Enhanced metal loading in SBA-15-type catalysts facilitated by salt addition: Synthesis, characterization and catalytic epoxide alcoholysis activity of molybdenum incorporated porous silica

Article abstract of DOI:10.1016/j.apcata.2014.01.055

We report a novel method to increase the metal loading in SBA-15 silica matrix via direct synthesis. It was demonstrated through the synthesis and characterization of a series of molybdenum containing SBA-15 mesoporous silica catalysts prepared with and without diammonium hydrogen phosphate (DHP) as an additive. Catalysts prepared with DHP show a 2-3 times increase in incorporation of molybdenum in the silica matrix and pore size enlargement. The synthesized catalysts were characterized using nitrogen sorption, X-ray diffraction (XRD), Raman spectroscopy, scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS) and inductively coupled plasma-optical emission spectroscopy (ICP-OES). The catalytic activity of catalysts prepared with DHP for alcoholysis of epoxides was superior than the catalyst prepared without DHP. Alcoholysis of epoxides was demonstrated for a range of alcohols and epoxides under ambient conditions in as little as 30 min with high selectivity.

Full text of DOI:10.1016/j.apcata.2014.01.055

Applied Catalysis A: General 475 (2014) 469–476
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Enhanced metal loading in SBA-15-type catalysts facilitated by salt
addition: Synthesis, characterization and catalytic epoxide alcoholysis
activity of molybdenum incorporated porous silica
Sridhar Budhia,b, Chorthip Peeraphatdita, Svitlana Pylypenkoc, Vy H.T. Nguyena,d
,
Emily A. Smitha,d, Brian G. Trewyna,b,∗
a Department of Chemistry, Iowa State University, Ames, IA 50011, United States
b Department of Chemistry and Geochemistry, Colorado School of Mines, Golden, CO 80401, United States
c Department of Metallurgical and Materials Engineering, Colorado School of Mines, Golden, CO 80401, United States
d U.S. Department of Energy, Ames Laboratory, Ames, Iowa 50011, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
We report a novel method to increase the metal loading in SBA-15 silica matrix via direct synthesis.
It was demonstrated through the synthesis and characterization of a series of molybdenum contain-
ing SBA-15 mesoporous silica catalysts prepared with and without diammonium hydrogen phosphate
(DHP) as an additive. Catalysts prepared with DHP show a 2–3 times increase in incorporation of molybde-
num in the silica matrix and pore size enlargement. The synthesized catalysts were characterized using
nitrogen sorption, X-ray diffraction (XRD), Raman spectroscopy, scanning electron microscopy (SEM),
transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS) and inductively cou-
pled plasma–optical emission spectroscopy (ICP–OES). The catalytic activity of catalysts prepared with
DHP for alcoholysis of epoxides was superior than the catalyst prepared without DHP. Alcoholysis of
epoxides was demonstrated for a range of alcohols and epoxides under ambient conditions in as little as
30 min with high selectivity.
Received 28 December 2012
Received in revised form 23 January 2014
Accepted 30 January 2014
Available online 7 February 2014
Keywords:
SBA-15
Alcoholysis
Epoxide
Molybdenum
Salt effect
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
templates under strong acidic conditions. However, unlike MCM-
41-type synthesis, incorporation of metal ions in the framework
Mesoporous materials have gained momentum after the dis-
covery of ordered mesoporous silicates (M41S) by scientists at
Mobil Corporation [1] two decades ago. Since then it has gained
global interest by addressing pressing problems of society such as
energy and environment. Metal incorporated porous silica materi-
als have been tested in applications such as catalysis [2], hydrogen
energy [3], etc. as efficient and reusable catalysts. The most com-
prehensively studied porous silica material for various applications
(catalysis, host–guest chemistry, chromatographic separation) is
MCM-41 since it has high specific surface area with uniform
mesoporous channels [4]. However, small pore size and limited
hydrothermal stability has limited its applications.
of this mesoporous silica support for downstream applications
is challenging because of strong acidic synthetic conditions. The
highly acidic synthetic conditions of SBA-15 are detrimental for
incorporation of metal ions through co-condensation [6] as it
breaks the Si–O–metal bond. Additionally, under SBA-15 synthetic
conditions, highly solubilized metal ions fail to precipitate and
incorporation in the silica framework is not effective. Thus, post-
synthetic grafting [7] is the widely used technique for doping metal
ions in SBA-15 framework. Owing to these difficulties there have
only been a few reports in the literature for direct incorporation
of metal ions in SBA-15 framework. Vinu and coworkers [8] have
optimized the synthetic conditions to synthesize SBA-15 under
relatively low acidic conditions for improved incorporation of
metal ions. It was later determined that metal incorporation
efficiency was better under these less acidic conditions [9].
Improvements and tailoring of mesoporous silica for specific
applications are often done by adding additives to the reaction
mixture. Hydrocarbons added during the synthesis of mesoporous
materials influences average pore size, crystallinity and pore size
distribution [10]. Addition of inorganic salts is another approach
to modify the textural properties of mesoporous silica. Addition
of salts such as MgCl2, Ni(CH3COO)2, and Mg(CH3COO)2 increased
Stucky and coworkers [5] developed a new class of mesoporous
silica called SBA-15 containing uniform cylindrical pores with
tunable pore size (5–30 nm) and thick pore walls using environ-
mentally benign non-ionic block copolymers as structure-forming
∗
Corresponding author at: Colorado School of Mines, Department of Chemistry
and Geochemistry, 152 Coolbaugh Hall, 1500 Illinois St, Golden, CO 80401, United
States. Tel.: +1 11 303 384 2224.
E-mail address: btrewyn@mines.edu (B.G. Trewyn).
0926-860X/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2014.01.055

470
S. Budhi et al. / Applied Catalysis A: General 475 (2014) 469–476
the periodicity of mesophase structure as demonstrated by Wang
et al. [11]. Enlargement of the average pore size was observed when
NaCl was added to reaction mixture when Krämer et al. [12] syn-
thesized cubic Ia3d mesoporous silica. Tunable morphologies were
reported by employment of K2SO4 and Na2SO4 by the Stucky and
Zhao research groups [13,14]. Despite these advancements, to the
best of our knowledge, there are no literature reports for the role
of salts in dictating the incorporation of metals in a silica matrix.
Epoxidation, one of the most studied reactions in the litera-
ture, is of academic and industrial importance. It is a valuable
intermediate to yield a range of products with applications in
the pharmaceutical, polymer and agrochemical industries through
regioselective ring opening. Nucleophiles such as alcohols, amines,
cyanides, hydroxides, halides to name a few, can open epox-
ides. Ring opening of epoxides by alcohols (alcoholysis) yields
␤-alkoxyalcohols, which are precursors for mandelic acid and
antibacterial agents including ␤-lactam antibiotics. Ring opening
of epoxides are chemically cleaved by acid or base catalysts under
elevated temperatures. In addition to acids and bases, several metal
ions such as Al(III), Sn(II), Sn(IV), Co(III), triflates, Cr(III), and Lewis
acid supported metal–organic frameworks were recently reported
as catalysts for ring opening of epoxides [15]. However, these cat-
alysts are either toxic, less abundant in nature, involve complex
preparatory procedures for synthesis, energy-intensive or require
prolonged reaction times.
Next to titanium, molybdenum is the most studied transition
metal through incorporation into silica matrix due to its wide cat-
alytic applications. Molybdenum is widely studied in oxidation
reactions, petroleum chemistry [16] and recently in the conversion
of biomass into renewable energy [17]. A few examples of reac-
tions catalyzed by molybdenum are epoxidation of olefins [18],
decomposition of NOx [19], hydrodeoxygenation, hydrodenitro-
genation, hydrodesulfurization, alkane oxidation [20], oxidative
dehydrogenation [21], metathesis [22] and transesterification [23].
To the best of our knowledge, there have been only two reports
where Mo was used to study the ring opening of epoxides on
alumina support [24,25]. However, their studies failed to distin-
guish whether the catalytic activity was due to Mo or Al(III) ions
on the support. Graham et al. [26] demonstrated ring opening of
epoxides using aluminosilicates where Al(III) was the active cat-
alyst. Thus, the reports of catalytic activity by molybdenum on
alumina for ring opening of epoxides could potentially be due to the
presence of both molybdenum and alumina. These reports clearly
indicate that participation of supports on catalytic activity cannot
be ignored.
the molybdenum precursor and additive, respectively. Triblock
copolymer poly(ethylene glycol)-block-poly(propyleneglycol)-
block-poly(ethylene
glycol)
(Pluronic
P123,
MW = 5800,
EO20PO70EO20 Aldrich) was used as the structure-directing
,
template. Tetraethylorthosilicate (Aldrich) was used as the silica
source for the synthesis of SBA-15. HCl (2 M) solution was prepared
from 37 wt% HCl purchased from Fisher Chemical. All epoxides and
alcohols tested for catalytic activity were purchased from Fisher
and Aldrich. All reagents were used as received without further
purification.
Synthesis of Molybdenum incorporated SBA-15 catalyst: In a
typical synthesis, 4 g of pluronic P123 was added to 30 mL nanop-
ure water in a polypropylene bottle at 313 K and stirred at 600 rpm
for 3 h. A solution of previously mixed HCl (2 M, 10 mL) in 60 mL
of water was also added to the dissolved template solution and
stirred for another 1 h. TEOS (9 g, 4.3 mmol) was added drop wise
to the reaction mixture, followed by quick addition of the required
amount of ammonium heptamolybdate tetrahydrate. The samples
were labeled as Xg-Mo-SBA-15-TTT. Xg denotes amount of molyb-
denum precursor added and TTT represents the hydrothermal
treatment temperature in degrees Celsius. For the syntheses that
involved addition of the additive, diammonium hydrogen phos-
phate, labels were Xg-MoP-SBA-15-TTT, where Xg is the amount in
grams of molybdenum and DHP added and TTT as defined above.
In cases where different amounts of molybdenum precursor and
DHP were added it was denoted by XgP-YgMo-SBA-15-TTT where
Xg and Yg is amount of DHP and molybdenum precursor in grams,
respectively. Whenever the additive was included, it was added
along with the molybdenum precursor. After addition of all com-
ponents, the reaction mixture was stirred for another 24 h at 313 K
and then subjected to hydrothermal treatment at the desired tem-
perature for an additional 48 h. The reaction mixture was cooled,
filtered and washed with methanol and water. The resulting sample
was oven dried overnight at 373 K and then followed by calcina-
tion at 550 ◦C for 8 h to remove the template in presence of flowing
air.
2.2. Characterization
X-ray diffractograms (XRD) of synthesized samples were
recorded utilizing a Rigaku Ultima (IV) diffractometer using Cu K␣
radiation source. The diffractograms were recorded from 0.5◦ to
10◦ with a step size of 0.02◦. Wide angle XRD was recorded for
certain samples from 20◦ to 80◦ at a rate of 1◦ min−1. Surface anal-
yses of samples to determine surface area and pore sizes were
measured utilizing nitrogen sorption analysis in a Micromeritcs
ASAP 2020 analyzer. The samples were degassed for 6 h at 373 K
prior to measurements. The Brunauer–Emmett–Teller (BET) and
the Barrett–Joyner Halenda (BJH) equations were used to calcu-
late specific surface area and pore size distributions, respectively.
Transmission electron microscopy (TEM) imaging was done using
a Tecnai F2 microscope. Particle morphology was determined by
scanning electron microscopy (SEM) using a Hitachi S4700 FE-
SEM system with 10 kV accelerating voltage. Raman spectra were
collected using a previously described instrument with 785 nm
excitation and a 10×, 0.3 numerical aperture objective to collect
the Raman scatter [27]. The laser power at the sample was 6 mW.
Spectra were collected with a 30-s acquisition, and were back-
ground subtracted using a spectrum collected with no sample in
the sample holder. XPS was used to characterize the oxidation
states of Mo in the catalysts on a Kratos Nova X-ray photoelec-
tron spectrometer supplied with a monochromatic Al K␣ source
operating at 300 W. Casa XPS was used for analysis and quantifica-
tion of spectra, sensitivity factors were supplied by manufacturer.
A linear background was applied to C 1s, O 1s, and Si 2p regions,
and a Shirley background was applied to Mo 3d region. Analysis
Mo incorporated SBA-15 is often prepared by post-synthesis
impregnation and grafting rather than direct synthesis due to poor
metal incorporation. Apart from highly acidic conditions, the oxida-
tion state of molybdenum ions (+6) make isomorphic substitution
by Si (+4) challenging. Thus, we need a comprehensive prepara-
tory procedure for direct synthesis of metal incorporated SBA-15.
Herein, for the first time, we report the synthesis of Mo-SBA-
15 through co-condensation using DHP as an additive added in
the reaction mixture. Addition of DHP was found to influence the
incorporation of molybdenum along with other textural proper-
ties. These catalysts were successfully tested for alcoholysis under
ambient conditions for a wide range of alcohols.
2. Experimental
2.1. Materials
Ammonium heptamolybdate tetrahydrate [(NH4)6Mo7O24
·4H2O] and diammonium hydrogen phosphate [(NH4)2HPO4,
DHP] were purchased from Fisher Chemicals, and were used as

S. Budhi et al. / Applied Catalysis A: General 475 (2014) 469–476
471
included charge referencing to the adventitious carbon signal at
3.2. Textural properties
285 eV. After calibration, Si 2p spectrum is centered at 103.5 eV,
binding energy typical for SiO2. Metal loading was quantified on
a Perkin Elmer ICP-OES model Optima 2100DV. Approximately
5 mg of catalyst was dissolved in 10 mL aliquot of 500 mL solu-
tion prepared from mixing 50 ␮L of 36% HF and 500 ␮L of aqua
regia.
Nitrogen physisorption isotherms of samples synthesized with
and without DHP are seen in Fig. 2. Table 1 summarizes textural
properties of all synthesized materials along with metal loading
and incorporation efficiency. All synthesized samples except for
2g-Mo-SBA-15 have type (IV) adsorption isotherm characteris-
tic of mesoporous materials according to IUPAC classification and
H1 type broad hysteresis loop typical for large pore mesoporous
solids. The amount of nitrogen adsorbed for samples prepared with-
out DHP decreased with increasing amount of metal precursor
indicating that the surface area decreased with increasing addi-
tion of metal precursor. However, for the samples prepared with
DHP the amount of nitrogen adsorbed decreased until 1g-MoP-
SBA-15 and further increases in metal precursor did not result in
increased metal incorporation and led to increased surface area.
Surface area values of samples are the inverse to the amount of
molybdenum incorporated in the final catalyst. The amount of
metal incorporated in samples synthesized with DHP is 2–3 times
greater than the corresponding samples synthesized without salt
additive. Hence, surface area of samples prepared with DHP is
less than the corresponding samples synthesized without DHP.
Increasing hydrothermal temperature during synthesis decreased
the molybdenum loading, however the surface area was not mea-
surably affected. The metal loading wt% for samples synthesized
under different hydrothermal temperatures with and without DHP
was only marginally affected. In an effort to understand the role
of DHP in metal loading, we synthesized three additional sam-
ples varying the added salt amount (0.25 g P, 0.5 g P and 1.5 g
P) while keeping the amount of molybdenum precursor constant.
The surface area of 1g-Mo-SBA-15-100 dropped from 628 m2/g to
441 m2/g when 0.25 g of DHP was added during synthesis, which
led to increased metal loading (Table 1). When the DHP amount
was increased to 1 g the metal incorporation reached 12.8 wt%
and the surface area decreased to 229 m2/g and further addition
of DHP led to a drop in metal loading and an increase in surface
area. In general, of the samples we synthesized it was generally
observed that increased metal loading led to a decrease in surface
area.
Except for 2g-Mo-SBA-15-100, all other samples that we syn-
thesized, we observed a steep rise in the nitrogen physisorption
isotherm at a relative pressure of 0.6 indicating a narrow pore
size distribution. The reported pore sizes in Table 1 are based
on BJH calculations. Pore diameter increased with increasing
hydrothermal temperature and amount of metal loading except
for 2g-Mo-SBA-15-100 for samples prepared without DHP. The 2g-
MoP-SBA-15-100 has very low pore volume and small pore size,
which is different than the trend observed for the rest of the sam-
ples, potentially due to pore blocking caused by excessive metal
loading. However, there is no clear trend in pore sizes between sam-
ples synthesized with DHP as we observed with samples prepared
without DHP. Interestingly, the pore sizes for samples containing
similar amounts of metal precursor prepared with DHP are larger
than samples prepared without DHP.
2.3. Catalytic studies
In a typical experiment, a 20 mL sample vial was charged with
a certain amount of desired catalyst followed by 1.1 mmol of epox-
ide and 3 mL of alcohol. The reaction was stirred using magnetic
stir bar at ambient temperature (396 K 2 K). The reaction mixture
was withdrawn at regular intervals and injected into a gas chro-
matograph (HP 5890, FID detector and DB-5 capillary column) to
monitor the progress of the reaction. Reactant conversion was cal-
culated with respect to epoxide since alcohols were used in excess
as the solvent and nucleophile.
3. Results and discussion
3.1. X-ray diffraction measurements
Fig. 1 shows the XRD of Mo incorporated SBA-15 samples con-
taining different amounts of molybdenum with and without DHP.
Samples synthesized with 0.25 g molybdenum precursor with and
without DHP (Fig. 1a) show (1 0 0), (1 1 0), and (2 0 0) reflections
characteristic of SBA-15 as described by Zhao et al. [5]. The sam-
ple synthesized with DHP with 0.25 g of molybdenum precursor
shows an additional reflection (2 1 0) indicating higher degree
of periodicity. However, (1 0 0), (1 1 0), and (2 0 0) reflections are
observed in the sample containing 0.5 g of molybdenum precur-
sor without DHP, while the corresponding sample with DHP has
a broad, unresolved reflection representing superimposed (1 1 0)
and (2 0 0) peaks apart from an intense (1 0 0) reflection. Samples
containing 1 g of molybdenum precursor with and without DHP
have a (1 0 0) reflection while low intense peaks are not clearly
observed suggesting a loss in periodicity with increased amounts
of molybdenum precursor and DHP during synthesis. When 2 g of
molybdenum precursor was added without DHP a complete loss
of structure was observed by XRD, while the XRD of the corre-
sponding sample with DHP suggests that it has undergone a phase
transformation similar to Ia3d structure since it has reflections
(2 1 1) and (2 0 0) similar to MCM-48 [28]. This supports a report
by Flodström et al. that suggests the addition of salt to SBA-15 syn-
thesis can induce phase transformation to Ia3d [29]. The XRD peak
intensity of all samples with DHP was greater than samples with-
out DHP indicating that the inclusion of salt brings greater pore
uniformity. Similar observations were made by Newalkar et al.
when they synthesized salt added silica SBA-15 [30]. The d100
reflection of samples with DHP appeared at lower 2ꢀ than with-
out DHP except for samples with the greatest amount of added
molybdenum (2g-MoP-SBA-15 and 2g-Mo-SBA-15-100) indicating
that adding DHP increases pore size. All the synthesized samples
were subjected to high angle X-ray diffraction analysis between
20◦ and 80◦. All samples prepared without DHP gave spectra that
correspond to the background for amorphous silica irrespective
of molybdenum loading. For the samples prepared with DHP, one
sample (1g-MoP-SBA-15-100) gave peaks (2ꢀ = 23.4, 25.8 and 27.5),
which are characteristic for molybdenum trioxide in an orthorhom-
bic phase as previously reported in the literature [31]. However,
the 2g-Mo-SBA-15-100 sample, despite the increased molybdenum
loading lacked the peaks characteristic for molybdenum triox-
ide.
3.3. Raman studies
Raman studies were performed for samples synthesized with
a 373 K hydrothermal treatment with and without DHP. Repre-
sentative Raman spectra for samples synthesized without DHP are
shown in Fig. 3. Samples prepared without addition of DHP and
less than 2 g Mo precursor did not show any discernible Raman
peaks, potentially due to a low metal loading that is below the
detection limit or highly dispersed metal species in the silica
matrix. The bands characteristic of ␣-MoO3 in Raman spectra are
161, 285, 293, 339, 381, 666, 819 and 996 cm−1 [32]. In sample

472
S. Budhi et al. / Applied Catalysis A: General 475 (2014) 469–476
Fig. 1. Powder XRD patterns of SBA-15 samples prepared in the presence and absence of additives with (a) 0.25 g, (b) 0.5 g, (c) 1 g, (d) 2 g, and (e) 1 g (wide angle) of
molybdenum precursor. The spectra in (e) are offset for clarity.
Fig. 2. Nitrogen sorption isotherms of metal incorporated SBA-15 with different amounts of molybdenum precursor in (a) absence of additives, (b) presence of additives.
Spectra are offset on the y-axis for clarity.
2g-Mo-SBA-15-100, intense peaks corresponding to ␣-MoO3 at
996, 819 and 667 cm−1 were observed. The band at 996 cm−1 cor-
responds to terminal ꢁ(Mo=O) of MoO3 and the band at 819 cm−1
is assigned to ꢁ(OMo2). The band at 667 cm−1 corresponds to
ꢁ(OMo3) [33]. Similar bands were observed for 1g-MoP-SBA-15-
100 (only, since it has highest loading for samples prepared by
adding DHP) with a higher intensity than 2g-Mo-SBA-15-100 (data
not shown). In previous reports, it was observed that bands at
910, 847, and 704 cm−1 correspond to the presence of ammonium
heptamolybdate [9]. The absence of ammonium heptamolybdate
bands in any of the analyzed samples (Fig. 3) indicates that there is
no unreacted metal precursor present and further, suggests that the
Table 1
Textural properties of Mo incorporated SBA-15 samples prepared in presence and absence of additives.
Sample name
Si/Mo (added)
Si/Mo (Expt)
Metal loading
(wt%)
Loading
efficiency (%)
Surface area
(m2/g)
Pore volume
(cm3/g)
Pore diameter
(nm)
0.25g-Mo-SBA-15-100
0.5g-Mo-SBA-15-100
1g-Mo-SBA-15-100
1g-Mo-SBA-15-130
1g-Mo-SBA-15-150
36.7
18.3
9.2
9.2
9.2
437
150
107
165
127
26
149
81
38
44
57
113
42
43
1.1
3.3
4.5
4.0
3.8
18.8
3.3
6.1
12.8
11.1
8.5
11.9
17.3
12.1
10.7
10.2
24.3
35.0
32.3
33.9
29.4
22.5
4.3
875
744
628
480
367
234
804
590
227
246
329
694
441
408
552
1.3
1.1
1.3
1.2
1.2
0.4
1.2
1.3
0.8
1.0
1.2
0.9
1.0
0.9
0.9
8.6
8.8
10.0
10.9
11
2g-Mo-SBA-15-100
6.1
7.6
9.0
0.25g-MoP-SBA-15-100
0.5g-MoP-SBA-15-100
1g-MoP-SBA-15-100
1g-MoP-SBA-15-130
1g-MoP-SBA-15-150
2g-MoP-SBA-15-100
0.25gP-1gMo-SBA-15-100
0.5gP-1gMo-SBA-15-100
1.5gP-1gMo-SBA-15-100
36.7
18.3
9.2
9.2
9.2
6.1
9.2
9.2
9.2
10.7
10.2
12
10.7
9.0
7.4
9.0
9.2
4.4
11.4
11.3
9.4
30.1
30.0
24
52

S. Budhi et al. / Applied Catalysis A: General 475 (2014) 469–476
473
OH
OR
Mo-SBA-15
Room Temperature
+
ROH
O
Scheme 1. General alcoholysis reaction catalyzed by Mo-SBA-15.
Table 2. Initially, we studied activity of catalysts prepared with-
out DHP. It was observed that the TOF decreased with increasing
molybdenum content. The selectivity of the desired product was
above 95% in all cases. The catalyst in entry 4 took two hours to
reach 98% conversion, nearly twice the time the catalyst in entry
3 took, despite the greater molybdenum loading in the former.
The molybdenum loading of 2g-Mo-SBA-15-100 is 18.8 wt% by
ICP, however for 1g-Mo-SBA-15-100 the molybdenum loading is
4.5 wt%, a two-fold Mo loading increase but the measured activ-
ity (TOF = 10) decreased significantly. We hypothesize that upon
increasing the amount of molybdenum salt added to the catalyst
synthesis the occurrence of clustering or sintering of the molyb-
denum increases leading to less catalytically active sites. Reactions
using catalysts prepared with DHP followed the same trend as with-
out DHP of decreasing TOF (entries 5–7), except for entry 8 where
the observed TOF increased. As more molybdenum is loaded in
the catalysts, the TOF decreases indicating a reduced quantity of
catalytic active sites. When DHP is added to the catalyst synthesis
reaction, a measured increase in the TOF occurs in the correspond-
ing catalyst (i.e. TOF for 0.25g-Mo-SBA-15–100 is 123 h−1 and for
0.25g-MoP-SBA-15-100 is 142 h−1).). We hypothesize that DHP,
while clearly increasing the metal loading as measured by ICP,
also may lead to fewer occurrences of active metal clustering or
aggregation. Clustering of the active species leads to fewer surface
accessible sites and, ultimately, reduced TOF. The increased TOF of
entry 8 was possibly due to decreased molybdenum loading and
more accessible catalytic centers. Reaction times were reduced by
more than half using catalysts prepared with DHP. Other alcoholy-
sis reports [37] of cyclohexene oxide under ambient conditions in
30 min are limited. Once we confirmed that the catalysts prepared
with DHP have better catalytic activity than catalysts prepared
without DHP, we conducted further experiments using catalysts
prepared with DHP. We investigated the alcoholysis using cata-
lysts prepared under different hydrothermal treatments (entries 7,
9, and 10). Results indicate that with increasing hydrothermal treat-
ment the TOF increased, however the reaction time and selectivity
has typically the same for these three catalysts. Further catalytic
studies were carried out using the catalyst in entry 10 because of its
superior performance. Slightly higher TOF of the catalyst in entry 8
may be due to different silica phases as is suggested in XRD analysis;
more investigation in this is needed. As a control, we have carried
out reactions using just silica (entry 11) without molybdenum and
no conversion of starting material was observed for 2 h. To verify
the influence of salt on catalytic activity, we synthesized a cata-
lyst with only DHP and no molybdenum precursor. These prepared
materials have no effect in catalysis (entries 11 and 12) suggesting
that catalytic activity was exclusively due to molybdenum in the
silica support.
Fig. 3. Raman spectra of molybdenum incorporated SBA-15 samples with different
amounts of loading without additive. Spectra are offset on the y-axis for clarity.
polymolybdate ions were completely converted to monomeric
molybdenum species.
3.4. XPS analyses and electron microscopy
To support the Raman data, XPS analysis was performed to
determine the oxidation state of molybdenum. Fig. 4a shows
Mo 3d spectrum typical of samples synthesized without DHP.
The 3d5/2 component of Mo species in this sample is located at
232.5 0.1 eV, which is typical for Mo(VI) oxide [34–36]. In con-
trary, spectra acquired from samples synthesized with DHP and
shown in Fig. 4b–e are quite wide and indicate presence of more
than one type of molybdenum species. In addition to MoO3 com-
ponent, these samples demonstrate additional component at BE
∼231.4 0.2 eV. The binding energy is higher than values typically
reported for Mo(IV) and therefore this peak cannot be assigned to
MoO2 [35,36]. Most likely, the appearance of this peak in all samples
synthesized with DHP is due to incorporation of Mo into the silica
framework [34]. Comparing samples with different molybdenum
loading, it is clear that the ratio of these species to the unmodi-
fied MoO3 is highest in the sample with lowest concentration of
molybdenum and lowest in the sample with highest molybdenum
loading. The Mo:Si ratio obtained from XPS can be found in the
supporting information.
Scanning electron microscopy (SEM) images of the series of sil-
ica supported Mo catalysts (different Mo loadings and with and
without DHP) can be seen in the supporting information. It appears
that the amount of Mo loading and the presence or absence of
DHP does not affect the morphology of the catalysts, all materi-
als appear to have a typical SBA-15 morphology. The transmission
electron microscopy (TEM) images of Mo-SBA-15 and MoP-SBA-15
catalysts shown in Fig. 5 reveal the highly ordered pore structure
of each of the catalysts, regardless of the Mo loading or the pres-
ence or absence of DHP. Additionally, energy dispersive X-ray (EDX)
spectroscopy results show the presence of Mo and Si when a sin-
gle catalyst particle was analyzed (Supporting information). This
suggests a homogeneous mixture of Mo and Si within individual
particles.
Different alcohols were evaluated for alcoholysis with cyclo-
hexene oxide as one substrate and 1g-MoP-SBA-15-150 as the
catalyst under ambient conditions and results are presented in
Table 3. Primary alcohols up to five-carbon chain length underwent
100% conversion with high selectivity for the expected product in
30 min. With further increase in carbon chain length (entries 6–10)
it took longer reaction times to achieve conversions of less than
100%, however; the selectivity remains fairly constant. Unsatu-
rated primary alcohols such as crotyl alcohol and allyl alcohol had
reactivities similar to short chained, saturated alcohols, however,
3.5. Catalytic studies
The catalytic activity of molybdenum incorporated SBA-15 cat-
alysts prepared with and without DHP was investigated. First, we
studied the alcoholysis of cyclohexene oxide with ethanol using
catalysts containing different molybdenum loadings prepared with
and without DHP. The general reaction is shown in Scheme 1
and the selectivity and turnover frequency (TOF) are tabulated in

474
S. Budhi et al. / Applied Catalysis A: General 475 (2014) 469–476
a)
b)
d)
240
240
240
238
238
238
236
234
232
230
230
230
228
228
228
Binding Energy (eV)
c)
236
234
232
240
238
236
234
232
230
228
Binding Energy (eV)
Binding Energy (eV)
e)
236
234
232
240
238
236
234
232
230
228
Binding Energy (eV)
Binding Energy (eV)
Fig. 4. High resolution XPS of Mo 3d spectra (a) 0.25g-Mo-SBA-15-100, (b) 0.25g-MoP-SBA-15-100, (c) 0.5g-MoP-SBA-15-100, (d) 1g-MoP-SBA-15-100, and (e) 2g-MoP-SBA-
15-100.
Table 2
Alcoholysis of cyclohexene oxide with ethanol catalyzed by catalyst synthesized under different hydrothermal treatments and containing different amounts of molybdenum.
a
Entry
Catalyst
Catalyst amount (mg)
Molybdenum mol% by substrate
Time (min)
Conv (%)
Sel (%)
TOF (h−1
)
1
2
3
4
5
6
7
8
0.25g-Mo-SBA-15-100
0.5g-Mo-SBA-15-100
1g-Mo-SBA-15-100
2g-Mo-SBA-15-100
0.25g-MoP-SBA-15-100
0.5g-MoP-SBA-15-100
1g-MoP-SBA-15-100
2g-MoP-SBA-15-100
1g-MoP-SBA-15-130
1g-MoP-SBA-15-150
1g-P-SBA-15-100
20
20
20
20
20
20
20
20
23
30
20
20
0.2
0.6
0.8
3.4
0.6
1.1
2.4
1
2.4
2.4
NA
NA
120
120
60
120
45
45
30
30
30
72
98
92
98
92
96
96
95
96
96
95
94
95
96
95
95
0
123
58
45
10
142
82
60
88
70
90
100
100
100
100
0
9
10
11
12
30
120
120
NA
NA
Si-SBA-15-100
0
0
Reaction condition: 1.1 mmol cyclohexeneoxide, 3 mL ethanol, temperature 396 2 K.
a
Turnover frequency = turnover number/time (h). Turnover number = mmols of reactant converted/mmols of active species in catalyst.

S. Budhi et al. / Applied Catalysis A: General 475 (2014) 469–476
475
Fig. 5. Transmission electron micrographs (TEM) of Mo and MoP SBA-15 catalysts: (a) 1.0g-MoP-SBA-15-100, (b) 0.5g-MoP-SBA-15-100, (c) 1.0g-Mo-SBA-15-100, and (d)
0.5g-Mo-SBA-15-100. All scale bars represent 200 nm.
the selectivity of crotyl alcohol was very low. Aralkyl alcohols
such as benzyl alcohol and 2-phenylethanol had typical activities
similar to small chain, saturated alcohols with high conversion and
selectivity. A secondary alcohol, 2-propanol, converted completely
to the corresponding ether but the bulkier cyclohexanol took twice
the time as 2-propanol for 98% conversion. Tertiary alcohols were
found to have the poorest reactivity of all the alcohols toward the
cyclohexene oxide. For example, only 75% of the t-butyl alcohol
converted to the corresponding ether in 1 h with poor selectivity
(75%). The low reactivity for t-butyl alcohol is likely due to the
hydroxyl group being sterically hindered by alkyl groups. This
hypothesis was strengthened further when reactivity was dropped
further from 75% to 32% when one of the methyl groups was
replaced by an ethyl group (t-amyl alcohol) and selectivity too
decreased from 75% to 65%. The poor reactivity of tertiary alcohols
has been observed elsewhere [38]. Despite the poor reactivity
of tertiary alcohols, the conversion we observed is better than
previous reports in terms of kinetics [38].
We extended our catalytic studies further to study ethanol ring
opening with different epoxides as shown in Table 4. Styrene oxide,
cyclohexene oxide and epichlorohydrin have gone to 100% com-
pletion in just 30 min with selectivity greater than 90%. Except for
cyclohexene oxide, the rest of the epoxides tested are asymmetri-
cal with the possibility of two products. All previous reports in the
literature for asymmetrical epoxide ring opening agree that solid
acid catalysts will cleave the epoxide ring in such a way to yield a
more stable carbocation followed by attack of the alcohol to form
alkoxyalcohols. Our selectivity results for different epoxides pre-
dominantly give the expected alkoxyalcohols consistent with the
Table 3
Alcoholysis of cyclohexene oxide with different alcohols catalyzed by 1g-MoP-SBA-
15-150.
Entry
Alcohol
Time (min)
Conv (%)
Sel (%)
TOF (h−1
)
1
2
3
4
5
6
7
8
Methanol
Ethanol
30
30
30
30
30
30
60
60
60
30
30
30
30
30
60
60
60
100
100
100
100
100
100
72
81
86
99
93
100
97
96
98
75
97
95
94
91
93
94
82
89
91
96
77
94
97
90
90
75
65
90
90
90
90
90
90
32
37
39
89
84
90
87
86
44
34
15
1-Propanol
1-Butanol
iso-Butanol
1-Pentanol
1-Hexanol
1-Octanol
Table 4
9
1-Dodecanol
Allyl alcohol
Crotyl alcohol
Benzyl alcohol
2-Phenylethanol
2-Propanol
cyclohexanol
t-Butyl alcohol
t-Amyl alcohol
Alcoholysis of different epoxides with ethanol using 1g-MoP-SBA-15-150 as catalyst.
10
11
12
13
14
15
16
17
Entry
Epoxide
Time (min)
Conv (%)
Sel (%)
TOF (h−1
)
1
2
3
4
5
Cyclohexene oxide
Styrene oxide
Epichlorohydrin
1,2-epoxyhexane
1,2-epoxydodecane
30
30
30
60
60
100
100
100
85
95
96
93
95
94
90
90
90
39
27
32
60
Reaction condition: 1.1 mmol cyclohexene oxide, 2.4 mol% catalyst (1g-MoP-SBA-
15-150) by substrate, 3 mL alcohol, temperature 396 2 K.
Reaction condition: 1.1 mmol epoxide, 2.4 mol% catalyst (1g-MoP-SBA-15-150) by
substrate, 3 mL ethanol, temperature 396 2 K.

476
S. Budhi et al. / Applied Catalysis A: General 475 (2014) 469–476
mechanism of epoxide ring opening solid acid catalysts at very high
ratio.
References
[1] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992)
710–712.
[2] H. Subramanian, E.G. Nettleton, S. Budhi, R.T. Koodali, Journal of Molecular
Catalysis A: Chemical 330 (2010) 66–72.
3.6. Recycling studies
[3] D. Zhao, S. Budhi, A. Rodriguez, R.T. Koodali, International Journal of Hydrogen
Energy 35 (2010) 5276–5283.
Recycling studies are meant for ring opening of epoxide using
cyclohexene oxide as the substrate and ethanol as the nucleophile.
The experiment was repeated four times and the conversions are
100%, 82%, 60% and 30%, respectively. The molybdenum loaded
in the catalyst was reduced from 8.5 wt% to 6 wt% as determined
by ICP-AES analysis after four cycles. The surface area decreased
from 329 m2/g to 214 m2/g and pore volume decreased from 1.2
to 0.716 cm3/g, however, pore diameter remained the same. The
decrease in pore volume can be attributed to clogging of the pore
channels by reaction species in the mixture. It has been reported
in literature that metal incorporated on silica support leaches for a
variety of reasons including reaction parameters such as solvent,
temperature, polarity of reactants, etc [39]. Alcohols, especially
smaller carbon chain ones that are highly polar in nature, are
reported to facilitate leaching of metal ions supported on silica.
In our case, probably alcohols that we use may be a source of metal
leaching. However, a detailed investigation is required to conclude
if leaching is substrate specific or due to polarity of alcohols used
as nucleophiles, since we used only ethanol in our recyclability
experiments.
[4] A. Chenite, Y. Le Page, A. Sayari, Chemistry of Materials
1015–1019.
7 (1995)
[5] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky,
Science 279 (1998) 548.
[6] Z. Luan, E.M. Maes, P.A.W. van der Heide, D. Zhao, R.S. Czernuszewicz, L. Kevan,
Chemistry of Materials 11 (1999) 3680–3686.
[7] Z. Luan, M. Hartmann, D. Zhao, W. Zhou, L. Kevan, Chemistry of Materials,
American Chemical Society (1999) 1621–1627.
[8] A. Vinu, V. Murugesan, W. Böhlmann, M. Hartmann, The Journal of Physical
Chemistry B 108 (2004) 11496–11505.
[9] J.A. Melero, J. Iglesias, J.M. Arsuaga, J. Sainz-Pardo, P. de Frutos, S. Blazquez,
Applied Catalysis A: General 331 (2007) 84–94.
[10] S. Budhi, H.S. Kibombo, D. Zhao, A. Gonshorowski, R.T. Koodali, Materials Letters
65 (2011) 2136–2138.
[11] Y. Wang, B. Zibrowius, C.-m. Yang, B. Spliethoff, F. Schuth, Chemical Commu-
nications (2004) 46–47.
[12] E. Krämer, S. Förster, C. Göltner, M. Antonietti, Langmuir 14 (1998)
2027–2031.
[13] C. Yu, B. Tian, J. Fan, G.D. Stucky, D. Zhao, Journal of the American Chemical
Society 124 (2002) 4556–4557.
[14] C. Yu, J. Fan, B. Tian, D. Zhao, Chemistry of Materials 16 (2004) 889–898.
[15] B.M. Reddy, M.K. Patil, B.T. Reddy, S.-E. Park, Catalysis Communications 9 (2008)
950–954.
[16] E. Laurent, B. Delmon, Industrial & Engineering Chemistry Research 32 (1993)
2516–2524.
[17] V.M.L. Whiffen, K.J. Smith, Energy & Fuels 24 (2010) 4728–4737.
[18] C.D. Nunes, A.A. Valente, M. Pillinger, A.C. Fernandes, C.C. Romao,
J. Rocha, I.S. Goncalves, Journal of Materials Chemistry 12 (2002)
1735–1742.
[19] S. Higashimoto, R. Tsumura, M. Matsuoka, H. Yamashita, M. Che, M. Anpo,
in: C.C.A. Gamba, S. Coluccia (Eds.), Studies in Surface Science and Catalysis,
Elsevier, Amsterdam, 2001, pp. 315–321.
[20] L.-X. Dai, Y.-H. Teng, K. Tabata, E. Suzuki, T. Tatsumi, Microporous and Meso-
porous Materials 44–45 (2001) 573–580.
[21] P.S. Raghavan, M.P. Vinod, Journal of Molecular Catalysis A: Chemical 135
4. Conclusions and future studies
We have successfully synthesized molybdenum incorporated
SBA-15 by direct synthesis using diammonium hydrogen phos-
phate as an additive. Addition of diammonium hydrogen phosphate
increased the pore size of the materials. Most importantly, for the
first time, we have observed that the addition of a salt facilitates the
incorporation of metal on silica support by 2–3 times more than
the corresponding catalysts synthesized without DHP. However,
we have limited our investigation to just one additive and its effect
on loading. The catalyst we synthesized was used for effective ring
opening of epoxides in a short amount of time at ambient temper-
ature for a wide range of alcohols and epoxides. We are currently
investigating the applicability of additive enhanced metal loading
for different metal incorporated SBA-15, different pH and range of
additives and its influence in metal loading and catalytic activity.
(1998) 47–53.
ˇ
ˇ
[22] P. Topka, H. Balcar, J. Rathousky´, N. Zilková, F. Verpoort, J. Cejka, Microporous
and Mesoporous Materials 96 (2006) 44–54.
[23] T.M. Sankaranarayanan, A. Pandurangan, M. Banu, S. Sivasanker, Applied Catal-
ysis A: General 409–410 (2011) 239–247.
[24] S. Ramesh Kumar, P. Leelavathi, Journal of Molecular Catalysis A: Chemical 266
(2007) 65–68.
[25] S. Singhal, Chemistry Letters 37 (2008) 620.
[26] M.W.C. Robinson, D.A. Timms, S.M. Williams, A.E. Graham, Tetrahedron Letters
48 (2007) 6249–6251.
[27] C.-J. Shih, E.A. Smith, Analytica Chimica Acta 653 (2009) 200–206.
[28] B. Boote, H. Subramanian, K.T. Ranjit, Chemical Communications (2007)
4543–4545.
[29] K. Flodström, V. Alfredsson, N. Källrot, Journal of the American Chemical Society
125 (2003) 4402–4403.
Acknowledgments
[30] B.L. Newalkar, S. Komarneni, Chemistry of Materials 13 (2001) 4573–4579.
[31] X. Ma, J. Gong, S. Wang, N. Gao, D. Wang, X. Yang, F. He, Catalysis Communica-
tions 5 (2004) 101–106.
[32] S.B. Umbarkar, T.V. Kotbagi, A.V. Biradar, R. Pasricha, J. Chanale, M.K. Dongare,
A.-S. Mamede, C. Lancelot, E. Payen, Journal of Molecular Catalysis A: Chemical
310 (2009) 150–158.
[33] L. Seguin, Spectrochimica Acta Part A: Molecular and Biomolecular Spec-
troscopy 51 (1995) 1323.
[34] M. Anwar, C.A. Hogarth, R. Bulpett, Journal of Materials Science 25 (1990)
1784–1788.
[35] J.G. Choi, L.T. Thompson, Applied Surface Science 93 (1996) 143–149.
[36] P.A. Spevack, N.S. McIntyre, Journal of Physical Chemistry 97 (1993)
11020–11030.
[37] M.W.C. Robinson, R. Buckle, I. Mabbett, G.M. Grant, A.E. Graham, Tetrahedron
Letters 48 (2007) 4723–4725.
[38] S. Das, T. Asefa, ACS Catalysis 1 (2011) 502–510.
[39] Y. Deng, C. Lettmann, W.F. Maier, Applied Catalysis A: General 214 (2001)
31–46.
S.B. and B.G.T. are supported by ConocoPhillips and the Colorado
School of Mines start-up funding. V.H.T.N. and E.A.S. are supported
by the U.S. Department of Energy, Office of Basic Energy Sci-
ences, Division of Chemical Sciences, Geosciences, and Biosciences
through the Ames Laboratory. The Ames Laboratory is operated
for the U.S. Department of Energy by Iowa State University under
Contract No. DE-AC02-07CH11358.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.apcata.
2014.01.055.