Precursor effect on the property and catalytic behavior of Fe-TS-1 in butadiene epoxidation

Article abstract of DOI:10.1134/S0036024417110140

The effect of iron precursor on the property and catalytic behavior of iron modified titanium silicalite molecular sieve (Fe-TS-1) catalysts in butadiene selective epoxidation has been studied. Three Fe-TS-1 catalysts were prepared, using iron nitrate, iron chloride and iron sulfate as precursors, which played an important role in adjusting the textural properties and chemical states of TS-1. Of the prepared Fe-TS-1 catalysts, those modified by iron nitrate (FN-TS-1) exhibited a significant enhanced performance in butadiene selective epoxidation compared to those derived from iron sulfate (FS-TS-1) or iron chloride (FC-TS-1) precursors. To obtain a deep understanding of their structure-performance relationship, X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), Temperature programmed desorption of NH₃ (NH₃-TPD), Diffuse reflectance UV–Vis spectra (DR UV–Vis), Fourier transformed infrared spectra (FT-IR) and thermal gravimetric analysis (TGA) were conducted to characterize Fe-TS-1 catalysts. Experimental results indicated that textural structures and acid sites of modified catalysts as well as the type of Fe species influenced by the precursors were all responsible for the activity and product distribution.

Full text of DOI:10.1134/S0036024417110140

ISSN 0036-0244, Russian Journal of Physical Chemistry A, 2017, Vol. 91, No. 11, pp. 2103–2109. © Pleiades Publishing, Ltd., 2017.
CHEMICAL KINETICS
AND CATALYSIS
Precursor Effect on the Property and Catalytic Behavior
of Fe-TS-1 in Butadiene Epoxidation1
Mei Wua, Huahua Zhaob, Jian Yangb, Jun Zhaob, Huanling Songb,*, and Lingjun Choub,**
aFaculty of Chemical Engineering, Huaiyin Institute of Technology, Key Laboratory for Palygorskite Science
and Applied Technology of Jiangsu Province, Huaian, 223003 P.R. China
bState Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of Chemical Physics,
Chinese Academy of Sciences, Lanzhou, 730000 P.R. China
*e-mail: songhl@licp.cas.cn
**e-mail: ljchou@licp.cas.cn
Received April 18, 2016
Abstract—The effect of iron precursor on the property and catalytic behavior of iron modified titanium sili-
calite molecular sieve (Fe-TS-1) catalysts in butadiene selective epoxidation has been studied. Three Fe-TS-1
catalysts were prepared, using iron nitrate, iron chloride and iron sulfate as precursors, which played an
important role in adjusting the textural properties and chemical states of TS-1. Of the prepared Fe-TS-1 cat-
alysts, those modified by iron nitrate (FN-TS-1) exhibited a significant enhanced performance in butadiene
selective epoxidation compared to those derived from iron sulfate (FS-TS-1) or iron chloride (FC-TS-1) pre-
cursors. To obtain a deep understanding of their structure-performance relationship, X-ray diffraction
(XRD), X-ray photoelectron spectroscopy (XPS), Temperature programmed desorption of NH3 (NH3-
TPD), Diffuse reflectance UV–Vis spectra (DR UV–Vis), Fourier transformed infrared spectra (FT-IR) and
thermal gravimetric analysis (TGA) were conducted to characterize Fe-TS-1 catalysts. Experimental results
indicated that textural structures and acid sites of modified catalysts as well as the type of Fe species influ-
enced by the precursors were all responsible for the activity and product distribution.
Keywords: butadiene epoxidation, iron, precursors effect, TS-1 properties
DOI: 10.1134/S0036024417110140
INTRODUCTION
ethane to acetonitrile. Research revealed that final
ZSM-5 catalysts contain different types of cobalt spe-
cies and new acid sites depending on the nature of the
cobalt precursors, which exhibited different catalytic
properties [8]. Furthermore, four different copper
precursors were applied to adjust the textural proper-
ties and chemical states of the CuY catalysts. The
experimental results indicated that the textural struc-
tures of the catalysts and the acidity and type of Cu+
species influenced by the precursors were all responsi-
ble for the activity and product distribution [9]. The
above research work indicated the precursor effect on
pure silicate based zeolite, while TS-1 has framework
substituted Ti active center, whose chemical status
play key role in TS-1 catalytic performance. Up to
now, effects of precursors on TS-1 property has not
been studied. Studies of precursor’s effect on the
chemical state of Ti center, the acidic property of TS-1
surface and the TS-1 texture structure are of great
value.
Selective catalytic epoxidation of olefins is an
industrially valuable reaction widely used for interme-
diates production in the fine chemicals and pharma-
ceuticals. Titanium silicalite molecular sieve (TS-1)
has been reported as the most important heteroge-
neous catalysts for olefins selective epoxidation [1–5].
Nevertheless, the low single-pass yield of the obtained
epoxides limits the large scale applications of TS-1.
For the past few years, transition metals modification
has become an effective way to improve TS-1 catalytic
performance, especially, the single-pass yield [6, 7].
Transition metals modified TS-1 catalysts have been
verified to be effective for olefin selective epoxidation,
while studies it regard as rarely involve with precursors
effect.
Appropriate precursor is one of the key points to
prepare supporting catalyst with high performance.
Co-ZSM-5 catalysts prepared by different cobalt pre-
cursors (acetate, chloride, nitrate or formate) were
studied in the selective ammoxidation of ethylene and
We propose to modify TS-1 with iron. Fe-contain-
ing-molecular sieves are well known active/selective
catalysts for oxidation reactions using H2O2 or N2O as
1
The article is published in the original.
2103

2104
MEI WU et al.
oxidant [10]. Ahn et al. [11] introduced Fe into MCM- as FNTS-1, FSTS-1, and FCTS-1, respectively. In all
41 an reported that trivalent iron incorporated into tet- samples, Fe loading was 1.5 wt %.
rahedral silicate matrixes exhibited high catalytic
activity for phenol hydroxylation using H2O2 as oxi-
Physicochemical Characterization
dant, giving phenol conversion of ca. 60% at 50°C
(phenol : H2O2 = 1 : 1, water solvent). Adam et al. [12]
used heterogeneous iron silica as catalysts for phenol
oxidation, and high phenol conversion of 95.2% and
The XRD measurements were performed using an
X’Pert Pro multipurpose diffractometer (PANalytical,
Inc.) with Ni-filtered CuKα radiation (0.15046 nm)
selectivity
of
100%
to
dihydroxybenzenes from 5.0° to 80.0°. Measurements were conducted
(phenol : H2O2 = 1 : 2) were obtained. Berenguer– using a voltage of 40 kV, current setting of 40 mA, step
size of 0.02°, and count time of 4 s.
Murciaa and co-workers [13] synthesized iron silica
catalysts via a one-step synthesis procedure and
revealed that ferrosilicate catalysts were active in pro-
pylene epoxidation with dioxygen as oxidant. Previ-
ously, our group studied the catalytic performance of
Fe, Ni, Cr, V, etc. treated TS-1 catalysts for butadiene
selective epoxidation. Fe modification was found to
enhance TS-1 performance significantly on account
of enhancing the eletrophilicity of Ti active center and
promoting the formation of five-member transition
state [14].
Herein, we have investigated the effects of the iron
precursor on the physicochemical properties and cat-
alytic performance of Fe-TS-1 catalysts in the selec-
tive epoxidation of butadiene. Three different salts
(iron nitrate, chloride, and sulfate) were used to adjust
the textural properties and acid sites as well as chemi-
cal states of Fe-TS-1 catalysts. X-ray diffraction
(XRD), X-ray photoelectron spectroscopy (XPS),
Temperature programmed desorption of NH3
(NH3-TPD), Diffuse reflectance UV–Vis spectra
(DR UV‒Vis), Fourier transformed infrared spectra
(FT-IR) and thermal gravimetric analysis (TGA) were
used to characterize the treated catalysts for a deeper
understanding of their structure-performance rela-
tionship.
The N2 adsorption and desorption isotherms at
‒196°C were recorded on an Autosorb-iQ analyzer
(Quantachrome Instruments U.S.). Prior to the tests,
samples were degassed at 200°C for 4 h. The specific
surface areas were calculated via the BET method in
the relative pressure range of 0.05–0.30. Micropore
volumes were calculated using adsorption branches of
nitrogen adsorption-desorption isotherms by Saito-
Foley (SF) methods.
The oxidation states of Ti and Fe were character-
ized over a Thermo Fisher Scientific K-Alpha XPS.
The catalyst was pressed to self-supporting wafer prior
to analysis.
FT-IR characterizations of catalysts were recorded
on FT-IR spectrometer (Nicolet Nexus 870) with a
resolution of 4 cm–1 and 64 scans in the region of
4000–500 cm–1.
DR UV–Vis results were obtained on a PE Lambda
650S spectrometer with BaSO4 as standard.
TGA measurements were rendered on
a
NETZSCH STA 449F3 thermogravimetric analyzer
from room temperature to 600°C at the rate of
10 K/min.
NH3-TPD adsorbed on Fe-TS-1 samples was car-
ried out in the conventional flow system equipment
(Zeton-Altamira instrument, AMI-100) with AME-
TEK (LC-D-200 Dycor AMETEK) mass spectrum.
After being pretreated at 400°C under He flow
(50 mL/min), the catalyst (100 mg) was cooled to
100°C under He flows, and then adsorbed to satura-
tion by NH3 at 100°C for 15 min physically adsorbed
NH3 on catalyst was removed by flushing the sample
with He gas at the adsorption temperature. The heat-
ing rate was 10 K/min.
EXPERIMENTAL
Materials and Catalysts Preparation
H2O2 (40 wt %; local vender) concentrations were
determined by iodometrictitration prior to use. Tita-
nium silicalite 1 (TS-1, SiO2/TiO2 = 40) was prepared
according to the method described by Thangaraj et al.
[15]. All Fe salts (local vendor) and methanol
(CH3OH, local vendor) were all A.R. grade without
further purification.
Fe-TS-1 catalysts were obtained by supersonic
impregnation method. The typical synthesis process
was as follows: 1.00 g of TS-1 was dispersed in distilled
water, followed by adding into designed amount of Fe
salts solution under vigorous stirring. The mixture was
treated under supersonic for 3 h and infrared drying
until it changed to dry powder. Finally it was further
dried up overnight at 120°C. The output powder was
calcined at 550°C for 6 h. Three Fe salts were applied
to prepare the catalysts, which were nitrate, sulfate and
Catalytic Evaluation
The epoxidation of butadiene was performed in a
100 mL stainless-steel autoclave with the magnetic
stirrer. Typically, 0.2 g catalyst and designed amount
of H2O2 were dispersed in 25.00 mL methanol solvent
followed by introducing 0.15 MPa butadiene. Butadi-
ene was in excess amount during reaction process. The
reaction continued for 60 min at 40°C. Remaining
H2O2 concentration was determined by standard iodo-
chloride. Accordingly, the final catalyst was denoted metrictitration. Chromatography-mass spectroscopy
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 91
No. 11
2017

PRECURSOR EFFECT ON THE PROPERTY AND CATALYTIC BEHAVIOR OF Fe-TS-1
2105
(GC-MS 5973 equipment from Agilent Technology
Company) was used to confirm liquid products. Viny-
loxirane (VO) was confirmed as the main product, and
main by-products were small amount of tetrahydro-2-
furanmethanol and 4-ethenyl-cyclohexene. Quanti-
ties of products were performed on a gas chromato-
graph (SP-6800A GC) equipped with a flame ioniza-
tion detector and OV-1701 capillary column. VO
yields were determined relating to methanol via exter-
nal standard method based on the GC results, while
concentrations of by-products were too little to deter-
mine. VO turnover numbers (TON, determined as the
molar VO obtained per molar Ti atom:
TON = YVO/nTi), and VO concentrations in the final
reaction mixtures were utilized to evaluate activities of
different catalysts. Conversions and utilizations of H2O2
were calculated by means of next equations, respec-
tively:
FCTS-1
FNTS-1
FSTS-1
TS-1
10
20
30
40
50
60
70
80
2θ, deg
Fig. 1. XRD patterns of all catalysts.
H O (conversion), % = 100[(
)
0
n
2
2
H2O2
umes around 490 m2/g and 0.19 cm3/g. It reveals that
the modification of different Fe precursors has litter
effect on the texture property of TS-1. And it further
testifies Fe species are highly dispersed on TS-1 sup-
porter.
– (
)]/[(
)0],
n
n
H2O2
H2O2
H2O2(utilization), % = 100[YVO]/[(
)0].
n
H2O2
In above equations, nVO stands for moles of VO yield
after epoxidation reaction; nTi stands for molar of Ti
To elucidate the origin for variation of electronic
performance as a function of Fe precursors, the chem-
ical structure of Fe-TS-1 catalysts and as made-TS-1
were analyzed by XPS. As shown in Fig. 2a, peaks cen-
tered at 460.3 eV are assigned to the Ti 2p3/2 core-level
spectra of TS-1, confirming the presence of frame-
work Ti in tetrahedral coordination. Besides, other
features at 459.1 eV are attributed to framework Ti cen-
ter with higher coordination number. It can be sup-
posed that the additive Fe coordinates with Ti active
site via O, thus makes the Ti coordination number
increase. The higher electron density enhances the
screening effect on the Ti nuclei, which decreases the
effective nuclear charge, leading to a shift in the peak
toward lower binding energy. So to that extent, frame-
work Ti–O bond is weakened due to Fe modification.
The as formed Fe–O–Ti species can facilitate Ti
active center to be attacked by CH3OH and H2O2 and
promote the formation of five-membered cyclic inter-
mediate. And thus, catalytic performance will be
improved with Fe modification. In addition, Intensity
of band at 460.3 eV reflected the content of tetrahedral
atom in catalyst; besides, (
) and
stand for
n
n
0
H2O2
H2O2
the initial molar content and the remaining molar
content of H2O2 after reaction, respectively.
RESULTS AND DISCUSSION
Crystal structures of TS-1, FNTS-1, FSTS-1, and
FCTS-1 samples were exhibited in Fig. 1. According
to characterization results, all samples possess XRD
diffraction patterns identical to characteristic TS-1,
confirming the presence of orthorhombic MFI-type
crystalline phase [16]. High crystallinity retains with
the modification of different Fe precursors. Free iron
oxides or other Fe species are not observed in Fe
treated samples, indicating that the particle size of iron
oxides is too small to be detected [17], and it further
explains the high dispersion of Fe species on the TS-1
supporter. Diffracted peaks intensities of all Fe treated
catalysts decreased faintly compared to TS-1. It indi-
cated that the relative amount of TS-1 crystalline
phase in Fe modified catalysts declines [18].
N2 adsorption–desorption had been performed to
determine the textural properties of Fe-TS-1 catalysts.
All modified catalysts including as-made TS-1 exhib-
its typical type-I profiles, which are characteristic for
microporous structures. The BET surface area, aver-
age micropore size and micropore volume of Fe-TS-1
catalysts were listed in Table 1. As can be shown, in
comparison with the as made-TS-1, average microp-
ore sizes show a moderate increase with the modifica-
tion of various precursors. It is mainly due to the effect
of ultrasonic when immersing. But beyond that, all
sampled have similar BET surface area and pore vol-
Table 1. Physicochemical properties of TS-1 and Fe-TS-1
catalysts from N2 adsorption–desorption (d is pore size, V is
pore volume)
SBET, m2/g
V, cm3/g
Catalysts
d, nm
TS-1
493.75
488.74
492.32
500.52
0.57
0.62
0.62
0.62
0.191
0.188
0.189
0.193
FNTS-1
FSTS-1
FCTS-1
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 91 No. 11 2017

2106
MEI WU et al.
(b)
460.3
711
(a)
459.1
714
FCTS-1
FSTS-1
TS-1
FSTS-1
FCTS-1
FNTS-1
FNTS-1
450
455
460
465
470
475
705 710 715 720 725 730 735 740
Binding energy, eV
Binding energy, eV
Fig. 2. The XPS spectra of TS-1, FNTS-1, STS-1, and FCTS-1 samples: (a) Ti 2p, (b) Fe 2p.
Ti (Si–O–Ti), and intensity of band at 459.1 eV sug- with the Ti 2p3/2 results. It can be found that character-
gested the content of Ti sites which interacted with Fe
species (Fe–O–Ti). The ratio of the two band inten-
sity reflects the relative content of Fe–O–Ti and Si–
O–Ti. Making a comparison among FC-TS-1, FS-
TS-1, and FN-TS-1, it can be found that the content
of Fe–O–Ti in FN-TS-1 is much higher than the
other two catalysts. This implies that more iron species
interacting with TS-1 were formed by using iron
nitrate as Fe precursor.
For Fe 2p XPS spectra (Fig. 2b), two well-defined
peaks ascribed to Fe 2p3/2 (711 eV) and Fe 2p1/2
(724 eV) show up. Binding energies of these two peaks
correspond well with characteristic Fe3+ according to
literature [19, 20]. Compared with pure Fe2O3, Fe
2p3/2 peaks for all Fe-TS-1 samples show a little posi-
tive shift in binding energy from 711 to 714 eV. This
could be due to the electron transfer from Fe3+ to Ti4+.
Therefore, Fe species in Fe-TS-1 have lower density of
electron cloud. This further confirms that Fe coordi-
nated with Ti active site via O and is in accordance
istic peak at 714 eV for FN-TS-1sample had the high-
est intensity, FS-TS-1 takes second place, and FC-
TS-1 has the lowest intensity. This indicates the Fe–
O–Ti content of FN-TS-1 is higher than the other two
samples, again confirmed the Ti 2p3/2 XPS results.
The XPS characterization results including Ti 2p3/2
and Fe 2p1/2 core-level spectra elucidate that Fe pre-
cursors play a significant role in Fe and Ti electronic
performance. The effect of precursors on metal elec-
tronic status also influences interaction ways between
Fe and Ti active center. Specifically, catalyst derived
from iron nitrate precursors offers most amount of
Ti–O–Fe, which can improve catalytic performance.
The local environment of Ti and Fe in FNTS-1,
FSTS-1, and FCTS-1 samples as well as TS-1 were
analyzed by DR UV–Vis spectroscopy, as shown in
Fig. 3. Only one absorption peak at 210 nm was
recorded in TS-1 sample, which is originated from the
electronic transfer of the pπ–pπ transitions between
titanium and oxygen in the framework titanium spe-
cies [21]. However, in comparison with TS-1, other
additional spectra bands are exhibited in Fe-TS-1
samples. As testified in some references, bands at
220 nm are ascribed to tetra-coordinate Fe3+, bands at
350 nm are assigned to iron oligomeric clusters, bands
at 537 nm belong to Fe2O3 particles [22, 23]. Differ-
ences of the band intensity and peak shape at 350 or
537 nm between three Fe-TS-1 catalysts illustrated
that Fe coordination mode was variable according to
Fe precursors. As can be seen, iron oligomeric clusters
are dominantly formed for FSTS-1. Fe2O3 particles
are generated in FCTS-1 and FNTS-1 samples. While
FNTS-1 offers the most amount of tetra-coordinate
Fe3+ species among three modified samples.
210 nm
TS-1
220 nm
537 nm
FSTS-1
FCTS-1
FNTS-1
200
300
400
500
600
700
As can be suggested above, Fe precursor affects the
chemical state of Fe species in Fe-TS-1 catalysts. Of
three precursors, A positive effect on the catalytic per-
formance can be taken by FN-TS-1 due to its most
λ, nm
Fig. 3. The UV–Vis spectra of various catalysts.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 91
No. 11
2017

PRECURSOR EFFECT ON THE PROPERTY AND CATALYTIC BEHAVIOR OF Fe-TS-1
content of tetra-coordinate Fe3+. This conclusion is
2107
c
squared with above XPS results.
b
a
350°C
H2-TPR profiles of FNTS-1, FSTS-1, and FCTS-1
samples are showed in Fig. 4. All peaks detected
should be caused by the reduction of iron species
introduced in TS-1 TPR plots of FNTS-1 and FCTS-
1 samples exhibit a single reduction peak at 550°C,
attributed to the reduction of Fe3+ in TS-1 to Fe2+
[24]. However, two consumptions are observed for
FSTS-1 sample. The first low-temperature reduction
peak centered at 350°C was attributed to the reduction
of bulk or dispersed extra framework iron oxide spe-
cies deposited in the channels or outside the zeolite,
the second at 800°C was assigned to the drastic reduc-
tion of Fe-S oligomeric clusters inside the zeolite to Fe
[25]. H2-TPR results reconfirmed that the main Fe
550°C
800°C
a: FNTS-1
b: FSTS-1
c: FCTS-1
200 300 400 500 600 700 800
T, °C
species for FNTS-1 was the framework Fe3+, while
that for FSTS-1 sample was Fe–S oligomeric clusters
inside the zeolite.
Fig. 4. H -TPR curves of samples.
2
550°C heat treatment. This is in accordance with the
XRF results, in which there is S residual. Accordingly,
the main Fe specie in FSTS-1 catalyst is Fe–S oligo-
meric clusters.
NH3-TPD is widely used to determine the total
acidity of solid surface. TS-1 and Fe-TS-1 samples
were characterized by NH3-TPD to investigate the
surface acidic properties. As shown in Fig. 6, it is
observed that two weak stripping peaks are presented
The element content in bulk and surface for all of
the catalysts were analyzed by XRF and XPS, respec-
tively (Table 2). The bulk content of Si and Ti for TS-1
remained the theoretical content, even after modified
with various Fe precursors, the atom ratio of Si/Ti still
maintained around 44. However, the surface Si/Ti
ratio for Fe-TS-1 was larger than TS-1 sample, espe-
cially FNTS-1. Probably, a part of Si atoms immigrate
from bulk to the surface during modifying TS-1 using
iron salts and the reconstruction on the surface was at 220 and 720oC due to TS-1, which are ascribed to
occurred. Meanwhile, parts of Fe interacted with TS-1 weak acid sites and strong acid sites, respectively. The
and incorporated into the skeleton. The fact of surface surface acidic sites of TS-1 come from two aspects.
Fe/Ti ratio more than bulk indicated that Fe2O3 parti- Firstly, Ti species in TS-1 zeolite skeleton structure
presented as tripodal [Ti(OSi)3(OH)] develop weak
It was notable that Fe/Ti ratio of FNTS-1 was the low- acid centers. Secondly, unsaturated framework Si
cles enriched on the outer surface of Fe-TS-1 samples.
est in three Fe-TS-1 samples, which predicated Fe3+
species inside TS-1framework for FNTS-1 was higher
than the others. In addition, 0.4% of residual S was
detected in FSTS-1 and no Cl was found in FCTS-1,
which accounted that the sulfate precursor was not
thoroughly decomposed. The existence of sulfates
would enhance the surface acidity of the catalysts.
coordinating with surface hydroxyl presented as
[Si(OSi)3(OH)] produce strong acid centers [25, 28].
Whereas, saturated [Ti(OSi)4] and [Si(OSi)4] species
form the majority of Ti and Si species in TS-1 frame-
work. As a result, the NH3 desorption curves of TS-1
are very weak. As can be seen, precursors make nota-
ble variation of TS-1 surface acidic properties. Of
desorption curves, it is shown that FN-TS-1 has a
wider low-temperature desorption peak as well as a
similar high-temperature desorption peak compared
As shown in the Fig. 5, the TGA analysis of FNTS-1,
FSTS-1, and FCTS-1 samples were carried out,
respectively. Weight loss steps in the temperature
ranges 20–150 and 550–850°C are found in all sam-
ples. The weight loss at 20–150°C is attributed to the
emission of H2O. And that at 550–850°C is attributed
to the Curie point of Fe3O4 [26, 27]. Specifically, in
the case of FNTS-1, the weight loss at 180–300°C is
ascribed to the decomposition of Fe(NO3)3. For FC-
TS-1 sample, the obvious weight loss at 20–350°C is
ascribed to the decomposition of FeCl3. Generally
speaking, the decomposition temperature of
Fe2(SO4)3 is about 500°C. According to the FS-TS-1
TGA pattern, weight loss continuously take place
around 150–500°C. This may be attributed to residual
Fe-S species that cannot completely decompose under
Table 2. Bulk and surface atomic ratios of all catalysts
Surface atomic
Bulk atomic ratios
ratios
Catalyst
SO3,
wt %
Cl,
wt %
Si/Ti Fe/Ti
Si/Ti Fe/Ti
TS-1
FNTS-1 42.4
FSTS-1 42.4
FCTS-1 44.0
44.3
–
–
–
0.4
–
–
–
–
–
65.7
86.7
72.0
78.8
–
0.25
0.20
0.24
1.5
1.7
2.3
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 91 No. 11 2017

2108
MEI WU et al.
100
99
98
97
96
95
94
220°C
280°C
720°C
FSTS-1
FCTS-1
FNTS-1
c
a
b
200
400
600
800
1000
TS-1
600
T, °C
200
400
800
1000
T, °C
Fig. 5. TGA curve of the as-prepared FNTS-1 (a), FCTS-
1 (b), and FSTS-1 (c).
Fig. 6. The NH -TPD spectra of TS-1, FNTS-1, FSTS-1,
3
and FCTS-1.
to TS-1. It indicated that weak acid sites of TS-1 are
enhanced by iron nitrate precursors treated. While the
strong acid sites of it are not influenced. In compari-
son with FNTS-1 catalysts, even broader bands
including low-temperature and high-temperature
desorption peak are observed in FCTS-1 and FS-TS-1
samples. It is revealed the total amount of acid sites
either weak or strong in the case of FSTS-1 and
FCTS-1 is much more than that in FNTS-1.
extra-framework Fe2O3 presented in modified catalyst
abundantly in FCTS-1 sample, while a fraction of iron
interacted with Ti centers, and weak acid site on the
TS-1 surface derived from extra-framework Fe2O3
increased. Thus, H2O2 was partly decomposed rather
than turned into epoxidation product. Although H2O2
conversion increased from 83 to 91%, VO yield and
H2O2 utilization were depressed. Especially sulfate will
generate a lot of strong acid sites on the surface of TS-1
owing to the incomplete decomposition of sulfate at
According to TGA and element analyzed results,
Fe(NO3)3 is entirely decomposed with no N residuals
after calcining. As a result, the faint enhancement of
weak acid sites is on account of Fe species instituted
into TS-1 framework. While in the case of FCTS-1, the calcination temperature. These strong acids would
the increase of weak and strong acid sites is attributed
to the presence of Fe2O3 species. Also for FSTS-1,
impregnated sulfate ions have the most significant
influences on the TS-1 total surface acidity [29]. Acid
sites was generally thought as centers to activate H2O2
[30]. However, the appropriate acid sites and acidity
strength will be facilitated to oxidation and excess will
initiate the invalid decomposition of H2O2 or epoxides
solvolysis in methanol. From the above NH3-TPD
profiles, the acid sites in Fe-TS-1 catalyst modified by
iron nitrate were slightly higher than TS-1, which could
be response for the better activity for epoxidation.
result in side reactions such as epoxides solvolysis and
the quick decomposition of H2O2 et al. As a result, VO
yield and H2O2 utilization of FSTS-1 catalyst dramat-
ically decreased.
100
VO yield
H2O2 conversion
H2O2 utilization
0.5
0.4
0.3
0.2
0.1
0
80
60
40
Figure 7 depicted results of butadiene epoxidation
over TS-1, FNTS-1, FSTS-1, and FCTS-1 catalysts.
It can be seen that Fe-TS-1 catalyst treated by iron
nitrate exhibited higher activity, showing the positive
effect of Fe. Butadiene epoxidation experiments con-
firmed that interaction between Fe and Ti centers acti-
vated TS-1 catalytic performance. As exhibited in Fig. 6,
H2O2 conversion, VO yield, and H2O2 utilization were
effectively improved. This result was in accordance
with that of NiNO3 modified TS-1 catalyst [31]. How-
ever, the chloride and sulfate induced a negative func-
tion on the catalytic performance of the final Fe-TS-1
catalysts. The catalysts characterizations showed that
20
0
TS-1
FSTS-1 FN-TS-1 FC-TS-1
Fig. 7. Catalytic performance of TS-1, FNTS-1, FSTS-1,
and FCTS-1 samples on the butadiene epoxidation (reac-
tion conditions: 35°C; 60 min; butadiene, 0.15 MPa; H O
2
2
(40 wt %), 0.55 mol/L; catalysts, 0.20 g; CH OH,
3
25.00 mL).
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A
Vol. 91
No. 11
2017

PRECURSOR EFFECT ON THE PROPERTY AND CATALYTIC BEHAVIOR OF Fe-TS-1
2109
9. S. Huang, J. Zhang, Y. Wang, P. Chen, S. Wang, and
CONCLUSIONS
X. Ma, Ind. Eng. Chem. Res. 53, 5838 (2014).
Iron nitrate, iron chloride and iron sulfate were
employed as Fe precursors to modify TS-1. Butadiene
epoxidation over these Fe-TS-1 catalysts was tested,
which indicated the iron precursors affected the cata-
lytic performance of the final catalysts greatly. The
characterizations such as H2-TPR, UV–Vis spectra,
FT-IR spectra, and NH3-TPD point out that the
interaction between iron and TS-1 and the surface
acid sites can be responsible for the different activity.
Abundant Fe2O3 particles are formed to give rise to an
increase of weak acid sites due to iron chloride precur-
sor, which was unfavorable to the epoxidation of buta-
diene. FSTS-1 behaves a large amount of acid sites
including strong acid owing to the residue of the sul-
fate precursor and exhibits poor catalytic perfor-
mance. It further confirmed electronic effect of tran-
sition metal on Ti centers would activate TS-1 activity.
10. A. Ribera, I. W. C. E. Arends, S. de Vries, J. Perez-
Ramírez, and R. A. Sheldon, J. Catal. 195, 287 (2000).
11. J. S. Choi, S. S. Yoon, S. H. Jang, and W. S. Ahn, Catal.
Today 111, 280 (2006).
12. F. Adam, J. Andas, and I. Ab Rahman, Chem. Eng. J.
165, 658 (2010).
13. J. Garcia-Aguilar, I. Miguel-Garcia, J. Juan-Juan,
I. Such-Basanez, E. San Fabian, D. Cazorla-Amoros,
and A. Berenguer-Murcia, J. Catal. 338, 154 (2016).
14. M. Wu, L. Chou, and H. Song, Chin. J. Catal. 34, 789
(2013).
15. A. Thangaraj, R. Kumar, and P. Ratnasamy, Appl.
Catal. 57, L1 (1990).
16. T. Armaroli, F. Milella, B. Notari, R. J. Willey, and
G. Busca, Top. Catal. 15, 63 (2001).
17. C. Zhang, B. Lu, X. G. Wang, J. X. Zhao, and Q. H. Cai,
Catal. Sci. Technol. 2, 305 (2012).
18. D. Prasetyoko, Z. Ramli, S. Endud, and H. Nur,
ACKNOWLEDGMENTS
J. Mol. Catal. A: Chem. 241, 118 (2005).
The authors sincerely acknowledge the financial
support from the National Nature Science Founda-
tion of China (No.21606098, 51574130). Key Labora-
tory for Palygorskite Science and Applied Technology
of Jiangsu Province (HPK201402), key scientific and
technical tackle-key-problem project of Huai’an city
(HAG2015011), policy guidance program of Jiangsu
province (BY2016064-01) Colleges Natural Science
Foundation of the Jiangsu province (15KJB530003),
and State Key Laboratory of Materials-Oriented
Chemical Engineering (No.KL14-14).
19. A. Gervasini, C. Messi, P. Carniti, A. Ponti, N. Rava-
sio, and F. Zaccheria, J. Catal. 262, 224 (2009).
20. S. Roy, B. Viswanath, M. S. Hegde, and G. Madras,
J. Phys. Chem. C 112, 6002 (2008).
21. J. Zhuang, D. Ma, Z. Yan, X. Liu, X. Han, X. Bao,
Y. Zhang, X. Guo, and X. Wang, Appl. Catal. A: Gen.
258, 1 (2004).
22. M. Schwidder, M. S. Kumar, K. Klementiev, M. M. Pohl,
A. Bruckner, and W. Grunert, J. Catal. 231, 314 (2005).
23. L. Ma, J. H. Li, R. Ke, and L. X. Fu, J. Phys. Chem. C
115, 7603 (2011).
24. J. Perez-Ramirez, F. Kapteijn, G. Mul, and J. A. Mou-
lijn, J. Catal. 208, 211 (2002).
REFERENCES
25. C. Lamberti, S. Bordiga, D. Arduino, A. Zecchina,
F. Geobaldo, G. Spano, F. Genoni, G. Petrini, A. Car-
ati, F. Villain, and G. Vlaic, J. Phys. Chem. B 102, 6382
(1998).
1. M. G. Clerici and P. Ingallina, J. Catal. 140, 71 (1993).
2. S. Krijnen, P. Sanchez, B. T. F. Jakobs, and J. H. C. van
Hooff, Microporous Mesoporous Mater. 31, 163
(1999).
26. K. S. Suslick, S.-B. Choe, A. A. Cichowlas, and
3. L. Wang, Y. Liu, H. Zhang, H. Wu, Y. Jiang, P. Wu, and
M. W. Grinstaff, Nature 353, 414 (1991).
M. He, Chin. J. E: Catal. 27, 656 (2006).
27. R. Vijayakumar, Y. Koltypin, I. Felner, and A. Gedan-
4. A. Wroblewska, A. Fajdek, J. Wajzberg, and
ken, Mater. Sci. Eng. A 286, 101 (2000).
E. Milchert, J. Adv. Oxid. Technol. 11, 468 (2008).
28. C. Lamberti, S. Bordiga, A. Zecchina, A. Carati,
A. N. Fitch, G. Artioli, G. Petrini, M. Salvalaggio, and
G. L. Marra, J. Catal. 183, 222 (1999).
5. X. M. Zhang, Z. R. Zhang, J. S. Suo, and S. B. Li,
Catal. Lett. 66, 175 (2000).
6. V. Arca, A. Boscolo Boscoletto, N. Fracasso, L. Meda,
and G. Ranghino, J. Mol. Catal. A: Chem. 243, 264
(2006).
29. D. Nedumaran and A. Pandurangan, Microporous
Mesoporous Mater. 169, 25 (2013).
30. G. Bellussi and M. S. Rigutto, Adv. Zeolite Sci. Appl.
7. J. Jiang, R. Li, H. L. Wang, Y. F. Zheng, H. N. Chen,
85, 177 (1994).
and J. T. Ma, Catal. Lett. 120, 221 (2008).
8. M. Mhamdi, S. Khaddar-Zine, and A. Ghorbel, Appl. 31. M. Wu, L. Chou, and H. Song, Catal. Lett. 142, 627
Catal. A: Gen. 357, 42 (2009). (2012).
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 91 No. 11 2017