Enhanced Friedel-Crafts benzylation activity of bimetallic WSn-KIT-6 catalysts

Article abstract of DOI:10.1016/j.jcat.2020.07.001

A series of W and Sn-containing KIT-6 mesoporous silicates (designated as WxSn5) with similar Sn (~5 wtpercent) and varying W (x = 3–23 wtpercent) loadings were prepared by a one-pot hydrothermal synthesis method. No detectable bulk WO3 species were observed for W loading up to 14 wtpercent with preferential dispersion of WOx species near SnO2 sites. The WxSn5 materials exhibited superior performance for Friedel-Crafts anisole benzylation with intrinsic rate constants for all WxSn5 catalysts being greater than those without Sn. Based on results from DR-UV–Vis, XRD, XPS, EXAFS techniques, we postulate that the WxSn5 materials contain both isolated [WO4]^2-/SnO2 and isolated [WO4]^2-/SiO2 as active species, with the former being more active. The relative populations of these two highly dispersed species dictate the overall catalyst activity. The intrinsic activities of the WxSn5 catalysts are found to scale linearly with total acidity but are mainly dictated by the Lewis acid sites.

Full text of DOI:10.1016/j.jcat.2020.07.001

Journal Pre-proofs
Enhanced Friedel-Crafts Benzylation Activity of Bimetallic WSn-KIT-6 Cata-
lysts
Rajamanickam Maheswari, Anand Ramanathan, Hongda Zhu, Aline Araújo
do Nascimento Araújo, Clint Chapman, Yu Tang, Franklin (Feng) Tao, Bala
Subramaniam
PII:
S0021-9517(20)30275-X
https://doi.org/10.1016/j.jcat.2020.07.001
YJCAT 13813
DOI:
Reference:
To appear in:
Journal of Catalysis
Received Date:
Revised Date:
Accepted Date:
17 March 2020
28 June 2020
2 July 2020
Please cite this article as: R. Maheswari, A. Ramanathan, H. Zhu, A. Araújo do Nascimento Araújo, C. Chapman,
Y. Tang, F. (Feng) Tao, B. Subramaniam, Enhanced Friedel-Crafts Benzylation Activity of Bimetallic WSn-
KIT-6 Catalysts, Journal of Catalysis (2020), doi: https://doi.org/10.1016/j.jcat.2020.07.001
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
©
2020 Elsevier Inc. All rights reserved.

Enhanced Friedel-Crafts Benzylation Activity of Bimetallic WSn-KIT-6 Catalysts
a
a
a,b
Rajamanickam Maheswari , Anand Ramanathan , Hongda Zhu , Aline Araújo do Nascimento
a
b
a
a
a,b*
Araújo , Clint Chapman, Yu Tang, Franklin (Feng) Tao and Bala Subramaniam
aCenter for Environmentally Beneficial Catalysis, The University of Kansas, Lawrence, KS
6047, USA
bDepartment of Chemical and Petroleum Engineering, The University of Kansas, Lawrence, KS
6
6
6045, USA
*
Corresponding author. E-mail: bsubramaniam@ku.edu, Tel.: +1-785-864-2903; fax: +1-785-
64-6051.
8
1

Abstract
A series of W and Sn-containing KIT-6 mesoporous silicates (designated as W Sn ) with
x
5
similar Sn (~5 wt%) and varying W (x = 3 – 23 wt%) loadings were prepared by a one-pot
hydrothermal synthesis method. No detectable bulk WO species were observed for W loading
3
up to 14 wt% with preferential dispersion of WO species near SnO sites. The W Sn materials
x
2
x
5
exhibited superior performance for Friedel-Crafts anisole benzylation with intrinsic rate
constants for all W Sn catalysts being greater than those without Sn. Based on results from DR-
x
5
UV-Vis, XRD, XPS, EXAFS techniques, we postulate that the W Sn materials contain both
x
5
isolated WO /SnO and isolated WO /SiO as active species, with the former being more active.
4
2
4
2
The relative populations of these two highly dispersed species dictate the overall catalyst
activity. The intrinsic activities of the W Sn catalysts are found to scale linearly with total
x
5
acidity but are mainly dictated by the Lewis acid sites.
Keywords: tungsten; tin; acidity; mesoporous; Friedel–Crafts benzylation
2

1
. Introduction
The addition of tungsten to ZrO , SnO , TiO and Fe O creates solid superacids that are
2
2
2
2
3
active for isomerization of butane, pentane and isopentane [1-3]. In the case of tungstated
zirconia, the activity is related to the formation of subnanometer WO species incorporated with
x
2
zirconium cations [4]. These supported metal oxides often possess a low surface area (< 50 m /g)
and correspondingly low dispersion of WOx species. To increase the dispersion of active sites, it
is desirable to support them in high surface area supports such as ordered mesoporous silicates
(OMS). Other advantages of OMS materials include tunable pore size in a narrow distribution
and synthesis using direct one-pot methods. In recent years, we have incorporated various mono
and bimetallic combinations (such as W, Zr, Nb, WZr, WNb and WTi) into mesoporous silicates
via one-pot hydrothermal synthesis methods [5-14]. For instance, W and Zr were incorporated in
KIT-6 support and a synergistic effect was observed for dehydration of ethanol due to enhanced
dispersion of W in the presence of isolated Zr4+ sites as well as on nanodomains of ZrO [12].
2
Similarly, the introduction of Nb as either isolated sites or nanodomains of octahedral Nb O
2
5
resulted in the formation of new and more active sites, resulting in increased propene yield in the
metathesis of ethylene and 2-butene [13]. A simple impregnation of Na WO on Ti-TUD-1 was
2
4
also reported to enhance catalytic activity during the esterification of acetic acid and n-butanol
14].
Recently, enhanced oxidation activity of WO /SnO prepared by either co-precipitation or
[
3
2
impregnation methods has been reported for the oxidation of 1,2-benzenedimethanol to phthalide
by H O [15]. The enhanced activity is attributed to the high dispersion of tungsten species on
2
2
6
+
the support upon calcination with few W ions incorporated into the SnO lattice. MoO /SnO
2
2
3
and WO /SnO prepared by impregnation method were also reported to be an effective catalyst
3
2
3

for acetalization of glycerol [16]. Sn–W mixed oxide prepared by calcination of the Sn–W mixed
hydroxide precursor has been reported to show a higher yield (39 – 70%) towards the synthesis
of 5-hydroxymethylfurfural (HMF) from hexoses compared to Amberlyst-15 and H-mordenite
catalysts [17]. The similar catalyst system has also been reported for the synthesis of glycerol
carbonate from glycerol and urea exhibiting 52% glycerol conversion with >95% selectivity
towards glycerol carbonate [18]. However, such catalysts are reported to deactivate during the
transesterification of soybean oil with methanol [19]. Given that W-Sn mixed oxides are
effective for both oxidation and acid catalyzed reactions, it is desirable to incorporate both W
and Sn in KIT-6 support similar to WZr-KIT-6 and WNb-KIT-6 catalysts.
In this work, we extend the one-pot synthesis procedure to incorporate Sn and W into KIT-6
mesoporous silicate. Being a bulkier element compared to Zr and Nb, Sn forms predominantly
isolated Sn4+ sites when incorporated in KIT-5 and KIT-6 silicates [10, 20] at a loading of <4
wt%. At much higher loadings (> 7 wt% Sn), bulky SnO particles with a broad distribution of
x
sizes are formed. Hence, as previously shown, we chose a Sn loading of 5 wt% to form nano-
domains of SnO particles. The W precursor solution is added to the synthesis mixture after 24 h
2
following the formation of the SnO nanodomains, in order to maximize the dispersion and
2
interaction of W with Sn species. The catalysts prepared in this manner are designated as W Sn ,
x
y
where x and y denote the wt% of W and Sn, respectively, in the synthesis mixture.
The alkylation of anisole with benzyl alcohol is often used as a test reaction for catalysts
containing Lewis and/or Brønsted acid sites (Scheme 1). For example, Lewis acidic γ-alumina
[21] yields mainly dibenzyl ether (~32% yield), the condensation product of benzyl alcohol, at
1
60 °C after 5 h. On the other hand, Brønsted acidic zeolites such as ZSM-5 and beta zeolites
require either higher reaction temperature (~170 C) [22] or the presence of mesopores [23] to
4

form benzylated anisoles. Hence, this reaction is ideal for investigating the performance of
W Sn KIT-6 materials that contain both Lewis and Brønsted acid sites, and probing possible
x
y
synergistic effects. Herein, we present results of kinetic analysis showing that benzylation
activity is enhanced in W Sn materials (compared to Sn-free materials at identical W loadings).
x
y
This is attributed to the enhanced dispersion of monomeric WO species on the catalyst surface
x
as well as the formation of more active species, as evidenced by complementary catalyst
characterization results from techniques such as X-ray absorption fine structure (XAS), X-ray
photoelectron spectroscopy (XPS) and X-ray powder diffraction (XRD).
2
2
. Experimental
.1. Chemicals
Tri-block co-polymer (Pluronic P123, EO PO EO , with an average molecular weight of
2
0
70
20
~
5,800, Aldrich), ammonium metatungstate [(NH ) H W O ·xH O, Aldrich], Tin (IV) chloride
4 6 2 12 40 2
pentahydrate (98%, Strem Chemicals), tetraethyl orthosilicate (TEOS) (98% Acros Organics)
and hydrochloric acid (37%, Fisher) were used as received.
2
.2. Synthesis of W and Sn containing mesoporous silicate (KIT-6)
In a typical synthesis procedure, 5 g of P123 were dissolved in 185 mL of 0.5 N hydrochloric
acid solution under stirring at 35 °C. Then, 5 g of n-butanol were added and stirred for another 1
h. Following this step, 10.8 g TEOS and the required amount of SnCl ·5H O to yield 5 wt% Sn
4
2
in the final solid was added. After stirring for 24 h at 35 °C, an appropriate quantity of
ammonium metatungstate was added to yield final W wt% of 3, 9, 14 and 23. The mixture was
stirred at 35 °C for 24 h and heated at 100 °C for another 24 h. The resulting precipitate was
filtered, dried and calcined in flowing air at 550 °C for 5 h at a heating rate of 1.5 °C/min. The
5

synthesized samples are denoted as W Sn where x and y represent the wt% of W and Sn (y is
x
y
either 0 or 5). The wt% of W and Sn measured by XRF in the prepared catalysts are given in
Table 1. For catalytic activity comparison, monometallic W-KIT-6 (represented as W Sn ) and
1
4
0
Sn-KIT-6 (represented as W Sn ) materials were also prepared as reported elsewhere [5, 10]
0
5
employing ammonium metatungstate and stannic chloride as W and Sn sources, respectively.
2
.3. Catalyst characterization
Small Angle X-ray Scattering (SAXS) and wide angle XRD patterns were recorded on a
PANalytical Empyrean instrument with CuKα radiation in transmission and reflection modes
operating at 45 kV and 40 mA, respectively. The bulk elemental composition of the powdered
samples was determined using XRF (PANalytical Zetium) instrument. The specific surface area,
pore volume and pore size distributions were calculated from the N sorption isotherms
2
measured at liquid N temperature using a Quantachrome NOVA 2200e analyzer. Prior to the
2
physisorption experiment, the samples were heated in vacuum at 200 °C for 1 h. The diffuse
reflectance UV–vis spectra were collected on a Perkin Elmer Lambda 850 spectrophotometer
equipped with a diffuse reflectance integrating sphere and Spectralon as the reference.
Temperature programmed desorption of ammonia (NH -TPD) and temperature programmed
3
reduction (H -TPR) profiles were obtained using a Micromeritics Autochem 2910 instrument.
2
FTIR spectra of adsorbed pyridine were obtained on a Tensor-27 instrument with Pike diffuse
reflectance accessory, as reported elsewhere [9]. The concentrations of Brønsted and Lewis sites
−
1
−1
were estimated from their respective bands at 1545 cm and 1450 cm by employing reported
−
1
−1
adsorption coefficients (0.078 cm μmol for Brønsted acid sites and 0.165 cm μmol for Lewis)
[24].
6

EXAFS studies of W L edge and Sn K edge were performed at beamline 2-2 at Stanford
3
Synchrotron Radiation Laboratory in California, USA. The X-ray was monochoromated by Si
(220) double crystal. The catalyst samples were pressed into round pellets (13 mm diameter and
~
0.1 mm thickness) and mounted on the sample stage. The X-ray absorption signal was
measured by gas ionic chamber detectors under transmission mode. The absorption spectrum of
the corresponding metal foil reference was also collected and utilized as the energy calibration
reference. The XAS data files were analyzed and fitted using the Demeter software package [25].
2
The amplitude factor S was obtained by fitting the metal foil spectrum with full coordination in
0
st
the 1 shell and used in subsequent data fitting wherein the coordination number, bond length
−
1
and the Debye Waller factor were the fitted variables. The fitted k range was 3~12 Å and the
fitted R range was 1~4 Å.
X-ray photoelectron spectroscopy (XPS) was measured on a PHI 5000 Versa Probe II
instrument employing monochromated Al Kα (50 W, 15 kV) X-ray with a measurement spot size
(beam diameter) of 200 μm. The charging effect was eliminated by a dual-beam charge
neutralizer. The spherical capacitance analyzer was operated with a pass energy of 46.95 eV for
+
detailed scan. Prior to the analysis, the surface of the catalyst was sputtered with Ar ions (2.0
keV) for 5 min to expose the elements that reside deep in the pores of the catalyst. The following
spectra were recorded: survey spectrum, Si 2p, Sn 3d , and W 4f. The resulting spectral data
5
/2
were processed by MultiPak software. The W 4f doublet splitting was constrained to 2.18 eV
26]. All XPS peaks were referenced to the Si 2p peak at 103.5 eV [27].
[
2
.4. Friedel–Crafts benzylation of anisole with benzyl alcohol
Catalytic testing was performed in a 25 mL two-necked round bottom flask equipped with a
magnetic stirrer and a reflux condenser. In a typical experiment, benzyl alcohol (1.08 g, 10
7

mmol) or dibenzylether (0.99 g, 5 mmol), anisole (10.8 g, 100 mmol), dodecane (0.34 g, 2 mmol,
internal standard) and catalyst (100 mg) were charged into the reactor. The reaction mixture was
heated to the desired temperature, typically 130 °C, while being mixed with a magnetic stirrer
(
1000 rpm). The headspace was blanketed with flowing N (50 std cc/min) at 1 atm and the
2
reaction was carried out for 2 h. A small volume (100 µL) of the reaction mixture was
periodically sampled and analyzed with a Varian CP-3800 gas chromatograph equipped with a
ZB-Wax capillary column (25 m × 0.32 mm × 0.2 μm) and a flame-ionization detector. The C
balance, based on analysis of the reactants and liquid phase products was 100 ± 3%.
The following definitions for substrate conversion (푋sub) and selectivities (푆 for product i)
푖
are used to evaluate the performance of the various catalysts, including W Sn .
x
y
푛s,0 ― 푛s
푛s,0
푋s =
× 100%
(1)
(2)
푛푖
푆푖 = 푛sub,0 ― 푛sub × 100%
where 푛s,0 [mol] represents the initial moles of the limiting substrate (benzyl alcohol or dibenzyl
ether) while 푛 and 푛 [mol] represent the moles of the limiting substrate and products in the
s
푖
reaction mixture at a given reaction time, respectively.
The turnover frequency (TOF ) is calculated using the following equations based on the
0
conversion of the limiting substrate.
푚s,0
+
′
(
)
푇푂퐹 = 푚c푀s∑
푋 푡→0
(3)
푖
푥
m
(
푀
m)
where 푚s,0 [g] is the initial mass of the limiting substrate (benzyl alcohol or dibenzyl ether), 푚c
g] is the mass of catalyst, 푥 is the mass fraction of W, Nb or Zr in the catalyst, 푀 and 푀 are
[
m
m
s
the molecular weights of the metal (W, Nb, Sn or Zr) and limiting substrate, respectively, and 푋′
(
푡) is the initial slope of the substrate conversion vs time profile, where the conversion range is <
8

2
0%, estimated with MATLAB R2016a software. It must be noted that since the Sn content in
the W Sn KIT-6 catalysts is constant, the TOF values are reflective of the changes in the W
x
y
content in the bimetallic catalysts. Further, under our reaction conditions, mass transfer
limitations are insignificant (see ESI for calculations).
To probe the reaction pathways further and to compare the decoupled reaction rates on
materials with various metal loadings, reactions using W Sn and W Sn materials were carried
x
0
x
5
out at 110 °C, such that the relatively slower reactions allow kinetic modeling with finite
conversion during 20 – 40 min runs. The temporal solutions to the ordinary differential equations
(ODEs) describing a stirred batch reactor model containing reaction rate expressions were fit to
the substrate and product concentrations estimated from GC results.
d푐푖
=
푟푖
d푡
where 푐 is the concentration of species 푖, 푟 is the net formation rate of species 푖. Several sets of
푖
푖
rate expressions were derived from possible reaction pathways, as discussed in detail in section
3
3
3
.2 and the ESI.
. Results and discussion
.1. Characterization of W Sn Catalysts
x
y
Figure 1 shows the XRD patterns of W Sn compared with W-KIT-6 (W Sn ) and Sn-KIT-6
x
5
14
0
(
W Sn ) samples. All W Sn materials show well defined reflections of SnO (ICSD 98-016-
0 5 x 5 2
9
032) that clearly become narrower with increasing W content. The average crystallite sizes,
estimated from the Debye–Sherrer equation, are 5, 6.8, 8.8, 11.6 and 53 nm, respectively, for
W Sn , W Sn , W Sn , W Sn and W Sn materials. At the maximum W loading of 23 wt%,
0
5
3
5
9
5
14
5
23
5
intense peaks characteristic of crystalline WO (ICSD 98-003-6168) are observed. Interestingly,
3
9

XRD reflections characteristic of WO are not observed in other W Sn catalysts, even at the
3
x
5
relatively high W loadings in W Sn . This may indicate the presence of highly dispersed WO
x
1
4
5
species and/or amorphous WO nanoclusters that are < 3 nm, the lower detection limit of the
x
instrument. However, EXAFS results (discussed later) rule out the presence of WO3
nanoclusters in the W Sn materials. Interestingly, when using Na WO as the W source
1
4
5
2
4
(instead of ammonium metatungstate used in the present study), crystalline tungsten oxide
nanoclusters are reported to form on silicates (KIT-6, KIT-5 and SBA-16) at approximately 12
wt% W loadings [5, 28, 29].
Nitrogen adsorption isotherms of W Sn samples along with pore size distributions are
x
5
shown in Figure 2. All the bimetallic catalyst formulations exhibit Type IV isotherm with a H1
hysteresis loop characteristic of mesoporous materials containing large mesopores [30]. Table 1
lists the physicochemical characteristics of W Sn catalysts. For tungsten loadings up to 14 wt%,
x
y
2
the surface areas and pore volumes decrease while being in the ranges of 585 – 712 m /g and
3
pore 0.73 – 0.98 cm /g, respectively. The corresponding pore size distributions are narrowly
centered at approximately 9.2 nm. However, a further increase in the W loading to 23 wt% leads
2
3
to a significant decrease in surface area and pore volume to 411 m /g to 0.51 cm /g, respectively.
This decrease may result from the formation of nanoclusters of bulk oxides of tin and tungsten
within the pores, as also borne out by the XRD patterns at a W loading of 23 wt% (Fig. 1).
Representative TEM images (Figure 3a) of the W Sn sample reveal uniform hexagonal
1
4
5
mesopores. In addition, crystallites and aggregates of SnO particles, ranging in sizes between 20
2
and 100 nm, appear to be distributed throughout the silicate network as evidenced from TEM
image (Figure 3a) and High-Angle Annular Dark-Field (HAADF) STEM images (Figure 3b).
The presence of nanocrystalline SnO is consistent with wide angle XRD spectra (Figure 1).
2
1
0

EDX elemental mapping data clearly show that the W species are preferentially distributed on Sn
species (Figure 3c-f). Further, the surface elemental composition was also probed on a larger
area (scale bar 20 µm) by scanned X-ray images (SXI) images. The XPS mapping revealed a
homogenous distribution of W and Sn species in the silica matrix (see Figure S1).
Figure 4 summarizes the SAXS pattern of W Sn samples. As reported for other metal
x
5
incorporated KIT-6 silicates [5, 7, 9, 10], a characteristic intense reflection peak due to the (211)
plane and a distinct (220) plane of Ia3d space group are observed. The (211) peak becomes
progressively weaker and broader at higher W loadings, indicative of a decrease in the
mesostructured ordering.
Diffuse reflectance UV-Vis spectra (Figure S2) of the Sn-KIT-6 (W Sn ) sample (Fig. S2, e)
0
5
shows an unresolved broad absorption band with inflections at 210 and 270 nm. The band around
10 nm is attributed to ligand-to-metal charge transfer from an O2− to an isolated Sn4+ in a
2
tetrahedral configuration. Hexa-coordinated polymeric Sn–O–Sn-type species (detected in wide
angle XRD and TEM images) are typically observed around 270 nm [31, 32]. W Sn samples
x
5
(Fig. S2, a-d) show an absorption peak between 227 and 235 nm that increases with W loading
up to ~9 wt% W and decreases slightly with a further increase in W loading, indicating that this
peak represents the intrinsic nature of W coordination in the W Sn samples. The band around
x
5
2
−
2
27-235 nm is generally attributed to isolated [WO ] species [33]. The shoulder peaks that are
4
evident between 250–270 nm at the highest W loading in W Sn (Fig. S2, d) and in tungsten-
2
3
5
only (W Sn ; Fig. S2, f) samples are attributed to octahedral polytungstate species [5, 8, 23, 33].
1
4
0
Temperature-programmed desorption of ammonia (NH -TPD, Figure S3) profiles of Sn-KIT-
3
6
(Figure. S3, e) show a weak and broad peak centered around 300 °C with a relatively low
acidity of 0.04 mmol NH /g (Table 1). The total acidity increases with W loading in W Sn
5
3
x
1
1

materials (Fig. S3, a–d), with a sharp increase from 0.09 to 0.50 mmol NH /g when the W
3
content is increased from 3 to 14 wt%. However, a further increase in W loading from 14 to 23
wt% resulted in only a marginal increase in acidity by 0.06 mmol NH /g (Table 1). This implies
3
that the crystallite WO sites observed from XRD (Figure 1) are weakly acidic. Interestingly,
3
despite their similar W contents, the total acidity of W-KIT-6 (W Sn ) (0.31 mmol NH /g) is
1
4
0
3
significantly lower than that of W Sn (0.50 mmol NH /g). These findings suggest that the
1
4
5
3
interactions between W and Sn species in W Sn samples may be enhancing catalyst acidity.
x
5
Figure 5 shows the FT-IR spectra of adsorbed pyridine species over W Sn catalysts
x
y
measured at 100 °C. The relative ratios of Lewis (L) and Brønsted (B) sites are given in Table 1.
The presence of Brønsted sites protonates the pyridine molecule resulting in their adsorption as
pyridinium ions observed around 1545 cm−1 whereas coordinatively bound pyridine on Lewis
acid sites are observed at 1450 cm−1 [21]. As seen from the spectra in Figure 5, Sn-KIT-6
(
W Sn ) and W Sn samples display very little if any Brønsted acid sites, while all the other
0 5 3 5
materials exhibit Brønsted acidity that increases significantly with W loading. In contrast, all the
materials possess Lewis acidity which is less sensitive to metal loading. In combination with
NH -TPD, it can be inferred that W Sn and W Sn materials possess more acid sites and
3
23
5
14
5
increased acidity compared to W Sn (W-KIT-6) sample.
1
4
0
Temperature programmed reduction (H -TPR, Figure S4) profiles of Sn-KIT-6 (W Sn )
2
0
5
show gradual H consumption up to 415 °C followed by broad and distinct reduction peaks
2
occurring between 440 and 620 °C. Usually a bulk SnO reduction happens around 760 C [34,
2
3
5] and the lack of such peak clearly indicates the presence of isolated sites of tin oxide
nanoparticles that get reduced around 550 °C[35]. In the case of W Sn catalysts, this peak is
x
5
progressively shifted to higher temperatures (602, 625, 714 and 735 °C, respectively) with
1
2

increasing W content (3, 9, 14 and 23 wt%). We attribute this increase in the SnO reduction
2
temperatures to either strong interaction with WO species and/or an increase in the size of the
x
SnO particles (evidenced in XRD). The W-KIT-6 (W Sn ) material shows two broad reduction
2
14
0
6
+
peaks one around 804 °C and the other around 990 °C due to the stepwise reduction of W →
W4+ and W4+ → W , respectively [16, 28, 29]. Further, the progressive shift of the WO
0
x
reduction peak in W Sn materials to higher temperatures indicates possible interactions between
x
5
the WO and SnO species at the higher W loadings.
x
x
3
.2. Friedel-Crafts benzylation of anisole
As shown in Scheme 1, the Friedel-Crafts benzylation of anisole with benzyl alcohol over
W Sn catalysts results in the formation of ortho-benzylated anisole (o-BA), para-benzylated
x
5
anisole (p-BA) and dibenzyl ether (DBE). It has been reported that the self-etherification step of
benzyl alcohol leading to the formation of the DBE is faster than the alkylation reaction [21, 36,
3
7]. However, DBE also acts as an alkylating agent and gets eventually consumed at longer
reaction times to produce the benzylated products [38].
Figure 6a compares the temporal BzOH conversion profiles of W Sn catalysts with
x
5
monometallic W Sn (W-KIT-6) and W Sn (Sn-KIT-6) catalysts for the benzylation of anisole
1
4
0
0
5
at similar operating conditions (130 °C). The temporal product distribution obtained with the
different catalysts is given in Table 2. W Sn (Sn only) showed < 9% BzOH conversion after 2 h
0
5
with DBE as the main product (> 95%, Table 2, Entry 2). In contrast, W Sn materials show
x
5
significant BzOH conversion. While approximately 30% BzOH conversion was observed over
W Sn catalyst in 2 h (Table 2, Entry 3), nearly complete BzOH conversion was achieved in less
3
5
than 30 min over W Sn catalyst (Table 2, Entry 5). As the W loading was increased to 23%
1
4
5
1
3

(
W Sn ), a moderate decrease in the BzOH conversion rate was observed (Fig. 6a and Table 2,
23 5
Entry 7).
The TOFs computed from the linear portion of the temporal conversion curves show that Sn-
−
1
based silicate provides relatively low TOF of 0.14 min (Table 2, Entry 2). In contrast, the TOFs
are greater when W is present, increasing by more than an order of magnitude as the W loading
−
1
−1
is increased from 3 wt% (TOF = 0.26 min ) to 14 wt% (TOF = 11.07 min ). Interestingly, the
−
1
TOF for the W Sn catalyst without Sn (TOF = 3.53 min ; Table 2, Entry 6) is approximately
1
4
0
−
1
three times lower than the W Sn catalyst at similar W loading (TOF = 11.07 min ; Table 2,
1
4
5
Entry 5).
The temporal product distribution reveals significant differences in the product selectivity
among the various catalysts. Even though the W Sn catalyst shows facile BzOH conversion, it
1
4
0
mainly forms DBE (~80%) after 20 min and with time, DBE reacts with anisole yielding
benzylanisoles. Yet nearly 15% DBE remains unconverted after 60 min (Table 2, Entry 6). In
sharp contrast, W Sn catalyst shows only ~20% DBE selectivity after 10 min and the reaction
1
4
5
is complete after 30 min forming only the ring alkylated products, o-BA and p-BA (Table 2,
Entry 5).
As shown in Table S1, various other monometal-incorporated mesoporous silicates such as
Zr-KIT-5 (Entry 1), Zr-KIT-6 (Entries 2–4), Nb-KIT-6 (Entries 5–7) showed generally lower
conversion rates and higher DBE formation compared to the bimetallic W Sn catalysts (Table
x
5
2
). In addition, the TOFs for these monometallic silicates (ranging from 0.11 min−1 to 2.18
−
1
min ) are several-fold lower compared to the maximum TOF value observed with the W Sn
5
1
4
−
1
catalyst (TOF = 11.07 min , Table 2, Entry 5). These results clearly divulge the beneficial role
of Sn in significantly enhancing W utilization in the mixed metal oxide (W Sn ) catalysts.
x
y
1
4

To probe the reaction pathway further, DBE was tested as an alkylating agent. The temporal
DBE conversion profiles over W Sn catalysts are depicted in Figure 6b. The product
x
5
distribution along with the TOFs are given in Table 3. As expected, the major products are o-BA
and p-BA. As with BzOH as alkylating agent, the TOFs for W Sn catalysts display a maximum
x
5
−
1
at a W loading of 14 wt% (TOF = 1.96 min , Table 3, Entry 3) which is greater than the value
−
1
(
1.62 min ) observed with W-KIT-6 catalyst with similar W loading (W Sn ; Table 3, Entry 1).
14 0
However, the TOFs are much lower than those observed with BzOH as the benzylating agent
suggesting that over W Sn catalysts, the direct alkylation pathway involving BzOH is much
x
y
faster (Scheme 1).
To decouple the kinetics of competing reactions of BzOH condensation and alkylation,
several kinetic models (Scheme S1 and Equation Sets S6 – S10 in the ESI) were fitted to the
experimental data acquired at 110 °C. The W Sn sample show negligible activity during the 40
3
0
minutes run and was therefore excluded from the kinetic modeling. Also, on the relatively more
active materials, W Sn , W Sn , and W Sn , DBE forms predominantly benzylanisoles,
1
4
5
23
0
23
5
resulting in the apparent rate constants being zero for BzOH formation from DBE, suggesting
that DBE alkylation and DBE hydrolysis may share a common surface species produced by the
rate-determining step, DBE adsorption. The best-fitting reaction scheme is as shown in Scheme
2
, corresponding to the rate expressions Eq. 4a – 4c for the net formation rates of the species (푟푖
d푐푖
= d푡). The model confirms (a) the BzOH condensation is reversible, and (b) one mole of DBE
reacts with two moles of anisole without forming free BzOH in the bulk liquid phase. The
inclusion of water concentration into the hydrolysis of DBE deteriorated the fitting quality,
suggesting that the adsorption of BzOH and DBE was likely the rate determining step for BzOH
condensation and DBE hydrolysis, respectively.
1
5

2
푟(BzOH) = 푘 [DBE] ― 푘 [BzOH] ― 푘 [BzOH]
(4a)
―
1
1
2
1
2
1
2
2
푟(DBE) = ― 푘 [DBE] + 푘 [BzOH] ― 푘 [DBE]
(4b)
(4c)
―1
1
3
푟(BA) = 푘 [BzOH] +2푘 [DBE]
2
3
The intrinsic rate constants obtained using the foregoing model were compared for catalysts
with 5% Sn (W Sn catalysts) and without any Sn content (W Sn catalysts) at identical W
x
5
x
0
loadings and operating temperature (Table 4). This comparison clearly shows that the intrinsic
rate constants (any k value) are greater in the case of Sn-containing (W Sn ) catalysts, providing
x
5
compelling evidence for the beneficial role of Sn.
Further, based on a linear-dependence
analysis of the reaction rates at 130 °C and 110 °C against Lewis, Brønsted and total acid
amounts (Figures S6 and S7), it is concluded that the observed activity is mainly dictated by the
Lewis acid sites.
The reusability of W Sn catalysts was also investigated. Following a reaction run, the
1
4
5
catalyst was filtered, washed with acetonitrile, dried and calcined at 550 °C prior to its use in the
next run. The nearly overlapping temporal conversion profiles (Figure S8) suggest that the active
sites are preserved even after three reuses. This implies that insights gained into the active sites
from catalyst characterization (such as EXAFS) with used catalysts are useful.
3
.3. Nature of active species
The activity enhancement displayed by the W Sn catalysts correlates well with an increase in
x
5
the number and nature of acid sites (Figures. S3 and 5) compared to the W Sn samples. Such
x
0
2
-
increase could partially be attributed to the formation of isolated [WO ] species as evidenced by
4
EXAFS results. The presence of tin oxide could modify the chemical environment around the
1
6

WO species in a manner similar to recently reported metal dopants such as Zr, Nb, Hf, and Ta
x
[12, 13], resulting in enhanced activity.
As a bulk characterization technique, XAS reveals the local coordination environment of
target element atoms. All the target atoms in the sample contribute to the overall XAS signal.
EXAFS of three spent catalysts (after runs at 130 °C) including W Sn , W Sn and W Sn were
1
4
0
0
5
14
5
investigated to examine the coordination environment. The results of least-square fits in the r-
space spectra are given in Table 5. Figure 7a represents the experimental data of the r-space of
Sn-K edge (black line) of W Sn and its fitted curve (red line). For the spent W Sn catalyst (Sn-
0
5
0
5
KIT-6), the coordination number of O to Sn in the first shell is 6.24 with an average Sn–O
distance of 2.04 Å (Table 5). The CN(Sn–O) of 6.24 is consistent with the ideal number of
oxygen atoms around a Sn atom in bulk SnO crystal [39]. The second shell could be interpreted
2


by two types of Sn atoms, Sn and Sn which are reflected by two separated peaks as marked in
the r-space (Figure 7a). The fitting and analysis of the r-space suggest that the coordination


numbers of Sn and Sn to the center Sn atom are 1.74 and 5.44, respectively. Clearly, these
EXAFS results suggest the formation of SnO nanoclusters, which is also supported by the XRD
2
pattern of W Sn in Figure 1.
0
5
The experimental and fitted curves of W-L edge of W Sn are shown as black and red lines
3
14
0
(Figure 7b), respectively. The peak centered at 1.3 Å stems from two different types of oxygen
atoms which coordinate with the central W atoms with distances of 1.797 and 2.410 Å. The two
peaks at 2.6 and 3.2 Å were assigned to Si and W atoms in the second shells of the central W
atom. The distances from the Si and W atoms to the central W atoms are 3.384 and 3.82 Å,
respectively, and are consistent with literature reports [40-43].
1
7

Figure 7c presents the experimental (black) and fit (red) r-space of Sn-K edge of W Sn .
1
4
5
Similar to W Sn , the Sn atoms were identified in both the first and second shells in the r-space
0
5
spectrum suggesting the existence of SnO nanoclusters in W Sn . However, Sn-(O)-W
2
14
5
coordination is hard to be deconvoluted in the Sn K edge spectrum. This is because the XAS is a
bulk measurement technique and hence the overall Sn K edge spectrum cannot sensitively
capture the signal from a relatively small portion of Sn atoms potentially coordinated in the form
of Sn-(O)-W. Further, from the XRD pattern of W Sn samples, it is evident that Sn exists as
x
5
SnO and crystalline WO is evident only in the W Sn sample (Figure 1). Hence, the formation
2
3
23
5
of tin tungstate (SnWO ) is excluded. Figure 7d represents the r-space of W-L edge of W Sn .
4
3
14
5
The r-space was deconvoluted into three peaks including a W–O of the first shell and W–(O)–Si
and W–(O)–Sn in the second shell (Figure 7d). The W atoms in W Sn most likely form WO
x
1
4
5
monomer species on the surface of the mesopores. These WO monomer species are linked via
x
W-O-Si bonds with the SiO surface in the pores and via W-O-Sn bonds on the SnO nano-
2
2
domains.
Based on the foregoing discussion, a catalyst model and probable geometries of tungsten
coordination with SiO and SnO are presented in Figures 8a and 8b, respectively. In this
2
2
postulated model, SnO species exist in the form of nanoclusters/nanoparticles on the SiO
x
2
support while the W species are dispersed as even smaller particles on the surfaces of SnO and
x
the SiO support. Hence, the number of Sn atoms at the Sn-O-W interface is a relatively small
2
fraction of the total Sn atoms. On the other hand, W is highly dispersed, resulting in a high
fraction of W atoms at both the Sn-O-W and Si-O-W interfaces. Thus, while the Sn-O-W
coordination is easily visible in W L3 edge EXAFS spectra, it is hard to distinguish in the Sn K
1
8

edge EXAFS spectra. The increased TOF of W Sn catalysts suggests that the WO species
x
y
x
associated with Sn species enhances catalysis.
Compared to bulk techniques such as XAS and XRD, XPS provides more surface sensitive
chemical information of the W Sn catalyst (Figure S5). The photoemission feature of W 4f (31
1
4
5
–
4
42 eV) appears as a well resolved doublet attributed to W 4f7/2 and W 4f5/2 components [44,
5]. The peaks at lower binding energies (BE), 34.70 and 36.88 eV can be attributed to W–O–Si
interactions, whereas those at higher BE, 37.52 and 39.70 eV are due to either W–O–W or W–
O–Sn [46]. Since XAS studies suggest that there are two types of W atoms based on the average
coordination number of O atoms around a W atom, we assign the peaks at higher BE to W–O–Sn
interactions. Further, the spectra of Sn 3d5/2 could be deconvoluted to two components, which is
assigned to two Sn chemical states viz., Sn-O-Sn and Sn-O-W. Thus, the XPS results
complement XAS analysis and provide credence for the postulated catalyst structure (Figures 8a
and 8b).
The catalyst model, supported by EXAFS, XPS and XRD results, suggests that the overall
activity of the W Sn KIT-6 catalysts is dictated by the relative populations of isolated W-O-Sn
x
y
and isolated W-O-Si species (Figure 8b). From Table 4, it is clear that at identical W loadings,
the intrinsic rate constants obtained with catalysts containing 5% Sn are greater than those
without Sn. This implies that the W-O-Sn species are more active than the W-O-Si species.
4
. Conclusions
We demonstrate that highly dispersed WO species can be generated by consecutive
x
additions of Sn and W sources during the one-pot synthesis of KIT-6. Based on complementary
catalyst characterization results (Diffuse Reflectance UV-Vis, XRD, XPS, EXAFS), we postulate
that the overall activity of the W Sn KIT-6 catalysts in the Friedel-Crafts benzylation of anisole
x
y
1
9

is dictated by the relative populations of the W-O-Sn and W-O-Si species. At identical W
loadings, the intrinsic rate constants obtained with catalysts containing 5% Sn are greater than
those without Sn, implying that the W-O-Sn species are more active than the W-O-Si species.
The intrinsic activity correlates linearly with total acidity but is mainly dictated by the Lewis
acid sites. By facilitating the selective formation of the active W species (isolated WO species,
x
with strong interaction with the Sn dopants), Sn doping enhances W utilization. These findings
open new avenues for better understanding and exploiting bimetallic mesoporous silicates in
various chemistries.
Acknowledgments
This research was supported from joint National Science Foundation and Environmental
Protection Agency program Networks for Sustainable Material Synthesis and Design (NSF-EPA
1
339661). The authors thank Dr. Prem Thapa of the Microscopy and Analytical Imaging
Laboratory, University of Kansas for TEM characterizations. Use of the Stanford Synchrotron
Radiation Light Source, SLAC National Accelerator Laboratory, is supported by the U.S.
Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No.
DE-AC02-76SF00515.
References
2
0

[1] M. Hino, K. Arata, Synthesis of solid superacid of tungsten oxide supported on zirconia and
its catalytic action for reactions of butane and pentane, Journal of the Chemical Society,
Chemical Communications, (1988) 1259-1260.
[2] M. Hino, K. Arata, Synthesis of Highly Acidic Catalysts of Tungsten Oxide Supported on
Tin Oxide, Titanium Oxide, and Iron Oxide, Bulletin of the Chemical Society of Japan, 67
(1994) 1472-1474.
[3] S. Kuba, P. Lukinskas, R.K. Grasselli, B.C. Gates, H. Knözinger, Structure and properties of
tungstated zirconia catalysts for alkane conversion, Journal of Catalysis, 216 (2003) 353-361.
4] W. Zhou, E.I. Ross-Medgaarden, W.V. Knowles, M.S. Wong, I.E. Wachs, C.J. Kiely,
Identification of active Zr–WO clusters on a ZrO support for solid acid catalysts, Nature
[
x
2
Chemistry, 1 (2009) 722-728.
5] A. Ramanathan, B. Subramaniam, D. Badloe, U. Hanefeld, R. Maheswari, Direct
[
incorporation of tungsten into ultra-large-pore three-dimensional mesoporous silicate framework:
W-KIT-6, J. Porous Mater., 19 (2012) 961-968.
[
6] Q. Pan, A. Ramanathan, W.K. Snavely, R.V. Chaudhari, B. Subramaniam, Synthesis and
Dehydration Activity of Novel Lewis Acidic Ordered Mesoporous Silicate: Zr-KIT-6, Industrial
Engineering Chemistry Research, 52 (2013) 15481-15487.
7] A. Ramanathan, B. Subramaniam, R. Maheswari, U. Hanefeld, Synthesis and
&
[
characterization of Zirconium incorporated ultra large pore mesoporous silicate, Zr–KIT-6,
Microporous and Mesoporous Materials, 167 (2013) 207-212.
[8] Q. Pan, A. Ramanathan, W.K. Snavely, R. Chaudhari, B. Subramaniam, Intrinsic Kinetics of
Ethanol Dehydration Over Lewis Acidic Ordered Mesoporous Silicate, Zr-KIT-6, Topics in
Catalysis, (2014) 1-5.
[9] A. Ramanathan, R. Maheswari, D.H. Barich, B. Subramaniam, Niobium incorporated
mesoporous silicate, Nb-KIT-6: Synthesis and characterization, Microporous and Mesoporous
Materials, 190 (2014) 240-247.
[
10] R. Rajalakshmi, R. Maheswari, A. Ramanathan, Characterization and activity of novel tin
incorporated ordered cubic mesoporous silicate, Sn-KIT-6, Materials Research Bulletin, 75
2016) 224-229.
(
2
1

[11] H. Zhu, R.V. Chaudhari, B. Subramaniam, A. Ramanathan, J.-F. Wu, Effects of tunable
acidity and basicity of Nb-KIT-6 catalysts on ethanol conversion: Experiments and kinetic
modeling, AIChE Journal, 63 (2017) 2888-2899.
[12] H. Zhu, A. Ramanathan, J.-F. Wu, B. Subramaniam, Genesis of Strong Brønsted Acid Sites
in WZr-KIT-6 Catalysts and Enhancement of Ethanol Dehydration Activity, ACS Catalysis, 8
(2018) 4848-4859.
[13] J.-F. Wu, A. Ramanathan, R. Kersting, A. Jystad, H. Zhu, Y. Hu, C.P. Marshall, M.
Caricato, B. Subramaniam, Enhanced Olefin Metathesis Performance of Tungsten and Niobium
Incorporated Bimetallic Silicates: Evidence of Synergistic Effects, ChemCatChem, doi:
1
0.1002/cctc.201902131 (2020).
14] M.P. Pachamuthu, R. Maheswari, A. Ramanathan, Synthesis and characterizations of
isolated WO4 anchored on mesoporous TiTUD-1 support, Applied Surface Science, 402 (2017)
86-293.
15] Z. Zhang, Q. Zhu, J. Ding, X. Liu, W.-L. Dai, Effect of calcination temperature of the
[
2
[
support and the catalyst of WO3/SnO2 on the catalytic oxidation of 1,2-benzenedimethanol by
H2O2, Applied Catalysis A: General, 482 (2014) 171-178.
[16] B. Mallesham, P. Sudarsanam, G. Raju, B.M. Reddy, Design of highly efficient Mo and W-
promoted SnO2 solid acids for heterogeneous catalysis: acetalization of bio-glycerol, Green
Chemistry, 15 (2013) 478-489.
[17] K. Yamaguchi, T. Sakurada, Y. Ogasawara, N. Mizuno, Tin–Tungsten Mixed Oxide as
Efficient Heterogeneous Catalyst for Conversion of Saccharides to Furan Derivatives, Chemistry
Letters, 40 (2011) 542-543.
[18] K. Jagadeeswaraiah, C.R. Kumar, P.S.S. Prasad, S. Loridant, N. Lingaiah, Synthesis of
glycerol carbonate from glycerol and urea over tin-tungsten mixed oxide catalysts, Applied
Catalysis A: General, 469 (2014) 165-172.
[19] W. Xie, T. Wang, Biodiesel production from soybean oil transesterification using tin oxide-
supported WO catalysts, Fuel Processing Technology, 109 (2013) 150-155.
3
[20] R. Rajalakshmi, V.V. Srinivasan, M.P. Pachamuthu, R. Maheswari, Characterizations of tin
(
SnO ) doped KIT-5 by direct synthesis, Materials Chemistry and Physics, 154 (2015) 164-169.
2
2
2

[21] M.H.C. de la Cruz, M.A. Abdel-Rehim, A.S. Rocha, J.F.C. da Silva, A. da Costa Faro, E.R.
Lachter, Liquid phase alkylation of anisole by benzyl alcohol catalyzed on alumina-supported
niobia, Catal. Commun., 8 (2007) 1650-1654.
[22] H. Jin, M.B. Ansari, E.-Y. Jeong, S.-E. Park, Effect of mesoporosity on selective
benzylation of aromatics with benzyl alcohol over mesoporous ZSM-5, Journal of Catalysis, 291
(2012) 55-62.
[23] Z. Miao, H. Song, H. Zhao, L. Xu, L. Chou, One-pot synthesis of mesoporous ZrPW solid
acid catalyst for liquid phase benzylation of anisole, Catalysis Science & Technology, 4 (2014)
38-850.
24] J. Datka, B. Gil, A. Kubacka, Heterogeneity of OH groups in NaH-mordenites: Effect of
Na/H exchange degree, Zeolites, 18 (1997) 245-249.
25] B. Ravel, M. Newville, ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray
absorption spectroscopy using IFEFFIT, Journal of Synchrotron Radiation, 12 (2005) 537-541.
26] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photoelectron
8
[
[
[
Spectroscopy: A Reference Book of Standard Spectra for Identification and Interpretation of
XPS Data, Perkin-Elmer Coporation, Eden Prairie, MN, 1992.
[
27] NIST X-ray Photoelectron Spectroscopy Database, in, National Institute of Standards and
Technology, Gaithersburg, MD.
28] R. Maheswari, M.P. Pachamuthu, A. Ramanathan, B. Subramaniam, Synthesis,
[
Characterization, and Epoxidation Activity of Tungsten-Incorporated SBA-16 (W-SBA-16),
Industrial & Engineering Chemistry Research, 53 (2014) 18833-18839.
[29] A. Ramanathan, R. Maheswari, B.P. Grady, D.S. Moore, D.H. Barich, B. Subramaniam,
Tungsten-incorporated cage-type mesoporous silicate: W-KIT-5, Microporous and Mesoporous
Materials, 175 (2013) 43-49.
[30] M. Thommes, K. Kaneko, A.V. Neimark, J.P. Olivier, F. Rodriguez-Reinoso, J. Rouquerol,
K.S.W. Sing, Physisorption of gases, with special reference to the evaluation of surface area and
pore size distribution (IUPAC Technical Report), Pure and Applied Chemistry, 87 (2015) 1051.
[31] Z.C. Liu, H.R. Chen, W.M. Huang, J.L. Gu, W.B. Bu, Z.L. Hua, J.L. Shi, Synthesis of a
new SnO /mesoporous silica composite with room-temperature photoluminescence, Microporous
2
and Mesoporous Materials, 89 (2006) 270-275.
2
3

[32] S.-Y. Chen, H.-D. Tsai, W.-T. Chuang, J.-J. Lee, C.-Y. Tang, C.-Y. Lin, S. Cheng, Direct
Preparation of Thermally Stable Sn-Incorporated SBA-15 Mesoporous Materials in the Self-
Generated Acidic Environment, The Journal of Physical Chemistry C, 113 (2009) 15226-15238.
[33] Q. Zhao, S.-L. Chen, J. Gao, C. Xu, Effect of tungsten oxide loading on metathesis activity
of ethene and 2-butene over WO /SiO catalysts, Transition Met. Chem., 34 (2009) 621-627.
3
2
[
34] X. Wang, Y.-c. Xie, Preparation and Characterization of SnO2-Based Composite Metal
Oxides: Active and Thermally Stable Catalysts for CH4 Oxidation, Catalysis Letters, 75 (2001)
3-80.
35] E.A. Alarcón, A.L. Villa, C.M.d. Correa, Characterization of Sn- and Zn-loaded MCM-41
catalysts for nopol synthesis, Microporous and Mesoporous Materials, 122 (2009) 208-215.
36] A.B. Deshpande, A.R. Bajpai, S.D. Samant, The enhanced activity of Sb after supporting on
7
[
[
K10 in the benzylation of benzene using benzyl chloride and benzyl alcohol, Applied Catalysis
A: General, 209 (2001) 229-235.
[37] C.R. Kumar, P.S.S. Prasad, N. Lingaiah, Heteropoly tungstate supported on tin oxide
catalysts for liquid phase benzylation of anisole with benzyl alcohol, Applied Catalysis A:
General, 384 (2010) 101-106.
[38] A. Ramanathan, H. Zhu, R. Maheswari, B. Subramaniam, Novel zirconium containing cage
type silicate (Zr-KIT-5): An efficient Friedel–Crafts alkylation catalyst, Chemical Engineering
Journal, 278 (2015) 113-121.
[39] J. Rockenberger, U. zum Felde, M. Tischer, L. Tröger, M. Haase, H. Weller, Near edge X-
ray absorption fine structure measurements (XANES) and extended X-ray absorption fine
structure measurements (EXAFS) of the valence state and coordination of antimony in doped
nanocrystalline SnO , The Journal of Chemical Physics, 112 (2000) 4296-4304.
2
[
40] T. Pauporté, Y. Soldo-Olivier, R. Faure, XAS Study of Amorphous WO Formation from a
3
Peroxo-Tungstate Solution, The Journal of Physical Chemistry B, 107 (2003) 8861-8867.
[41] L.A. Morton, M. Miao, T.M. Callaway, T. Chen, S.-J. Chen, A.A. Tuinman, X. Yu, Z. Lu,
0
Z.-L. Xue, Reactions of d tungsten alkylidyne complexes with O or H O. Formation of an oxo
2
2
siloxy complex through unusual silyl migrations, Chemical Communications, 49 (2013) 9555-
557.
42] Y. Bouhoute, D. Grekov, K.C. Szeto, N. Merle, A. De Mallmann, F. Lefebvre, G. Raffa, I.
Del Rosal, L. Maron, R.M. Gauvin, L. Delevoye, M. Taoufik, Accessing Realistic Models for the
9
[
2
4

WO –SiO Industrial Catalyst through the Design of Organometallic Precursors, ACS Catalysis,
3
2
6
(2016) 1-18.
43] G. Chen, M. Dong, J. Li, Z. Wu, G. Wang, Z. Qin, J. Wang, W. Fan, Self-metathesis of 1-
butene to propene over SBA-15-supported WO , Catalysis Science & Technology, 6 (2016)
[
3
5
515-5525.
[44] X.-L. Yang, W.-L. Dai, R. Gao, K. Fan, Characterization and catalytic behavior of highly
active tungsten-doped SBA-15 catalyst in the synthesis of glutaraldehyde using an anhydrous
approach, Journal of Catalysis, 249 (2007) 278-288.
[45] T.Y. Kim, D.S. Park, Y. Choi, J. Baek, J.R. Park, J. Yi, Preparation and characterization of
mesoporous Zr-WO /SiO catalysts for the esterification of 1-butanol with acetic acid, Journal of
x
2
Materials Chemistry, 22 (2012) 10021-10028.
46] H. Shao, M. Huang, H. Fu, S. Wang, L. Wang, J. Lu, Y. Wang, K. Yu, Hollow WO /SnO
[
3
2
Hetero-Nanofibers: Controlled Synthesis and High Efficiency of Acetone Vapor Detection,
Frontiers in Chemistry, 7 (2019) 785.
2
5

Scheme 1. Influence of acid sites on product selectivity during benzylation of anisole with
benzyl alcohol
2
6

O
H O
2
OH
k2
O
k–1
k3
H O
2
OH
k1
H O
2
O
O
2
Scheme 2. The scheme of reaction pathways fit best to the kinetic data
2
7

Table 1. Metal content, textural properties and total acidity of W Sn samples
x
y
b
SBETc Vp, BJH d dP, BJH e
Catalyst
M, wt %
Mole ratio
Total
acidity
L/B
f
a
2
3
ratio g
(
W Sn )
W
Sn SiO /WO SiO /SnO m /g cm /g
nm
NH3
x
y
2
3
2
2
mmol/g
0.09
0.23
0.50
0.56
0.31
0.04
0.03
W Sn
3.5
9.2
4.3
5.3
79
27
17
9
41
712
641
585
411
0.98
0.81
0.73
0.51
9.2
9.2

11.5
3.8
1.7
3.4

3
5
5
W Sn
30
30
23
-
9
W Sn
5
13.9 5.0
23.0 5.4
9.2
1
4
W Sn
11.8
2
3
5
0
W Sn
14.0
-
-
18
-
1
4
W Sn
5.5
33
-
0
5
0
W Sn
-
1013
1.30
9.3
-
0
a
x and y represents wt% of Wand Sn; b actual elemental wt% measured by XRF; SBET
c
=
d
e
Specific Surface Area; V
= Total Pore Volume measured at P/P = 0.99; d
= BJH
P, BJH
o
P,BJH
f
g
adsorption Pore Diameter; Estimated from NH -TPD. Calculated from relative peak areas
3
corresponding to Lewis (L) and Brønsted (B) acid sites, using published adsorption coefficients
−
1
−1
(
0.165 cm μmol for Lewis and 0.078 cm μmol for Brønsted acid sites) [24].
2
8

Table 2. Comparison of anisole benzylation activity for W Sn catalysts (Reaction conditions:
x
y
anisole = 100 mmol, BzOH = 10 mmol, dodecane = 2 mmol, T = 130 °C, 1000 rpm)
Selectivity (%)
TOF a
min-1
0
TOF b
min-1
0
t,
min
Entry
Catalyst
XBzOH
o-BA p-BA DBE
1
2
3
4
None
360
120
120
20
0
–
–
–
W Sn
8.2
-
-
95.1
40.9
33.7
24.9
14.3
23.6
0
0.14
0.26
2.46
0
5
5
5
W Sn
29.2
42.8
97.2
100
89.6
100
100
100
62.0
100
30.9
35.3
37.7
42.1
39.1
48.7
11.4
41.7
38.3
46.3
26.9
31.0
37.3
43.4
37.3
49.2
10.4
44.0
35.0
48.1
0.74
4.67
3
W Sn
9
6
0
1
20
10
5
6
W Sn
5
11.07
3.53
3.18
17.24
3.53
4.34
1
4
3
0
W Sn
20
78.2
14.1
26.6
5.3
1
4
0
5
6
0
7
W Sn
10
2
3
6
0
a
b
TOF is defined is given by Eq. 3. TOF calculated based on W only.
2
9