.
1998, 120, 6024–6036.
[17] A. Ungureanu, B. Dragoi, A. Chirieac, S. Royer, D. Duprez, E. Dumitriu, J.
Mater. Chem. 2011, 21, 12529.
[18] J. R. A. Sietsma, J. D. Meeldijk, M. Versluijs-Helder, A. Broersma, A. J.
van Dillen, P. E. d. Jongh, K. P. d. Jong, Chem. Mater. 2008, 20, 2921–
2931.
[19] R. Huirache-Acuña, B. Pawelec, E. Rivera-Muñoz, R. Nava, J. Espino,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
J. L. G. Fierro, Appl. Catal. B 2009, 92, 168–184.
[20] S. Brunauer, P. H. Emmett, E. Teller, J. Am. Chem. Soc. 1938, 60, 309–319.
[21] E. P. Barrett, L. G. Joyner, P. P. Halenda, J. Am. Chem. Soc. 1951, 73, 373–
80.
3
[22] M. A. Smith, R. F. Lobo, Microporous Mesoporous Mater. 2010, 131, 204–
209.
[
[
[
23] G. Koch, L. Schmack, T. Ressler, ChemistrySelect 2016, 1, 2040–2049.
24] P. Pfeifer, Phys. Rev. Lett. 1989, 62, 1997–2000.
25] N. S. Genz, D. Baabe, T. Ressler, J. Anal. Methods Chem. 2017, 2017, 1–13.
[26] T. Abe, Y. Tachibana, T. Uematsu, M. Iwamoto, J. Chem. Soc., Chem.
Manuscript received: June 26, 2019
Revised manuscript received: July 13, 2019
Commun. 1995, 1617–1618.
ChemistryOpen 2019, 8, 1133–1142
www.chemistryopen.org
1142
© 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA