Steam-Induced Coarsening of Single-Unit-Cell MFI Zeolite Nanosheets and Its Effect on External Surface Br?nsted Acid Catalysis

Article abstract of DOI:10.1002/anie.202000395

Commonly used methods to assess crystallinity, micro-/mesoporosity, Br?nsted acid site density and distribution (in micro- vs. mesopores), and catalytic activity suggest nearly invariant structure and function for aluminosilicate zeolite MFI two-dimensional nanosheets before and after superheated steam treatment. Yet, pronounced reaction rate decrease for benzyl alcohol alkylation with mesitylene, a reaction that cannot take place in the zeolite micropores, is observed. Transmission electron microscopy images reveal pronounced changes in nanosheet thickness, aspect ratio and roughness indicating that nanosheet coarsening and the associated changes in the external (mesoporous) surface structure are responsible for the changes in the external surface catalytic activity. Superheated steam treatment of hierarchical zeolites can be used to alter nanosheet morphology and regulate external surface catalytic activity while preserving micro- and mesoporosity, and micropore reaction rates.

Full text of DOI:10.1002/anie.202000395

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Steam-Induced Coarsening of Single-Unit-Cell MFI Zeolite
Nanosheets and its Effect on External Surface Brønsted Acid
Catalysis
Authors: Yasmine Guefrachi, Geetu Sharma, Dandan Xu, Gaurav
Kumar, Katherine Vinter, Omar Abdelrahman, Xinyu Li, Saeed
Alhassan, Paul Dauenhauer, Alexandra Navrotsky, Wei
Zhang, and Michael Tsapatsis
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202000395
Angew. Chem. 10.1002/ange.202000395
Link to VoR: http://dx.doi.org/10.1002/anie.202000395
http://dx.doi.org/10.1002/ange.202000395

10.1002/anie.202000395
Angewandte Chemie International Edition
RESEARCH ARTICLE
Steam-Induced Coarsening of Single-Unit-Cell MFI Zeolite
Nanosheets and its Effect on External Surface Brønsted Acid
Catalysis
Yasmine Guefrachi,[a] Geetu Sharma,[b] Dandan Xu,[a] Gaurav Kumar,[a] Katherine P. Vinter,[a] Omar A.
Abdelrahman,[a] Xinyu Li,[a] Saeed Alhassan,[c] Paul J. Dauenhauer,[a] Alexandra Navrotsky,[b] Wei
Zhang,[d],[e] and Michael Tsapatsis*[a],[f],[g]
[a]
Y. Guefrachi, D. Xu, G. Kumar, K. P. Vinter, O. A. Abdelrahman, X. Li, Prof. P. J. Dauenhauer, Prof. M. Tsapatsis
Department of Chemical Engineering and Materials Science
University of Minnesota
421 Washington Avenue SE, Minneapolis, MN 55455 (USA)
E-mail: tsapatsis@jhu.edu
[*]
[b]
Dr. G. Sharma, Prof. A. Navrotsky
Peter A. Rock Thermochemistry Laboratory, NEAT-ORU
University of California Davis
Davis, CA 95616 (USA)
[c]
[d]
Prof. S. Alhassan
Department of Chemical Engineering
Khalifa University of Science and Technology
Habshan Building, Sas Al Nakhl Campus, Abu Dhabi (UAE)
Prof. W. Zhang
Department of Diagnostic and Biological Sciences
University of Minnesota
515 Delaware St SE, Minneapolis, MN 55455 (USA)
Characterization Facility
University of Minnesota
312 Church St, Minneapolis, MN 55455 (USA)
Department of Chemical and Biomolecular Engineering and Institute for NanoBioTechnology
Johns Hopkins University
[e]
[f]
3400 N. Charles Street, Baltimore, MD 21218 (USA)
Applied Physics Laboratory
[g]
Johns Hopkins University
11100 Johns Hopkins Road, Laurel, MD 20723 (USA)
Supporting information for this article is given via a link at the end of the document.
Abstract: Commonly used methods to assess crystallinity, micro-
/mesoporosity, Brønsted acid site density and distribution (in micro-
vs. mesopores), and catalytic activity suggest nearly invariant
structure and function for aluminosilicate zeolite MFI two-
dimensional nanosheets before and after superheated steam
treatment. Yet, pronounced reaction rate decrease for benzyl alcohol
alkylation with mesitylene, a reaction that cannot take place in the
zeolite micropores, is observed. Transmission electron microscopy
images reveal pronounced changes in nanosheet thickness, aspect
ratio and roughness indicating that nanosheet coarsening and the
associated changes in the external (mesoporous) surface structure
are responsible for the changes in the external surface catalytic
activity for benzyl alcohol alkylation with mesitylene. It is
demonstrated that superheated steam treatment of hierarchical
zeolites can be used to alter nanosheet morphology and regulate
external surface catalytic activity while preserving micro- and
mesoporosity, and micropore reaction rates.
water vapor-induced structural rearrangements[15–23] at the
nanometer (single-unit-cell) level is of particular significance for
two-dimensional (2D) zeolites[24,25] and thicker nanosheets[26] that
constitute an emerging class of catalysts, adsorbents and
membranes.[27–34] Here, we demonstrate that an all-silica single-
unit-cell meso/microporous MFI-type zeolite (SPP: self-pillared
pentasil)[35–38] retains its crystallinity and micro- and mesoporosity
o
under steaming at 350 C, while small but detectable changes
take place in the content of silanol groups and the enthalpy of
transition ( 훥퐻푇푟푎푛푠푖푡푖표푛 ) relative to α-quartz (the most stable
polymorph of silica under ambient conditions). Electron
microscopy reveals major changes in the nanosheet dimensions:
increase in thickness along the b-direction (straight pore
channels) and reduction of basal ((010) plane) dimensions along
the c-direction. Implications of these changes are shown to be
significant for the catalytic performance of aluminosilicate SPP
providing a method for controlling external surface catalytic
activity without interfering with catalysis in the micropores.
Introduction
Results and Discussion
During their use, zeolites are typically exposed to
hydrothermal treatments which may alter their structure at the
atomic to the nanometer scale with desirable or undesirable
effects on performance.[1–14] Understanding and controlling the
All-silica SPP was synthesized based on the reported
procedure[35] and its evolution was monitored during exposure to
an equimolar mixture of superheated steam and nitrogen at 350
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202000395
Angewandte Chemie International Edition
RESEARCH ARTICLE
oC for up to 30 days (Section SI and Figure S1 in the Supporting
Information).
determining the dimensions of one hundred domains in the
starting and the 7-day steamed SPP (Figure 3a). The dimensions
along the b-axis and the a-axis were measured using the
tomogram cross-sections made along the [001] zone axis, while
the length along the c-axis was measured using the cross-
sections made along the common [100]/[010] zone axes. The
relative changes in the dimensions of the domains are evident by
inspection of the corresponding histograms shown in Figure 3a.
Despite possible deviations of the domain dimensions from a
Gaussian distribution, we adapt it here in order to estimate
representative mean values. The thickness along b evolved from
2.1 nm (a single MFI unit-cell dimension along b) to an average
thickness of 8.5 nm (ca. four-unit-cell). The average length along
the a-axis before and after steam treatment was found to be 17.6
nm and 14.5 nm, respectively, with the Gaussian fits to the length
distributions nearly indistinguishable. The average length along
the c-axis becomes significantly shorter (22.5 nm from 44.6 nm)
with a narrower length distribution. This analysis establishes that
the main characteristic of the evolution during steam treatment is
thickening of the nanosheets perpendicular to their basal plane
(along the b-axis) and shortening of the basal dimensions mostly
along the c-axis. The TEM images show that the evolved domains
adopt curved edges, exhibit increased roughness and become
more globular compared to the well-defined 2D MFI nanosheets
in SPP before steam treatment. Apparently, local rearrangements
of silica within individual SPP particles are responsible for these
changes and the preservation of the intergrown architecture. The
aligned tilt-series of steamed SPP showing the above described
morphology in 3D is shown in Multimedia S2.
The morphology of calcined SPP (before steaming) has
been described in detail elsewhere.[35,36] For completeness, we
provide TEM images in Figure S2a-g and a detailed discussion in
Section SII in the Supporting Information. Also, a 3D view
perspective of the starting calcined SPP is obtained through the
aligned tilt-series displayed in Multimedia S1 in the Supporting
Information. Briefly, SPP consists of intergrown nanosheets that
are 2 nm-thick (i.e., one unit-cell-thick) along their b-axis and ca.
20x50 nm2 in their basal plane. This is illustrated by the
representative images obtained by 3D TEM tomography[39–41]
shown in Figure 1. The 2D TEM images collected in the tilt-series
are used along with a numerical algorithm (IMOD software
package, Version 4.9, University of Colorado, Boulder, CO,
USA)[42] to reconstruct the real space structure and map it into a
3D tomogram. The generated tomograms are presented as
sequential numerical cross-sectional images (0.52 nm-thick)
made along a certain zone axis. These cross-sections are
regarded as virtual slices obtained from running a sharp virtual
knife through the 3D reconstructed model. Figure 1 confirms the
2D uniformity of the domains. They are single-unit-cell throughout
the length of [010].
Figure 1. 3D TEM tomography of calcined SPP with a schematic representation
of a portion of SPP particle consisting of intergrown single-unit-cell-thick MFI
nanosheets. a) Sequential (numbered 1-4) cross-sectional images along [001].
b) Sequential (numbered 5-7) cross-sectional images along [100]/[010].
Thickness of each image is 0.52 nm. Scale bars are 50 nm.
Figure 2. 3D TEM tomography of 7-day steamed SPP with a schematic
representation of
a portion of a coarsened SPP particle. a) Sequential
(numbered 1-4) cross-sectional images along [001]. b) Sequential (numbered
5-7) cross-sectional images along [100]/[010]. Thickness of each image is 0.52
nm. Scale bars are 50 nm.
SPP exposed to steam loses the characteristic thin
dimension of the MFI nanosheets along the b-axis. TEM images
are presented in Figure S2h-n and discussed in Section SII in the
Supporting Information. Representative tomograms are
presented in Figure 2. The observed changes in the relative
dimensions of SPP nanosheets upon steaming are quantified by
SPP evolution during steaming was followed by the
thickening along the b-direction (using TEM images down the c-
axis) and was found to proceed faster at the beginning and very
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202000395
Angewandte Chemie International Edition
RESEARCH ARTICLE
slow after 5.25 days of steaming (Figure 3b); for example, the
thickness of the 7-day steamed SPP sheets shown in Figure 3d
is comparable to that of the 11-day steamed SPP in Figure 3e.
Despite the major morphological changes described above,
steamed SPP exhibited indistinguishable X-ray diffraction (XRD)
patterns compared to the starting material irrespective of the
treatment duration (Figure 4a). This finding confirms retention of
crystallinity (i.e., no evidence of amorphization) and preservation
of the crystallographic alignment of the domains within the
individual SPP particles, (i.e., preservation of the original 90o
intergrown architecture which coherently connects all individual
domains aligned as parts of a single crystal). Argon (Ar)
physisorption isotherms also remain invariant upon steam
treatment (Figure 4b) revealing that although there are drastic
changes in the domain morphology, the microporosity and
mesoporosity within SPP particles remain unaltered (Figure S3).
Apparently, any increase in microporous volume and decrease in
mesoporosity due to the zeolite domain thickening along [010] is
exactly offset by the contraction experienced mostly along [001]
and somewhat along [100] (Figure 3a). Use of the invariant XRD
and Ar physisorption data as criteria to monitor the effect of steam
treatment would have mistakenly implied stability of 2D
nanosheets in SPP, whereas the TEM and 3D tomography results
discussed above, demonstrate that the nanosheets lose their 2D
(i.e., single-unit-cell) morphology.
Figure 4. Evolution of crystallinity, porosity and silica connectivity in steamed
SPP. a) Powder XRD patterns of calcined single-unit-cell SPP (black), 1-
(purple), 2- (brown), 5.25- (arctic), 7- (red), 11- (blue) and 30-day (green)
steamed SPP as measured under the same conditions and compared in the
same vertical scale. b) Ar adsorption and desorption isotherms of calcined
single-unit-cell SPP (black), 1- (purple), 2- (brown), 5.25- (arctic), 7- (red), 11-
(blue) and 30-day (green) steamed SPP and calcined (directly-synthesized)
four-unit-cell SPP (orange). The isotherms of the calcined single-unit-cell SPP
and the steamed SPP are identical; they are offset by 200, 400, 600, 800, 1000
and 1200 mL/g, respectively for better visualization. c) Solid-state 29Si MAS
NMR spectrum of calcined single-unit-cell SPP. Shown are the measured
spectrum (black), Q4 (red), Q3 (neon) and Q2 (blue) deconvoluted peaks fitted
to Gaussian functions and the cumulative fit (arctic). d) Solid-state 29Si MAS
NMR spectrum of 30-day steamed SPP. Shown are the measured spectrum
(green), Q4 (red), and Q3 (neon) deconvoluted peaks fitted to Gaussian
functions and the cumulative fit (arctic). e) Enthalpy of transition relative to α-
quartz at 25 oC (훥퐻푇푟푎푛푠푖푡푖표푛): 훥퐻푇푟푎푛푠푖푡푖표푛 of starting calcined single-unit-cell
SPP (black) = 30.45 ± 1.07 kJ/mol SiO2, 훥퐻푇푟푎푛푠푖푡푖표푛 of four-unit-cell 30-day
steamed SPP (green) = 26.09 ± 1.10 kJ/mol SiO2, 훥퐻푇푟푎푛푠푖푡푖표푛 of four-unit-cell
(directly-synthesized) SPP (orange) = 12.82 ± 0.93 kJ/mol SiO2, 훥퐻푇푟푎푛푠푖푡푖표푛 of
silicalite-1[45] (pink) = 8.56 ± 0.72 kJ/mol SiO2 and 훥퐻푇푟푎푛푠푖푡푖표푛 of α-quartz[45]
(grey) = 0.00 ± 0.40 kJ/mol SiO2.
Changes, albeit small, after steam treatment are also
evident in the 29Si Magic Angle Spinning (MAS) NMR spectrum
shown in Figure 4d. A comparison of the corresponding spectra
before and after steam treatment, shown in Figure 4c,d, shows a
clear but small (ca.10 %) increase of the Q4 fraction, which can be
attributed to defect (e.g., Q2 and Q3 sites at nanosheets edges,
internal silanol defects, etc.) elimination and/or differences in
surface density of silanol groups among the expressed crystal
facets within SPP particles.[43,44] Consistent with earlier work,[45]
oxide melt drop solution calorimetry indicates that the reduced
silanol group density in steamed SPP is associated with an
enthalpic stabilization. Figure 4e shows the 훥퐻푇푟푎푛푠푖푡푖표푛 for the
starting single-unit-cell SPP and the steamed SPP along with the
reported value for conventional silicalite-1.[45] Among the listed
materials, SPP consisting of single-unit-cell nanosheets is the
least energetically stable. It is destabilized by an enthalpy of 30.5
kJ/mol SiO2 relative to α-quartz, while the 30-day steamed SPP is
26.1 kJ/mol SiO2 less stable in enthalpy than α-quartz. Although
Figure 3. Evolution of domains in steamed SPP. a) Dimensions of one hundred
MFI domains of calcined SPP (black) and 7-day steamed SPP (red) determined
using 3D TEM tomography along [100], [010] and [001] fitted to Gaussian
distributions. b) Gaussian distributions of nanosheet thickness along the b-axis
of one hundred MFI domains of calcined SPP (black), 1- (purple), 2- (brown),
5.25- (arctic), 7- (red), 11- (blue) and 30-day (green) steamed SPP as
determined by TEM; inset is the time evolution of nanosheet thickness along
[010]. The black line is a tracking line for better visualization. c) Single-unit-cell
SPP particle viewed down [001]. d) 7-day steamed coarsened SPP particle
viewed down [001]. e) 11-day steamed coarsened SPP particle viewed down
[001]; insets in c-e are idealized schematics of the house-of-cards architecture.
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202000395
Angewandte Chemie International Edition
RESEARCH ARTICLE
more enthalpically stable than the starting single-unit-cell SPP,
steamed SPP remains far from the microporous most stable
zeolitic form of the MFI topology, silicalite-1,[46] which was found
to be only 8.6 kJ/mol SiO2 less stable than α-quartz.[45]
The marginal enthalpic stabilization and consistently small
changes in Qn distribution, as well as the invariant XRD and Ar
physisorption data, are not representative of the pronounced
coarsening of the single-unit-cell domains within SPP caused by
steam treatment.
In order for the observed coarsening to take place, silica
bonds should break and reform and silicate surface species
should diffuse along the rearranging surfaces. It is remarkable
that these processes do not lead to the creation of amorphous or
dense silicates and that the evolution ceases while the material
remains enthalpically destabilized retaining its original
microporosity and mesoporosity.
A dynamic picture of the zeolite crystal in the presence of
superheated steam is emerging from these observations; the
crystalline framework rearranges itself, on timescales similar to
those typically encountered in catalyst operation, to reach a
kinetically or entropically stabilized hierarchical structure with
similar initial crystallinity, meso- and microporosity. Although the
zeolite domains within SPP particles remain nanosized, they do
not retain their 2D (single-unit-cell-thick) morphology. This
morphological evolution of domains within SPP yields a new
material and offers a new degree of freedom in controlling their
properties.
We first contrast (i) steamed SPP, coarsened from single-
unit-cell to become four-unit-cell-thick along [010] with (ii)
untreated, i.e., directly-synthesized SPP, composed of four-unit-
cell-thick nanosheets. Synthesis of the later is described in
Section SI in the Supporting Information. Figure S4 shows TEM
images that reveal significant differences; the directly-synthesized
material consists of relatively densely-packed high-aspect-ratio
nanosheets compared to more globular and less densely-packed
domains in steamed SPP. Consistent with the observed SPP
particle morphologies, the Ar physisorption isotherm of material
(ii) (included in Figure 4b) shows a slightly higher adsorbed
volume in the microporous range and a significantly lower one in
the mesoporous range relative to that of material (i). Moreover,
the less mesoporous four-unit-cell directly-synthesized SPP
(material (ii)) is found to be much more enthalpically stabilized
compared to material (i) (with 훥퐻푇푟푎푛푠푖푡푖표푛 of 12.8 kJ/mol SiO2 vs.
26.1 kJ/mol SiO2, as shown in Figure 4e). These findings suggest
that the two materials, despite having the same characteristic
length along the b-axis, have very different hierarchical
architecture: the directly-synthesized material (ii) is less
mesoporous with its high-aspect-ratio intergrown nanosheets
exposing well-defined crystal facets (dominated by (010)
surfaces), while the steam-coarsened SPP material (i) has higher
mesoporosity with lower aspect ratio nanosheets exhibiting not as
well-defined faces with increased roughness.
site densities of these four aluminosilicate SPP catalysts (thin-
SPP, steamed-thin-SPP, thick-SPP and steamed-thick-SPP,
denoted as SPP, s-SPP, tk-SPP, s-tk-SPP, respectively) as
measured by the Hofmann elimination of tert-butylamine via
reactive gas chromatography,[47] along with external Brønsted
acid site fraction (fext) determined by ethanol dehydration using
2,6-di-tert-butylpyridine (DTBP) as a titrant of external Brønsted
acid sites.[48–50] Total acid site densities change by less than 25 %
upon steam treatment indicating moderate dealumination. Turn
over frequency (TOF) of ethanol dehydration to diethyl ether in
the absence of DTBP remains similar across all materials tested
indicating that the nature of active sites as probed by this reaction
remains unaltered (Table S3). The fraction of external sites of
SPP (47 %) is much higher than that of tk-SPP (21 %) and both
conform with the expected (developed earlier based on geometric
arguments)[37] dependence on high-aspect-ratio nanosheet half-
thickness along b (xp) shown in Figure 5a. The external Brønsted
acid site fractions of SPP and tk-SPP increase upon steam
treatment from 47 % to 51 % and from 21 % to 27 %, respectively.
This nearly invariant fext is consistent with the observed
preservation of micro- and mesoporosity upon steam-induced
coarsening of SPP.
Although ethanol dehydration indicates unaltered catalytic
activity upon steam treatment, a drastically different outcome is
observed when we use benzyl alcohol etherification to dibenzyl
ether and benzyl alcohol alkylation with mesitylene as probe
reactions under the reaction conditions specified in Section SI in
the Supporting Information. The first reaction takes place both in
micro- and mesopores of SPP, while alkylation can only take
place in the mesopores (external surfaces of nanosheets).[37,51,52]
In earlier work, we developed a reaction-diffusion mathematical
model (Section SIII in the Supporting Information) for these
reactions and obtained the reaction rate and equilibrium
constants that determine the observed etherification and
alkylation rates.[37] This was accomplished using experimental
data from directly-synthesized SPP catalysts with different
nanosheet thicknesses. The alkylation and etherification rates
can be written as shown in Eq. S1 and Eq. S2 in the Supporting
Information, respectively, while the selectivity SB/P (etherification
rate over alkylation rate) is given in Eq. S3 in the Supporting
Information, which with the parameters of Table S2 becomes Eq.
S4 in the Supporting Information that is plotted in Figure 5b.[37]
We have shown that the etherification rate in SPP is free of
micropore diffusion effects (i.e., the micropore effectiveness
factor ηm is ca. 1) for nanosheet thicknesses below ca. 20 nm,
while diffusion limitations emerge for crystals with larger
characteristic diffusion lengths.[37] As discussed in Xu et al.[37] and
shown in Figure 5b for the SPP catalysts (points 1, 2 and 3),
etherification over alkylation selectivity SB/P increases as the
nanosheet thickness increases due to the reduction of the
external acid site fraction fext with nanosheet thickness (Figure 5a).
However, this trend does not continue. For larger crystals (points
4, 5 and 6 in Figure 5b), selectivity decreases, despite their even
lower fext, due to the onset of diffusion limitations reflected in ηm
decrease. As a result, the maximum selectivity that can be
obtained with directly-synthesized catalysts before steam
treatment at the specific reaction conditions cannot exceed ca. 5.
However, as shown in Figure 5b, s-SPP and s-tk-SPP exhibit
selectivities of 6.8 and 20.2, respectively, which greatly exceed
the selectivities of their parent materials before steam treatment
(2.7 and 4.2, respectively). Another SPP catalyst with
We examine next the effect of steam treatment on two
aluminosilicate SPP catalysts with Si/Al ratio of ca. 120 and 100,
one with thin (3.8 ± 1.7 nm along b) and one with thicker (7.3 ±
2.6 nm) intergrown nanosheets, which after steam treatment
evolve to (7.5 ± 2.3 nm) and (9.2 ± 2.1 nm), respectively. As with
the all-silica SPP, these morphological changes take place while
XRD patterns and Ar physisorption isotherms remain invariant
(Figure S5), and only small changes are observed by 29Si and 27Al
MAS NMR (Figure S6). In Table 1, we include total Brønsted acid
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202000395
Angewandte Chemie International Edition
RESEARCH ARTICLE
Table 1. Effect of steam-induced coarsening on Brønsted acid catalysis by aluminosilicate SPP. Brønsted acid site density, mass-normalized rate of diethyl
ether formation from ethanol dehydration, mass-normalized rate of dibenzyl ether formation from benzyl alcohol etherification, mass-normalized rate of 1,3,5-
trimethyl-2-benzylbenzene formation from benzyl alcohol alkylation with mesitylene, external Brønsted acid sites’ fraction (fext), half-thickness along the b-axis
(xp) and selectivity (SB/P) of thin-SPP (SPP), steamed-thin-SPP (s-SPP), thick-SPP (tk-SPP) and steamed-thick-SPP (s-tk-SPP) materials.
[b]
[c]
[d]
Material
Brønsted acid
site density[a]
[µmol g-1]
Initial rate of diethyl ether
formation
[µmol g-1 min-1]
Rate of dibenzyl ether
formation
[µmol L-1 s-1 g-1]
Rate of 1,3,5-trimethyl-2-benzylbenzene
formation
[µmol L-1 s-1 g-1]
fext
xp
SB/P
[%]
[nm]
SPP
135
106
158
133
26.0
15.3
26.1
22.6
15.6
9.7
13.1
7.6
5.8
1.4
3.1
0.4
47
51
21
27
1.9
3.8
3.7
4.6
2.7
6.8
4.2
s-SPP
tk-SPP
s-tk-SPP
20.2
[a] Determined by Hofmann elimination of tert-butyl amine via reactive gas chromatography.[47] [b] Determined by ethanol dehydration using 2,6-di-tert-
butylpyridine (DTBP) as a titrant of the external Brønsted acid sites.[48] fext = (1 – ((rate of diethyl ether formation with DTBP after saturation)/(steady state rate
of diethyl ether formation without DTBP))) x 100 %. [c] Determined using TEM for one hundred MFI domains and corresponds to the average of the half-
thickness along the b-axis of the nanosheets. [d] Determined as the ratio of dibenzyl ether formation rate at 5 % conversion of benzyl alcohol by 1,3,5-trimethyl-
2-benzylbenzene formation rate at 5 % conversion of benzyl alcohol.[37]
intermediate nanosheet thickness (i-SPP) is also included in
Figure 5b and follows a similar trend with the selectivity increasing
from 4.0 to 8.6 after steam treatment. This significant increase in
selectivity upon steam treatment is due to the 75 % to 90 % redu-
ction in alkylation mass-normalized rate, while the decrease in the
rate of etherification ranges from 38 % to 42 % (Table 1). Part of
these losses in catalytic activity (up to 25 %) can be attributed to
the overall reduction in the acid site concentration as determined
by the Hofmann elimination of tert-butylamine. The activity losses
for alkylation greatly exceed 25 % and can only be attributed to
loss of external (mesopore) catalytic activity since this reaction
only takes place on the external surface of the nanosheets.
Figure 5. Effect of steam-induced coarsening on Brønsted acid catalysis by
Similar loss of external catalytic activity should also be
experienced for the etherification rate. However, based on a
model developed earlier,[37] the major fraction (ca. 75 %) of
etherification reaction is being contributed by catalysis in the
micropores even for the thinnest nanosheets; therefore, the
overall (due to external and micropore acid sites) etherification
rate does not decrease as much as the alkylation rate.
aluminosilicate SPP. a) Fraction of external Brønsted acid sites of thin-SPP
(SPP; black), steamed-thin-SPP (s-SPP; red), thick-SPP (tk-SPP; black) and
steamed-thick-SPP (s-tk-SPP; red) zeolites (fext
)
determined by ethanol
function of the
dehydration using 2,6-di-tert-butylpyridine (DTBP)[48] as
a
characteristic diffusion length corresponding to the nanosheet half-thickness
along the b-axis (xp) as determined by TEM for one hundred MFI domains. 1, 2
and 3 correspond to previously reported[37] SPP (without steam treatment) with
xp = 2 ± 1 nm, xp = 3 ± 1 and xp = 4 ± 1 nm, respectively. Error bars represent
the standard deviation of the mean of measured xp. The solid straight line (black)
corresponds to the geometric model developed earlier[37] for high-aspect-ratio
nanosheets: fext = d/xp, where d = 0.8 ± 0.4 nm (the accessible thickness by
DTBP assuming that the Brønsted acid sites are randomly distributed). The grey
region represents the error in the determination of the fraction of external
Brønsted acid sites resulting from the error in the accessible thickness by DTBP.
b) Selectivity (SB/P) (etherification rate over alkylation rate) as a function of the
external acid site fraction (fext) and the effectiveness factor (ηm) of thin-SPP
(SPP; black), steamed-thin-SPP (s-SPP; red), thick-SPP (tk-SPP; black),
steamed-thick-SPP (s-tk-SPP; red), SPP with intermediate nanosheet thickness
before (i-SPP; black) and after steam treatment (s-i-SPP; red) zeolites. 1, 2 and
3 are the same as those in Figure 5a and 4, 5 and 6 correspond[35,37] to calcined
untreated conventional MFI zeolites with a nominal particle size of 0.2 µm, 1.4
µm and 17 µm, respectively. The solid line (black) corresponds to the selectivity
(SB/P) expression developed earlier[37] and given in Eq. S4. The grey region
represents the error in the selectivity determination resulting from the 95 %
confidence intervals of the estimated modeled parameters listed in Table S2.[37]
What causes the large reduction of catalytic alkylation rates
on the external surface of nanosheets upon their steam-induced
coarsening, given that the nature of active sites remains unaltered
as probed by the ethanol dehydration to diethyl ether in the
absence of DTBP (Table S3) and confirmed by the results of
infrared (IR) spectroscopy using pyridine as a base probe shown
in Figure S7? In our earlier work, we demonstrated that the
alkylation rate constant depends strongly on the structure of the
external surface of the zeolite by quantifying a 10-fold increase in
alkylation rate on MWW nanosheets compared to that on the
surface of MFI nanosheets.[37] It is plausible that the reduced
abundance of (010) faces and their replacement with other not
well-defined facets create topologically and chemically distinct
environments around the external acid sites that although do not
affect the catalytic activity for reactions like ethanol dehydration,
they compromise it for reactions involving larger and less polar
molecules like mesitylene. The rough facets of the coarsened
external surfaces of the steamed catalysts may have affected the
accommodation of mesitylene and probably the stabilization of
the bulky transition state of the alkylation reaction resulting in
reduced rates.
Conclusion
Steam-induced coarsening of single-unit-cell nanosheets
reduces the external surface catalytic alkylation rates while
maintains kinetically-controlled micropore etherification catalysis
unaltered. This finding is of practical significance as it provides a
new method to regulate external surface catalytic activity while
preserving micropore reaction rates in zeolite nanosheets, for
which, due to their nm-thickness, conventional approaches, like
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202000395
Angewandte Chemie International Edition
RESEARCH ARTICLE
[15] G. T. Kerr, J. Phys. Chem. 1967, 71, 4155–4156.
aluminum-zoning, are difficult to implement. Of fundamental
significance is that aluminosilicate MFI nanosheets catalysts can
exhibit external surface structure sensitivity, e.g., with (010) facets
being more reactive than coarser MFI surfaces, for certain
reactions. Electron microscopy proves to be the necessary
characterization method to link these pronounced catalytic
performance changes to nanoscale coarsening. Data from other
characterization techniques, including porosimetry, XRD and
NMR remain invariant, while probe reactions like Hofmann
elimination of tert-butylamine and ethanol dehydration in the
absence and presence of DTBP do not exhibit external surface
structure sensitivity.
[16] C. J. Heard, L. Grajciar, C. M. Rice, S. M. Pugh, P. Nachtigall, S. E.
Ashbrook, R. E. Morris, Nat. Commun. 2019, 10, 4690.
[17] S. Malola, S. Svelle, F. L. Bleken, O. Swang, Angew. Chemie Int. Ed.
2012, 51, 652–655.
[18] M. C. Silaghi, C. Chizallet, E. Petracovschi, T. Kerber, J. Sauer, P.
Raybaud, ACS Catal. 2015, 5, 11–15.
[19] R. Gounder, A. J. Jones, R. T. Carr, E. Iglesia, J. Catal. 2012, 286, 214–
223.
[20] L. R. Aramburo, J. Ruiz-Martínez, J. P. Hofmann, B. M. Weckhuysen,
Catal. Sci. Technol. 2013, 3, 1208–1214.
[21] L. R. Aramburo, E. de Smit, B. Arstad, M. M. van Schooneveld, L.
Sommer, A. Juhin, T. Yokosawa, H. W. Zandbergen, U. Olsbye, F. M. F.
de Groot, et al., Angew. Chemie Int. Ed. 2012, 51, 3616–3619.
[22] D. E. Perea, I. Arslan, J. Liu, Z. Ristanović, L. Kovarik, B. W. Arey, J. A.
Lercher, S. R. Bare, B. M. Weckhuysen, Nat. Commun. 2015, 6, 7589.
[23] C. A. Fyfe, J. L. Bretherton, L. Y. Lam, J. Am. Chem. Soc. 2001, 123,
5285–5291.
Acknowledgements
[24] M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki, R. Ryoo, Nature 2009,
461, 246–249.
This work was supported by the Catalysis Center for Energy
Innovation, an Energy Frontier Research Center funded by the
U.S. Department of Energy, Office of Science, Basic Energy
Sciences under Award DE-SC0001004; and, in its earlier stages,
by ADMIRE (Abu Dhabi–Minnesota Institute for Research
Excellence) through the Petroleum Institute (now Khalifa
University of Science and Technology). Parts of this work were
carried out in the Characterization Facility, University of
Minnesota, which receives partial support from NSF through the
MRSEC program. Y.G. acknowledges support from ADNOC (Abu
Dhabi National Oil Company). We thank Dr. S. Hwang (Caltech)
for performing the solid-state NMR experiments.
[25] K. Kim, R. Ryoo, H. D. Jang, M. Choi, J. Catal. 2012, 288, 115–123.
[26] W. J. Roth, P. Nachtigall, R. E. Morris, J. Čejka, Chem. Rev. 2014, 114,
4807–4837.
[27] M. E. Davis, Nature 2002, 417, 813–821.
[28] M. A. Snyder, M. Tsapatsis, Angew. Chem. Int. Ed 2007, 46, 7560–7573.
[29] C. M. Lew, R. Cai, Y. Yan, Acc. Chem. Res. 2009, 43, 210–219.
[30] M. V. Opanasenko, W. J. Roth, J. Čejka, Catal. Sci. Technol. 2016, 6,
2467–2484.
[31] M. Hartmann, A. G. Machoke, W. Schwieger, Chem. Soc. Rev. 2016, 45,
3313–3330.
[32] M. S. Holm, E. Taarning, K. Egeblad, C. H. Christensen, Catal. Today
2011, 168, 3–16.
[33] J. Pérez-Ramírez, C. H. Christensen, K. Egeblad, C. H. Christensen, J.
C. Groen, Chem. Soc. Rev. 2008, 37, 2530–2542.
Keywords: catalysis • electron tomography • stability • two
dimensional materials • zeolites
[34] L. H. Chen, X. Y. Li, J. C. Rooke, Y. H. Zhang, X. Y. Yang, Y. Tang, F.
S. Xiao, B. L. Su, J. Mater. Chem. 2012, 22, 17381–17403.
[35] X. Zhang, D. Liu, D. Xu, S. Asahina, K. A. Cychosz, K. V. Agrawal, Y. Al
Wahedi, A. Bhan, S. Al Hashimi, O. Terasaki, M. Thommes, M. Tsapatsis,
Science 2012, 336, 1684–1687.
[1]
[2]
[3]
B. Elyassi, X. Zhang, M. Tsapatsis, Microporous Mesoporous Mater.
2014, 193, 134–144.
S. van Donk, A. H. Janssen, J. H. Bitter, K. P. de Jong, Catal. Rev. 2003,
45, 297–319.
[36] D. Xu, G. R. Swindlehurst, H. Wu, D. H. Olson, X. Zhang, M. Tsapatsis,
Adv. Funct. Mater. 2014, 24, 201–208.
T. Ennaert, J. Geboers, E. Gobechiya, C. M. Courtin, M. Kurttepeli, K.
Houthoofd, C. E. A. Kirschhock, P. C. M. M. Magusin, S. Bals, P. A.
Jacobs, et al., ACS Catal. 2015, 5, 754–768.
[37] D. Xu, O. Abdelrahman, S. Hyun Ahn, Y. Guefrachi, A. Kuznetsov, L.
Ren, S. Hwang, M. Khaleel, S. Al Hassan, D. Liu, et al., AIChE J 2019,
65, 1067–1075.
[4]
[5]
[6]
A. Vjunov, J. L. Fulton, D. M. Camaioni, J. Z. Hu, S. D. Burton, I. Arslan,
J. A. Lercher, Chem. Mater. 2015, 27, 3533–3545.
K. Chen, J. Kelsey, J. L. White, L. Zhang, D. Resasco, ACS Catal. 2015,
5, 7480–7487.
[38] L. Ren, Q. Guo, P. Kumar, M. Orazov, D. Xu, S. M. Alhassan, K. A.
Mkhoyan, M. E. Davis, M. Tsapatsis, Angew. Chemie Int. Ed. 2015, 54,
10848–10851.
[39] P. A. Midgley, R. E. Dunin-Borkowski, Nat. Mater. 2009, 8, 271–280.
[40] H. Friedrich, P. E. de Jongh, A. J. Verkleij, K. P. de Jong, Chem. Rev.
2009, 109, 1613–1629.
L. R. Aramburo, L. Karwacki, P. Cubillas, S. Asahina, D. A. M. de Winter,
M. R. Drury, I. L. C. Buurmans, E. Stavitski, D. Mores, M. Daturi, et al.,
Chem. A Eur. J. 2011, 17, 13773–13781.
[41] Y. Wei, T. E. Parmentier, K. P. de Jong, J. Zečević, Chem. Soc. Rev.
2015, 44, 7234–61.
[7]
[8]
M. C. Silaghi, C. Chizallet, P. Raybaud, Microporous Mesoporous Mater.
2014, 191, 82–96.
[42] J. R. Kremer, D. N. Mastronarde, J. R. McIntosh, J. Struct. Biol. 1996,
116, 71–76.
L. Karwacki, D. A. M. de Winter, L. R. Aramburo, M. N. Lebbink, J. A.
Post, M. R. Drury, B. M. Weckhuysen, Angew. Chemie Int. Ed. 2011, 50,
1294–1298.
[43] A. Zecchina, S. Bordiga, G. Spoto, L. Marchese, G. Petrini, G. Leofanti,
M. Padovan, J. Phys. Chem. 1992, 96, 4991–4997.
[9]
D. Mehlhorn, R. Valiullin, J. Kärger, K. Cho, R. Ryoo, Microporous
Mesoporous Mater. 2012, 164, 273–279.
[44] V. Nikolakis, M. Tsapatsis, D. G. Vlachos, Langmuir 2003, 19, 4619–
4626.
[45] P. M. Piccione, C. Laberty, S. Yang, M. A. Camblor, A. Navrotsky, M. E.
Davis, J. Phys. Chem. B 2000, 104, 10001–10011.
[10] L. R. Aramburo, Y. Liu, T. Tyliszczak, F. M. F. de Groot, J. C. Andrews,
B. M. Weckhuysen, ChemPhysChem 2013, 14, 496–499.
[11] S. M. T. Almutairi, B. Mezari, E. A. Pidko, P. C. M. M. Magusin, E. J. M.
Hensen, J. Catal. 2013, 307, 194–203.
[46] J. D. Rimer, O. Trofymluk, A. Navrotsky, R. F. Lobo, D. G. Vlachos,
Chem. Mater. 2007, 19, 4189–4197.
[12]
M. Milina, S. Mitchell, N. L. Michels, J. Kenvin, J. Pérez-Ramírez, J.
Catal. 2013, 308, 398–407.
[47] O. A. Abdelrahman, K. P. Vinter, L. Ren, D. Xu, R. J. Gorte, M. Tsapatsis,
P. J. Dauenhauer, Catal. Sci. Technol. 2017, 7, 3831–3841.
[48] D. Liu, A. Bhan, M. Tsapatsis, S. Al Hashimi, ACS Catal. 2011, 1, 7–17.
[13] L. Zhang, K. Chen, B. Chen, J. L. White, D. E. Resasco, J. Am. Chem.
Soc. 2015, 137, 11810–11819.
[49] A. Corma, V. Fornés, L. Forni, F. Márquez, J. Martınez-Triguero, D.
́
[14] R. M. Ravenelle, F. Schüβler, A. D’Amico, N. Danilina, J. A. van
Bokhoven, J. A. Lercher, C. W. Jones, C. Sievers, J. Phys. Chem. C
2010, 114, 19582–19595.
Moscotti, J. Catal. 1998, 179, 451–458.
[50] K. Gꢀragꢀra-Marek, K. Tarach, M. Choi, J. Phys. Chem. C 2014, 118,
12266–12274.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202000395
Angewandte Chemie International Edition
RESEARCH ARTICLE
[51]
D. Liu, X. Zhang, A. Bhan, M. Tsapatsis, Microporous Mesoporous
Mater. 2014, 200, 287–290.
[52] A. Korde, B. Min, Q. Almas, Y. Chiang, S. Nair, C. W. Jones,
ChemCatChem 2019, 11, 4548–4557.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202000395
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
A two-dimensional zeolite undergoes pronounced steam-induced coarsening detectable only by electron microscopy, while it evades
detection by commonly used diffraction, porosimetry and catalytic probe reaction methods. For a reaction involving bulky molecules,
the observed coarsening remarkably alters external surface catalytic activity, while preserving microporous catalysis unaltered,
demonstrating a new method to fine-tune selectivity.
8
This article is protected by copyright. All rights reserved.