Creation of Al-Enriched Mesoporous ZSM-5 Nanoboxes with High Catalytic Activity: Converting Tetrahedral Extra-Framework Al into Framework Sites by Post Treatment

Article abstract of DOI:10.1002/ange.202002416

ZSM-5 zeolite nanoboxes with accessible meso-micro-pore architecture and strong acid sites are important in relevant heterogeneous catalysis suffering from mass transfer limitations and weak acidities. Rational design of parent zeolites with concentrated and non-protective coordination of Al species can facilitate post-synthetic treatment to produce mesoporous ZSM-5 nanoboxes. In this work, a simple and effective method was developed to convert parent MFI zeolites with tetrahedral extra-framework Al into Al-enriched mesoporous ZSM-5 nanoboxes with low silicon-to-aluminium ratios of ≈16. The parent MFI zeolite was prepared by rapid ageing of the zeolite sol gel synthesis mixture. The accessibility to the meso-micro-porous intra-crystalline network was probed systematically by comparative pulsed field gradient nuclear magnetic resonance diffusion measurements, which, together with the strong acidity of nanoboxes, provided superb catalytic activity and longevity in hydrocarbon cracking for propylene production.

Full text of DOI:10.1002/ange.202002416

Angewandte
Eine Zeitschrift der Gesellschaft Deutscher Chemiker Chemie
www.angewandte.de
Akzeptierter Artikel
Titel: Creation of Al-Enriched Mesoporous ZSM-5 Nanoboxes with
High Catalytic Activity: Converting Tetrahedral Extra-Framework
Al into Framework Sites via Post Treatment
Autoren: Yilai Jiao, Luke Forster, Shaojun Xu, Huanhao Chen, Jingfeng
Han, Xuqing Liu, Yangtao Zhou, Jinming Liu, Jinsong Zhang,
Jihong Yu, Carmine D’Agostino, and Xiaolei Fan
Dieser Beitrag wurde nach Begutachtung und Überarbeitung sofort als
"
akzeptierter Artikel" (Accepted Article; AA) publiziert und kann unter
Angabe der unten stehenden Digitalobjekt-Identifizierungsnummer
DOI) zitiert werden. Die deutsche Übersetzung wird gemeinsam mit der
(
endgültigen englischen Fassung erscheinen. Die endgültige englische
Fassung (Version of Record) wird ehestmöglich nach dem Redigieren
und einem Korrekturgang als Early-View-Beitrag erscheinen und kann
sich naturgemäß von der AA-Fassung unterscheiden. Leser sollten
daher die endgültige Fassung, sobald sie veröffentlicht ist, verwenden.
Für die AA-Fassung trägt der Autor die alleinige Verantwortung.
Zitierweise: Angew. Chem. Int. Ed. 10.1002/anie.202002416
Angew. Chem. 10.1002/ange.202002416
Link zur VoR: http://dx.doi.org/10.1002/anie.202002416
http://dx.doi.org/10.1002/ange.202002416

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
Creation of Al-Enriched Mesoporous ZSM-5 Nanoboxes with High
Catalytic Activity: Converting Tetrahedral Extra-Framework Al
into Framework Sites via Post Treatment
[
a,b]
+[b]
+[b]
[b]
[c]
[d]
Yilai Jiao* , Luke Forster , Shaojun Xu , Huanhao Chen , Jingfeng Han , Xuqing Liu , Yangtao
[a]
[a]
[a]
[e,f]
[b]
[b]
Zhou , Jinming Liu , Jinsong Zhang , Jihong Yu , Carmine D’Agostino* , Xiaolei Fan*
[a]
Shenyang National Laboratory for Materials Science, Institute of Metal Research, Chinese Academy of Sciences, 72 Wenhua Road, Shenyang 110016,
China
E-mail: yljiao@imr.ac.cn (YJ)
[
b]
c]
Department of Chemical Engineering and Analytical Science, School of Engineering, The University of Manchester, Oxford Road, Manchester, M13 9PL,
United Kingdom
E-mail: carmine.dagostino@manchester.ac.uk (CD); xiaolei.fan@manchester.ac.uk (XF)
National Engineering Laboratory for Methanol to Olefins, Dalian National Laboratory for Clean Energy, Dalian Institute of Chemical Physics, Chinese
Academy of Sciences, Dalian 116023, China
[
[d]
[e]
[f]
Department of Materials, School of Natural Sciences, The University of Manchester, Oxford Road, Manchester, M13 9PL, United Kingdom
State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, College of Chemistry, Jilin University, Changchun 130012, China
International Center of Future Science, Jilin University, 2699 Qianjin Street, Changchun 130012, China
Supporting information and he ORCID identification number(s) for the author(s) of this article can be found under: xxx
[+] These authors contributed equally to this work.
Abstract: ZSM-5 zeolite nanoboxes with accessible meso-micro-pore
architecture and strong acid sites are important to heterogeneous
catalysis suffering from mass transfer limitations and weak acidities.
Rational design of parent zeolites with concentrated and non-
protective coordination of Al species can facilitate the post-synthetic
treatments to produce mesoporous ZSM-5 nanoboxes. In this work, a
simple and effective method was developed to prepare and convert
parent MFI zeolites with tetrahedral extra-framework aluminium
treatments (via selective desilication) using sodium hydroxide
NaOH)[
2-3]
and sodium carbonate (Na
CO
)
[7]
are generally
(
2
3
effective to create mesoporous structures in ZSM-5, such as
mesopores and hollow crystals. Notice that SAR values of parent
ZSM-5 zeolites are the critical parameter in the controlled
dissolution of Si species from the framework of the parent zeolites
to form accessible meso-micro-porosity.
Previous research has confirmed the protective role of the
tetrahedral framework Al for the Si extraction in alkaline media.
For parent ZSM-5 zeolites with low SAR values of <20 (i.e. Al-
rich), post-synthetic alkaline treatments are not effective for the
formation of mesoporous features, and sequential fluorination-
desilication and steaming-desilication for creating mesoporous
structures in the ZSM-5 zeolites (via tandem dealumination and
desilication treatments) are necessary.[ However, the use of
corrosive chemicals and steam makes the sequential treatment
processes dangerous and energy-intensive, and thus industrially
unfavourable. Conversely, for highly siliceous ZSM-5 with SAR
values >50, excessive desilication occurs during post-synthetic
alkaline treatments, leading to the uncontrollable mesopores
formation.[ Accordingly, the optimal framework SAR of 25–50
was identified for forming mesoporous structures effectively and
controllably.[
(EFAL) into Al-enriched mesoporous ZSM-5 nanoboxes with low
silicon-to-aluminium ratios (SARs) of ~16. The parent MFI zeolite was
prepared by a strategy of rapid ageing of the zeolite sol gel synthesis
mixture. The accessibility to the well-developed meso-micro-porous
intra-crystalline network was probed systematically by comparative
pulsed field gradient nuclear magnetic resonance (PFG-NMR)
diffusion measurements, which, in synergy with the high concentration
of strong acidity of nanoboxes, provided superb catalytic activity and
catalyst longevity in hydrocarbon cracking for producing propylene.
1]
2]
1-3]
Introduction
The use of structural directing agents (SDAs), especially
tetrapropylammonium hydroxide (TPAOH), in the modification of
MFI zeolites (such as TS-1,[ silicalite-1 and ZSM-5 ) under
alkaline conditions can facilitate the recrystallisation of leached
species from dissolution (i.e. the dissolution-recrystallisation
8]
[9]
[10]
Hierarchical zeolites with mesoporous features are a class of
materials
with high industrial relevance,
particularly
heterogeneous catalysis for both conventional petrochemical and
the emerging biomass conversions, due to the improved diffusion
properties of reactants and/or products within their crystalline
domains.[ MFI-type zeolites (with a nearly circular cross section
of ~0.54 nm diameter), particularly ZSM-5, are one of the most
important active components (along with FAU-type Y zeolites) in
petroleum refining and petrochemical catalysts, and have seen
significant growth in recent years for developing propylene-
selective processes such as catalytic cracking and methanol-to-
olefins (MTO).[ The limited accessibility of guest molecules to
intrinsic micropores of ZSM-5 is still very problematic for realising
its full catalytic potential. ZSM-5 has medium to high silicon-to-
[
8]
mechanism ), and hence enabling the controlled formation of
hollow yet mesoporous zeolite crystals (i.e. nanoboxes). Such
zeolite nanoboxes are important for catalysis, not only being able
to allow percolation diffusion through short paths but also
4]
providing the opportunity for functionalisation (e.g. encapsulation
[10]
of metal nanoparticles).
Since recrystallisation on the outer
surface of parent zeolite crystals occurs only with the pre-
selective dissolution of unprotected Si species, the dissolution-
recrystallisation route is again limited by the precise location of Al
in the framework of the parent zeolite; this is critical to the success
of post-synthetic desilication methods aiming to improve
molecular transport through the zeolite crystals. Currently, to
5]
aluminium ratios (SAR ≥~12) with heterogeneous Al distribution
across its framework,[
3a, 6]
and the post-synthetic alkaline
1
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
enable the post-synthetic modification using SDAs for making
mesoporous ZSM-5 nanoboxes, the use of parent zeolites such
temperature for 24 h) shows the sole peak at ca. 24.4° which is
assigned to the typical orthorhombic crystal symmetry of ZSM-
[9]
[15]
as siliceous silicalite-1 (in presence of Al sources) or ZSM-5 with
5. Comparably, after TPAOH treatment of AS-MFI (pH ≈ 13, at
SAR values >40[
5d, 9c, 10-11]
(ideally 40–74 ) is the essential
[10b]
160 °C under static conditions), the resulting zeolites (denoted as
prerequisite. Therefore, although ZSM-5 nanoboxes with a low
framework SAR of <25 (and hence high Brønsted acidity) are
significantly beneficial to zeolite catalysis (especially diffusion-
limited reactions), one-step post-synthetic alkaline treatments of
the homologous parent zeolites to obtain such materials is
challenging, regardless with or without SDAs or additional Al
sources. Therefore, rational design of the parent zeolite with high
concentration yet non-protective coordination of Al species (which
facilitates dissolution of Si and recrystallisation of Si and Al to
crystallographic T-sites during the post-synthetic treatment) can
be the solution to develop novel mesoporous ZSM-5 nanoboxes
with the combination of meso-micro-pore architecture and
concentrated Brønsted acidity.
Herein, we report a simple yet effective method to synthesise
mesoporous ZSM-5 nanoboxes with the low SAR value of ~16.
The method involves (i) the synthesis of a parent zeolite with
tetrahedral extra-framework Al (EFAL) and (ii) post-synthetic
treatment of the parent zeolite with TPAOH solution. The novel
feature of the method lies in the rapid ageing of zeolite sol gel
synthesis mixture at 40 °C (i.e. 30 min), which enables the
synthesis of the parent zeolite with tetrahedral EFAL. The
subsequent treatment using TPAOH solution allows the
recrystallisation and redistribution of dissolved Si and EFAL to
occur, forming highly crystalline hollow ZSM-5 zeolites with Al-rich
shell (SAR = ca. 16, shell thickness = ca. 20 nm), and hence
affording both improved acidity and diffusivity for catalysis. The
mechanism of the new method is elaborated based on the
comprehensive characterisation of materials using X-ray powder
ZSM-5-P-x-y, where
P
represents via post-treatment,
x
represents the TPAOH concentration in M and y represents post-
treatment time in h, respectively) show the single peak at about
24.4° as well (Figures 1a and S2a), suggesting the re-insertion of
the tetrahedral EFAL in AS-MFI into the newly formed framework
(due to dissolution and recrystallisation mechanism)[ during the
post-treatment. We found that a 6 h post-treatment of AS-MFI with
0.1 M TPAOH was sufficient to produce ZSM-5 nanoboxes with
excellent crystallinity (relative crystallinity, RC, >98%, Figure S2b)
and mesoporous features (Figure S3 and Table S1). By extending
the treatment time beyond 12 h, excessive dissolution occurred,
damaging the intactness of the hollow structure to certain extents
(as evidenced by the transmission electron, TEM, and scanning
electron microscopy, SEM, analysis, Figures S4 and S5) and
variation in mesoporosity of ZSM-5-P zeolites (Table S1). The
excessive dissolution due to the prolonged treatment time (>12 h)
was also reflected by the reduced RC values of the relevant ZSM-
5-P zeolites (at ~81%, Figure S2b).
8]
Solid-state 27Al magic angle spinning (MAS) NMR analysis
(
Figure 2b) shows the presence of tetrahedral (chemical shift
centres at ~58 ppm) and octahedral Al species (at ~0 ppm) for all
2
7
zeolites. However, one-dimensional Al MAS NMR is not
possible to elucidate the local symmetry and coordination state of
Al species in zeolites.[ Previous research has confirmed that the
change of peak width at ~58 ppm indicates the variation of local
16]
[
9b, 9c, 17]
Al coordination after post-treatments of MFI zeolites.
in this
work, it was measured that the peak width of the tetrahedral 27Al
in AS-MFI is the broadest among the samples under investigation
diffraction (XRD), nitrogen (N
magnetic resonance (NMR) spectroscopy and ammonia
temperature programmed desorption (NH -TPD) analysis at
2
) physisorption, solid-state nuclear
(
as the shaded orange area in Figure 2b), suggesting the
presence of EFAL in AS-MFI. The post-treatment using TPAOH
changed the local Al coordination substantially, and the peak
width of the resulting ZSM-5-P-0.1-6 was narrowed noticeably,
being comparable to that of the highly crystallised C-ZSM-5. Such
variation of the NMR signal of tetrahedral Al may be attributed to
3
different stages of the synthesis. Notably, the obtained ZSM-5
nanoboxes possess significantly high Al concentration (SAR of
∼
16) in the shell (∼20 nm) and mesoporous features (e.g. specific
3
−1
mesopore volume = ∼0.26 cm g ). The percolation diffusion of
probing molecules within the materials was assessed by pulsed-
field gradient NMR (PFG-NMR) measurements (Supporting
Information, Figure S1). Previous research has shown that PFG-
NMR is a powerful tool of investigating the mass transport in
[9b, 9c, 17]
the reintegration of EFAL in the recrystallised framework.
Comparison of 29Si MAS NMR spectra of AS-MFI, ZSM-5-P-0.1-
6
and C-ZSM-5 is shown in Figure 2c. Only the resonances above
−
110 ppm were measured for AS-MFI, which are assigned to
Si(4Si) sites,[
18]
proving the insignificant presence of the
zeolites with mesoporous features[
12]
[13]
and cracking catalysts,
framework Al in AS-MFI. Conversely, for C-ZSM-5, the chemical
such as intracrystalline diffusivity. However, the interpretation of
the diffusion data in relation to the relevant catalytic data has not
yet been reported. Benefits of the mesoporous ZSM-5 nanoboxes
to catalytic cracking (of model naphtha and aromatic compounds)
is demonstrated, showing excellent activity and selectivity to
propylene due to the unique combination of pore structural
features and chemical properties (i.e. the low SRA and
percolating porous network). The simple and effective strategy
solves the challenge of preparing mesoporous ZSM-5 zeolites
with low SAR values.
shifts of lower intensities at −103–−108 ppm (corresponding to
Si(1Al)) and <−100 ppm (corresponding to Si(2Al))[
18]
were
identified. After the TPAOH treatment, the band at −105.4 ppm
was measured for ZSM-5-P-0.1-6, evidencing the conversion of
EFAL into framework T-sites.
Post-treatments using SDAs, especially TPAOH, are known to
be effective to recover the dissolved species to a certain extent
(
reducing the loss of materials) and to form hollow MFI zeolites
[5d, 6, 8-9, 10]
with controlled properties such as wall thickness.
However, for ZSM-5, SAR value of the parent zeolite is the most
important parameter to ensure the successful dissolution-
recrystallisation process. Accordingly, the formation of ZSM-5
Results and Discussion
with regular hollow structures is only likely with the parent ZSM-5
having the SAR value >40,[
5d, 10-11]
leading to hollow materials with
For the first time, the rapid ageing of the sol gel synthesis
mixture (at 40 °C for 30 min, thereby increasing the rate of
hydrolysis of the silica source, i.e. tetraethylorthosilicate, TEOS)
was explored to produce the as-synthesised parent MFI zeolite
the Si-rich shell. According to the energy dispersive X-ray
spectrometry (EDX) and inductively coupled plasma-atomic
emission spectrometry (ICP-AES), the parent AS-MFI has a bulk
SAR of ∼12 (Table S1). A conventional ZSM-5 with the low SAR
of about 12 is not appropriate as the parent zeolite for preparing
ZSM-5 with mesoporous hollow structures via the controlled
dissolution and recrystallisation approach. However, as
discussed above, AS-MFI inherently possesses tetrahedral EFAL
which does not interfere with the Si dissolution during the post-
treatment. More importantly, the co-recrystallisation of dissolved
Si and EFAL facilitated by TPAOH (0.1 M at 160 °C) on the
(
AS-MFI) with tetrahedral EFAL (Supporting Information).
Comparison of X-ray powder diffraction (XRD) patterns of AS-MFI
and conventional ZSM-5 (C-ZSM-5) is shown in Figure 2a,
showing the distinct difference at 2θ of ~24.4°. AS-MFI shows the
characteristic of monoclinic phase for MFI zeolites,[14] i.e. the split
twin peaks, proving the presence of tetrahedral EFAL. Conversely,
C-ZSM-5 (prepared via conventional ageing at ice bath
2
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
−
1
external surface during the post-treatment (6 h to 96 h) produced
zeolite nanoboxes with Al-rich walls (SARs of ∼16, Table S1). X-
ray photoelectron spectroscopy (XPS) also shows that the
surface SARs of ZSM-5-P zeolites are lower than the according
bulk ones by EDX and ICP (Table S1), suggesting the occurrence
of Al redistribution during the TPAOH treatment.
404.9±33.4 µmol g ), reflecting the reconstruction of EFAL to the
tetra-coordinated framework Al during the post-treatment using
TPAOH. Regarding the acidity of C-ZSM-5, it is comparable to
that of ZSM-5-P nanoboxes (Figure S14c and Table S3). However,
the post-treatment of C-ZSM-5 (to P-C-ZSM-5-0.1 (6)) reduced
−
1
the strong acidity by ca. 28% (from 393.6 to 282.3 µmol g ).
AS-MFI is primarily microporous (specific external surface
ZSM-5-P nanoboxes combine the mesoporous structure with
the low SAR value (~16, corresponding to a high concentration of
framework Al) favour the relevant zeolite catalysis such as
propylene-selective catalytic cracking. ZSM-5 is the widely used
additives in cracking catalysts for improving propylene selectivity,
2
−1
area, SBET = 375 m g ) with uniform sizes of about 300–500 nm
as shown by the High resolution TEM (HRTEM) and scanning
transmission electron microscopy (STEM) analysis, Figures 3a1,
b1 and c). After the post-treatment, zeolite nanoboxes were
(
3
[
5b]
formed with comparable crystal sizes (Figures 3a1, 3b1, 3c, 3d
due to his pore diameter which provide size/shape selectivity.
and S4). The post-treatment condition used (i.e. 0.1 M TPAOH, at
However, the conventional microporous ZSM-5 zeolite is prone to
deactivate due to carbon deposition which is mainly the results of
160 °C for <12 h) was appropriate to produce the regular ZSM-5
nanoboxes with uniform cavities, which can be attributed to the
preferential desilication of siliceous part of AS-MFI and
recrystallisation of dissolved Si and EFAL. The shell thickness of
ZSM-5-P zeolites is about 20 nm (Figures 3e, 3f and S4). The
shell of ZSM-5-P-0.1-6 is well crystallised, as evidenced by the
fast Fourier transform (FFT) of HRTEM (inset in Figure 3f5) and
XRD analysis (Figure S2). The Al-rich walls of ZSM-5 nanoboxes
are also confirmed by EDX, as shown in Figure 3. ZSM-5-P
nanoboxes show the significant mesoporous features as revealed
long diffusion pathway, as well as the redundant strong acid
sites.[
19]
Although strong acid sites are crucial for catalytic
reactions, especially for hydrocarbon cracking, the reactants
could over-react on the acid sites and convert to aromatic
hydrocarbons or carbon deposition, resulting in the deactivation
of catalyst. Therefore, the development of zeolite catalysts
combing the strong acidity with improved accessibility to active
sites will be highly beneficial to address the challenge.
by the nitrogen (N
and S3, Table S1) with specific external surface areas (Sext.) =
2
) adsorption-desorption analysis (Figures 4a
Comparative catalytic evaluation of ZSM-5-P-0.1-6 along with the
control catalysts, i.e. AS-MFI, C-ZSM-5 and a conventional hollow
ZSM-5 nanoboxes with a SAR value of ~45 (i.e. C-HO-ZSM-5,
Figures S16 and S17, Table S4, Supporting Information) was
performed using cracking reactions with n-octane (kinetic
1
13–149 m g and mesopores volumes (Vmeso.) = 0.16–0.26 cm3
2
−1
−
1
g . Using concentrated SDA (i.e. 0.3 M or 0.5 M TPAOH,
Supporting Information) in the post-treatment (24 h) was not
beneficial to the formation of mesoporous hollow structures (e.g.
ext. <86 m g ), as well as reducing the crystallinity of the
resulting zeolites (i.e. ZSM-5-P-0.3-24 and ZSM-5-P-0.5-24, RC
56%), which is evidenced by various characterisation data of the
[
20]
[21]
diameter, KD, = 0.43 nm) and cumene (KD = 0.68 nm) as the
model naphtha and aromatic compounds (Figures 5 and S18,
Tables S5 and S6). Catalysis was carried out in a fixed bed
reactor (I.D = 10 mm) with 1 g pelletised zeolite catalysts (particle
size = 1.6–1.8 mm). Figure 5 presents the conversion of n-octane
over different zeolites at 540 °C as a function of time-on-stream
(ToS). AS-MFI shows insignificant activity compared to other
catalysts due to the lack of framework Al, and thus Brønsted
acidity (Figures S14a and S15). Although the microporous C-
ZSM-5 presented the highest initial activity (with an initial n-octane
conversion of ∼90%), it deactivated gradually and significantly
over time (the final n-octane conversion and dropped to ∼48%
after 25 h on stream). The deactivation of C-ZSM-5 was due to
the coke formation on the external surface of the crystals, which
was the result of the diffusion resistance caused by the pure
microporous framework of C-ZSM-5, leading to the loss of
accessibility and acidity (i.e. SBET and strong acidity dropped by
about 73% and 57%, respectively, according to the post-reaction
2
−1
S
<
materials (Figures S6–S9 and Table S2). This is again due to the
fast and excessive dissolution, which suppresses the
recrystallisation rate, making the formation of mesoporous hollow
structures challenging. RC value of ZSM-5-P-0.5-24 from the
treatment using 0.5 M TPAOH aqueous solution was only ~51%
(
Table S2), suggesting the significant loss of crystallinity due to
the fast dissolution. N physisorption analysis also shows that the
2
hysteresis loops of ZSM-5-P-0.3-24 and ZSM-5-P-0.5-24 zeolites
are less significant (Figure S8) in comparison to that of ZSM-5-P
zeolites (Figure S3). In summary, the post-treatment with TPAOH
solution is effective to revive EFAL in the parent AS-MFI,
converting it into framework Al in ZSM-5-P zeolites. However, the
balance of dissolution and recrystallisation needs to be regulated
(
by varying the treatment time and the concentration of aqueous
TPAOH solution) in order to obtain well-defined crystalline
nanoboxes. C-ZSM-5 shows the typical features of conventional
ZSM-5 zeolites which were characterised as presented in Table
S2. Comparably, as discussed above, the post-synthetic
treatment of C-ZSM-5 (with 0.1 M TPAOH for 6 h, Supporting
Information) was not able to develop mesoporous structures (e.g.
characterisation of the used catalysts using N
2
physisorption and
NH -TPD analyses, Tables S7 and S8). Conversely, the ZSM-5-
3
P-0.1-6 nanoboxes promoted the diffusion of n-octane through
their newly formed percolation pore network, being highly stable
regarding both n-octane conversion (at ca. 73%) and selectivity to
propylene (at ~30%, as shown in Figure S18). C-HO-ZSM-5 with
the mesoporous hollow structure (Figures S16 and S17) showed
a stable catalytic performance in cracking n-octane as well.
However, due to the low concentration of strong acidity in C-HO-
2
−1
Sext. = ~76 m g , Figures S10–S12, Table S2) due to the
presence of abundant framework Al in its framework (Figures 2b
and S13), inhibiting the effective dissolution of Si, and resulting in
the post-treated ZSM-5 zeolite (i.e. P-C-ZSM-5-0.1 (6)) with
limited and irregular mesopores (by SEM and TEM, Figure S11).
−1
ZSM-5 (at 214.4 µmol g ), it was outperformed by ZSM-5-P-0.1-
nanoboxes by ∼130% regarding n-octane conversion. The used
ZSM-5-P-0.1-6 (denoted as ZSM-5-P-0.1-6-U) can be
regenerated (by calcination at 550 °C under 10 vol.% O in N ),
6
Al-rich ZSM-5-P nanoboxes (i.e. with low SARs) show the
improved acidity, especially strong acidity corresponding to
Brønsted acidity (concentration of strong acid sites, as
2
2
and the regenerated ZSM-5-P-0.1-6 showed the comparable
chemical, physical and catalytic properties, as shown in Figures
S19–S21 and Tables S7–S8).
3 3
determined by NH -TPD for NH desorption at 300–500 °C, Figure
S14 and Table S3), which is beneficial to catalysis. Due to the
absence of Al-O-Si sites in AS-MFI, its strong acidity is
The accessibility issue was substantial when relatively bulky
−
1
insignificant at 24.1 µmol g , as shown in Figure 4b and Table S3
and infrared (IR) study of pyridine adsorption on zeolites (Figure
S15). It should be noted that the lack of strong acidity in AS-MFI
also serves as the evidence of the absence of framework
tetrahedral Al in it. Conversely, ZSM-5-P nanoboxes show a
significantly enhanced acidity (with an average strong acidity of
[21]
cumene (KD = 0.68 nm)
was cracked (at 320 °C), the
advantages of Al-rich ZSM-5-P-0.1-6 nanoboxes over the
conventional microporous C-ZSM-5 and conventional C-HO-
ZSM-5 (SAR = ~45) are highly recognisable. Severe deactivation
(
i.e. cumene conversion dropped by ~88% within 10 h) was
3
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
measured for C-ZSM-5 due to coke deposition (as shown by the
thermogravimetric, N physisorption and NH -TPD analysis of the
2 3
used zeolite catalysts, Figure S22, Tables S9–S10). By
comparing the two ZSM-5 nanoboxes under study, ZSM-5-P-0.1-
coefficient due to increased collisions with the intra-crystalline
pore walls. Conversely, for the microporous AS-MFI and C-ZSM-
5, the probing molecules diffuse primarily within the inter-
crystalline pore space (since the intra-crystalline space is less
accessible due to the much smaller pore size of 0.54 nm). To
prove this further, PFG-NMR experiments were carried out using
6
showed remarkably better activity than C-HO-ZSM-5 on stream
of 70 h (e.g. deactivation rate regarding the cumene conversion:
−
1
−1
[25]
−
0.31% h for ZSM-5-P nanoboxes vs. −0.48% h for C-HO-
a bulky molecule, TIPB (kinetic diameter = 0.94 nm), which is
ZSM-5). More importantly, although with a low SAR value of ~16,
ZSM-5 nanoboxes remained stable as well, as evidenced by the
not able to enter the micropores of the intra-crystalline space (i.e.
the intrinsic micropores of ZSM-5 zeolites). It was found that,
when bulky TIPB was used as the probing species, (i) unlike n-
octane and cumene, PFG-NMR plots of the materials are
comparable for all the zeolite samples (Figure 6c) and (ii) the rate
of diffusion within the ZSM-5-P-0.1-6 sample measured across
the whole zeolite particle increases, leading to a significant
decrease in τ values, which becomes comparable with values
measured for the parent and conventional zeolites (Table S13).
Such finding can be clearly explained by considering the larger
size of TIPB (in comparison with n-octane and cumene) hinders
the access to newly formed percolating network inside the
crystalline space of ZSM-5-P-0.1-6 zeolites, hence the probe
molecules experience a faster diffusion, lower tortuosity, within
the inter-crystalline space only. This is also confirmed by the
comparable D and τ values of the zeolites under investigation
when TIPB was used as the probing molecules in PFG-NMR
measurements (Table S13). In comparison with the state-of-the-
art post-synthetic alkaline treatments (with or without SDA, Tables
S14 and S15), the method produced ZSM-5 nanoboxes with (i) a
well-developed percolating meso-micro-porous network and (ii)
high concentration of strong acid sites (i.e. low SAR) in zeolite
crystals.
comparable XRD and N
2
physisorption analysis of the used ZSM-
-P-0.1-6 before and after steam ageing (at 500 °C for 10 h with
0% water in N , Figure S23 and Table S11). The specific
5
5
2
micropore surface area (Smicro) of the used ZSM-5-P-0.1-6 after
2
−1
2
−1
steam ageing dropped by ca. 6% (from 415 m g to 391 m g ),
while the RC values remained comparable at ~85%.
As previously mentioned, it is likely that the creation of an intra-
crystalline accessible meso-micro-porous structure of ZSM-5-P
nanoboxes contributes to the measured catalytic activity; this was
experimentally confirmed by PFG-NMR measurements carried
1
out at a H frequency of 43 MHz, with a diffusion probe capable
−
1
of producing magnetic field gradient pulses up to 163 mT m , at
atmospheric pressure and 25 °C. The mass transport properties
of the zeolites under study measured by PFG-NMR (using probing
molecules of n-octane, cumene and 1,3,5-triisopropylbenzene,
TIPB) are shown in Figure 6, together with PFG-NMR plots of the
bulk liquid of probing molecules (i.e. purple inverted triangle
symbols in Figures 6a–6c) as
a
reference. Diffusion
measurements were performed using the pulsed-field gradient
stimulated echo sequence (PGSTE sequence).[ The sequence
is made by combining a series of radiofrequency pulses (RF) with
magnetic field gradients (g, Figure S1). According to Eq. S1, the
NMR signal attenuation of PFG-NMR experiments as a function
of the gradient strength, E(g), is related to the experimental
variables and the diffusion coefficient (D, calculated by fitting Eq.
S1 to the experimental data). PFG-NMR plots, i.e. log-attenuation
plots as shown in Figures 6a–6c, provide a visual representation
of the diffusion properties of the guest molecules within the porous
media being studied. The lack of any evident curvature for PFG-
NMR plots, as well as the relatively large root-mean-square
22]
Conclusion
ZSM-5 zeolites are important catalysts for many catalytic
conversions such as catalytic cracking (to increase propylene
selectivity by cracking gasoline range molecules selectively),
MTO, alkylation and ethanol dehydration. Accessibility issues and
mass transfer limitations in ZSM-5’s microporous framework
commonly affect the outcome of the catalysis to a great extent.
Post-synthetic desilication treatment of ZSM-5 is the practical way
to introduce mesoporosity to ZSM-5 zeolites but is limited by SAR
of the parent zeolites. This work presents a simple and novel
strategy via rapid ageing of the sol gel mixture to prepare the
parent zeolite with the tetrahedral EFAL and low SAR of about 12,
which can be subsequently reconstructed (during the post-
synthetic treatment using SDA, i.e. TPAOH) to give mesoporous
hollow ZSM-5 nanoboxes with low SAR of ~16. The developed
protocol removed the constraint of SAR of the parent zeolite on
properties of the post-treated ZSM-5. The unique combination of
the hierarchical meso-micro-porous thin shell and low SAR of
such ZSM-5 nanoboxes led to (i) the significantly improved
accessibility of guest molecules to the active sites (evidenced by
PFG-NMR measurements) and (ii) comparably high yet stable
catalytic performance in cracking reactions, which is significantly
important for the development of specific propylene-selective
catalysts aiming to improve the current on-purpose propylene
production technologies.
displacement (RMSD) values (Table S12),[
23]
which are much
greater than the average crystal size of the zeolite particles,
indicates that the guest molecules have the time to explore the
overall pore structure of the solid samples (both inter- and intra-
crystalline space) during the observation times used for the
experiments (i.e. 200 ms for n-octane/cumene and 500 ms for
TIPB). Therefore, the calculated diffusion coefficients (D)
represent the averaged molecular diffusivity across the whole
zeolite particle.[
24]
Consequently, the diffusion measured is not
solely intracrystalline diffusion but rather, an effective diffusivity
comprised of diffusion within the intracrystalline pore space and
the pore space between crystallites, commonly referred to as
long-range diffusion. By fitting PFG-NMR plots using Eq. S1, the
relevant diffusion coefficient of the systems under investigation
was obtained (the negative value of the slope is equivalent to the
numerical value of D of the guest molecules being studied), as
shown in Table S13. Interestingly, PFG-NMR measurements
showed that D values of the probing molecules in ZSM-5-P-0.1-6
were the smallest in comparison with the microporous AS-MFI
and C-ZSM-5. As a consequence, the pore network tortuosity,
defined as the ratio of the bulk diffusivity of the guest molecule
and that of the same molecule within the pore space, (τ, Figure
Acknowledgements
6d, calculated using Eq. S3) of the ZSM-5-P-0.1-6 sample is the
largest. Considering the kinetic diameter n-octane (0.43 nm) and
cumene (0.68 nm), for ZSM-5-P-0.1-6, the comparatively small
value of D and large value of τ, suggest that the developed
method created a new percolating network within ZSM-5-P-0.1-6
zeolite crystals. As a result, the probing molecules gain access to
the newly formed percolating network in the intra-crystalline pores,
which is more tortuous than the inter-crystalline space, hence
leading to lower values of the averaged measured diffusion
This project has received funding from European Union’s
Horizon 2020 research and innovation programme under grant
agreement No. 872102. YJ thanks the China Scholarship Council
(
CSC) for his academic visiting fellowship in the UK (file no.
201604910181) and financial support from the Shenyang
4
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
Kočiřík, A. Zikánová, H. Jirglová, C. Berger, R. Gläser, J.
Weitkamp, E. W. Hansen, Chemistry of Materials 2005,
National Laboratory for Materials Science for his research
(
Y8L6641161). HC thanks the financial support from the
17, 2466-2474.
European Commission Marie Skłodowska-Curie Individual
[14]
E. Duprey, P. Beaunier, M. A. Springuel-Huet, F. Bozon-
Verduraz, J. Fraissard, J. M. Manoli, J. M. Brégeault,
Journal of Catalysis 1997, 165, 22-32.
M. Ardit, A. Martucci, G. Cruciani, The Journal of Physical
Chemistry C 2015, 119, 7351-7359.
Z. Yu, A. Zheng, Q. Wang, L. Chen, J. Xu, J.-P.
Amoureux, F. Deng, Angewandte Chemie International
Edition 2010, 49, 8657-8661.
aD. Verboekend, S. Mitchell, M. Milina, J. C. Groen, J.
Pérez-Ramírez, The Journal of Physical Chemistry C
Fellowship (grant agreement No. 748196). JY and XF thank the
1
11 Project of China (Grant No. B17020) for the collaborative
research. CD thanks EPSRC New Investigator Award
EP/S019138/1) for supporting his research. The authors thank
[
15]
16]
[
(
Ms Xi Chen for her creative contribution to the design of the
artwork.
[17]
2
011, 115, 14193-14203; bA. Erigoni, S. H. Newland, G.
Paul, L. Marchese, R. Raja, E. Gianotti, ChemCatChem
016, 8, 3161-3169.
Keywords: ZSM-5 • post-synthetic treatment • mesoporous
nanoboxes • silicon-to-aluminium ratio (SAR), PFG-NMR •
diffusion
2
[
18]
S. Sklenak, J. Dědeček, C. Li, B. Wichterlová, V. Gábová,
M. Sierka, J. Sauer, Physical Chemistry Chemical Physics
2
009, 11, 1237-1247.
[
[
[
1]
2]
3]
D. Zhang, C. Jin, M. Zou, S. Huang, Chemistry – A
European Journal 2019, 25, 2675-2683.
J. C. Groen, J. C. Jansen, J. A. Moulijn, J. Perez-Ramirez,
J. Phys. Chem. B 2004, 108, 13062-13065.
aJ. C. Groen, T. Bach, U. Ziese, A. M. Paulaime-van
Donk, K. P. de Jong, J. A. Moulijn, J. Pérez-Ramírez,
Journal of the American Chemical Society 2005, 127,
[19]
[20]
Y. Ji, H. Yang, W. Yan, Catalysts 2017, 7, 367.
H. H. Funke, A. M. Argo, J. L. Falconer, R. D. Noble,
Industrial & Engineering Chemistry Research 1997, 36,
137-143.
M. Jahandar Lashaki, M. Fayaz, S. Niknaddaf, Z.
Hashisho, Journal of Hazardous Materials 2012, 241-242,
154-163.
E. O. Stejskal, J. E. Tanner, The Journal of Chemical
Physics 1965, 42, 288-292.
M. Dvoyashkin, R. Valiullin, J. Kärger, Physical Review E
2007, 75, 041202.
Q. Zhu, G. D. Moggridge, M. Ainte, M. D. Mantle, L. F.
Gladden, C. D’Agostino, Chemical Engineering Journal
2016, 306, 67-76.
N. Hosseinpour, Y. Mortazavi, A. Bazyari, A. A.
Khodadadi, Fuel Processing Technology 2009, 90, 171-
179.
[21]
1
0792-10793; bJ. C. Groen, J. A. Moulijn, J. Perez-
Ramirez, Industrial & Engineering Chemistry Research
007, 46, 4193-4201.
[22]
[23]
[24]
2
[
[
4]
5]
aR. Chal, C. Gérardin, M. Bulut, S. van Donk,
ChemCatChem 2011, 3, 67-81; bY. Li, L. Li, J. Yu, Chem
2017, 3, 928-949; cR. Bai, Y. Song, Y. Li, J. Yu, Trends in
Chemistry 2019, 1, 601-611.
aA. V. Lavrenov, L. F. Saifulina, E. A. Buluchevskii, E. N.
Bogdanets, Catalysis in Industry 2015, 7, 175-187; bE. T.
C. Vogt, B. M. Weckhuysen, Chemical Society Reviews
[25]
2
015, 44, 7342-7370; cY. Jiao, X. Fan, M. Perdjon, Z.
Yang, J. Zhang, Applied Catalysis A: General 2017, 545,
04-112; dJ. Li, M. Liu, X. Guo, S. Zeng, S. Xu, Y. Wei, Z.
1
Liu, C. Song, Industrial & Engineering Chemistry
Research 2018, 57, 15375-15384; eY. Jiao, X. Ou, J.
Zhang, X. Fan, Reaction Chemistry & Engineering 2019,
4, 427-435.
[
[
[
[
6]
7]
8]
9]
C. Pagis, A. R. Morgado Prates, D. Farrusseng, N. Bats,
A. Tuel, Chemistry of Materials 2016, 28, 5205-5223.
C. Mei, Z. Liu, P. Wen, Z. Xie, W. Hua, Z. Gao, Journal of
Materials Chemistry 2008, 18, 3496-3500.
Y. Wang, M. Lin, A. Tuel, Microporous and Mesoporous
Materials 2007, 102, 80-85.
aY. Wang, A. Tuel, Microporous and Mesoporous
Materials 2008, 113, 286-295; bJ. Shao, T. Fu, Z. Ma, C.
Zhang, H. Li, L. Cui, Z. Li, Catalysis Science & Technology
2019, 9, 6647-6658; cD. Verboekend, J. Pérez-Ramírez,
Chemistry – A European Journal 2011, 17, 1137-1147; dZ.
Ma, T. Fu, Y. Wang, J. Shao, Q. Ma, C. Zhang, L. Cui, Z.
Li, Industrial & Engineering Chemistry Research 2019, 58,
2146-2158.
[
10]
aS. Li, A. Tuel, D. Laprune, F. Meunier, D. Farrusseng,
Chemistry of Materials 2015, 27, 276-282; bC. Dai, A.
Zhang, M. Liu, X. Guo, C. Song, Advanced Functional
Materials 2015, 25, 7479-7487.
[
[
11]
12]
Z. Hong, Z. Wang, D. Chen, Q. Sun, X. Li, Applied Surface
Science 2018, 440, 1037-1046.
aP. Kortunov, S. Vasenkov, J. Kärger, R. Valiullin, P.
Gottschalk, M. Fé Elía, M. Perez, M. Stöcker, B. Drescher,
G. McElhiney, C. Berger, R. Gläser, J. Weitkamp, Journal
of the American Chemical Society 2005, 127, 13055-
1
3059; bR. Valiullin, J. Kärger, K. Cho, M. Choi, R. Ryoo,
Microporous and Mesoporous Materials 2011, 142, 236-
44; cD. Mehlhorn, R. Valiullin, J. Kärger, K. Cho, R.
Ryoo, Microporous and Mesoporous Materials 2012, 164,
73-279.
2
2
[
13]
P. Kortunov, S. Vasenkov, J. Kärger, M. Fé Elía, M. Perez,
M. Stöcker, G. K. Papadopoulos, D. Theodorou, B.
Drescher, G. McElhiney, B. Bernauer, V. Krystl, M.
5
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
Entry for the Table of Contents
ZSM-5 nanoboxes with low SARs: Rapid ageing of the sol gel synthesis mixture produces the parent zeolite with the tetrahedral
extra framework Al and a low SAR value of ~12, which is transformed to mesoporous ZSM-5 nanoboxes with low SARs of ~16 during
the post-synthetic TPAOH treatment. Mesoporous ZSM-5 nanoboxes are comprehensively and systematically characterised, and
shows comparatively enhanced mass transfer ability, catalytic cracking activity and longevity.
6
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
Figure 1. Schematic illustration of the preparation of the mesoporous ZSM-5-P nanoboxes via the rapid ageing (of the precursor sol gel mixture) and post-
synthetic TPAOH treatment.
Figure 2. (a) XRD patterns, (b) 27Al MAS NMR spectra and (c) 29Si MAS NMR spectra of ZSM-5-P-0.1-6 (top), AS-MFI (middle) and C-ZSM-5 zeolites
(bottom).
7
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
Figure 3. HRTEM micrographs of (a1–a2) AS-MFI and (b1–b2) ZSM-5-P-0.1-6 zeolites; STEM micrographs of (a3) AS-MFI and (b3) ZSM-5-P-0.1-6 zeolites;
(b4–b5) HRTEM micrograph of ZSM-5-P-0.1-6 zeolite (Inset: the corresponding fast Fourier transform (FFT) of HRTEM).
8
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
Figure 4. (a) N
2
3
adsorption (solid symbols)/desorption (open symbols) isotherms and (b) NH -TPD curves of AS-MFI, C-ZSM-5 and ZSM-5-P-0.1-6.
9
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
Figure 5. Conversion profiles of different zeolite catalysts as a function of time-on-stream (ToS) in (a) catalytic n-octane cracking and (b) catalytic cumene
cracking.
1
0
This article is protected by copyright. All rights reserved.

Angewandte Chemie
10.1002/ange.202002416
RESEARCH ARTICLE
Figure 6. PFG-NMR log attenuation plots for (a) n-octane, (b) cumene and (c) TIPB within different zeolite samples under investigation (solid lines represent
the fittings using Eq. S1); (d) values of tortuosity of the probe molecules of n-octane, cumene and TIPB in different zeolite samples.
1
1
This article is protected by copyright. All rights reserved.