Hierarchically porous BEA stannosilicates as unique catalysts for bulky ketone conversion and continuous operation

Article abstract of DOI:10.1039/c5ta08709k

Pore size limitations typically limit the applicability of Lewis acidic zeolites, such as titano- and stanno-silicates, to catalytic processes based on small-to-mid sized substrates, and increase their rates of deactivation, prohibiting further exploitation. Herein, we demonstrate that tin-containing zeolites possessing modified hierarchical BEA matrices can be prepared. These hierarchical stannosilicates are able to mediate the catalytic conversion of bulky ketone substrates, a pertaining challenge in the field that purely microporous analogues are unable to mediate. Deactivation studies in the continuous regime also demonstrate the exceptional stability of hierarchical Sn-Beta compared to purely microporous Sn-Beta, with <20% loss of activity observed over 700 h on stream. In contrast, the purely microporous analogue lost ±70% activity in only 200 h. To the best of our knowledge, this is the first time a stannosilicate with a beneficial hierarchical BEA framework has been prepared, and the first evidence of cyclododecanone valorisation with stannosilicate catalysts.

Full text of DOI:10.1039/c5ta08709k

Journal of
Materials Chemistry A
View Article Online
View Journal
PAPER
Hierarchically porous BEA stannosilicates as unique
catalysts for bulky ketone conversion and
continuous operation†
Cite this: DOI: 10.1039/c5ta08709k
Abbas Al-Nayili, Keiko Yakabi and Ceri Hammond*
Pore size limitations typically limit the applicability of Lewis acidic zeolites, such as titano- and stanno-
silicates, to catalytic processes based on small-to-mid sized substrates, and increase their rates of
deactivation, prohibiting further exploitation. Herein, we demonstrate that tin-containing zeolites
possessing modified hierarchical BEA matrices can be prepared. These hierarchical stannosilicates are
able to mediate the catalytic conversion of bulky ketone substrates, a pertaining challenge in the field
that purely microporous analogues are unable to mediate. Deactivation studies in the continuous regime
also demonstrate the exceptional stability of hierarchical Sn-Beta compared to purely microporous Sn-
Beta, with <20% loss of activity observed over 700 h on stream. In contrast, the purely microporous
analogue lost ꢀ70% activity in only 200 h. To the best of our knowledge, this is the first time
a stannosilicate with a beneficial hierarchical BEA framework has been prepared, and the first evidence of
cyclododecanone valorisation with stannosilicate catalysts.
Received 28th October 2015
Accepted 17th December 2015
DOI: 10.1039/c5ta08709k
www.rsc.org/MaterialsA
Possessing a microporous structure, Sn-b is, however, very
susceptible to the limitations associated with this particular
Introduction
Lewis acid catalysts have been widely used for a range of degree of connement, such as (i) internal mass transfer limi-
essential catalytic transformations, including biomass valor- tations hindering catalytic performance through slow molecular
isation. Of particular interest are crystalline, porous inorganic diffusion, (ii) restricted diffusion of bulkier reactants, which
materials such as zeolites, which contain encapsulated Lewis limits their scope of reactivity, and (iii) increased rates of deac-
1–4
acidic centres, such as Al, Sn, Ti and Zr. These materials tivation through pore blocking (cf. fouling). These limitations are,
combine the advantages of molecular catalysts, such as site unfortunately, common to almost all microporous zeolites.
isolation and high intrinsic activity, along with the practical Indeed, some of us recently demonstrated that despite exhibiting
advantages of heterogeneous catalysts, which simplify down- high levels of stability during steady state (continuous) operation,
stream processing. Amongst these materials, Sn-b, a medium microporous Sn-b slowly deactivates with extended time on
pore zeolite possessing isolated Sn(IV) sites and BEA topology, stream through fouling of the micropores with reaction products
15
has garnered tremendous levels of academic and industrial and higher molecular weight carbonaceous residue.
interest. Over the last decade, it has been demonstrated that
Over the years, several approaches have been followed in
this material possesses an exceptional ability to catalyse various order to alleviate these limitations, such as (i) the synthesis of
1
7–19
transformations involving carbonyl compounds, such as the extra-large pore zeolites,
(ii) the development of ordered
5
–7
20
Baeyer–Villiger oxidation (BVO) of ketones, the Meerwein– mesoporous materials (e.g. MCM-41) and related compos-
Pondorf–Verley (MPV) transfer hydrogenation of ketones to ites, (iii) the preparation of nano-sized zeolite particles, and
21
22
8
–11
23–28
alcohols,
glucose,
and the isomerization of renewable sugars such as (iv) the development of 2D materials.
Indeed, several of
1
2–15
amongst several others.
these approaches have been attempted for a range of zeotype
In addition to high levels of activity, microporous crystalline materials, including silicates, aluminosilicate, titanosilicate,
zeolites, such as Sn-b, also provide excellent levels of hydro- and more recently, stannosilicate. Nevertheless, each of these
thermal stability, high surface area, and molecular sieving alternative materials experience particular disadvantages when
capability due to the uniform nature of their micropores. it comes to catalytic activity, and several hurdles and challenges
remain to be tackled. Decreased levels of specic active site
activity (e.g. turnover frequency), poorer levels of hydrothermal
Cardiff Catalysis Institute, Cardiff University, Main Building, Park Place, CF10 3AT,
stability and increased production costs (through the use of
UK. E-mail: hammondc4@cardiff.ac.uk; Web: http://blogs.cardiff.ac.uk/hammond;
costly surfactant templates) are just three of the known disad-
Tel: +44 (0)29 2087 4082
vantages of each of the alternative materials that have been
prepared to date.
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c5ta08709k
1
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
Hierarchical zeolites are those that possess levels of meso-
Motivated by this research challenge, we demonstrate in this
porosity imprinted upon the conventional microporous struc- work that hierarchically-modied zeolite beta, doped with 2
ture. In principle, these materials retain the advantages of wt% Sn through solid-state stannation, is an effective catalyst
purely microporous materials, such as increased intrinsic for the catalytic conversion of bulky ketones, such as cyclo-
reactivity and (hydro)thermal stability, whilst also improving dodecanone. These ketones are far too large to be catalytically
the particular disadvantages associated with overly conned converted in the micropores of even medium pore zeolites,
materials, such as retarded diffusion and increased rates of and thus offer an ideal challenge for their hierarchical matrices.
29–31
deactivation.
Consequently, these materials have experi- Kinetic and spectroscopic methodologies reveal that the active
enced an explosion of interest in recent years. Most of this sites present in the hierarchical matrix are comparable to
attention has focused upon the generation of hierarchical those found in the conventional microporous Sn-b despite
aluminosilicate materials, unsurprising given their widespread the modied porosity, and that the hierarchical material
32,33
use throughout the petrochemical industry.
Desilication, demonstrates signicantly higher levels of activity than ordered
a top down approach involving the treatment of a parent zeolite mesoporous analogues such as Sn-MCM-41. Furthermore,
in an alkaline medium under mild conditions, has been shown deactivation studies in continuous mode also demonstrate the
34–36
to be a particular efficient method of generating hierarchy.
supreme advantage of the hierarchical analogue when it comes
The role of the base is to remove silicon atoms from the to extended steady state operation, with high levels of activity
framework of the material (Fig. 1), thus resulting in partial observed even aer 700 h on stream. To the best of our
destruction of the micropores, and leading to the appearance of knowledge, this is the rst time a stannosilicate with a hierar-
mesoporosity. Under optimised conditions, this can lead to the chical BEA framework has been prepared, and the rst evidence
generation of a hierarchical material without causing excessive of cyclododecanone valorisation with stannosilicate catalysts.
destruction, i.e. collapse, of the zeolite framework.
Given their emerging status as sustainable heterogeneous
catalysts, it is unsurprising that attention is being turned to the
Results and discussion
Generation of hierarchical zeolite beta
generation of stannosilicates with improved diffusional prop-
37–39
erties.
To date, however, focus has primarily been on the Our typical solid-state stannation methodology involves the
synthesis of stannosilicates with a hierarchical MFI-type remetallation of dealuminated (deAl-) b with Sn(II) acetate
40
topology, and very little attention has been devoted to the (Fig. 1). Accordingly, we rst focused upon the desilication
synthesis of other hierarchical stannosilicates. Furthermore, in (deSi) of a pre-dealuminated (deAl) sample of H-b (denoted
all of these previous cases, catalytic studies have only focused [deAl,deSi]-b), which was previous prepared by dealumination
upon the conversion of relatively small substrates that do not in HNO . We note here that for materials with multiple pre-
3
present a signicant hindrance for conventional medium pore treatment steps, the nomenclature used follows the order of
zeolites, such as zeolite beta. Examples include the isomer- treatment i.e. [deAl,deSi]-b indicates dealumination, followed
isation of dihydroxyacetone (C
3
) and the catalytic conversion of by desilication. The subsequent desilication protocol employed
glucose (C ). As such, full appreciation of the potential advan- was based on a modied protocol published by the group of
6
34
tages of hierarchical stannosilicates has yet to be realised.
Perez-Ramirez. The optimised protocol involves treatment
of the dealuminated solid in an aqueous solution of NaOH
ꢁ
(
0.2 M) for 0.5 h at 45 C. Although only minor changes to the
textural properties of H-b were observed aer the rst step i.e.
dealumination (Table 1, Entry 2), desilication of the already
dealuminated material ([deAl,deSi]-b) resulted in complete
structural collapse, as evidenced by the porosimetry data (Table
1, Entry 3) and XRD analysis (Fig. 2c). Clearly, the presence of
framework Al in the zeolite is essential for desilication to
proceed without complete destruction of the framework, in line
41
with previous research.
We subsequently reversed our preparation procedure, and

rst desilicated the parent zeolite directly (deSi-b, Table 1 Entry
4). The desilication procedure employed leads to several
changes in the properties of the parent zeolite. Substantial
increases in BET surface area (SBET), external surface area (Sext),
and total pore volume (Vtotal) are accompanied by decreased
micropore surface area (Smicro), and micropore volume (Vmicro).
More crucially, however, is the retained crystalline structure
(Fig. 2) and the substantial increase in mesopore volume (Vmeso)
following the desilication procedure. Having generated a clearly
hierarchical structure, we subsequently dealuminated the
material in line with our previous work, in order to generate the
Fig. 1 Postsynthetic desilication procedure for mesopores creation in
the zeolite b using NaOH as desilication agent (step 1), and subsequent
formation of stannosilicate via solid state stannation (step 2 and 3).
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
Journal of Materials Chemistry A
Table 1 Textural properties of sample used in this work
a
2
ꢂ1
b
2
ꢂ1
b
2
ꢂ1
c
3
ꢂ1
b
3
ꢂ1
d
3
ꢂ1
Sample
S
BET (m g
)
S
external (m g
)
S
micro (m g
)
V
total (cm g
)
V
micro (cm g
)
V
meso (cm g )
Parent H-b
deAl-b
498
541
243
805
691
614
528
832
61
437
417
47
323
233
259
412
0
0.35
0.53
0.40
0.95
0.84
0.67
0.42
0.85
0.23
0.23
0.0
0.19
0.15
0.15
0.23
0
0.11
0.22
0.4
0.76
0.61
0.52
0.20
0.85
124
197
483
458
356
115
832
[
deAl,deSi]-b
deSi-b
deSi,deAl]-b (h*b)
[
2
2
2
Sn-h*b
Sn-b
Sn-MCM-41
a
b
Brunauer–Emmett–Teller surface area (SBET) was calculated from BET method. External and micropore surface area (Sext and Smicro) and
c
d
micropore volume (Vmicro) derived from the t-plot method. Total pore volume (Vtotal) was evaluated at P/P
was calculated according to V
0
¼ 0.99. Meso pore volume (Vmeso
)
T
ꢂ Vmicro
.
display Type IV character, typically experience adsorption at
intermediate relative values. Although missing the low relative
pressure region (vide supra), the N adsorption isotherm ob-
2
tained for Al-b is clearly reminiscent of a Type I isotherm, and
Type IV character is clearly gained following dealumination
(deAl-b, open triangles), or more extensively by the [deSi,deAl]
treatment (h*b, open circles). This clearly demonstrates the
formation of mesopores, which is more readily visualised from
the BJH data. In line with its microporous structure, no meso-
pores are evident in Al-b. For deAl-b and h*b, the presence of
new mesopores with pore diameters of approximately 7–8 nm
are evident. These values are in excellent agreement to previous
30
work with hierarchical zeolites. The more extensive adsorption
observed for h*b indicates that the extent of mesopores
formation increases substantially with the added disilication
step. It can also be observed that the dealumination step alone
is able to introduce some minor mesoporosity, although to
a substantially lower degree than desilication (Fig. 3).
Fig. 2 XRD patterns for (a) commercial zeolite b, (b) deAl-b, (c)
deAl,deSi]-b, (d) [deSi,deAl]-b (h*b) and (e) 2Sn-h*b.
[
vacant framework sites required for the incorporation of tin
Scanning Electron Microscopy analysis of the nal catalysts
i.e. hierarchical and microporous Sn-b, demonstrated that
neither the deAl protocol (in the case of the microporous Sn-b)
nor the [deSi,deAl] protocol (as for the hierarchical material)
overly modied the zeolite crystals (Fig. 4a–c). No signicant
(
[deSi,deAl]-b, Table 1 Entry 5).
Despite the dealumination protocol being able to induce
some mesopore formation and increasing SBET and Sext on its
own (Table 1, Entry 2), dealumination of the desilicated struc-
ture ([deSi,deAl]-b, henceforth h*b) leads to minor decreases in
each of these parameters (Table 1, Entry 5) compared to desi-
lication alone (Table 1, Entry 4). The complete absence of Al
from the dealuminated material conrms that extra-framework
Al species are not present and are therefore not partially
blocking the pores (Table S1†). Accordingly, we attribute this
unexpected result to a slight relaxation of the framework upon
removal of the isomorphously substituted framework atoms.
Physical adsorption in micropores, such as those found in
the zeolites in this study, occurs at relative pressures substan-
tially lower than in case of adsorption in mesopores. We note,
2
therefore, that adsorption measurements with N at 77.4 K do
not directly provide information on the microporosity of the
samples. However, it remains an extremely useful methodology
for gaining insight into the mesoporosity of the materials,
which is clearly a key parameter for hierarchical materials. Type
I isotherms, as observed for purely microporous zeolites, exhibit
a large plateau region at relative pressures above values of
approximately 0.1, whereas mesoporous materials, which
Fig. 3 Porosimetry data for (black/squares) commercial zeolite b,
(hollow/triangles) deAl-b, and (hollow/circles) [deSi,deAl]-b (h*b).
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
changes in terms of crystal morphology or size are observed. sites in both 2Sn-b and 2Sn-h*b. CD
3
CN is an extremely useful
Spherical crystals between 0.2 to 0.9 mm were maintained aer probe molecule for Lewis acidic zeolites since it can interact
4
2,43
the post-synthetic treatments. Although the samples experience with the framework heteroatoms,
resulting in an acid–base
ꢂ1
some beam damage, preliminary TEM studies conrmed the adduct that displays an intense absorbance at 2311 cm . This
absence of any major mesopores in deAl-b, and conrms the particular feature is not observed for extra-framework Sn
presence of mesopores (5–10 nm) in the h*b material. The species nor SnO
2
, and its appearance can thus conrm the
(vide infra) domains is also evident presence of framework Sn(IV) atoms. As can be seen (Fig. 5A),
upon dosing 2Sn-h*b with CD CN, two intense features are
observed in the DRIFTS spectrum.
absence of major bulk SnO
from this analysis.
2
3
44,45
ꢂ1
The rst (2275 cm
)
arises from physisorbed CD CN, which is readily desorbed from
Preparation and characterisation of 2Sn-h*b, following solid
state stannation
3
the sample upon heat treatment (see desorption prole,
ꢁ
ꢁ
temperature increasing from top (50 C) to bottom (200 C)).
Subsequently, we focused upon the solid-state incorporation of
ꢂ1
The second feature (2311 cm ), arises from the CD
3
CN mole-
Sn(IV) into the vacant tetrahedral framework sites of h*b,
cules interacting with framework Sn(IV) atoms, and clearly
demonstrates the presence of isomorphously substituted Sn
atoms in our material, in excellent agreement to our prior work.
The comparative spectra of purely microporous 2Sn-b are also
displayed in Fig. 5B. The presence of Lewis acidic Sn hetero-
atoms is also evident in this material. Similar observations are
11
according to our recently published and patented solid-state
stannation methodology. The preparation typically involves the
dealumination of zeolite beta with HNO , prior to remetallation
3
with tin. In this case, our typical solid-state stannation proce-
dure was preceded by the desilication step (Fig. 1). In contrast to
approaches whereby pre-synthesised stannosilicates are con-
verted into hierarchical matrices, incorporating Sn post-
synthetically into a pre-modied hierarchical structure ensures
the presence of accessible Sn(IV) sites in the mesopores, in line
found in the DRIFT spectra of CD CN-dosed 2Sn-MCM-41
3
(Fig. S3†).
37
with the observations of Dapsens et al.
Tin(IV) was introduced via a tin(II)acetate precursor. In line
with our recent work, we focused upon the preparation of
materials with a high degree of site isolation, and conse-
quently incorporated 2 wt% of Sn into the material, as this is
a loading we have shown to lead to the highest levels of
intrinsic activity i.e. TOF per Sn atom. The materials are
henceforth denoted as 2Sn-b (microporous) and 2Sn-h*b
(
hierarchical). A multitude of spectroscopic techniques (XAS,
MAS NMR, DRIFTS) were subsequently employed to verify the
speciation of the incorporated Sn(IV) atoms, and to deduce
whether Sn(IV) had incorporated into the vacant framework
sites of h*b. A control sample of microporous 2Sn-b was also
prepared, whereby 2 wt% Sn was incorporated into a deal-
uminated (but not desilicated) material. We have recently
benchmarked the activity, selectivity and Sn site speciation of
this control sample in depth.
Fig. 5 In situ DRIFTS spectra of (A) 2Sn-h*b, and (B) 2Sn-b, following
dashed) 10 minute treatment with CD CN at room temperature.
(
3
We rst employed in situ DRIFT spectroscopy (with CD CN
3
Various desorption temperatures (50 (red), 100 (blue), 150 (purple) and
ꢁ
probe) to verify the presence of Lewis acidic, framework Sn(IV) 200 C (green)) are also displayed.
Fig. 4 SEM images of (A) commercial zeolite b, (B) 2Sn-b, and (C) 2Sn-h*b.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
Journal of Materials Chemistry A
Although there are some minor differences in absorbance for the hierarchical material is in excellent agreement to that
intensity and desorption rate for both of these samples, DRIFT obtained for our microporous Sn-b at similar metal loadings,
spectroscopy is not sufficiently quantitative to differentiate and indicates that Sn(IV) is primarily present in the zeolite
between the total number of Lewis acidic (framework) and framework.
spectator sites (extra-framework) in each of these materials.
To verify the presence of extra-framework Sn species in these
Accordingly, we turned our attention to spectroscopic tech- samples, we also probed the Sn site speciation with X-ray
niques that provide more detailed insight of the Sn site speci- Absorption Spectroscopy. Various elements of the EXAFS
ation, such as X-ray Absorption Spectroscopy (XAS) and Magic spectra of 2Sn-h*b and the conventional microporous analogue
Angle Spinning (MAS) NMR.
are shown in Fig. 7A and B. The spectra are of excellent quality
ꢂ1
˚
Despite the low S/N ratio (materials were prepared with up to a distance of 14 A . The rst feature present arises from
1
18
naturally abundant
Sn), the ambient temperature and pres- Sn–O scattering interactions, whilst the second feature (ca. 2.5–
˚
sure MAS NMR spectra of 2Sn-b and 2Sn-h*b primarily consists 4 A) arises from Sn–Sn scattering interactions, which are
of an intense resonance at ꢂ688 ppm, which is indicative of indicative of (i) extra-framework Sn atoms that are oligomeric or
44
framework Sn(IV) species in their hydrated state, in excellent oxidic in nature, and/or (ii) Sn–Sn pairing. It is clear from this
agreement with our recent work (Fig. 6). The absence of analysis neither material possesses a substantial quantity of
a substantial resonance at ꢂ602 ppm in the spectrum of 2Sn- extra-framework Sn, given the relatively minor scattering
˚
b indicates that extra-framework Sn species are a very minor intensity observed between 2.5 and 4 A, particularly when
component of this catalytic material. The presence of a shoulder compared to a reference sample of SnO . However, the intensity
2
at ꢂ602 ppm does, however, indicate that a minor quantity of of the second shell region clearly differs between both the
extra-framework Sn is present in this material, both in line with microporous and hierarchical analogues, with Sn–Sn scattering
11
our previous study, and that of Bare and co-workers focusing interactions clearly being more intense in the hierarchical
46
on conventional i.e. hydrothermally prepared, Sn-b.
material. We conclude that this conrms the presence of an
In good agreement to the spectrum of 2Sn-b, the MAS NMR extra-framework Sn component in the hierarchical matrix,
spectrum of 2Sn-h*b is also dominated by a resonance at ꢂ688 suggesting that the modied porosity following the generation
ppm, indicative of Sn being primarily present within the zeolite of mesopores leads to a very slight increase in difficulty for
4
7–49
framework.
least one additional resonance at ꢂ591 ppm is present. The agreement to the MAS NMR analysis of 2Sn-b and 2Sn-h*b.
chemical shi of this resonance overlaps with the resonance In summary, the spectroscopic study undertaken indicates
observed for extra-framework Sn sites, such as SnO
(ꢂ602 that Sn(IV) is almost exclusively incorporated into the vacant
In contrast to the spectrum of 2Sn-b, however, at incorporating Sn into the framework successfully. This is in full
2
ppm), but with a +11 ppm shi. This shi, coupled with the framework sites of h*b, although a non-negligible fraction of
pronounced asymmetry of the resonance, causes some ambi- extra-framework Sn species appear to be present at very low
guity as to whether this resonance is associated with a minor concentrations. The fraction of extra-framework Sn appears to
contribution of extra-framework Sn species, or isolated Sn sites be somewhat larger in the hierarchical Sn-b than for purely
present in a different geometrical position or hydrated state to microporous Sn-b, that the modied hierarchical topology has
those typically found in Sn-b. Given the XAS analysis (vide infra) some impact on the solid-state stannation step. Nevertheless,
and the observation that a dehydration/rehydration protocol the relatively low quantity of extra-framework Sn observed in
did not affect the intensity or position of this resonance, we both materials, as observed spectroscopically and through TEM
tentatively attribute it to extra-framework Sn sites in a slightly
perturbed environment. Even so, the MAS NMR data obtained
Fig. 7 (A) Magnitude of the Fourier transform spectra, and (B) EXAFS
3
Fig. 6 MAS NMR spectra of (A) SnO
mixture), (C) 2Sn-b, and (D) 2Sn-h*b.
2
, (B) 10% SnO
2
/deAl-b (physical k -weighted x data, of 2Sn-b, (black) and 2Sn-h*b (blue). The corre-
2
sponding spectra of SnO (green dots) are also displayed.
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
11
analysis (Fig. S1†), agrees both with our previous studies, and
those published for conventional Sn-b, and provides us with two
suitable materials from which the benet(s) of a hierarchical
topology may be studied.
Catalytic studies
To probe the catalytic properties of our hierarchical stannosi-
licates, particularly in reference to purely microporous Sn-b, we
focused upon the Meerwein–Pondorf–Verley (MPV) transfer
hydrogenation of various cyclic ketones to alcohols (Scheme 1).
We chose this probe reaction for two main reasons; (i) we have
recently shown it to be an excellent model reaction for Sn-
b catalysis, matching the TOF trends observed for other Sn-b the
11
isomerisation of glucose to fructose; (ii) in contrast to several
other Sn-b catalysed reactions (e.g. glucose–fructose isomer-
Fig. 8 Catalytic activity for 2Sn-b (open triangles), 2Sn-h*b (open
isation, BVO), the MPV process is an intermolecular reaction, circles) and 2Sn-MCM-41 (filled squares) for the MPV reduction of (A)
cyclohexanone and (B) cyclooctanone.
which requires coordination of both the ketone substrate and
the alcohol solvent. It is, therefore, a very sterically challenging
reaction, and provides an excellent case study for the hierar-
chical materials. The activity of both zeolite materials i.e. 2Sn-
h*b and 2Sn-b, were compared to an authentic sample of Sn- MPV transfer hydrogenation of cyclohexanone are extremely
Fig. 8 shows that the pseudo-rst order rate constants for the
ꢂ3 ꢂ1
ꢂ3
MCM-41, an amorphous but ordered, mesoporous Sn-contain- similar for both 2Sn-b (2.5 ꢃ 10
s
) and 2Sn-h*b (2.2 ꢃ 10
ꢂ1
ing silicate containing 2 wt% Sn. The characterisation of this
s
), although there is a minor decrease in activity for the
material is provided in the ESI Fig. S2–S4.†
hierarchical material. We attribute this minor decrease in
To fully evaluate the performance of 2Sn-h*b and 2Sn-b, and activity to the presence of additional extra-framework Sn species
to determine the positive (or negative) impact of the hierar- in this material, which decreases the relative amount of active
chical matrix, we performed the MPV transfer hydrogenation of Sn per gram of catalyst. Nevertheless, the comparable catalytic
various cyclic ketones, containing between 6 and 12 carbon rates obtained over these two materials indicates that (a) the
atoms (i.e. cyclohexanone to cyclododecanone). To date, only C6 generation of a hierarchical structure does not excessively
substrates have been efficiently converted by MPV transfer impact the intrinsic activity of the isomorphously substituted
hydrogenation under the mediation of purely microporous Sn- Sn atoms within the matrix, a key result given the decreased
b, with diffusion limitations largely accounting for the poor intrinsic activity observed in other reports (vide supra), and (b)
activity observed for bulkier substrates. Indeed, whilst cyclo- that increased micropore accessibility is not essential for
˚
6
enhanced catalytic rates to be observed for C substrates. This is
hexanone (kinetic diameter, 6.0 A, relative diameter 1.0) is on
the borderline of being able to freely diffuse in the pores of in line with our observation that the C
6
reaction experiences no
zeolite beta (free diffusion limit of a sphere, 5.95 A˚ ), cyclo-
octanone (relative diameter 1.1) and cyclododecanone (1.4)
should be too large to freely diffuse through the micropores of
zeolite beta (ESI, Fig. S5†), even before the need for solvent
molecule coordination is considered (vide supra). The three
substrates therefore provide a useful span of molecular sizes to
truly evaluate the performance of the hierarchical material. We
50
internal mass transfer limitations (Fig. 9).

rst focused upon the MPV transfer hydrogenation of cyclo-
hexanone (Fig. 8A), which can be effectively catalysed by purely
microporous Sn-b. The initial reaction conditions chosen for
this reaction were identical to those recently published by our
11
team. Initial rate data was used to accurately compare the
intrinsic activity of each catalytic material without excessive
inuence from (product induced) deactivation.
Fig. 9 Catalytic activity of 2Sn-b for the MPV hydrogenation of
cyclohexanone as a function of catalyst loading. Sn content was varied
by changing the mass of 2Sn-b present in the reactor. Both the initial
Scheme 1 Generalised scheme for the MPV transfer hydrogenation of rate data (main figure) and the time online analysis as a function of Sn
various ketones.
content within the reactor (inset) are presented.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
However, as the size of the cyclic ring is increased beyond six
Journal of Materials Chemistry A
carbon atoms, the signicant advantages of the hierarchical
BEA framework become apparent. For the MPV reduction of
cyclooctanone (Fig. 8B), the hierarchical material is by far
the most active catalyst, demonstrating a catalytic rate (1.4 ꢃ
ꢂ4 ꢂ1
10
s ) over one order of magnitude greater than that
observed over the conventional microporous analogue (1.6 ꢃ
ꢂ5 ꢂ1
1
(
0
s
), and also the mesoporous material 2Sn-MCM-41
ꢂ5 ꢂ1
1.2 ꢃ 10
s ). Increasing the carbon chain length of the
ketone substrate is highly detrimental to catalysis in the
microporous analogue, demonstrating that pore size limita-
tions are a major disadvantage of the conventional BEA
framework structure when bulkier substrates are undergoing
study. To further evaluate the performance of these three
materials, we also examined the comparative performance of
each catalyst as a function of conversion and selectivity against
time. The superior performance of the hierarchical analogue for
Fig. 10 (A) Initial rate data for 2Sn-b (open triangles), 2Sn-h*b (open
circles) and 2Sn-MCM-41 (filled squares) for the MPV reduction of
cyclododecanone, and (B) extended time online profile for the same
the larger substrate is clearly evident, particularly at extended reactions.
reaction times (Table 2). In line with previous research, neither
deAl-b, nor SnO2 dispersed upon deAl-b behave as suitable
catalysts for these reactions.
To further probe this, we also explored the MPV reduction of
cyclododecanone (Fig. 10A). In this case, the catalytic rate of
substantially lower, typically <50%, thus indicating the further
favourability of the hierarchical material over the purely mes-
oporous analogue. At this time, we also draw attention to
surprising activity exhibited by microporous Sn-b for the
transfer hydrogenation of cyclododecanone (Fig. 10B). The
conversion of this bulky substrate by this microporous material
indicates that a certain fraction of Sn-b catalysis must occur on
the external surface of the material, potentially as a conse-
quence of external enrichment of the zeolite crystals, which is
ꢂ4 ꢂ1
2
Sn-h*b (1.0 ꢃ 10
s ) is again over one order of magnitude
greater than that observed over the microporous material (8.5 ꢃ
ꢂ6 ꢂ1
ꢂ6 ꢂ1
10
s
) and 2Sn-MCM-41 (6.9 ꢃ 10
s ). To the best of our
knowledge, this is the rst time that an efficient heterogeneous
catalyst for the catalytic valorisation of cyclododecanone in the
liquid phase has been reported, a key result given that this
substrate is a precursor to laurolactam and hence, polyamide
51
known to occur for these materials.
12. This clearly demonstrates that the presence of a hierarchical
structure signicantly improves the general applicability of Sn-
b, is extremely benecial to Sn-mediated catalysis in general.
In all cases, the ordered mesoporous analogue, Sn-MCM-41,
is the least active catalyst for these catalytic transformations,
although its performance relative to microporous Sn-
b improves substantially as the carbon chain length is
increased. This lower activity of 2Sn-MCM-41 relative to the
hierarchical material is likely due to the lower levels of Lewis
acidity exhibited by heteroatoms when buried in the amor-
phous walls. We note that for both zeolitic materials, the
selectivity to the primary MPV product (alcohol) remains largely
constant at above 90% regardless of the size of the substrate.
However, the selectivity observed for 2Sn-MCM-41 is
Deactivation studies
Operational lifetime is a critical, but oen overlooked, key
performance indicator a heterogeneous catalyst. Due to their
52
microporous structure, zeolites such as Sn-b are particularly
prone to some deactivation events, such as fouling, which can
signicant impact their commercial feasibility. Some of us
recently demonstrated that Sn-b possesses high levels of
stability during Lewis acid catalysis (MPV transfer hydrogena-
tion), but steadily deactivates over the course of steady state
operation. Our spectroscopic and kinetic analysis revealed that
product absorption and coking were largely responsible for this
slow but steady decrease in catalytic activity, causing fouling i.e.
Table 2 Catalytic performances of microporous and hierarchical Sn-b, and Sn-MCM-41, for the MPV hydrogenation of cyclohexanone and
cyclooctanone
a
b
Catalyst
Cyclohexanone converted (cyclohexanol selectivity)
Cyclooctanone converted (cyclooctanol selectivity)
2
2
2
Sn-b
Sn-h*b
Sn-MCM-41
97.5 (>95)
93.9 (>95)
8.2 (49)
0.8 (26)
1.2 (10)
17.9 (90)
83.4 (94)
12.5 (45)
1.1 (29)
1.0 (24)
deAl-b
2
SnO
2
/deAl-b
a
ꢁ
b
Reaction conditions: cyclohexanone (0.2 M) in 2-butanol, 10 mL; 100 C; 1.0 mol% Sn relative to CyO; 750 rpm; 30 minutes. Reaction conditions:
ꢁ
cyclooctanone (0.2 M) in 2-butanol (0.2 M), 10 mL; 100 C; 1.0 mol% Sn relative to cyclooctanone; 750 rpm; 240 minutes.
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
fouling. In this manuscript, we show that hierarchically porous
BEA stannosilicates are able to mediate the catalytic conversion
of bulky ketone substrates that purely microporous analogues
are unable to catalyse. Deactivation studies in the continuous
regime also demonstrate the exceptional stability of hierar-
chical Sn-b compared to purely microporous Sn-b, with limited
(<20%) activity loss observed over 700 h on stream. In contrast,
microporous Sn-b lost ꢀ70% activity in only 200 h on stream. To
the best of our knowledge, this is the rst time a stannosilicate
with a hierarchical BEA topology has been prepared, the rst
evidence of condensed phase cyclododecanone valorisation
with stannosilicates, and the rst full study of condensed phase
deactivation kinetics for hierarchical zeolite materials.
Fig. 11 Catalytic activity for 2Sn-b and 2Sn-h*b for the MPV reduction
of cyclohexanone in the continuous regime.
Experimental
Catalyst synthesis
Commercial zeolite Al-b (Zeolyst, NH -form, SiO /Al O ¼ 38)
4
2
2 3
3 3
was dealuminated by treatment in HNO solution (13 M HNO ,
1
6
blocking, of the micropores. At this stage, we hypothesised
that the improved diffusional properties of the hierarchical
matrix would not only allow us to study new reactions with
bulky substrates, but may also lead to decreased rates of deac-
tivation through fouling. Consequently, we examined the steady
state activity of both 2Sn-h*b and 2Sn-b for the MPV transfer
hydrogenation of cyclohexanone in continuous ow. Details of
our reactor are described in the ESI.†
ꢁ
ꢂ1
1
00 C, 20 h, 20 mL g zeolite). The dealuminated powder was
ꢂ1
washed extensively with water (ꢀ500 mL g catalyst), and dried
ꢁ
overnight at 110 C. Desilicated-dealuminated zeolite b ([deSi,-
deAl]-b, referred to as h*b throughout the manuscript) was
prepared as follows: commercial Al-b zeolite (CP814E, NH4-
form, SiO /Al O ¼ 38, Zeolyst International) was rst converted
2
2 3
ꢁ
ꢁ
into protonic form by calcination in static air at 550 C (5 C
ꢂ1
min ) for 5 h. The H-form zeolite was subsequently suspended
in an aqueous solution of NaOH (0.2 M, 318 K, 0.5 h, 30 mL g
Fig. 11 demonstrates the steady state activity of 2Sn-h*b and
Sn-b for the MPV transfer hydrogenation of cyclohexanone
ꢂ1
2
zeolite) to obtain a desilicated zeolite b (deSi-b). The reaction
was stopped immediately by cooling the container in an ice
bath. The remaining solid product was ltered, thoroughly
washed with deionized water until neutral pH was attained, and
over a 700 h period. The contact time for each catalyst was
optimised so that similar levels of conversion (80% for 2Sn-h*b
and 92% for 2Sn-b) were achieved, in order to allow accurate
comparison of lifetime. As can be seen, the hierarchical
analogue possess substantially improved lifetime compared to
the microporous analogue. Whereas the conversion obtained
over 2Sn-b decreases from 92% to 33% over a 200 h period, 2Sn-
h*b retains over 80% of its original activity over 420 h on
stream. With a single regeneration procedure under the
conditions employed during the nal calcination step (3 h,

nally dried at 373 K overnight. Subsequently, deSi-b was
dealuminated following the same procedure described above.
Aer ltering and drying the sample, the sample was recon-
verted into NH
4
-form via ion exchange in a solution of NH
4
NO
3
ꢁ
(1 M, 2 h, 85 C, 30 mL solution per g (catalyst)).
Solid-state stannation of both dealuminated zeolite b (deAl-
ꢁ
ꢂ1
b) and [deSi,deAl]-b (h*b) was performed the procedure re-
ported in ref. 11 and 53, by grinding the appropriate amount of
tin(II) acetate with the necessary amount of dealuminated or
desilicated-dealuminated zeolite for 10 minutes in a pestle and
mortar. Following this procedure, the sample was heated in
5
50 C, 60 mL min air), 2Sn-h*b is able to catalyse the MPV
reaction at >80% of its initial activity for over 700 h on stream,
a remarkable improvement in long-term stability. Although
periodic regeneration can also restore activity of 2Sn-b,
increasing the time between essential regenerations provides
added benets for both reactor design and process economics.
We note here that both the selectivity to CyOH and the carbon
balance remained >95% throughout for both catalysts during
steady state operation.
ꢁ
ꢁ
a combustion furnace (Carbolite MTF12/38/400) to 550 C (10 C
ꢂ1
min ) rst in a ow of N
2
(3 h) and subsequently air (3 h) for
a total dwell time of 6 h. Gas ow rates of 60 mL min were
ꢂ1
employed at all times.
Catalyst characterisation
Conclusions
A PANalytical X'PertPRO X-ray diffractometer was employed for
Lewis acidic zeolites have rapidly emerged as unique hetero- powder XRD analysis. A CuKa radiation source (40 kV and 30
geneous catalysts for a range of sustainable liquid phase, cata- mA) was utilised. Diffraction patterns were recorded between 5–
ꢁ
ꢁ
lytic applications. However their microporous structure means 55 2q (step size 0.0167 , time/step ¼ 150 s, total time ¼ 1 h).
that they lack general applicability (typically they may only be Porosimetry measurements were performed on a Quantach-
used for reactions involving small to medium sized substrates), rome Autosorb, and samples were degassed prior to use (277 C,
ꢁ
and makes them particularly prone to deactivation through 6 h). Adsorption isotherms were obtained at 77 K and various
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
Journal of Materials Chemistry A
analysis methods were employed (see Table 1). SEM Images of (University College London/UK Catalysis Hub) is thanked for
the catalysts were obtained using a Hitachi TM3030Plus, at 15 measurement of the XAS spectra and for useful discussions.
kV and at EDX observation conditions. The EDX tool was uti- Diamond Light Source and the Research Complex at Harwell are
lised to analyse the elemental composition and distribution. A thanked for the provision of beamtime (SP8071). Dr Nikolaos
Bruker Tensor spectrometer equipped with a Harrick praying Dimitratos is thanked for continued support and discussion.
mantis cell was utilised for DRIFT measurements. Spectra were
ꢂ1
ꢂ1
recorded between 4000–650 cm at a resolution of 2 cm
.
CD CN measurements were performed on pre-treated zeolite
3
References
ꢁ
ꢂ1
powders (550 C, 1 h under owing air, 60 mL min ) according
to the method described in ref. 10. MAS NMR experiments were
performed at Durham University through the EPSRC UK
National Solid-state NMR Service. Samples were measured
under conditions identical to those reported by Bermejo-Deval
1 Y. Roman-Leshkov and M. E. Davis, ACS Catal., 2011, 1,
1566–1588.
2 P. Y. Dapsens, C. Mondelli, B. T. Kusema, R. Verel and
J. Perez-Ramirez, Green Chem., 2014, 16, 1176–1186.
3 P. Y. Dapsens, C. Mondelli and J. P. Ramirez, Chem. Soc. Rev.,
2015, 44, 7015–7430.
4
7
11
et al., and Hammond and co-workers. Sn K-edge XAS data
11
was collected by the methods described in detail elsewhere.
4
5
M. Moliner, Dalton Trans., 2013, 43, 4197–4208.
A. Corma, L. T. Nemeth, M. Renz and S. Valencia, Nature,
Kinetic evaluation and analytical methods
2001, 412, 423–425.
MPV reactions with cyclohexanone were performed in a 100 mL
round bottom ask equipped with a reux condenser, which
was thermostatically controlled by immersion in a silicon oil
bath. The vessel was charged with a 10 mL solution of cyclo-
hexanone in 2-butanol (0.2 M), which also contained an internal
standard (biphenyl, 0.01 M), and was subsequently heated to
6
7
8
9
C. Hammond, S. Conrad and I. Hermans, Angew. Chem., Int.
Ed., 2012, 51, 11736–11739.
Z. Kang, X. Zhang, H. Liu, J. Qiu and K. L. Yeung, Chem. Eng.
J., 2013, 218, 425–432.
A. Corma, M. E. Domine, L. Nemeth and S. Valencia, J. Am.
Chem. Soc., 2002, 124, 3194–3195.
A. Corma, M. E. Domine and S. Valencia, J. Catal., 2003, 215,
ꢁ
the desired temperature (100 C internal temperature). The
reaction was initiated by addition of an appropriate amount of
catalyst, corresponding to 1 mol% Sn relative to cyclohexanone.
The solution was stirred at ꢀ750 rpm with an oval magnetic
stirrer bar. MPV reactions with cyclooctanone were performed
with the same methodology and conditions. MPV reactions with
cyclododecanone were performed in a 15 mL thick walled glass
reactor, which was thermostatically controlled by immersion in
a silicon oil bath. The vessel was charged with a 5 mL solution of
cyclododecanone in 2-butanol (0.2 M), which also contained an
internal standard (biphenyl, 0.01 M). The appropriate amount
of catalyst, corresponding to 2.5 mol% Sn relative to cyclo-
dodecanone, was also added. The reactor was subsequently
2
94–304.
0 P. Wolf, C. Hammond and I. Hermans, Dalton Trans., 2014,
3, 4514–4519.
1
1
4
1 C. Hammond, D. Padovan, A. Al-Nayili, P. P. Wells,
E. K. Gibson and N. Dimitratos, ChemCatChem, 2015, 7,
3
322–3332.
1
2 J. Dijkmans, D. Gabriels, M. Dusselier, F. de Clippel,
P. Vanelderen, K. Houthoofd, A. Maliet, Y. Pontikes and
B. F. Sels, Green Chem., 2013, 15, 2777–2785.
3 R. Bermejo-Deval, M. Orazov, R. Gounder, S. Hwang and
M. E. Davis, ACS Catal., 2014, 4, 2288–2297.
4 Y. R. Leshkov, M. Moliner, J. A. Labinger and M. E. Davis,
Angew. Chem., Int. Ed., 2010, 49, 8954–8957.
5 M. Moliner, Y. R. Leshkov and M. E. Davis, Proc. Natl. Acad.
Sci. U. S. A., 2010, 107, 6164–6168.
1
1
1
1
1
1
1
ꢁ
heated to the desired temperature (130 C internal tempera-
ture), and the reaction was initiated by stirring at ꢀ750 rpm
with an oval magnetic stirrer bar. Aliquots of both reaction
solutions were taken periodically for analysis, and were centri-
fuged prior to injection into a GC (Agilent 7820, 25m CP-Wax 52
CB). Reactants were quantied against a biphenyl internal
standard. Continuous ow experiments were performed as
described in the ESI.†
6 D. Padovan, M. S. Grasina, C. Parsons and C. Hammond,
2
015, submitted manuscript.
7 J. Jiang, J. Yu and A. Corma, Angew. Chem., Int. Ed., 2010, 49,
120–3145.
3
8 C. C. Freyhardt, M. Tsapatsis, R. F. Lobo, K. J. Balkus Jr and
M. E. Davis, Nature, 1996, 391, 295–298.
9 A. Corma, M. J. Diaz-Caba n˜ as, J. Martinez-Triguero, F. Rey
and J. Rius, Nature, 2002, 418, 514–517.
Acknowledgements
CH gratefully appreciates the support of The Royal Society for 20 C. Perego and R. Millini, Chem. Soc. Rev., 2010, 42, 3956–
the provision of a University Research Fellowship (UF140207) 3976.
and further research funding (RG140754). AA-N is indebted to 21 P. S. Niphadkar, A. C. Garade, R. K. Jha, C. V. Rode and
Al-Qadisiya University (Iraq) for the provision of a PhD schol-
arship. KY wishes to thank the Catalysis CDT for the PhD
P. N. Joshi, Microporous Mesoporous Mater., 2010, 136, 115–
125.
scholarship. Dr David Apperley, Dr Fraser Markwell and Dr Eric 22 L. Tosheva and V. P. Valtchev, Chem. Mater., 2005, 17, 2494–
Hughes of the ESPRC UK National Solid-state NMR Service at 2513.
Durham University are thanked for performing the MAS NMR 23 M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki and R. Ryoo,
experiments, and for useful discussions. Dr Peter Wells
Nature, 2009, 461, 246–249.
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
24 K. Na, C. Jo, K. Kim, K. Cho, J. Jung, Y. Seo, R. J. Messinger, 40 J. Jin, X. Ye, Y. Li, Y. Wang, L. Li, J. Gu, W. Zhao and J. Shi,
B. F. Chmelka and R. Ryoo, Science, 2011, 333, 328–332. Dalton Trans., 2014, 43, 8196–8204.
25 A. Corma, V. Fornes, S. B. Pergher, T. L. M. Maesen and 41 J. C. Groen, L. A. A. Peffer, J. A. Moulijn and J. Perez-Ramirez,
J. G. Buglass, Nature, 1998, 396, 353–356. Chem.–Eur. J., 2005, 11, 4983–4994.
26 L. Ren, Q. Guo, P. Kumar, M. Orazov, D. Xu, S. M. Alhassan, 42 M. Boronat, P. Concepcion, A. Corma, M. Renz and
K. A. Mkhoyan, M. E. Davis and M. Tsapatsis, Angew. Chem.,
Int. Ed., 2015, 54, 10848–10851.
7 X. Ouyang, S.-J. Hwang, D. Xie, T. Rea, S. I. Zones and A. Katz,
S. Valencia, J. Catal., 2005, 234, 111–118.
43 S. Roy, K. Bakhmutsky, E. Mahmoud, R. F. Lobo and
R. J. Gorte, ACS Catal., 2013, 3, 573–580.
2
2
2
3
ACS Catal., 2015, 5, 3108–3119.
8 H. Y. Luo, L. Bui, W. R. Gunther, E. Min and Y. Roman-
Leshkov, ACS Catal., 2012, 2, 2695–2699.
44 A. G. Pelmenschikov, R. A. V. Santen, G. J. Pnchen and
E. Meijer, J. Phys. Chem. B, 1993, 97, 11071–11074.
45 J. Chen, J. M. Thomas and G. Sankar, J. Chem. Soc., Faraday
Trans., 1994, 90, 3455–3459.
46 S. R. Bare, S. D. Kelly, W. Sinkler, J. J. Low, F. S. Modica,
S. Valencia, A. Corma and L. T. Nemeth, J. Am. Chem. Soc.,
2005, 127, 12924–12932.
9 K. Li, J. Valla and J. Garcia-Martinez, ChemCatChem, 2014, 6,
46–66.
0 J. Perez-Ramirez, C. H. Christensen, K. Egleblad,
C. H. Christensen and J. C. Groen, Chem. Soc. Rev., 2008,
3
7, 2530–2542.
47 R. Bermejo-Deval, R. Gounder and M. E. Davis, ACS Catal.,
2012, 2, 2705–2713.
3
3
1 M. Hartmann, Angew. Chem., Int. Ed., 2004, 43, 5880–5882.
2 M. S. Holm, E. Taaring, K. Egeblad and C. H. Christensen, 48 R. Bermejo-Deval, M. Orazov, R. Gounder, S.-J. Hwang and
Catal. Today, 2011, 168, 3–16. M. E. Davis, ACS Catal., 2014, 4, 2288–2297.
3 T. C. Keller, J. Arras, S. Wershofen and J. Perez-Ram ´ı rez, ACS 49 W. R. Gunther, V. K. Michaelis, M. A. Caporini, R. G. Griffin
3
3
3
3
Catal., 2015, 5, 734–743.
and Y. Roman-Leshkov, J. Am. Chem. Soc., 2014, 136, 6219–
6222.
4 J. C. Groen, L. A. A. Peffer, J. A. Moulijn and J. Perez-Ramirez,
Microporous Mesoporous Mater., 2004, 69, 29–34.
5 M. S. Holm, M. K. Hansen and C. H. Christensen, Eur. J.
Inorg. Chem., 2009, 1194–1198.
6 L. Sommer, D. Mores, S. Svelle, M. St ¨o cker,
B. M. Weckhuysen and U. Olsbye, Microporous Mesoporous
Mater., 2010, 132, 384–394.
50 C. Baerlocher and L. B. McCusker, Database of Zeolite
Structures, 2015, http://www.iza-structure.org/databases/.
51 S. Tolberg, A. Katerinopolou, D. D. Falcone, I. Sadaba,
C. M. Osmundsen, R. J. Davis, E. Taarning, P. Fristrup and
M. S. Holm, J. Mater. Chem. A, 2014, 2, 20252–20262.
52 I. S ´a daba, M. L. Granados, A. Riisager and E. Taarning, Green
Chem., 2015, 17, 4133–4145.
3
3
3
7 P. Y. Dapsens, C. Mondelli, J. Jagielski, R. Hauert and
J. P ´e rez-Ram ´ı rez, Catal. Sci. Technol., 2014, 4, 2302–2311.
8 H. J. Cho, P. Dornath and W. Fan, ACS Catal., 2014, 4, 2029–
53 A. Parvulscu, U. M u¨ ller, J. H. Teles, N. Vautravers, G. Uhl,
I. Hermans, P. Wolf and C. Hammond, WO 2015/067654
A1, 2015.
2037.
9 J. C. Groen, S. Abello, L. A. Villaescusa and J. Perez-Ramirez,
Microporous Mesoporous Mater., 2008, 114, 93–102.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016