Post-synthesis Snβ: An exploration of synthesis parameters and catalysis

Article abstract of DOI:10.1016/j.jcat.2015.06.023

Snβ is probably one of the best water tolerating heterogeneous Lewis acids for liquid phase catalysis. Instead of applying the usual lengthy hydrothermal synthesis to prepare Snβ, this contribution uses a more hands-on two-step synthesis method, involving the grafting of Sn precursors in isopropanol under reflux conditions on a commercial β zeolite that was dealuminated in acid. Among several reference synthesis procedures, this Sn introduction method resulted in active Sn catalytic sites. Taking advantage of this practical method, several synthesis parameters were explored and their impact on the catalytic activity in four different Lewis acid catalyzed reactions is discussed. The adsorption isotherm of Sn^IV in isopropanol over a broad range of Sn salt concentrations at reflux temperature is presented and discussed in relation with FTIR spectroscopy, UV-vis absorption characteristics and the porosity of the materials. The study reveals a selective Sn uptake, up to 2 wt% Sn loading, into silanol nests of the dealuminated precursor, forming a diversity of mononuclear Sn^IV. Higher Sn loadings result in less active Sn (hydrous) extraframework oxide phases, which also cause partial blockage of the zeolite micropores. Depending on the reaction type under study, space time yield may increase with increasing Sn loading, but the activity per Sn is always lower. Therefore it is concluded that a preferred synthesis should form high contents of isolated Sn active sites, especially for sugar isomerization and intermolecular Meerwein-Ponndorf-Verley, while the other reaction types like Baeyer-Villiger is also sufficiently catalyzed by the small Sn oxide clusters, albeit less actively.

Full text of DOI:10.1016/j.jcat.2015.06.023

Journal of Catalysis xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Post-synthesis Snb: An exploration of synthesis parameters and catalysis
Jan Dijkmans a, Jan Demol a, Kristof Houthoofd a, Shuigen Huang b, Yiannis Pontikes b, Bert Sels a,
⇑
a Center for Surface Science and Catalysis, KU Leuven, Kasteelpark Arenberg 23, 3001 Heverlee, Belgium
b Department of Materials Engineering, KU Leuven, Kasteelpark Arenberg 44, 3001 Heverlee, Belgium
a r t i c l e i n f o
a b s t r a c t
Article history:
Snb is probably one of the best water tolerating heterogeneous Lewis acids for liquid phase catalysis.
Instead of applying the usual lengthy hydrothermal synthesis to prepare Snb, this contribution uses a
more hands-on two-step synthesis method, involving the grafting of Sn precursors in isopropanol under
reflux conditions on a commercial b zeolite that was dealuminated in acid. Among several reference syn-
thesis procedures, this Sn introduction method resulted in active Sn catalytic sites. Taking advantage of
this practical method, several synthesis parameters were explored and their impact on the catalytic
activity in four different Lewis acid catalyzed reactions is discussed. The adsorption isotherm of SnIV in
isopropanol over a broad range of Sn salt concentrations at reflux temperature is presented and discussed
in relation with FTIR spectroscopy, UV–vis absorption characteristics and the porosity of the materials.
The study reveals a selective Sn uptake, up to 2 wt% Sn loading, into silanol nests of the dealuminated
precursor, forming a diversity of mononuclear SnIV. Higher Sn loadings result in less active Sn (hydrous)
extraframework oxide phases, which also cause partial blockage of the zeolite micropores. Depending on
the reaction type under study, space time yield may increase with increasing Sn loading, but the activity
per Sn is always lower. Therefore it is concluded that a preferred synthesis should form high contents of
isolated Sn active sites, especially for sugar isomerization and intermolecular Meerwein–Ponndorf–
Verley, while the other reaction types like Baeyer–Villiger is also sufficiently catalyzed by the small Sn
oxide clusters, albeit less actively.
Received 6 April 2015
Revised 23 June 2015
Accepted 29 June 2015
Available online xxxx
Keywords:
Heterogeneous catalysis
Lewis acid catalysis
Sn-beta
Post-synthesis procedures
Synthesis parameters
Adsorption process
Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction
including hydrogen-shift, oxidation and carbon–carbon coupling
reactions [25,30,33,34]. Though there is a general agreement on
Sn-containing silica materials are widely studied in catalysis
owing to their Lewis acid properties [1,2]. Tetrahedral Sn is able
to expand its coordination with two extra ligands, forming an octa-
hedral configuration [3,4]. Such expansion of coordination is useful
for its interaction with organic reagents in a catalytic cycle.
Formation of stable octahedral Sn sites, like in the framework of
SnO2, prevents such coordinative interaction rendering such mate-
rials catalytically inactive [5]. Therefore, homogeneous [6–12] and
heterogeneous Sn-species with varying Sn geometry have been
subject of many catalytic studies. The heterogeneous catalysts span
from amorphous [13], over mesoporous [14–24] to highly crys-
talline zeolitic silica [1,2,25–29]. Among them, Snb type zeolites
show the highest catalytic activity, provided that not too large
crystals or too bulky substrates are used [18,30,31]. According to
literature, Snb contains (in dried state) isolated and tetrahedrally
coordinated SnIV, built into the framework [32]. The catalyst has
been used successfully in several chemical transformations,
the presence of Lewis acidity because of Sn incorporation, the exact
nature of the Sn active catalytic site is under debate. So far, two dif-
ferent SnIV active sites have been revealed showing a common 4-O
inner-sphere complexation [4]: a perfect framework substitution
leading to Sn(OSi)4, also denoted as ‘closed’ site and a partly
hydrated Sn species, formally written as Sn(OSi)3(OH) and termed
‘open’ Sn sites [3,35]. A schematic representation of both sites is
shown in Scheme 1. Multiple reports currently discuss the impor-
tance (or not) of SnOH and nearby SiOH groups for the catalytic
properties [4,36,37]. Aside from the framework Sn-species, cat-
alytic activity has also been ascribed to small extra-framework
(EFW) oxidic Sn particles within the hydrophobic pores of the zeo-
lite, whereas large Sn oxide particles on the crystal surface do not
show this activity [5].
Snb is usually synthesized by a traditional hydrothermal syn-
thesis, that is above the water boiling point, in a closed vessel,
reaching super-atmospheric pressure (Scheme 2, top pathway)
[25,32]. During the synthesis process, Sn is gradually incorporated
into the beta zeolite framework. Unfortunately, nucleation of Snb
from Sn-containing gels is difficult. To facilitate the crystallization
⇑
Corresponding author.
E-mail address: bert.sels@biw.kuleuven.be (B. Sels).
http://dx.doi.org/10.1016/j.jcat.2015.06.023
0021-9517/Ó 2015 Elsevier Inc. All rights reserved.
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

2
J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
ent reaction conditions. In addition, use of TOF values implies the
absence of any mass transfer rate limitation, causing underestima-
tion of the true potential of the Sn sites and the determination of
the concentration of the active site, provided the identity of the lat-
ter is known [14,31]. Also the heterogeneity of Sn sites makes the
determination of genuine turn-over frequencies rather challenging.
Alternatively, a property known as the space-time yield (STY) is
often used to evaluate the practical potential of a catalytic mate-
rial, as not the activity per site but per catalyst weight is involved.
Careful inspection of Snb literature indeed indicates that a high Sn
content is mostly not a guarantee for a high STY, as often TOF
gradually decreases with increasing Sn, even when the authors
correctly identified the exclusive presence of isolated tetrahedral
SnIV [46,49,51].
This work evaluates variously prepared Snb zeolites for four
different reaction types and their catalytic data, expressed in both
TOF and STY. The selected reactions are glucose isomerization
(ISOM) to fructose, Meerwein–Ponndorf–Verley (MPV) of cyclohex-
anone with 2-butanol to cyclohexanol, Baeyer–Villiger (BV) oxida-
tion of cyclohexanone to caprolactone and a multi-step conversion
(via dehydration – MPV (DH/MPV)) of 1,3-dihydroxyacetone in
ethanol to ethyl lactate (Scheme 3). The products of these reactions
have applications as sweetener (fructose) or polymer building
blocks [52,53,54,55]. First, a set of reported post-synthetic proce-
dures is evaluated and compared, and the best method is used to
exploit the effect of numerous synthetic parameters such as Sn pre-
cursor type and concentration and water content on the catalytic
properties. Interestingly, the results will show that the best param-
eter set differs for some of the tested reactions, pointing to a diver-
sity of Sn sites and different Sn active site requirements. Catalytic
data are finally linked with Sn active site characteristics, as provided
by electronic and vibrational spectroscopy, and with pore blockage
phenomena.
Scheme 1. Schematic representation of a (a) closed tetrahedral Sn-site and (b) an
open tetrahedral Sn-site.
process addition of seed crystals is often used [38]. In spite of suc-
cessful synthesis of Snb, long synthesis times and hazardous min-
eralizing agents like HF or NH4F are indispensable. More practical
synthesis methods are therefore proposed. Modification of seeding
procedures result in increased rate of crystallization, while other
syntheses use steam to convert dry Sn gels into Snb materials
[39,40]. Unfortunately, these methods maintain the need for
F-ions as mineralizing agents.
A different approach is followed in the so-called secondary (or
post-) synthesis procedures. Instead of bottom-up synthesis,
pre-formed zeolites like beta are doped with Sn (top-down
approach, Scheme 2 bottom pathway). Usually, the parent zeolites
are subjected to an acid treatment to remove (part of) framework
Al (or Si or B), leading to dealuminated zeolites rich in silanol nests
[41–44]. These silanol groups function as anchoring points for the
Sn precursor. Among the top-down synthesis procedures, there is a
large variety of ways to introduce Sn. A solid-gas reaction with gas-
eous SnCl4 at elevated temperatures for instance results in a 6 wt%
Snb containing substantial amounts of framework Sn active sites
but also catalytically inactive EFW SnO2 [45,46]. A practical alter-
native, first introduced by the Hermans’ group, describes
a
solid-state mixing method of the dealuminated beta zeolite with
Sn2+acetate, followed by calcination. They showed the synthesis
of Snb with extremely high Sn content, up to 10 wt%, avoiding sub-
stantial content of EFW SnO2 [47]. In a similar procedure, use of
dimethyltin dichloride resulted in a 3.8 wt% of Snb [48]. Finally,
an adsorption process of SnCl4 in dried isopropanol under reflux
conditions, was recently communicated by our group, leading to
an active 1.6 wt% Snb [49,50].
The notion of turnover frequency (TOF) is used as a measure of
the intrinsic activity of a catalytic site. In Sn-catalyzed materials
too, efforts are reported to achieve maximal TOF values [4,49,51].
Though comparison between different reports is not always
straightforward, as several reaction types are investigated at differ-
2. Experimental
2.1. Material synthesis
Commercial
b
zeolite (CP814e, Zeolyst International,
SiO2/Al2O3 = 25) was dealuminated by stirring the zeolite powder
Scheme 2. Schematic representation of synthesis pathways of Sn-containing
zeolites. The top pathway shows the traditional (bottom-up) hydrothermal
synthesis; a Sn-containing gel in an autoclave is placed in an oven for a period of
time and the molecular building blocks assemble to form the Sn-containing zeolite
at high temperature and pressure. The bottom pathway shows the post-synthetic or
secondary synthesis. An Al-containing zeolite is dealuminated during an acid
treatment or steaming operation. The dealuminated zeolite is then modified with a
Sn precursor.
Scheme 3. Overview of the reactions studied. (a) Isomerization of glucose to
fructose; (b) Meerwein–Ponndorf–Verley reaction of cyclohexanone and 2-butanol
forming cyclohexanol and 2-butanone; (c) Baeyer Villiger oxidation of
cyclohexanone with hydrogen peroxide yielding
e-caprolactone and water;
(d) transformation of 1,3-dihydroxyacetone in ethanol to ethyl lactate.
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
3
overnight in an aqueous HNO3 solution (55 ml/g of zeolite, over-
night at 353 K), an acid concentration of 7.2 M was used. Afterward
the solids were filtered, washed with water and dried at 333 K. All
of the post-synthesis procedures were preceded by an activation of
the dealuminated zeolite at 423 K to remove any physisorbed
water, followed by a calcination step in air at 823 K. Samples made
by an impregnation method (IMP) with SnCl4Á5H2O, containing
2 wt% of Sn w.r.t. dry zeolite powder, were slurried in dry iso-
propanol (5 ml of alcohol per gram of zeolite powder). The solvent
was slowly evaporated at 323 K under vigorous stirring, causing Sn
precipitation onto the zeolite [56]. For samples made by a chemical
vapor deposition (CVD), the procedure described by de Correa et al.
was used [24]. The required amounts of SnCl4 and dealuminated
beta zeolite were placed into a Teflon-lined stainless steel auto-
clave, that was heated at 373 K for 8 h. Materials made by a solid
mixing procedure (Solid-MIX), underwent manual grinding for
several minutes, using a mixture of dealuminated zeolite and
SnCl4Á5H2O powder in the required amounts [56]. Grafting of SnCl4
onto the dealuminated beta zeolite was tested in two different sol-
vents. For the H2O-graft, a procedure described by Zhang et al. was
followed [57]. The second solvent in the grafting procedure was
dried isopropanol (IPA-graft), using a method described by Wagner
et al. for grafting Sn onto mesoporous carrier materials [58]. The
dealuminated zeolite was added to SnCl4Á5H2O (27 mmol/g of zeo-
lite) in water or dry isopropanol (100 ml/g of zeolite) and placed in
a reflux setup under N2 atmosphere. After 7 h, the mixture was fil-
tered in air, rinsed with water or dried isopropanol and dried at
333 K. When varying the Sn-precursor, equimolar amounts of the
new precursor were used. All other synthesis parameters were
kept constant. Metallic Sn0 powder was converted in situ to a
Sn-salt by adding a 4.0 M HCl in 1,4-dioxane-solution (8 moles of
HCl per mole of Sn0) to the isopropanol reflux solution. As the
intended catalytic material is investigated as an alternative for
the hydrothermally (HT) synthesized Snb, this material was syn-
thesized using a procedure described in literature [59]. The
hydrothermally synthesized and post-synthetically treated beta
zeolite crystals used in this work have dimensions, as measured
with SEM of 1000–1500 and 10–30 nm, respectively [49]. This is
far below the limit at which pore diffusion limitations was shown
Agilent 1200 series HPLC equipped with Metacarb 67C column
and RI detector was used. Identification of the products was based
on retention time analysis and confirmed by GC–MS (Agilent 6890
GC with HP5-MS column and Agilent 5973 Mass Selective
À1
Detector). TOFs were calculated as moleproduct mole hÀ1, while
Sn
À1
STY was determined as g
g
hÀ1
.
product catalyst
2.3. Characterization techniques
The Sn-content of the materials was all determined by electron
probe microanalyzer (EPMA) analysis conducted on JEOL
a
JXA-8530F field emission microprobe using WDS. Samples were
embedded in a resin, and the surface was ground, polished and
coated with carbon before measurement. The microprobe was
operated at 10 kV with a probe current of 1.5 nA. The Sn L 1-signal
a
was detected using a PETH crystal and the Sn concentration was
quantified with a cassiterite standard. ZAF was used for the matrix
correction method. Absorption in the UV–vis region was recorded
on an Agilent Cary 5000 spectrophotometer. Samples were placed
in a quartz tube with window, and were dried at 823 K in a dry
N2-flown (unless stated otherwise) before measurement using a
heating rate of 5 K/min. Samples were corrected with a calcined
dealuminated beta zeolite to isolate Sn-related signals. FT-IR mea-
surements of the silanol groups were performed on a Nicolet 6700
Spectrometer equipped with DTGS detector. The materials were
pressed into self-supporting wafers and degassed at 673 K in vacuo
before measurement at 423 K. The broad silanol signal was decon-
voluted into the individual signals using OriginPro 8 software using
fixed FWHM values for the signals between the materials. Lewis
acid sites were probed with deuterated acetonitrile, cyclohexanone
and pyridine as probe molecule. Deuterated acetonitrile was
adsorbed at room temperature. The samples were exposed to
100 mbar of probe molecule for 10 min., afterward desorption in
vacuo of the molecule was followed at the same temperature. For
pyridine adsorption, the samples were subjected to 25 mbar of
the probe until saturation at 323 K. Spectra were recorded at
423 K in vacuo after equilibration for 20 min. 0.89 cm l
molÀ1 was
used as integrated molar extinction coefficient for calculation of
Lewis acid density [60]. For cyclohexanone the samples were
exposed to 6.5 mbar of probe molecule until saturation at room
temperature. Spectra were recorded at different temperatures in
vacuo after equilibration of 10 min. at each heating step. X-ray
diffraction patterns were recorded on a STOE Stadi P diffractometer
to take place (>7 lm [31]). Therefore the catalytic reaction rates
are assumed to be diffusion free and only determined by the
chemical transformations.
2.2. Catalytic tests
equipped with Cu Ka1 source and IP-PSD detector. Before measure-
ment, the samples were stored in a controlled humidity environ-
ment, ie. above an aqueous saturated NH4Cl solution (79%
humidity). 119Sn MAS NMR spectra were recorded on a Bruker
DSX400 spectrometer (B0 = 9.4 T) operating at a NMR frequency
of 149.21 MHz. Single-pulse excitation measurements with vary-
ing interscan delays between 5 and 40 s have been used. The sam-
ples were packed in 4 mm zirconia rotor; the spinning frequency of
the rotor was 12,000 Hz. The chemical shift is referenced with
respect to Sn(CH3)4. SnO2 was used as a secondary standard; the
chemical shift of its center band was set at À604.0 ppm [61]. N2
sorption measurements were performed using a Micromeritics
Instruments Tristar 3000 at 77 K. Samples were degassed under
nitrogen flow at 527 K overnight prior to measurement. Pore
volumes were calculated using the t-plot method.
Catalytic tests were performed in magnetically stirred and
closed glass reactors of 10 ml, which were placed in a copper block.
Temperature control is carried out in a reference glass reactor with
solvent. For isomerization reactions, 100 mg of catalyst was added
to 5 ml of a 10 wt% aqueous glucose solution. The reactions were
performed at 383 K. For Baeyer–Villiger reactions, 1.11 mmol of
ketone was added to 50 mg of catalyst in 5 ml of dioxane. 50 wt%
aqueous H2O2 solution was added in a molar H2O2/ketone-ratio
of 2. Ethylcyclohexane was used as the internal standard for chro-
matographic analysis. The reactions were performed at 363 K. Me
erwein–Ponndorf–Verley reactions were performed at 373 K.
15 mg of catalyst was added to a 2-butanol solution containing
1 mmol of ketone, the solvent was used in a 2-butanol/ketone ratio
of 50. 1,4-Dioxane was used as an internal standard. For the con-
version of dihydroxyacetone to ethyl lactate, 1 mmol of substrate
was dissolved in 5 ml of ethanol, 1,4-dioxane was used as internal
standard. 75 mg of catalyst was used. The reaction was performed
at 363 K. For each reaction, aliquots of the sample were taken
at regular time intervals through a rubber septum and were
quantitatively analyzed with an Agilent 6850 GC, equipped with
a HP-1 column and FID detector. For non-volatile products, an
3. Results and discussion
3.1. Catalytic exploration of various post-synthetic Snb materials
The catalysts studied in this work are all synthesized according
to (top down) post-synthesis routes. All procedures start with a
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

4
J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
commercial aluminum containing
b
zeolite (Zeolyst CP814E,
(compare Entries 5 and 6). Surprisingly high Sn activity in BV is
SiO2/Al2O3 = 25). Incorporation of Sn proceeds in two consecutive
steps. First, the zeolite material is dealuminated by an acid treat-
ment using a 7.2 M HNO3 aqueous solution. Such dealumination
of beta zeolites has been studied extensively in the past
[41,42,62–65]. Though the BEA structure is conserved during this
acid treatment, extensive removal of Al is observed, creating resid-
ual holes in the framework surrounded by silanol groups. In a sec-
ond step, a Sn-containing precursor interacts with these silanol
groups, forming a covalent SiAOASn bond introducing Sn in the
zeolite framework (Scheme 2, bottom pathway) [49]. Several pro-
cedures to introduce Sn onto carrier materials have been described
[16,24,47,48,56–58,66]. Inspired by them, we selected several syn-
thesis procedures to achieve the most active catalytic material.
In total, six materials were synthesized. Five materials were
made by using different Sn introduction methods and one refer-
ence material was synthesized using the traditional hydrothermal
(HT) procedure [59]. The Sn introduction methods, which were
used, are as follows: impregnation (IMP) [56], chemical vapor
deposition (CVD) [16,24], solid mixing (Solid-MIX) [56] and graft-
ing in water (H2O-graft) [57] and dried isopropanol (IPA-graft)
[58]).
Results of the elemental analysis of the synthesized materials
are displayed in Table 1. All samples show Sn-contents ranging
from 1 to 3 wt%, except for the material synthesized by grafting
of SnCl4Á5H2O in water, which showed an elevated Sn-content of
27.6 wt%. This sample colored yellow, whereas all other samples
were white. EPMA analysis of this water-grafted sample clearly
showed two distinct regions. The first was the zeolite with high
amount of Sn (20 wt%), corresponding to a Si/Sn-ratio of 7.3. A sec-
ond region was identified as SnO2 (see Fig. S1). The use of water as
a solvent in the grafting procedure thus leads to formation of a
large SnO2-phase, a catalytically inactive oxide. EPMA of the other
synthesis procedures shows a homogeneous phase of zeolite parti-
cles, containing Sn according to the aforementioned elemental
compositions.
observed with the Solid-MIX, almost equally comparable with
the IPA-graft (47 and 49 hÀ1, respectively), whereas this catalyst
shows only moderate Sn activity for the other reaction types.
Similarly, the Sn active sites in IMP are not very active for most
reactions except for DH/MPV. Here, the Sn active sites perform
nearly as good as those in IPA-graft, showing TOFs of 87 and
103 hÀ1, respectively (compare Entries 1 and 5).
The conclusions are slightly different by comparing STYs. The
following activity order can be extracted: for ISOM,
HT > IPA-graft > CVD = IMP > H2O-graft > Solid-MIX;
IPA-graft > HT = IMP > CVD > Solid-MIX > H2O-graft;
IPA-graft > CVD > Solid-MIX > HT > IMP > H2O-graft;
for
for
and
MPV,
BV,
for
DH/MPV, IMP > IPA-graft > HT = CVD > Solid-MIX > H2O-graft. As
before, IPA-graft performs very good in most reaction types. Most
notable is probably the successful use of IMP for DH/MPV catalysis,
showing very high STY of about 1737 hÀ1 and the disappointing
results with HT in catalyzing this reaction (Entry 1 vs. 6). The
changes in the activity order between TOF and STY are due to dif-
ferences in Sn content. Literature data and results in this work (vide
infra) reveal that a simple increase of Sn not necessarily means a
linear increase of STY, pointing to the formation of various Sn
species with different catalytic activity. For ISOM, it was demon-
strated that active Sn sites in Snb synthesized by grafting or solid
mixing procedures provide high activity sites at low Sn-content.
Maximal STY however was often reached at moderate Sn-content
[49,51]. A similar conclusion may be extracted from reported data
for a gaseous grafting procedure; a constant STY value was
observed despite a rising Sn-content, as confirmed by a decreased
TOF [46].
The initial screening of the different Sn-incorporation proce-
dures thus demonstrates that the synthesis of catalytically active
materials using straightforward and practical post-synthesis pro-
cedures is possible, in line with other literature reports [43–49].
Depending on the reaction type, however, different optimal active
sites are found and consequently, variation in synthesis method
and parallel characterization is advisable. As the highest activity
for most reactions was reached with IPA-graft, this synthesis
method will be applied in the continuation of this work, describing
the impact on the catalytic activity of various synthesis parameters,
like Sn precursor type, Sn loading and water content.
Data of the catalytic tests with the five catalysts for the four
reactions are collected in Table 1. A tentative catalyst ranking
based on the average activity per Sn site shows the following
order:
Solid-MIX; for MPV: IPA-graft > HT > IMP > Solid-MIX > CVD >
H2O-graft; for BV: IPA-graft = Solid-MIX > HT > CVD =
for
ISOM:
HT > IPA-graft > CVD = IMP > H2O-graft >
IMP > H2O-graft; and for DH/MPV: IPA-graft > IMP > HT > CVD >
Solid-MIX > H2O-graft.
3.2. Effect of precursor concentration
Clearly, the order of preferred Sn active sites is different for each
reaction type, pointing to different Sn active site requirements. As
expected, the highly Sn-loaded H2O-graft performs inferior in all
reactions, due to the presence of inert SnO2. In general, IPA-graft
performs best for most reaction types. For instance, TOF values of
96, 49 and 103 hÀ1 were measured for MPV, BV and DH/MPV,
respectively (Entry 5). Only for ISOM the Sn active sites of the orig-
inal HT are preferred, despite having the largest crystals of the five
materials tested [49,50]. A high TOF value of 306 hÀ1 was obtained
with HT, whereas the second best IPA-graft has a TOF of 133 hÀ1
First, the effect of varying SnCl4Á5H2O concentrations on the Sn
loading and Sn oxide speciation of the final material was studied.
The adsorption isotherm at reflux temperature, displayed in
Fig. 1a, show two different Sn adsorption sites on the dealuminated
beta zeolite, one at low (site A) and one at high Sn loading (site B).
The isotherm shows a clear Sn adsorption at low SnCl4Á5H2O con-
centrations with moderate affinity, corresponding to an uptake
efficiency of 8.3 lmol Sn per soluble mmol Sn per gram of dealumi-
nated zeolite. The adsorption site is saturated at 0.3 M equilibrium
Sn concentration, corresponding to a 1.6 wt% or 0.13 mmol gÀ1 Sn
Table 1
Comparison of the Sn-content and catalytic performance in the ISOM, MPV, BV and HS reactions catalyzed with materials synthesized according to different synthesis procedures.
Entry
Synthesis method
Sn-content
ISOM
MPV
BV
DH/MPV
STY
TOF
hÀ1
STY
TOF
hÀ1
STY
TOF
hÀ1
STY
TOF
hÀ1
wt%
hÀ1 (Â1000)
hÀ1 (Â1000)
hÀ1 (Â1000)
hÀ1 (Â1000)
1
2
3
4
5
6
IMP
CVD
Solid-MIX
H2O-graft
IPA-graft
HT
2
2.9
1
27.6
1.6
1.3
702
722
0
280
3640
5880
23
16
–
0.7
152
306
758
350
246
57
1277
824
45
14
29
0.3
96
78
402
606
462
156
742
448
21
22
47
0.6
49
37
1737
1069
269
69
1621
1131
87
37
27
0.3
103
72
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
5
loading. This value is unexpectedly low taking into account that
about 0.9 mmol gÀ1 Al is removed. This low uptake, which is
that these silanols are involved in site A adsorption. As higher Sn
uptake occurs without further consuming silanols, site B adsorp-
tion is not associated with the silanol nests, but rather implies a
structure independent extraframework (EFW) species. The exter-
nal silanols appear unreactive during the Sn grafting procedure.
As the FTIR analysis suggests a different identity of Sn in the
two adsorption sites, a similar detailed study was carried out with
diffuse reflectance UV–vis spectroscopy (Fig. 1b). DR UV–vis on
dried samples was used to investigate the electronic properties
of the adsorbed Sn species, through the charge transfer absorption
of SnAO. Absorption in the low energy UV range (280–350 nm) is
indicative of an octahedral coordination in SnIV species, often of
polymeric nature like in bulk SnO2. Isolated tetrahedrally coordi-
nated SnIV absorbs high energy UV light at 200 nm [5]. The inten-
sity at 200 nm was used as indicator of isolated Sn sites, whereas
the intensity at 255 nm shows the presence of oligomeric, assum-
edly tetrahedrally coordinated, Sn oxide species. The intensities
were plot against the Sn loading and superimposed onto Fig. 1a.
Except for the sample free of Sn, all catalysts show clear absor-
bance at 200 nm and this signal rises linearly with the adsorption
of Sn on the internal silanols (site A). This site matches with the
currently difficult to rationalize, could be indicative of
a T
site-specific adsorption process in the dealuminated Beta zeolite.
After site A saturation, a linear increase of Sn content is set in with
increasing SnCl4Á5H2O equilibrium concentration above 0.30 M.
The affinity of the site B for Sn, corresponding to 5.2 lmol adsorbed
Sn per mmol Sn in solution per gram, is somewhat lower than that
of site A. Absence of a saturation point is noticed at least up to
8 wt% Sn loading, indicative of
mechanism.
a non-specific adsorption
More proof of a dual SnCl4 adsorption mechanism is given by
inspection of the FTIR spectra in the silanol vibrational range.
Two silanol species were therefore monitored after varying Sn con-
tent uptakes and sample calcination; the internal silanols, formed
by the dealumination procedure, show a characteristic absorption
around 3730 cmÀ1, while external silanols at crystal surfaces and
defects are visible at 3740 cmÀ1 [41]. The absorption area of both
silanol species, relative to these in a Sn-free dealuminated material
is plotted against the Sn-content in Fig. 1c. Clearly, internal silanol
groups are consumed up to about 2 wt% Sn content, suggesting
Fig. 1. (a) Left and bottom axis, d: equilibrium concentration of SnCl4Á5H2O used in the grafting procedure and the Sn-content of the final Sn-grafted materials. The lines are
guide to the eyes. A standard deviation error bar for the Sn-content of samples grafted with 0.27 M SnCl4Á5H2O (based on 5 separate samples) is shown next to the data point.
Left and top red axis, red hollow data points: Intensity of the 200 ( ) and 255 nm ( ) signal as recorded with UV–vis for dried samples with varying Sn-content. The
intersection of the red line ( ) and y-axis, around 2 wt%, indicates the maximal tetrahedrally coordinated isolated Sn-content possible without substantial presence of other
Sn oxide species. (b) UV spectra with varying Sn content (increasing Sn-content from bottom to top. The spectra are corrected with a calcined dealuminated zeolite to isolate
the Sn-related signals. (c) Normalized concentration of internal ( ) and external ( ) silanol groups in function of the Sn-content of the materials. The sum of both species (D)
is shown with dotted line. Lines are guide to the eye. The data were obtained by deconvolution of the FTIR signal for the 3730 (internal) and 3740 (external) cmÀ1 signals.
Signal was scaled to weight. d. FTIR spectra of samples with varying Sn content with the deconvolution of the signals for internal and external silanol groups. The red line
indicates the 3740 cmÀ1 signal for external silanols, the blue line shows the 3730 cmÀ1 signal for internal silanol and the dotted black line shows the recorded signal. The
broad shoulder at lower frequencies is caused by hydrogen bonded silanol groups, the fitting being omitted in the graph for reasons of clarity. (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

6
J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
formation of the tetrahedral Sn sites, pointing out that about 2 wt%
of Sn can be selectively accommodated in this particular site.
Further increase of the Sn loading is no longer accompanied with
an increase of the 200 nm signal. Instead, a 255 nm feature
becomes more dominant with increasing Sn loadings above
2 wt%. Though its exact molecular nature is unknown, its low
energy absorption suggests an octahedral mono- or oligo-meric
or a tetrahedral oligomeric Sn oxide speciation, which agrees with
the EFW location, suggested by the IR analysis [5,67,68]. Sizes of
SnO2 may be obtained from the wavelength at maximum absorp-
tion, showing a shift to higher wavelengths for increasing oxide
particle size [5,69]. As no shift in the absorption maximum is
observed, size of the oxides particles is assumed to be constant,
and the higher absorption intensity is caused by an increased
concentration.
Further information about the identity of the 255 nm signal was
obtained with X-ray diffraction. The patterns of three grafted
samples, having resp. 1.6, 4.5 and 8.5 wt% Sn, are illustrated in
Fig. 2(c–e) together with the Sn-free dealuminated beta (b) and a
SnO2À reference (a). The characteristic d302 reflection (2h = 22.4°)
of the BEA topology is visible in all samples, while d101 and d211
reflections (2h = 33.8° and 51.7° resp.) of bulk SnO2 show only
Fig. 3. Plot of TOF and STY (Â1000, inset) in function of Sn-content for the ISOM
(black squares), MPV (red circles), BV (blue triangles) and DH/MPV (green
diamonds) reaction. For the STY-plot of the ISOM reaction, extra data points (gray)
from Ref. [49] are added for clarity. Sn-content was varied by varying the
SnCl4Á5H2O concentration in the IPA grafting liquid. The lines are guides to the
eye. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
weak intensities for the highest Sn loading (8.5 wt%). As
a
0.5 wt% detection limit was established for the presence of bulk
SnO2 with X-ray analysis [50], the present Sn oxide phase is likely
less crystalline in nature except for highest Sn-loading.
The influence of the two types of Sn sites on the catalytic prop-
erties (STY and TOF) is evaluated for the four model reactions in
function of the Sn loading (Fig. 3 and Table S1) In all reaction types,
the highest TOFs are found for samples with low Sn content.
Though UV–vis absorption demonstrated a single active site of
tetrahedral Sn below 2 wt% Sn loading, maximum turnovers per
Sn were always observed at the lowest Sn content (0.5 wt%). This
observation is in line with reported data for various
post-synthetic synthesis procedures [46,49,51]. Therefore, the
200 nm absorption, represented by the Sn adsorbed in site A, likely
represents a diversity of catalytic Sn sites with different activities.
Such heterogeneity in the catalytic properties of the 200 nm
absorbing Sn-species may be due to the varying site specifications,
but also its location and thus its steric demand in the zeolite
framework may cause dissimilarities in the kinetic parameters.
Numerous literature examples describe changed catalytic proper-
ties in enzymatic, homogeneous and heterogeneous catalysts due
to small structural alterations [35,70–73].
The STY of the four reactions in function of the Sn loading is dis-
played in the insert of Fig. 3. As STY values initially increases, the
lower activity per Sn (TOF) is largely compensated by the amount
of Sn. However, the optimal Sn loading to achieve maximum STY
depends on the reaction type. The catalyst containing about
1 wt% Sn is preferred for ISOM, whereas higher Sn loadings of 2
and 4 wt% are required for high STY in MPV and BV, respectively.
Higher Sn loadings cause a decrease in catalytic activity. Three
reasons may be invoked to explain this phenomenon: formation
of EFW SnO2 hides Sn atoms into its structure, Lewis acid sites of
the EFW are catalytically less active due to site-specific require-
ments, or EFW particles block micropores. Probing of Lewis acid
sites with pyridine is a useful method to evaluate Sn site accessibil-
ity. The data are presented in Fig. 4 and Table S2. A constant Lewis
acid-to-Sn molar ratio of 0.7 was monitored for Sn loadings below
2 wt%, corresponding to Sn in site A. Though a ratio of 1 was
expected, deviation is explained by either the presence of some
weaker Lewis acid sites, not probed by the method, or more likely
by the uncertainty of the molar extinction coefficient in the quan-
tification, as multiple values are reported [50,60,74–76]. Higher
Sn-loadings show more Lewis acid sites, likely due to additional
Lewis acidity of the EFW species, but the increase deviates from
linearity with the Sn content as lower Lewis acid-to-Sn molar
ratios were measured: for instance, 0.5 and 0.3 for 3.6 and
4.5 wt% Sn samples, respectively. Pore blockage (partially) explains
this deviation as a systematic drop in micropore volume was
noticed for the highly loaded Sn samples, in concert with the drop
in acid-to-Sn molar ratio (Fig. 4). For instance, 0.16 mL gÀ1 pore
volume was measured for the 1.6 wt% loading, whereas this value
drops to 0.12 mL gÀ1 for the 4.5 wt% sample. However, as normal-
izing the number of Lewis acid sites for total micropore volume is
insufficient to provide a linear correlation with the Sn content
(Fig. S2), inaccessibility of Sn atoms in the inner parts of the EFW
particles should also be invoked to explain the lower acid-to-Sn
molar ratios for the highly loaded samples.
Fig. 2. X-ray diffraction pattern of (a) SnO2, (b) Dealuminated Beta zeolite and (c)
1.6 wt% Sn (0.27 M SnCl4Á5H2O, H2O:Sn = 5), (d) 4.5 wt% Sn (0.81 M SnCl4Á5H2O,
H2O:Sn = 5), (e) 8.5 wt% Sn (1.35 M SnCl4Á5H2O, H2O:Sn = 5), (f) 7.1 wt% Sn (0.27 M
SnCl4Á5H2O, H2O:Sn = 12) and (g) 11.3 wt% (0.27 M SnCl4Á5H2O, H2O:Sn = 31)
grafted zeolite. The d101 and d211 SnO2 reflections are indicated with a vertical
dotted line.
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
7
and an elevated catalytic activity (in comparison with SnCl4) in an
epoxide hydration reaction [48].
Following the IPA grafting procedure, precursors with different
alkyl groups, such as tetrabutyltin (Bu4SnIV) (four alkyl groups),
dibutyltin dichloride (Bu2SnIVCl2), tin dichloride dihydrate
(SnIICl2Á2H2O) (two alkyl groups), tin tetrachloride pentahydrate
(SnIVCl4Á5H2O) and anhydrous SnIVCl4 were used. The elemental
analysis (Table 3) shows a higher Sn-content for the final material
for precursors with a higher number of Sn–Cl bonds, despite the
use of equimolar concentrations of the precursor during its synthe-
sis. This observation suggests that the driving force of the SnCl4
adsorption is governed by the surface reaction between silanol
and Sn–Cl. As significant amounts of Sn were found in IPA-graft
with Bu4Sn, also other adsorption pathways are present, presum-
ably proceeding via adsorption onto the zeolite during the grafting
procedure and disintegration of the molecule adsorbed species
during the calcination step.
Fig. 4. Lewis acidity (left axis, d) and the micropore volume (right axis, ) in
Metallic Sn powder, combined with an excess of 8 equivalents
of HCl was also used as a precursor, by converting the metal
in situ in the reflux solution to SnClx. This pathway was briefly
explored due to the low cost of metallic Sn, but also to create a pos-
sible synthesis route to yield isotope enriched materials, in order to
enhance the sensitivity of 119Sn-characterization techniques like
119Sn MAS NMR and 119Sn Möbbauer. Interestingly, use of Sn0/HCl
in the IPA-graft procedure results in a higher Sn-content (Table 2),
while no patches of undissolved Sn0 or formed SnO2 were found in
the EPMA and X-ray diffraction analyses (Figs. S4 and S5).
Grafting of SnCl4Á5H2O, SnCl4 and Bu2SnCl2 onto dealuminated
beta zeolite all result in an intense UV absorption around
200 nm, indicative of pure tetrahedral SnIVO4 moieties, recognized
as the catalytically active Sn species (Fig. 5) [5]. Whereas this
absorption is the single absorption feature for SnCl4Á5H2O, use of
the other precursors leads to additional signals. For materials
grafted with anhydrous SnCl4 and Bu2SnCl2 precursors, these low
UV energy absorption features are present at 263 nm and a shoul-
der at 245 nm (Fig. 5b and c). The observed absorptions are surpris-
ing, as formation of oligomeric or polymeric (hydrous) Sn oxides
requires the presence of water molecules and no X-ray reflections
of the oxide were found (Fig. S5). Residual SnACl or Sn-alkylic may
not be invoked since the first band shows UV absorption in the
120–210 nm range [78,79], while the latter SnAC bond type is
decomposed thorough calcination in air at high temperature
[48,77]. However, it has to be noted that hydrothermally synthe-
sized Snb, believed to contain isolated tetrahedral SnAO, also
shows broad absorption features, with shoulders up to 300 nm
[5,25,49,80]. Possibly, the secondary absorptive features are also
related to isolated tetrahedral Sn-species, but showing different
SnAO bond lengths due to dissimilarities in the Sn geometry. The
difference in nature and concentration of the Sn-species when
using SnCl4Á5H2O or its anhydrous equivalent is remarkable. Use
of the former leads to a lower Sn loading, showing just one UV
absorption feature, whereas in complete dry circumstances a
higher Sn loading is obtained, showing multiple absorption signals.
The difference in loading seems to be related to the strong interac-
tion of the water molecules with the empty silanol nests in the
dealuminated zeolite, as very strong interaction of water molecules
with such silanols in zeolites has been proven with NMR and ther-
mogravimetric analysis [81,82]. Occupation of the silanol nests
with strongly bound water complicates reaction with SnCl4, lead-
ing to lower Sn-loading.
function of the Sn-content of the materials. Lines are guide to the eye. Black dotted
line indicates
a linear increase in acidity, red dotted line indicates a stable
micropore volume.
Similar as for the Lewis acidity of the materials, normalizing STY
for total micropore volume (Fig. S3) provides insight into the
trends of changing STY’s with high Sn loadings. For ISOM and
MPV, a plateau in the normalized STY is reached at loadings of
respectively 0.4 and 2 wt%, indicating that the observed decrease
in STY, as presented in Fig. 4, is directly attributed to inaccessibility
of Sn sites by pore blockage. BV, and to lesser extent DH/MPV,
shows a linear increase for the normalized STY at Sn-contents
above 0.4 wt%, and therefore the reaction rates are not hampered
by the presence of pore-blocking EFW SnO2 species. Moreover, as
an increase in normalized STY is observed, the EFW species (site
B type) must possess some catalytic activity. Indeed the STY of
BV increases from 742 to 1202 hÀ1 with 1.6 and 3.6 wt% Sn, respec-
tively, whereas UV analysis clearly indicated no additional forma-
tion of tetrahedral isolated Sn species (Site A types). A further
comparison of the catalytic data with the UV features of the mate-
rials supports the idea that isolated Sn types (in site A) are very
active for all reactions, and required to obtain activity in MPV
and sugar isomerization. BV reactions can also be substantially
catalyzed by oligomeric clusters (in site B). The seemingly
independence of the DH/MPV rate for Sn loading (in the range of
0.5–8 wt%), despite a dropping accessibility of the tetrahedral
Sn-sites in the pores of the material suggests that this reaction is
also catalyzed by the EFW Sn species, albeit with a much lower cat-
alytic efficiency. More likely, the reaction rate of DH/MPV should
be taken with caution as DH rather than MPV is rate determining
in the cascade reaction, as was discussed in an earlier publication
[50].
3.3. Variation of Sn precursor
The identity of the Sn active site might change by varying the
Sn-precursor during the grafting procedure. As explained above,
SnCl4 reacts with the silanol groups of the zeolite to form SnAOASi
framework bonds, releasing HCl in the chemical process. A replace-
ment of a Cl-atom of the Sn-precursor molecule with an alkyl
group should easily prevent formation of such a framework bond.
The alkyl groups are removed in a calcination step and likely result
in SnOH formation. Note that the SnOH species, present in an
‘open’ Sn site, were proposed as an essential part of the active cat-
alytic species in BV and MPV reactions [3,35]. Use of dibutyltin
dichloride (Bu2SnCl2) onto MCM-41 was successfully tested [77],
and more recently a dry impregnation of dimethyltin on dealumi-
nated beta zeolite resulted in high concentrations of ‘open’ Sn sites
SnCl2Á2H2O and Bu4 Sn-grafted materials show absorption max-
ima at 211 and 218 nm, respectively, and some secondary signals
at 281 and 275–325 nm. The presence of SnII could be used to
explain the 281 nm signal, though high temperatures in the pres-
ence of oxygen during the calcination ensures total oxidation to
SnIV [51]. Weak signals of the d101 and d211 reflections of SnO2
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

8
J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
Table 2
Comparison of the Sn-content and catalytic data of ISOM, MPV, BV and HS reactions for Sn-grafted catalyst synthesized with varying Sn precursors.
Entry
Precursor
Sn-content
ISOM
MPV
BV
DH/MPV
STY
TOF
hÀ1
STY
TOF
hÀ1
STY
TOF
hÀ1
STY
TOF
hÀ1
wt%
hÀ1 (Â1000)
hÀ1 (Â1000)
hÀ1 (Â1000)
hÀ1 (Â1000)
1
2
SnCl4Á5H2O
SnCl4
1.6
2.9
2.1
0.8
0.7
0.3
2.6
3640
2585
3100
2740
1233
360
152
59
97
227
112
95
1277
1143
1087
1154
933
96
47
61
174
140
143
741
495
705
633
578
132
385
49
17
35
87
83
55
15
1621
1319
1274
1666
198
103
47
61
216
28
70
3a
4
SnCl4 + 5H2O
Bu2SnCl2
SnCl2Á2H2O
Bu4Sn
5
6
7
304
31
175
592
Sn0/HClb
2710
68
1.4
23
a
Anhydrous SnCl4 was mixed with 5 equivalents of water.
Converted in situ to SnClx.
b
show an absorption feature at 2308 cmÀ1, predominantly ‘closed’
Sn sites are expected to be present in both samples.
A similar FTIR experiment was performed with cyclohexanone
instead of acetonitrile. Here, the vibration frequency of the car-
bonyl group was taken as a measure of the Lewis acid strength:
the larger the wavenumber, the stronger the Lewis acid interaction.
The data of two Snb catalysts, grafted with Bu2SnCl2 and SnCl4, are
presented in Fig. 6(2). The vibration of an unperturbed carbonyl
group, likely of physisorbed cyclohexanone, is visible at
1714 cmÀ1. Upon adsorption on Sn this vibration shifts to higher
frequencies. As both SnCl4 and Bu2SnCl2 grafted materials show a
similar shift of the carbonyl vibration to 1652 cmÀ1 (at 423 K), it
is concluded that the Lewis acidity of both Sn-sites is similar. This
statement is also confirmed by a temperature programmed FTIR
study with pyridine, the data being presented in Fig. 6(3). The ratio
of pyridine sorbed onto the Lewis acid sites of both materials at
523 compared to 423 K is very similar, viz. 0.22 for SnCl4-grafted
materials and 0.24 for Bu2SnCl2 grafted materials, confirming that
the Sn-sites have very similar Lewis acid strengths. The lower
intensity of the signal for the Bu2SnCl2 grafted material points to
lower concentration of adsorbed pyridine, in line with its lower
Sn-content. A similar trend is visible in the deuterated acetonitrile
and cyclohexanone probed spectra.
Fig. 5. DR-UV–vis spectra of material grafted with (a) SnCl4Á5H2O, (b) SnCl4, (c)
Bu2SnCl2, (d) SnCl2Á2H2O, (e) Bu4Sn and (f) Sn0/HCl. Samples were dried in N2 at
823 K and corrected for dealuminated beta. Samples (e) and (f) are scaled with a
factor of 4 and 0.5, respectively.
In conclusion, although the nature of the Sn-species in the var-
ious materials may not be fully established by the characterization
techniques, it is clear that use of different Sn-precursors results in
varying Sn-species in the final Sn beta zeolites. However, irrespec-
tive of the differences in the electronic spectra, Sn sites with differ-
ent electronic properties may deliver comparable Lewis acid
properties, and therefore may lead to comparable catalytic results.
The catalytic data for the different reactions tested in terms of
TOF and STY values, are summarized in Table 2. For all reaction
types, the precursor Bu2SnCl2 showed the highest activity per Sn
atom, though the activity order depends on the reaction type.
The TOF-ranking for ISOM is Bu2SnCl2 > SnCl4Á5H2O > SnCl2 >
Bu4Sn > Sn0/HCl = SnCl4, while for MPV it is Bu2SnCl2 > SnCl2 =
Bu4Sn > SnCl4Á5H2O > SnCl4 > Sn0/HCl; for BV it is Bu2SnCl2 =
SnCl2 >Bu4Sn = SnCl4Á5H2O > SnCl4 > Sn0/HCl and for DH/MPV:
Bu2SnCl2 ꢀ SnCl4Á5H2O > Bu4Sn > SnCl4 > SnCl2 > Sn0/HCl. There-
fore, it seems that the occurrence of Sn after grafting with Bu2SnCl2
provides suitable active sites, whereas different, less active sites
are present on the other catalysts. Nevertheless, these sites show
diverse activity depending on the reaction type.
oxide were found in the X-ray diffraction pattern of the SnCl2
grafted material (Fig. S4), indicative of a preferable a-selective
deposition of SnAO species. No such reflections were found for
Bu4Sn-grafted materials, possibly due to low Sn-content (below
XRD bulk SnO2 detection limit of 0.5 wt%, the low content also
explains the low UV absorption intensity) or a more amorphous
nature of the (hydrous) oxides. Whereas for SnCl2, the secondary
species can be assigned to polymeric Sn oxide species, for Bu4Sn
their nature is unclear. The high energy SnAO species at about
200 nm are ascribed to tetrahedral Sn-species. Finally, the material
grafted with a metallic Sn0/HCl solution shows a broad absorption
band between 200 and 400 nm with a maximum around 272 nm,
whereas no Sn oxide related reflections were observed in X-ray
analysis.
Use of Bu2SnCl2 instead of SnCl4 was anticipated to generate
more ‘open’ Sn sites. Since it has been reported that BV and ISOM
reactions benefit from the presence of SnAOH (or SiAOH), one
could expect higher BV and ISOM activity of the Sn active sites of
grafted Bu2SnCl2. Therefore, the presence of ‘open’ sites was
evaluated by a reported FTIR study, using deuterated acetonitrile
[4]. Absorption bands at 2308 and 2316 cmÀ1 were assigned to
As in previous paragraphs the negative impact of Sn loading on
activity for a specific Sn precursor, viz. SnCl4Á5H2O, was demon-
strated, STY values should preferably be plotted against Sn loading
in order to judge the catalytic performance (Fig. 6). By filtering out
the Sn loading effect, ISOM catalysis best proceeds with the
SnCl4Á5H2O-grafted materials, irrespective of the Sn loading
(Fig. 7a). The catalyst is most productive with a Sn loading below
acetonitrile (mC„N) adsorbed on ‘closed’ and ‘open’ Sn sites,
respectively. The spectra of SnCl4Á5H2O and Bu2SnCl2 are shown
in Fig. 6(1). The signal at 2274 cmÀ1 is due to acetonitrile
interacting with silanol groups. Interestingly, since both materials
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
9
Fig. 6. (1) FTIR of deuterated acetonitrile desorbing of (a) SnCl4Á5H2O grafted, (b) Bu2SnCl2 grafted dealuminated beta zeolite. The black dotted line shows the main signal of
acetonitrile adsorbed onto Lewis acid sites. (2) FTIR of cyclohexanone desorbing of (a) SnCl4-grafted, (b) Bu2SnCl2-grafted dealuminated beta zeolite. The black dotted line
shows the main signal of cyclohexanone adsorbed onto Lewis acid sites. (3) FTIR spectrum of adsorbed pyridine on (a) SnCl4-grafted, (b) Bu2SnCl2-grafted, dealuminated beta
zeolite. Desorption of the probe molecule at 423 K (black), 523 K (red) and 623 K (blue) was investigated. Samples were evacuated before measurement and signals are scaled
to weight. Ratios of the surface area of the Lewis acid adsorbed pyridine vibration around 1451 cmÀ1 are given for 523 vs. 423 K. (For interpretation of the references to color
in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

10
J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
1 wt%, likely because of the requirement of very specific Sn isolated
tetrahedral sites with UV absorption features below 210 nm.
In case of MPV (Fig. 7b), STY shows independent behavior of the
precursor type, except for Sn0/HCl. Merely the Sn loading is the
determining factor for productivity, showing the most productive
material with a Sn loading of about 2 wt%. MPV therefore requires
preferably tetrahedral Sn sites with absorption features below
210 nm, which are less specific than those for ISOM. Clearly, the
Sn species formed by grafting in Sn0/HCl are undesired for MPV.
The low activity of this catalyst, containing many oligomeric SnAO
species and no isolated tetrahedral SnAO, emphasizes the impor-
tance of the isolated Sn species for catalytic MPV activity.
A similar trend is observed for BV (Fig. 7c). The relationship of
STY with Sn loading is similar for most Sn precursors, except for
the anhydrous SnCl4 and Sn0/HCl. The Sn loading for maximum
STY values though is clearly higher, corresponding to 4 wt%. Such
high value reveals that oligomeric Sn species also exhibit consider-
able BV activity, whereas their activity is very low for ISOM and
MPV. The underperformance of SnCl4 and Sn0/HCl demonstrates
specific electronic and geometric requirements of the oligomeric
active sites, though their nature is unclear at this stage.
as a result of Sn grafting [83], the presence of Al3+-sites [50] or oxy-
gen containing carbon groups [17], determines the rate, whereas
the Sn-catalyzed hydride-shift is not. Although unclear now, the
underperformance of the other catalysts should therefore be
caused by a deficit of Brønsted acidity on the final catalyst.
The difference between a SnCl4Á5H2O precursor and its anhy-
drous counterpart was investigated by carrying out a synthesis,
in which 5 molar equivalents of water were added to the anhy-
drous precursor. An increase in Sn-content but a lower catalytic
performance were found for this material (Entry 3 in Table 2 and
Fig. 7) in comparison with those made with the hydrated
SnCl4Á5H2O salt. This result suggests that (crystal) water plays an
important role in the grafting procedure (vide infra), which cannot
be simply replaced by free water molecules.
3.4. Effect of water concentration during the Sn grafting step
Finally, the influence of water on the grafting of Sn was investi-
gated. Assuming water may react with SnCl4, producing (hydrous)
oxide complexes of Sn and HCl already in the reflux isopropanol
solution, deposition of oligomeric EFW Sn oxide patches may be
expected when the water content is too high. In fact, the formation
of large Sn oxide phases were indeed observed by grafting with an
aqueous SnCl4 solution (vide supra, section on synthesis method
exploration). To understand the consequence of the water effect
on Sn catalysis, grafting in isopropanol was investigated in the
presence of varying amounts of water. Therefore, the dealuminated
zeolite powder was activated at 423 K under dry air to remove
Finally, the productivity of the catalysts was compared for
DH/MPV in Fig. 7d. This reaction type clearly prefers the catalyst
obtained from the SnCl4Á5H2O or Bu2SnCl2 precursor. The STY is
independent of the Sn concentration, indicating that the cascade
reaction probably needs a Sn in a reaction step that is not rate
determining. Literature describes that the dehydration, preferably
carried out with (mild) Brønsted acidity for instance originating
Fig. 7. STY (Â1000) in function of Sn loading for (a) ISOM, (b) MPV, (c) BV, and (d) DH/MPV reactions catalyzed by samples synthesized with changing Sn-precursor. The black
data points are from SnCl4Á5H2O, orange is anhydrous SnCl4, pink is mixture of anhydrous SnCl4 and 5 equivalents of water, red is Butyl2SnCl2, green is SnCl2, blue is Butyl4Sn
and yellow is metallic Sn. The black lines are guides to the eye and indicates the trend for SnCl4Á5H2O. (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
11
physisorbed water before being added to dry isopropanol. Known
aliquots of water were then added to the grafting solution contain-
ing a constant concentration of SnCl4Á5H2O. Elemental analysis of
obtained Snb samples is presented in Table 3.
The presence of some water results in a higher Sn adsorption.
Whereas a Sn loading of 1.6 wt% was found for water to Sn molar
ratio of 5, further increase this ratio leads to higher Sn loadings;
e.g., 3.8 and 11.3 wt% of Sn for a ratio of 6.5 and 31, respectively.
High water contents generate oligomeric (hydrous) oxide forms
of Sn as a result of Sn chloride salt hydrolysis, leading to nonselec-
tive adsorption in and on the zeolite crystals. Their formation is
clearly apparent in the XRD patterns, presented in Fig. 2(f), though
finely dispersed and invisible in SEM electrographs (Fig. S6).
Ultimately, at very high water contents or grafting in pure water,
separate Sn (hydrous) oxide phases are formed next to the zeolite
crystals, as indicated by micrographs and EPMA analysis (Fig. S1).
The samples were also investigated with UV–vis DR spec-
troscopy. Clearly, the low energy absorption band in the electronic
spectrum shifts to higher wavelengths for Snb samples prepared in
reflux solution with higher water content (Fig. 8). This shift indeed
points to the presence of larger Sn (hydrous) EFW oxide phases
with increasing water content in the reflux solution.
Fig. 8. UV-spectrum of samples synthesized with H2O:Sn-ratio of 5, 6.5, 12 and 31
(bottom to top at 280 nm). The secondary Sn-species maxima are labeled. Samples
were dried in N2 at 823 K and corrected for dealuminated beta.
A FTIR spectroscopic analysis of the silanol groups of a material
grafted in the presence of
a high concentration of water
(H2O:Sn = 31) shows only a slight drop in the concentration of
the internal silanols, whereas the external silanol groups remain
unchanged (data in Table S3). Though this trend is similar to sam-
ples grafted in the presence of low water concentration, the abso-
lute decrease of internal silanols is considerably less outspoken.
Nevertheless, the UV-absorption of the 200 nm signal, which was
assigned to Sn-species adsorbed in the internal silanol nests, of
the materials synthesized in the presence of water is similar to this
of materials made in water-limiting conditions, suggesting equal
amounts of such tetrahedral type A Sn-species. Possibly, the
presence of water generates additional silanol groups during
the grafting process, e.g., by siloxane hydration or additional
dealumination. Analysis of the micropore volume and the Lewis
acidity of materials prepared in high water to Sn molar ratios by
N2-sorption and pyridine probed FTIR respectively, suggests a
significant decrease in Sn accessibility. For instance, a significant
decrease in micropore volume, viz. from 0.16 to 0.05 ml gÀ1 for
materials grafted with H2O:Sn of 5 and 31 respectively was
measured, while the total Lewis acidity probed per Sn-atom
decreased from 0.7 to 0.2 (data in Table S3). These observations
point to a serious pore blockage by the present EFW SnO2 species.
119Sn MAS NMR of a sample prepared with a high H2O:Sn ratio
of 12 was measured. The spectrum (Fig. S7) only shows a broad
feature at À715 ppm, a diagnostic chemical shift of tetrahedrally
coordinated Sn, whereas a signal at À604 ppm of polymeric octa-
hedrally coordinated Sn, like in bulk SnO2, is absent, despite clear
presence of the EFW Sn species from UV absorption. A similar anal-
ysis with bulk SnO2 in beta zeolites shows a clear À604 ppm signal
after a short measuring time (Fig. S8). Therefore, the 280 nm Sn
oxide species is tentatively assigned to oligomeric tetrahedral Sn
sites. Unfortunately, as NMR spectroscopy of oxidic Sn species is
not as mature as NMR of other more abundant nuclei like Al and
Si, any assignment should be considered with caution. Indeed,
there are reports pointing to signal intensity reduction and broad-
ening of nano-sized oligomeric Sn oxides [84,85]. Improving 119Sn
MAS NMR resolution and better identification of chemical shifts is
currently under research using 119Sn enriched samples or by using
Dynamic Nuclear Polarization NMR techniques [51,86].
The different Snb catalysts were tested in the four reactions
(Table 3 and Fig. S9). As before, the highest TOFs for each reaction
type were found for the lowest Sn contents. In contrast to previous
catalytic tests with increasing Sn loadings (by increasing
SnCl4Á5H2O concentration in the reflux), the higher Sn loadings
do not lead to higher STYs. Obviously, the abundant presence of
oligomeric EFW Sn species blocks the pore entrance, as evidenced
by the micropore volume decrease, leading to a loss of catalyst effi-
ciency. This suggests that the Sn speciation is different, and thus
not only depends on the Sn content, but also on other grafting
parameters like the water content in the reflux solution. The best
catalysis both in terms of TOF and STY was achieved when Snb is
made by Sn adsorption in dry isopropanol with water to Sn molar
ratio of 5.
4. Conclusions
Contacting a Sn precursor with a dealuminated zeolite is a prac-
tical synthesis method to produce Sn containing Beta catalysts,
which are active in four different Lewis acid catalyzed reaction
types (ISOM, MPV, BV and DH/MPV). The occurrence of Sn is highly
Table 3
Comparison of the Sn-content and catalytic data of ISOM, MPV, BV and HS reaction for Sn-grafted catalyst synthesized in the presence of varying H2O:Sn-ratio.
Entry
H2O:Sn
Sn-content
ISOM
MPV
BV
DH/MPV
STY
TOF
hÀ1
STY
TOF
hÀ1
STY
TOF
hÀ1
STY
TOF
hÀ1
wt%
hÀ1 (Â1000)
hÀ1 (Â1000)
hÀ1 (Â1000)
hÀ1 (Â1000)
1
2
3
4
5
1.6
3.8
7.1
3640
2556
2098
1496
152
44
19
9
1277
1021
303
96
32
5
714
473
401
274
47
13
6
1621
1370
1187
935
103
36
17
8
6.5
12
31
11.3
173
2
3
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

12
J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
dependent on the Sn introduction method (solid mix,
impregnation, sorption) and on synthesis parameters like Sn
precursor, precursor and water concentration during the contact
process. Likely, though several methods may lead to active Snb cat-
alysts, they require different protocols to introduce the most active
Sn sites. This contribution elaborated on the sorption method in
liquid, that is grafting a Sn precursor into the silanol nests of the
dealuminated Beta zeolite. Two different sorption sites were
observed: one site-selective sorption in the internal silanol nests
(site A), taking place below 2 wt% Sn loading and creating isolated
Sn sites, and one a-selective sorption above 2 wt% of Sn, leading to
oligomeric EFW Sn species (site B), both sites showing low affinity
for the Sn precursor. To increase the affinity of the silanol site,
removal of water from the adsorption site and use of Sn precursor
with considerable chloride ligands are advised.
Efforts were carried out to clarify the identity of the two Sn sites
using various spectroscopic tools. The isolated Sn sites are
tetrahedral in coordination in line with its absorption feature
below 200 nm in the electronic spectrum. Higher Sn loadings showed
formation of oligomeric EFW species, usually with UV absorption
at about 255 nm, but also at higher wavelengths (280 nm) in case
of the presence of high concentrations of water in the grafting solu-
tion. Shoulders up to 300 nm were observed when the Sn loading is
above 10 wt%. The presence of the latter is often accompanied with
small X-ray reflections of SnO2 in the diffractogram, suggesting
they are octahedrally coordinated and polymerized SnAO like in
bulk SnO2, and a phase separation observable in EPMA/SEM graphs.
It is difficult to distinguish SnAO species absorbing at intermediate
UV wavelengths. Based on 119Sn MAS NMR, the 280 nm Sn
species and therefore the 255 nm is also tentatively considered
in tetrahedral coordination. Though its nuclearity is assumed to
be oligomeric, the 255 nm feature may also result from tetrahedral
species with one or multiple extended SnAO bonds.
Catalytic performance may be expressed as turn-over fre-
quency and space-time-yield. While the first measures the activ-
ity per Sn active site, assuming every Sn is accessible, the second
one refers to the activity per catalyst weight. Whereas general
agreement exists that isolated Sn sites have to be correlated to
the most active Lewis acid sites, the other Sn species also con-
tribute to Lewis acid catalyzed reactions, depending on the reac-
tion type. Increase of Sn loading may therefore improve the
productivity of the catalysis, albeit not linearly correlated to
the Sn loading. Since site requirements are most specific for the
aldo/keto isomerization of sugars and Meerwein–Ponndorf–Ver
ley, an increase of Sn loading by grafting is not very helpful to
increase product formation as at high Sn-loadings, formation of
inactive EFW species occurs, which additionally cause pore block-
age. High productivity will therefore only be achieved by search-
ing for new synthetic methods, accounting for high densities of
framework tetrahedral Sn active sites. However, oligomeric EFW
species show considerable catalytic contribution in the BV reac-
tion. This reaction type therefore benefits from increasing Sn
loadings, despite a constant lowering of activity per Sn site. Too
high Sn loadings by grafting though should be avoided because
of pore blocking, resulting into decreasing STYs. Conclusions on
the cascade reaction (DH/MPV) should be taken with caution, as
this test reaction preferably requires Brønsted acid sites in the
rate determining dehydration step.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2015.06.023.
References
[1] Y. Román-Leshkov, M.E. Davis, ACS Catal. 1 (2011) 1566.
[2] M. Moliner, Dalton Trans. 43 (2014) 4197.
[3] M. Boronat, A. Corma, M. Renz, J. Phys. Chem. B 110 (2006) 21168.
[4] M. Boronat, P. Concepcion, A. Corma, M. Renz, S. Valencia, J. Catal. 234 (2005)
111.
[5] R. Bermejo-Deval, R. Gounder, M.E. Davis, ACS Catal. 2 (2012) 2705.
[6] M. Dusselier, P. Van|Wouwe, F. de|Clippel, J. Dijkmans, D.W. Gammon, B.F. Sels,
ChemCatChem 5 (2013) 569.
[7] Y. Hayashi, Y. Sasaki, Chem. Commun. (2005) 2716.
[8] M. Dusselier, P. Van Wouwe, S. De Smet, R. De Clercq, L. Verbelen, P. Van
Puyvelde, F.E. Du Prez, B.F. Sels, ACS Catal. 3 (2013) 1786.
[9] S. Yamaguchi, K. Motokura, Y. Sakamoto, A. Miyaji, T. Baba, Chem. Commun. 50
(2014) 4600.
[10] S. Yamaguchi, T. Matsuo, K. Motokura, Y. Sakamoto, A. Miyaji, T. Baba,
ChemSusChem 8 (2015) 853.
[11] G. Tian, X. Tong, Y. Cheng, S. Xue, Carbohydr. Res. 370 (2013) 33.
[12] L. Zhou, L. Wu, H. Li, X. Yang, Y. Su, T. Lu, J. Xu, J. Mol. Catal. A: Chem. 388–389
(2014) 74.
[13] J. Venselaar, Gas-Phase Aldol Condensation over Tin on Silica Catalysts, Delft
University, Delft, 1980. pp. 140.
[14] L. Li, X. Collard, A. Bertrand, B.F. Sels, P.P. Pescarmona, C. Aprile, J. Catal. 314
(2014) 56.
[15] M. Saito, H. Ikeda, Y. Horiuchi, M. Matsuoka, Res. Chem. Intermed. 40 (2014)
87.
[16] A.L. Villa, L.F. Correa, E.A. Alarcón, Chem. Eng. J. 215–216 (2013) 500.
[17] F. de Clippel, M. Dusselier, R. Van Rompaey, P. Vanelderen, J. Dijkmans, E.
Makshina, L. Giebeler, S. Oswald, G.V. Baron, J.F.M. Denayer, P.P. Pescarmona,
P.A. Jacobs, B.F. Sels, J. Am. Chem. Soc. 134 (2012) 10089.
[18] C.M. Osmundsen, M.S. Holm, S. Dahl, E. Taarning, Proc. Roy. Soc. A 468 (2012)
2000.
[19] S. Sisodiya, S. Shylesh, A.P. Singh, Catal. Commun. 12 (2011) 629.
[20] A.C. Garade, P.S. Niphadkar, P.N. Joshi, C.V. Rode, Chem. Lett. 39 (2010) 126.
[21] U.S. Taralkar, P. Kalita, R. Kumar, P.N. Joshi, Appl. Catal. A 358 (2009) 88.
[22] S. Endud, K.-L. Wong, Microporous Mesoporous Mater. 101 (2007) 256.
[23] A. Corma, M. Renz, Collect. Czech. Chem. Commun. 70 (2005) 1727.
[24] A.L. Villa de P, E. Alarcon, C. Montes de Correa, Chem. Commun. (2002) 2654.
[25] A. Corma, L.T. Nemeth, M. Renz, S. Valencia, Nature 412 (2001) 423.
[26] N.K. Mal, A. Bhaumik, R. Kumar, A.V. Ramaswamy, Catal. Lett. 33 (1995) 387.
[27] N.K. Mal, A. Bhaumik, V. Ramaswamy, A.A. Belhekar, A.V. Ramaswamy, in: H.K.
Beyer, H.G. Karge, I. Kiricsi, J.B. Nagy (Eds.), Stud. Surf. Sci. Catal, Elsevier,
Amsterdam, 1995, p. 317.
[28] N.K. Mal, V. Ramaswamy, S. Ganapathy, A.V. Ramaswamy, J. Chem. Soc. Chem.
Commun. (1994) 1933.
[29] H.Y. Luo, L. Bui, W.R. Gunther, E. Min, Y. Román-Leshkov, ACS Catal. 2 (2012)
2695.
[30] E. Taarning, S. Saravanamurugan, M.S. Holm, J. Xiong, R.M. West, C.H.
Christensen, ChemSusChem 2 (2009) 625.
[31] S. Tolborg, A. Katerinopoulou, D.D. Falcone, I. Sadaba, C.M. Osmundsen, R.J.
Davis, E. Taarning, P. Fristrup, M.S. Holm, J. Mater. Chem. A 2 (2014) 20252.
[32] S.V. Valencia, A.C. Corma, US005968473A, 1999, to UOP.
[33] A. Corma, M.E. Domine, S. Valencia, J. Catal. 215 (2003) 294.
[34] S. Van de Vyver, C. Odermatt, K. Romero, T. Prasomsri, Y. Roman-Leshkov, ACS
Catal. (2015).
[35] M. Boronat, A. Corma, M. Renz, G. Sastre, P.M. Viruela, Chem. A Euro. J. 11
(2005) 6905.
[36] G. Li, E.A. Pidko, E.J.M. Hensen, Catal. Sci. Technol. (2014).
[37] G. Yang, E.A. Pidko, E.J.M. Hensen, ChemSusChem (2013).
[38] R. van Grieken, C. Martos, M. Sánchez-Sánchez, D.P. Serrano, J.A. Melero, J.
Iglesias, A.G. Cubero, Microporous Mesoporous Mater. 119 (2009) 176.
[39] C.-C. Chang, Z. Wang, P. Dornath, H. Je Cho, W. Fan, RSC Adv. (2012).
[40] Z. Kang, X. Zhang, H. Liu, J. Qiu, K.L. Yeung, Chem. Eng. J. 218 (2013) 425.
[41] A. Omegna, M. Vasic, J. Anton van Bokhoven, G. Pirngruber, R. Prins, Phys.
Chem. Chem. Phys. 6 (2004) 447.
[42] J.A. van Bokhoven, D.C. Koningsberger, P. Kunkeler, H. van Bekkum, A.P.M.
Kentgens, J. Am. Chem. Soc. 122 (2000) 12842.
[43] Q. Guo, F. Fan, E.A. Pidko, W.N.P. van der Graaff, Z. Feng, C. Li, E.J.M. Hensen,
ChemSusChem 6 (2013) 1352.
[44] P.Y. Dapsens, C. Mondelli, J. Jagielski, R. Hauert, J. Perez-Ramirez, Catal. Sci.
Technol. 4 (2014) 2302.
[45] P. Li, G. Liu, H. Wu, Y. Liu, J.-G. Jiang, P. Wu, J. Phys. Chem. C 115 (2011) 3663.
[46] W.N.P. van der Graaff, G. Li, B. Mezari, E.A. Pidko, E.J.M. Hensen, ChemCatChem
(2015). n/a.
[47] C. Hammond, S. Conrad, I. Hermans, Angew. Chem. Int. Edit. 51 (2012) 11736.
[48] B. Tang, W. Dai, G. Wu, N. Guan, L. Li, M. Hunger, ACS Catal. 4 (2014) 2801.
[49] J. Dijkmans, D. Gabriels, M. Dusselier, F. de Clippel, P. Vanelderen, K.
Houthoofd, A. Malfliet, Y. Pontikes, B.F. Sels, Green Chem. 15 (2013) 2777.
Acknowledgments
Prof. Bart Blanpain is gratefully thanked for the EPMA measure-
ments Hercules Foundation (project ZW09-09) is acknowledged for
the funding of the JEOL JXA-8530F EPMA. IOF-KP/14/003 is thanked
for financial support of the authors.
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023

J. Dijkmans et al. / Journal of Catalysis xxx (2015) xxx–xxx
13
[50] J. Dijkmans, D. Gabriëls, M. Dusselier, K. Houthoofd, P.C.M.M. Magusin, S.
Huang, Y. Pontikes, M. Trekels, A. Vantomme, L. Giebeler, S. Oswald, B.F. Sels,
ACS Catal. 5 (2015) 928.
[51] P. Wolf, M. Valla, A.J. Rossini, A. Comas-Vives, F. Núñez-Zarur, B. Malaman, A.
Lesage, L. Emsley, C. Copéret, I. Hermans, Angew. Chem. Int. Ed. 53 (2014)
10179.
[66] T. Witvrouwen, J. Dijkmans, S. Paulussen, B. Sels, J. Energy Chem. 22 (2013) 451.
[67] Z. Chen, J.K.L. Lai, C.H. Shek, H. Chen, J. Mater. Res. 18 (2003).
[68] O. Lupan, L. Chow, G. Chai, H. Heinrich, S. Park, A. Schulte, J. Cryst. Growth 311
(2008) 152.
[69] G. Pang, S. Chen, Y. Koltypin, A. Zaban, S. Feng, A. Gedanken, Nano Lett. 1
(2001) 723.
[52] M. Moliner, Y. Román-Leshkov, M.E. Davis, Proc. Natl. Acad. Sci. U. S. A. 107
(2010) 6164.
[53] M.T. Musser, Ullmann’s Encyclopedia of Industrial Chemistry, Wiley-VCH
Verlag GmbH & Co. KGaA, 2000.
[70] H.S. Taylor, Proc. Roy. Soc. A 108 (1925) 355.
[71] B. Peters, S.L. Scott, J. Chem. Phys. 142 (2015) 104708.
[72] R.E. Parales, K. Lee, S.M. Resnick, H. Jiang, D.J. Lessner, D.T. Gibson, J. Bacteriol.
182 (2000) 1641.
[54] J. Ritz, H. Fuchs, H. Kieczka, W.C. Moran, Ullmann’s Encyclopedia of Industrial
Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA, 2000.
[55] M. Dusselier, P. Van Wouwe, A. Dewaele, P.A. Jacobs, B.F. Sels, Science 349
(2015) 78.
[56] G. Ertl, H. Knözinger, F. Schüth, J. Weitkamp, Handbook of Heterogeneous
Catalysis: Preparation of Solid Catalysts, Wiley-VCH, 2008.
[57] Z. Kang, H.o. Liu, X. Zhang, Chin. J. Catal. 33 (2012) 898.
[58] M. Casagrande, E. Moretti, L. Storaro, M. Lenarda, J. Gersich, L. Stievano, F.E.
Wagner, Microporous Mesoporous Mater. 91 (2006) 261.
[59] L. Haiou, K. Zihua, Z. Xiongfu, Q. Jie-shan, CN102249258 A, 2011, to D.U.o.
Technology.
[73] A. Fürstner, L. Ackermann, B. Gabor, R. Goddard, C.W. Lehmann, R. Mynott, F.
Stelzer, O.R. Thiel, Chem. Euro. J. 7 (2001) 3236.
[74] C.A. Emeis, J. Catal. 141 (1993) 347.
[75] A. Vimont, F. Thibault-Starzyk, J.C. Lavalley, J. Phys. Chem. B 104 (1999) 286.
[76] S. Khabtou, T. Chevreau, J.C. Lavalley, Microporous Mater. 3 (1994) 133.
[77] A. Corma, M.T. Navarro, M. Renz, J. Catal. 219 (2003) 242.
[78] J. Fernandez, G. Lespes, A. Dargelos, Chem. Phys. 111 (1987) 97.
[79] M. Rothschild, Laser Microfabrication: Thin Film Processes and Lithography,
Elsevier, 1989.
[80] S.R. Bare, S.D. Kelly, W. Sinkler, J.J. Low, F.S. Modica, S. Valencia, A. Corma, L.T.
Nemeth, J. Am. Chem. Soc. 127 (2005) 12924.
[60] N.S. Nesterenko, F. Thibault-Starzyk, V. Montouilliout, V.V. Yushchenko, C.
Fernandez, J.P. Gilson, F. Fajula, I.I. Ivanova, Kinet. Catal. 47 (2006) 40.
[61] K. Chaudhari, T.K. Das, P.R. Rajmohanan, K. Lazar, S. Sivasanker, A.J.
Chandwadkar, J. Catal. 183 (1999) 281.
[62] M.D. González, Y. Cesteros, P. Salagre, Microporous Mesoporous Mater. 144
(2011) 162.
[81] J. Van Aelst, M. Haouas, E. Gobechiya, K. Houthoofd, A. Philippaerts, S.P. Sree,
C.E.A. Kirschhock, P. Jacobs, J.A. Martens, B.F. Sels, F. Taulelle, J. Phys. Chem. C
118 (2014) 22573.
[82] E. Verheyen, L. Joos, C. Martineau, C.J. Dawson, C. Weidenthaler, W. Schmidt, R.
Yuan, E. Breynaert, V. Van Speybroeck, M. Waroquier, F. Taulelle, M.M.J. Treacy,
J.A. Martens, C.E.A. Kirschhock, Mater. Horizons 1 (2014) 582.
[83] L. Li, C. Stroobants, K. Lin, P.A. Jacobs, B.F. Sels, P.P. Pescarmona, Green Chem.
13 (2011) 1175.
[63] E.B. Lami, F. Fajula, D. Anglerot, T. Des Courieres, Microporous Mater. 1 (1993)
237.
[64] J.P. Marques, I. Gener, P. Ayrault, J.C. Bordado, J.M. Lopes, F. Ramôa Ribeiro, M.
Guisnet, Microporous Mesoporous Mater. 60 (2003) 251.
[65] D.M. Roberge, H. Hausmann, W.F. Holderich, Phys. Chem. Chem. Phys. 4 (2002)
3128.
[84] T. Dey, P. Khuntia, A.V. Mahajan, Europhys. Lett. 96 (2011) 670081.
[85] D.P. Tunstall, S. Patou, R.S. Liu, Y.H. Kao, Mater. Res. Bull. 34 (1999) 1513.
[86] W.R. Gunther, V.K. Michaelis, M.A. Caporini, R.G. Griffin, Y. Román-Leshkov, J.
Am. Chem. Soc. 136 (2014) 6219.
Please cite this article in press as: J. Dijkmans et al., Post-synthesis Snb: An exploration of synthesis parameters and catalysis, J. Catal. (2015), http://
dx.doi.org/10.1016/j.jcat.2015.06.023