Kinetics and mechanism of benzene, toluene, and xylene methylation over H-MFI

Article abstract of DOI:10.1021/cs400377b

The methylation of benzene, toluene, para-xylene, and ortho-xylene over MFI structured H-ZSM-5 and mesoporous self-pillared pentasil (H-SPP) with dimethyl ether (DME) at low conversions (<0.1%) and high DME:aromatic ratios (>30:1) showed linear rate dependencies on aromatic pressure and zero dependence on DME pressure for benzene and toluene. These results are consistent with studies performed for olefin methylation, and are indicative of a zeolite surface covered in DME-derived species reacting with benzene or toluene in the rate-determining step. Saturation in the reaction rate was observed in xylene pressure dependence experiments (at 473 K, <5 kPa xylene); however, enhancement in the reaction rate was not observed when comparing ～1 μm crystallite H-ZSM-5 and 2-7 nm mesopore H-SPP, indicating that xylene methylation proceeds in the absence of diffusion limitations. Simultaneous zero-order rate dependencies on xylene and DME pressures are described by a model based on adsorption of xylene onto a surface methylating species. This model is consistent with observed secondary kinetic isotope effects (k _H/k_D = 1.25-1.35) and extents of d₀, d ₃, and d₆ DME formation in the effluent because of isotopic scrambling between unlabeled and d₆ DME when co-fed with aromatics over H-ZSM-5. Post-reaction titration of surface species with water after desorption of physisorbed intermediates showed a 1:1 evolution of methanol to Al present in the catalyst, indicating the presence and involvement of surface methoxides during steady-state methylation of aromatics species.

Full text of DOI:10.1021/cs400377b

Research Article
pubs.acs.org/acscatalysis
Kinetics and Mechanism of Benzene, Toluene, and Xylene
Methylation over H‑MFI
†
‡
,†
Ian Hill, Andre Malek, and Aditya Bhan*
†Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, Minnesota 55455, United States
‡Feedstocks and Energy, The Dow Chemical Company, Freeport, Texas 77541, United States
*
S Supporting Information
ABSTRACT: The methylation of benzene, toluene, para-
xylene, and ortho-xylene over MFI structured H-ZSM-5 and
mesoporous self-pillared pentasil (H-SPP) with dimethyl ether
(
DME) at low conversions (<0.1%) and high DME:aromatic
ratios (>30:1) showed linear rate dependencies on aromatic
pressure and zero dependence on DME pressure for benzene
and toluene. These results are consistent with studies
performed for olefin methylation, and are indicative of a
zeolite surface covered in DME-derived species reacting with
benzene or toluene in the rate-determining step. Saturation in
the reaction rate was observed in xylene pressure dependence
experiments (at 473 K, <5 kPa xylene); however, enhancement
in the reaction rate was not observed when comparing ∼1 μm crystallite H-ZSM-5 and 2−7 nm mesopore H-SPP, indicating that
xylene methylation proceeds in the absence of diffusion limitations. Simultaneous zero-order rate dependencies on xylene and
DME pressures are described by a model based on adsorption of xylene onto a surface methylating species. This model is
consistent with observed secondary kinetic isotope effects (k /k = 1.25−1.35) and extents of d , d , and d DME formation in
H
D
0
3
6
the effluent because of isotopic scrambling between unlabeled and d DME when co-fed with aromatics over H-ZSM-5. Post-
6
reaction titration of surface species with water after desorption of physisorbed intermediates showed a 1:1 evolution of methanol
to Al present in the catalyst, indicating the presence and involvement of surface methoxides during steady-state methylation of
aromatics species.
KEYWORDS: aromatic methylation, H-ZSM-5, mesoporous zeolite, Brønsted acid catalysis, methanol-to-gasoline conversion
1
. INTRODUCTION
The use of aromatic co-feeds as a means of studying the
evolution of the aromatic hydrocarbon pool has been extensively
reported, mainly in using isotopically labeled reagents to track
the pathways available for chemical transformations of aromatics,
and specifically their role in the formation of light
1
Since its discovery in 1977 by Mobil, −3 consensus on the
mechanism governing the methanol-to-hydrocarbons (MTH)
process over zeolite or zeotype catalysts has remained elusive.
Despite this, aspects of the MTH chemistry have garnered
agreement among researchers. Direct C−C coupling of
methanol in MTH is regarded as negligible compared to indirect
methylation of olefin and aromatic impurities, as shown via
increasing induction periods through rigorous reactant purifica-
tion, isotopic studies, and high activation barriers for direct
coupling inferred from computational chemistry studies.
The development of indirect mechanisms
widely accepted “hydrocarbon pool” mechanism,
describes the growth, dealkylation, and interconversion of
25,28,33−35
25
olefins.
For example, Mikkelsen et al. have shown
through observed isotopic distributions in product olefins in the
13
methylation of aromatics with C methanol that all ethene and
majority propene are formed through the cracking of arenes on
H-ZSM-5, H-MOR, and H-BEA catalysts, and not via the direct
4
5−7
coupling of C species. Studies switching from an unlabeled
1
8
−10
13
methanol feed to a C labeled feed at steady-state MTH
1
1−13
1
3
has led to the
conditions have shown that aromatics incorporate C at similar
14−16
13
which
rates to ethene and higher olefins incorporate C at similar rates
to propene indicating that ethene and aromatics, and propene
and higher olefins originate from similar reaction steps that are
1
7−20
5,21−30
olefin
and aromatic species
within the confined
33−36
inorganic framework as being responsible for observed product
distributions. Recent work has described the role and
contributions of constituent reactions pertinent to MTH
chemistry, indicating that the relative rates of these steps yield
particular termination products from olefin and aromatic
hydrocarbon pools, which can be quantified to determine
distinct from each other.
This observation shows that
ethene primarily originates from the aromatic hydrocarbon pool,
and that the methylation of propene is mostly responsible for the
Received: May 21, 2013
Revised: July 11, 2013
Published: August 5, 2013
3
1,32
which cycle is relatively more productive.
©
2013 American Chemical Society
1992
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
formation of higher olefins over SAPO-34, BEA,
Research Article
36
33,34
and
temperature ramp) for 4 h prior to cooling to reaction
temperatures.
3
5
MFI. The predominant methylbenzenes responsible for the
formation of light olefins have been shown to be zeolite
Steady-State Aromatic Methylation Reactions. Ben-
framework dependent, and specifically for MFI to comprise C −
zene, toluene, para-xylene, and ortho-xylene methylation
8
3
−1
C aromatics.
reactions with DME were performed using 1.67 cm s total
reactant flow rates with 0.68 bar DME (Matheson, CP grade),
0.01 bar aromatic (Sigma Aldrich, 99.9% purity), and 0.03 bar
10
The quantitative investigation of the kinetics of aromatic
methylation has been a relatively recent focus, and allows for the
direct comparison of reaction rates and activation energies
between the aromatic and olefin hydrocarbon pools. A
comparison of benzene methylation rate parameters with
CH (Minneapolis O , CP grade) as an internal standard.
4 2
Methylation reactions over H-ZSM-5 used 50 mg of catalyst at
373 K for benzene, 10 mg catalyst at 403 K for toluene, and 1 mg
catalyst at 473 K for xylene to maintain <0.1% aromatic
conversion. Methylation over H-SPP used 1 mg catalyst at 473 K
for benzene and xylene, and 433 K for toluene co-feeds. Samples
19
propene methylation run at similar reaction conditions has
−1
shown that activation energies (58 and 69 kJ mol ) and first-
−3
−1 −1
−1
order rate constants (4.8 and 4.5 × 10 mol g
h mbar ) are
similar, and thus the aromatic and olefin hydrocarbon pools are
of 1 mg of zeolite were achieved using 10 mg of a homogenized
1
1
00 mg g− mixture solid solution with quartz sand, followed by
both contributors to the product distribution observed over H-
3
7
further dilution with quartz sand to 1 g total mass. Pressure
dependence reactions were run from 0.002−0.05 bar aromatic
and 0.29 to 0.68 bar DME pressures while keeping the partial
pressure of the other constant by adjusting He flow rates to
ZSM-5. Recent studies have also quantified the effects of pore
size and degree of methylation of co-feed aromatics on the
38
formation rate of light hydrocarbons and higher aromatics.
Density functional theory (DFT) calculations using a
coadsorbed methanol and toluene precursor over 4T clusters
3
−1
maintain 1.67 cm s total flow. Products were monitored using
an Agilent 5890 gas chromatograph equipped with an HP-1
capillary column attached to an FID detector.
(one alumina and three silica tetrahedra to represent the zeolite)
have shown that activation barriers for toluene methylation are of
−
1
39−42
Post-Reaction Zeolite Surface Titration with H O.
the order of 180−195 kJ mol .
Barriers calculated for the
2
Steady-state methylation reactions were run over 200 mg H-
SPP (in the absence of quartz sand diluent) at 358 K for benzene,
addition of methanol to toluene using hybrid ONIOM methods
−1
on 46T clusters of MFI are ∼163 kJ mol , showing that
framework effects are important in stabilizing MTH inter-
3
43 K for toluene, and 353 K for para-xylene co-feeds. After 2 h
3
2
3
time-on-stream, flow over the catalyst was switched to 1.67 cm
mediates and transition states.
−
1
s
He for 0.5 h to purge residual reactant gases in the reactor
In this work, a quantitative comparison of the kinetics and
lines and catalyst bed and reactor temperature was increased to
423 K at a rate of 0.17 K s− to remove physisorbed species from
the catalyst surface. Flow to the reactor was then switched to 10
mechanisms of aromatic methylation to previous work
1
4
3−45
investigating the methylation of C −C olefins is provided.
2
4
Rate constants and activation energies have been determined for
benzene, toluene, and xylene (BTX) methylation with dimethyl
ether (DME) over H-ZSM-5 and mesoporous self-pillared
3
−1
kPa H O in He with a total flow of 1.67 cm s to remove
2
chemisorbed *CH species as methanol. Effluent compositions
3
46
were monitored and quantified using an online Cirrus MKS
quadrupole mass spectrometer using CH4 as an internal
standard.
pentasil (H-SPP). Based on the observed kinetics of aromatic
methylation we infer the consequences of alkyl substitution on
methylation rates in the limit where the coreactant with DME
approaches a commensurate size as the zeolite pore. The effects
of transport are discussed in light of increasing diffusion
limitations with increasing aromatic size. Elementary-step
aromatic methylation of benzene, toluene, para-xylene, and
ortho-xylene is investigated to also probe the identity and
reactivity of surface species present during aromatic methylation
reactions.
In-Situ d DME/DME Switching. Steady-state methylation
6
reactions over H-SPP were performed at identical catalyst
weights, total flow rates, and reaction temperatures as described
in the H O titration section, with the exception of using 0.20 bar
2
DME pressure instead of 0.68 bar DME. After 2 h time-on-
stream, DME composition was changed to 0.10 bar d DME and
0
0
.10 bar d DME for 1 h prior to returning to 0.2 bar d DME.
6
0
Reaction rates were monitored via GC, and effluent composition
was monitored via online MS.
2. MATERIALS AND METHODS
3
. RESULTS AND DISCUSSION
Catalyst Preparation. Zeolite H-ZSM-5 (MFI, Zeolyst, Si/
Al = 42.6) and self-pillared pentasil (MFI, Si/Al = 75, 2−7 nm
The kinetics of benzene, toluene, para-xylene, and ortho-xylene
methylation have been systematically studied to derive reactant
pressure and temperature dependencies of the reaction rate at
differential reaction conditions in the absence of secondary
reactions. Based on these pressure dependencies and rate
mesopores; sample preparation and characterization described
4
6
elsewhere ) were pressed, crushed, and sieved to 180 to 425 μm
3
aggregate sizes. Catalysts were then treated in dry air (1.67 cm
s
−1
NTP, ultrapure, Minneapolis Oxygen) at 773 K for 8 h.
4
3,44
Thermal treatment in air was repeated to regenerate self-pillared
pentasil (H-SPP) samples between reactions to conserve sample.
Catalysts were diluted in quartz sand (Sigma Aldrich) to 1 g
total mass and supported on a quartz frit within a 10 mm quartz
U-shaped reactor tube where isothermal reaction conditions
were maintained using a National Electric FA120 furnace
connected to a 96 Series Watlow temperature controller.
Catalyst temperatures were measured using a K-type thermo-
couple located in a thermowell penetrating the center of the
equations, in conjunction with isotopic
and surface
45
titration methods used previously, a description of the effect
of aromatic substitution on methylation rates has been developed
to describe how the mechanistic pathway changes with increasing
pressure to one that involves the co-adsorption of the aromatic
onto a surface methoxide.
The kinetic relevance of values derived from experiments using
1 mg of catalyst have been considered as reactant stream
bypassing catalyst particles can lead to deviations in chemical
conversion and, consequently, mechanistic information. Berger
3
−1
catalyst bed. Samples were treated in flowing He (1.67 cm s ,
ultrapure, Minneapolis Oxygen) at 773 K (0.033 K s−
1
et al. have shown that there is no measurable effect of dilution
47
1
993
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
Research Article
Figure 1. DME methylation of benzene over H-ZSM-5, 50 mg at 373 K, (A and B) and H-SPP, 1 mg at 473 K (C and D). Benzene pressure dependence
■) and DME pressure dependence (□) are shown in the left panels, temperature dependence is shown in the panels to the right.
(
for an irreversible first order reaction at conversions less than
0% through a systematic investigation of deviation of chemical
constant within a factor of 0.72 of studies that were performed
with 0.037 bar methanol and 0.017 bar benzene at 623 K over H-
ZSM-5, where parameters were derived by extrapolating
1
conversion caused by dilution of the catalyst bed with inert
particles in gas−solid systems. As the conversions in this study
were kept below 0.1% and previous work has shown that olefin
methylation systems proceed via irreversible first-order kinetics
37
conversions to zero contact time with the catalyst.
A comparison with DFT calculations performed using hybrid
4
8,49
19,43
functionals
and previous experimental studies
shows
4
3−45
47
that benzene methylation proceeds with similar apparent
over H-ZSM-5
the scenario detailed by Berger et al. is
−1
activation barriers (62−77 kJ mol ) and rate constants (3.7−
directly applicable to the kinetics discussed in this work.
+
−1
−1
5
.8 [C H ][H ] h bar at 373 K) as propene methylation. The
Benzene Methylation on H-ZSM-5. The methylation of
benzene with DME yielded first-order pressure dependence in
benzene and zero-order dependence in DME (Figure 1, panels A
and B). These pressure dependencies are consistent with the
rapid formation and predominant surface coverage of DME-
derived species that react with benzene in the rate-determining
step, as observed for C −C methylation with DME or
4 8
similar rate of propene and benzene methylation could be
explained through similarities in the local structure of the two
molecules, namely, the secondary substitution about the
carbocation formed upon methylation in the absence of
additional alkyl substituents to inductively donate electron
density. This result is not intuitive on the basis that apparent
barriers include adsorption enthalpies, which are calculated to be
2
4
1
9,20,43−45
methanol.
First-order rate constants and activation
−1
37
−
94 kJ mol at the wB97x-D/6-31 +g(d) level of theory for
energies (Table 1) show an identical activation energy and a rate
−1
benzene and −53 kJ mol using MP2/DFT with periodic
boundary conditions for propene on H-ZSM-5, indicating a
49
Table 1. Apparent Activation Energies and Rate Constants at
−1
greater intrinsic barrier for benzene methylation by ∼40 kJ mol
3
73 K for Aromatic Methylation Reactions over H-ZSM-5 and
compared to that of propene. Experimental enthalpies of
adsorption obtained by generating adsorption isochores in low
dead-volume systems estimate that benzene has a heat of
a
H-SPP
H-ZSM-5
H-SPP
−1
adsorption of −55 to −75 kJ mol on H-ZSM-5, depending on
Ea
kJ/mol)
k373
Ea
k373
(h bar)
50,51
(h bar)−
1
(kJ/mol)
−1
unit cell loading;
however these values represent an upper
(
bound compared to propene adsorption, and this study reflects
adsorption on a Brønsted acid site rather than co-adsorbing on
surface-bound methanol-derived species. The noted consistency
in benzene and propene methylation rates on HMFI results in
part because of the higher adsorption and hence, higher surface
coverage of benzene at comparable temperature and pressure
conditions.
benzene
toluene
58 ± 3
52 ± 4
6.8
58 ± 2
47 ± 3
44 ± 4
62 ± 3
34 ± 3
63 ± 4
8.3
48.1
31.4
12.5
10.1
3.0
para-xylene 0.05 bar
.003 bar
0.05 bar
.003 bar
0
62 ± 3
62 ± 4
13.4
3.1
ortho-
xylene
0
4.0
a
Listed pressures denote xylene reactant pressure at which temper-
ature dependence reactions were run.
Toluene Methylation on H-ZSM-5. Toluene methylation
with DME over H-ZSM-5 showed first-order dependence in
1
994
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
Research Article
Figure 2. DME methylation of toluene over H-ZSM-5, 10 mg at 403 K, (A and B) and H-SPP, 1 mg at 433 K (C and D). Toluene pressure dependence
■) and DME pressure dependence (□) are shown in the left panels, temperature dependence is shown on the panels to the right.
(
Figure 3. (A) Xylene pressure dependencies for para-xylene over H-MFI(□) and H-SPP(■), and ortho-xylene over H-MFI(Δ) and H-SPP(▲). (B)
DME pressure dependencies for para-xylene over H-MFI(□) and H-SPP(■), and ortho-xylene over H-MFI(Δ) and H-SPP(▲). (C) Temperature
dependence of DME methylation of para-xylene over 1 mg H-ZSM-5 (□) and ortho-xylene over 1 mg H-ZSM-5 (Δ). (D) Temperature dependence of
DME methylation over 1 mg H-SPP at 0.002 bar para-xylene (■), and 1 mg H-SPP at 0.05 bar para-xylene (□), and 1 mg SPP at 0.002 bar ortho-xylene
(
▲), and 1 mg SPP at 0.05 bar ortho-xylene (Δ).
1
995
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
Research Article
aromatic and zero-order dependence in DME partial pressures in
agreement with observed kinetics for benzene methylation,
indicating that the zeolite surface is predominantly covered with
DME-derived species reacting with toluene in the rate-
determining step. The first-order rate constant for toluene
methylation on H-ZSM-5 (Table 1) is a factor of 1.5 larger
compared to a first-order rate constant calculated from
saturate with respect to xylene partial pressure at the reaction
conditions used in this study (Figures 1−3). The origin of this
behavior could arise from (i) diffusion limitations, (ii) a change in
the rate-determining step, or (iii) a change in the predominant
surface species prior to the rate-determining step.
Diffusion of Aromatics in H-ZSM-5. The diffusion of
aromatic molecules in zeolites has been extensively studied using
52
55,56
50,51,57−64
previously reported rates at 373 K, where Vinek et al. co-fed
.5 kPa toluene and 3.5 kPa methanol over H-ZSM-5 at 373−773
spectroscopy,
experimental methods,
and theoreti-
The effect of coking on diffusion in zeolites
6
5,66
2
cal calculations.
5
2
K. Vinek et al. also reported activation energies of 52−85 kJ
mol , which is in good agreement with 68 kJ mol derived from
toluene methylation in a simulated riser reactor, 47 kJ mol
from a fixed fluidized bed reactor from 573−673 K, and 52 kJ
mol reported in this work (Table 1).
has shown that coking has little effect on the selectivity toward
para-xylene during toluene methylation reactions on H-ZSM-5
as inferred using infrared (IR) spectroscopy and reaction rate
−
1
−1
53
−1
54
55
measurements. A model for a zero-length column (ZLC)
−1
59
experimental system, developed by Ruthven and Vidoni, has
The methylation of toluene proceeds with a rate constant of 48
allowed for this experimental technique to account for combined
kinetic effects of surface resistance and internal diffusion
limitations to mass transfer. Diffusion coefficients for benzene,
ortho- and para-xylene over H-ZSM-5 and silicalite using ZLC
+
−1 −1
−1
[
Xylene][H ]
h bar at 373 K and agrees well with n-butene
+
−1 −1
−1 45
methylation rate constants of 41−90 [C +C ][H ]
and the activation barrier of 52 kJ mol agrees with butene
h
bar ,
5
6
−1
45
19
−1
57,58
methylation barriers of 44−49 and 48 kJ mol from previous
experimental studies. The effective change in kinetics from
benzene to toluene are similar to that of propene and n-butene,
respectively, indicating that the addition of a methyl group in
these cases has a similar effect on stabilizing reaction
intermediates, namely, inductive electron donation, despite the
different extended structure of aromatics compared to olefins.
Toluene heats of adsorption derived from equilibrium
have been calculated using this model for MFI frameworks.
Calculating Thiele moduli using these values, and toluene
64
diffusivities obtained via TGA, at relevant reaction temper-
atures and rate constants reported in Table 1, yield the most
diffusion-limited case of ortho-xylene methylation φ = 0.19 using
1 μm crystallite sizes, which indicates that the rate constants
reported in this work are dictated by reaction kinetics. Thiele
moduli calculations are confirmed by the experimental
observation that mesoporous and microporous H-SPP and H-
ZSM-5 samples saturate at nearly identical rates (2.7 compared
51
isochores over H-ZSM-5 range between −60 to −80 kJ
−1
mol , which is similar to trans-2-butene adsorption values of
1
−1
−
68 kJ mol− calculated using MP2/DFT with periodic
to 2.9 h ) for para-xylene methylation. With a >150 fold
4
9
boundary conditions. This agreement makes a direct
comparison of apparent barriers between toluene and butene
more relevant than for benzene and propene.
decrease in the critical diffusion length, the saturation rate
achieved with respect to xylene partial pressure indicates that this
reaction is operating in kinetically limited regimes.
Comparison of Benzene and Toluene Methylation
Rates on H-MFI and Mesoporous H-SPP. Aromatic and
DME pressure dependencies and temperature dependencies for
benzene and toluene methylation over H-SPP are shown in the
bottom panels of Figures 1 and 2 for benzene and toluene,
respectively. Rate constants agree within 50% and measured
activation energies within 10% (Table 1) between microporous
and mesoporous MFI samples. This indicates kinetic control for
these systems, and any effects of sterics or confinement are
strictly limited to the micropores for benzene and toluene
methylation.
Temperature Dependence for Ortho-Xylene and Para-
Xylene in Linear and Saturated Rate Regimes. Temper-
ature dependence studies were carried out on H-SPP at 0.003
and 0.05 bar para- and ortho-xylene pressures. These pressures
correspond to linear and saturated regimes in the xylene pressure
dependent data, which we ascribe to a zeolite surface
predominately covered in surface methoxides and xylene
molecules adsorbed on methoxides, respectively. These inferred
surface species and experimental evidence for their presence and
reactivity will be discussed in studies presented below. High-
pressure regimes yielded apparent activation barriers of 34 to 44
kJ mol− and low-pressure regimes 63 to 62 kJ mol for ortho-
and para-xylene, respectively (Table 1). While first-order rate
constants between high and low-pressure regimes are within 25%
at 373 K, the differences in activation barriers indicate that this
agreement is coincidental, as a different choice in reference
temperature would cause these values to diverge. Differing
apparent barriers for methylation at saturated pressures
compared to linear regimes implies either a change in the rate
determining step or a change in the predominant zeolite surface
species. Considering the same bond to be broken in both cases,
the intrinsic activation barrier would remain constant, requiring
the saturated regime to have a more stabilized surface species
prior to the rate-determining step compared to surface species
present in low-pressure regimes.
1
−1
Ortho- and Para-Xylene Methylation. The results of
xylene methylation are presented in Figure 3, with kinetic
parameters reported in Table 1. While methylation rate constants
increased from benzene to toluene methylation reactions over H-
ZSM-5, the marked decrease in xylene methylation rate
constants is not intuitive, as increasing methyl substitution
about the aromatic ring is expected to further increase
38
intermediate carbocation stability. Ahn et al. have recently
reported that the rate of methylation from toluene, to para-
xylene, to 1,2,4-trimethylbenzene decreases from 35, to 22, to 3.7
−2
−1
[
10 mol (mol H s) ] when feeding 1.2 kPa toluene and 0.3
kPa methanol, a 4:1 aromatic to methanol ratio, at 673 K over H-
ZSM-5 at 45−58% conversion. The decrease in methylation rate
was ascribed to diffusion limitations in xylenes and larger
aromatics in the medium pore zeolites, where these species
would mostly dealkylate to lower methylbenzenes with more
favorable diffusion properties as opposed to directly leaving the
zeolite framework. While reaction rates for para- and ortho-
xylene are independent of DME pressure, as in the case of
benzene and toluene methylation, xylene methylation rates
Post-Reaction Zeolite Surface Titration with H O for
2
Benzene, Toluene, and Para-Xylene Methylation. Resolv-
ing active surface derivatives of DME or methanol for
methylation reactions is a continuing debate in the MTH
literature, specifically, the formation of co-adsorption complexes
from physisorbed reactants or a stepwise mechanism through the
1
996
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
Research Article
Scheme 1. Three Steps Involved in the Titration of Surface Species Generated in Aromatic Methylation Experiments on H-SPP,
where n = 0,1,2
6
7
formation of a surface methoxide species. Computational
observed in effluent DME compared to benzene and toluene
methylation using 0.003 bar partial pressure of para-xylene, and
almost no scrambling is observed in the presence of 0.05 bar
para-xylene (Figure 4). This suppressed scrambling could
2
3,48,49,68
methods have reported consistent results
with kinetic
1
9,20
measurements
adsorption mechanism. However, FTIR and C MAS
72,73
by modeling methylation reactions via the co-
6
9
13
7
0,71
NMR
at 453 K and above and computational studies
evidence the existence of surface methoxides on the zeolite
surface. These surface methoxides are unable to desorb, because
of the absence of a β-hydrogen, except via reaction, and their
6
9,70,74
stability has been noted up to 673 K under vacuum.
The
titration of chemisorbed methoxides on the surface of H-SPP was
performed by isolating steady-state MTH surface species in a
flow of helium prior to heating to 423 K and introducing water to
45
remove any chemisorbed species. The experimental setup is
outlined in Scheme 1, and the results of these chemical titration
studies are reported in Table 2. Upon interaction with water,
Table 2. Results for (a) Post-Reaction Water Titration
Experiments and (b) Isotopic Switching Experiments over H-
a
SPP
effluent DME
content
Figure 4. Isotopic scrambling results upon introduction of 1:1 d :d
DME at steady-state methylation conditions for benzene, toluene, and
6
0
CH */Al
d0
d3
d6
k /k
D
para-xylene methylation.
3
H
benzene
toluene
0.95 ± 0.06
0.96 ± 0.07
0.98 ± 0.08
1
1
1
1
1.69
1.52
0.04
0.23
0.86
0.82
0.84
0.85
1.25 ± 0.08
1.35 ± 0.06
1.34 ± 0.04
1.31 ± 0.05
para-
xylene
0.05 bar
.003 bar
indicate that surface methoxides are hindered from reacting
with other gas-phase DME molecules to facilitate C−O
scrambling, which is consistent with para-xylene blocking these
sites through co-adsorption, which would be expected to
exacerbate at higher xylene partial pressures.
0
a
Deuterium distributions in effluent DME are normalized to d (m/z
46) abundances.
0
=
0
.95−0.98 CH OH:Al was evolved after benzene, toluene, and
3
Rate Equation for BTX Methylation Systems. The linear
pressure dependence of benzene and toluene methylation rates
on aromatic pressure and rate independence with respect to
DME over MFI indicates a mechanism similar to that previously
described for C −C olefins over MFI, BEA, MOR, and
para-xylene methylation at two different reactant partial
pressures, indicating that, at these conditions, the surface is
predominantly covered by surface methoxide species (*CH3).
In-Situ d DME/DME Switching for Benzene, Toluene,
6
2
4
and Para-Xylene Methylation. H-SPP samples were exposed
to a 50:50 mixture of d /d DME after steady-state operation in
43−45
FER.
toluene methylation are described in eq 1, and shown in Scheme
(top)
The elementary steps involved for benzene and
0
6
the absence of deuterated DME. Results for these studies are
shown in Table 2 and indicate that the rate-determining step for
all BTX methylation reactions involves the formation of a C−C
bond in the transition state as inferred from the observed
secondary kinetic isotope effect ranging from 1.25−1.35 k /k .
Benzene and toluene methylation reactions showed a near-
binomial distribution of d :d :d DME (1:2:1), consistent with
results from ethene methylation from a previous work. In these
instances, the binomial distribution of d , d , and d DME in the
2
(
i) CH OCH + H* ↔ H·CH OCH * K
3
3
3
3
ads
H
D
(ii) H·CH OCH * ↔ CH OH + CH * K
3
3
3
3
M
(iii)
CH * + Ar → Ar + H*
k
3
n
n+1
(1)
0
3
6
43
where Ar
represents an aromatic with n methyl groups, or 0, 1,
0
3
6
n
effluent indicates a rapid breaking/reforming of C−O bonds of
DME upon interaction with the zeolite surface. Para-xylene
methylation showed ∼15% of the C−O bond scrambling
and 2 for benzene, toluene, and xylene, respectively, and k is the
apparent rate constant. The related rate equation, normalized per
zeolite framework Al, for this sequence of elementary steps is
1
997
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
Research Article
Scheme 2. Proposed Mechanisms for Benzene and Toluene (top) and Para-Xylene and Ortho-Xylene (bottom) Methylation with
DME
[
DME]
Rate
Al
[DME]
K K k[Ar ]
ads
M
n
= K K K k[Xylene]
Rate
Al
[CH OH]
ads M c
3
= 1
[CH OH]
3
[
DME]
ads M
[CH OH]
+ K [DME] + K K
ads
3
(2)
⎡
/
[
DME]
⎢1 + K [DME] + K K
ads
ads M
⎣
[CH OH]
3
Assuming inhibition by methanol is negligible because of its trace
quantities observed compared to DME and that methanol is also
active for methylation reactions, and the surface is predominately
covered by surface methoxides as observed from surface titration
studies, eq 2 simplifies to
⎤
[DME]
+
K K K [Xylene]
⎥
ads
M c
[
CH OH] ⎦
3
(5)
Considering the negligible formation of methanol compared
to DME as observed in reactor effluents, and the predominant
surface species to be surface methoxides or coadsorbed
methoxides with xylene, eq 2 simplifies to:
Rate
Al
=
k[Arn]
(3)
Rate
Al
Kck[Xylene]
This model fits rate data presented in Figures 1 and 2 for all
partial pressures of DME and benzene/toluene measured, as well
as observed secondary kinetic isotope effects and surface titration
studies.
=
1 + K [Xylene]
(6)
c
Nonlinear fitting of eq 6 to the pressure dependent data are
plotted in Figure 3, and the parameter fits are presented in Table
Based on the observations that (i) the reaction rate
simultaneously saturates in xylene and DME pressures for both
para- and ortho-xylene, (ii) the saturated rate for para-xylene and
ortho-xylene methylation does not change appreciably from H-
ZSM-5 samples to H-SPP samples, and (iii) observed secondary
kinetic isotope effects, we postulate that this system is kinetically
limited and the rate-determining step involves a surface species
derived from DME and xylene, but still involves the breaking/
formation of a C−C bond. Equation 4 outlines a mechanism,
which involves the formation of a surface methoxide and the
adsorption of a subsequent xylene molecule prior to the rate-
determining step (Scheme 2, bottom).
3
. The zero-order rate constant for para-xylene saturates at 2.7−
Table 3. Parameter Fitting Results for Equation 3 Modeling
Xylene Pressure Dependent Data
Kc
k473 ([TMB][H+]−1 h−1
)
para-xylene
ortho-xylene
H-ZSM-5
H-SPP
130 ± 20
290 ± 70
17 ± 3
2.9 ± 0.5
2.7 ± 0.8
3.5 ± 0.7
3.5 ± 0.6
H-ZSM-5
H-SPP
24 ± 4
.9 [TMB][H+] h−1, and for ortho-xylene at 3.5 [TMB][H
−1
2
−1
−1
+
]
h . Co-adsorption equilibrium constants are higher for
(
(
(
(
i) CH OCH + H* ↔ H·CH OCH * K
3
3
3
3
ads
para-xylene compared to ortho-xylene, and H-SPP samples
increase this value compared to H-MFI samples.
ii) H·CH OCH * ↔ CH OH + CH * K
3
3
3
3
M
Adsorption Effects of Aromatics in MFI Structures. The
distinct aromatic pressure dependence behavior exhibited by
xylenes compared to benzene and toluene arises, in part, because
of stronger surface adsorption. Brogaard et al. have reported
using self-consistent DFT with BEEF-vdW and RPBE func-
tionals and periodic boundary conditions that for MFI structures
yield adsorption enthalpies of −59, −73, and −78 kJ mol− for
iii) CH * + Xylene ↔ Xylene·CH *
Kc
3
3
75
iv) Xylene·CH * → TMB + H*
k
3
(4)
The related rate equation for this sequence of elementary steps
1
is
1
998
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
Research Article
benzene, toluene, and para-xylene, respectively. When compared
to experimental adsorption enthalpies of −55, −80, and −96 kJ
observed increase in olefin methylation rates with increasing
substitution about the double bond, aromatic methylation rates
do not monotonically increase with increasing methyl sub-
stitution implying that the aromatic methylation cycle proceeds
under space limitation conditions relative to the olefin
methylation cycle.
−1
76
mol , the computational results progressively underestimated
75
adsorbants with higher degrees of substitution. Brogaard et al.
noted that this discrepancy could be possibly due to either
underestimates in the van der Waals interactions between methyl
groups and the inorganic framework or through effects specific to
the choice of Brønsted acid site location used for calculations.
Aromatic compounds studied (benzene through tetracene) were
found to prefer occupation of channel intersections, however,
benzene and toluene were the only adsorbents that had a relative
preference to adsorb into the more-restricted sinusoidal channels
4
. CONCLUSIONS
The rate of methylation of benzene and toluene with DME at
temperatures below 503 K at differential conversions is first-
order dependent in aromatic pressure and independent of DME
pressure, indicating the rapid formation of DME-derived species
covering the zeolite surface reacting with the aromatic in the rate-
determining step. The kinetics of benzene methylation are
consistent with previous studies at higher temperatures with rates
75
compared to adsorption in straight channels. This indicates
that for xylenes and larger aromatics, accessibility to sinusoidal
channels is restricted. These results are supported by IR
56
measurements by Armaroli et al. that show benzene, toluene,
para-xylene, and ortho-xylene completely occupy the acid sites of
H-ZSM-5, but meta-xylene only partially occupies surface acid
sites at room temperature, indicating restricted access to a
fraction of these sites, which would be expected to exacerbate
with bulkier surface methoxides in place of Brønsted acid sites
and could lead to similar behavior in para- and ortho-xylene.
This restricted accessibility has also been noted experimentally
by calorimetric studies obtaining adsorption isotherms for para-
xylene on H-ZSM-5, noting a phase transition in the zeolite
37
extrapolated to zero contact time, and activation energies for
toluene methylation are consistent with previous kinetic
5
2,53
studies.
Benzene and toluene methylation proceed with
similar rates and activation energies as propene and n-butene
methylation, respectively, on H-ZSM-5. A comparison of
measured rates of aromatic methylation with H-SPP show no
significant difference between microporous and mesoporous
samples (minimum factor of 150 difference in critical diffusion
length) thereby demonstrating that transport restrictions do not
play a role in benzene and toluene methylation under the
reaction conditions reported in this research.
77
structure at 300 K, thermogravimetric analysis citing 210 and
80 J mol−1
−1
1
K entropy differences for para-xylene and benzene
78 79
Methylation of para- and ortho-xylene show a shift in the
predominant surface species from surface methoxides to
coadsorbed xylene on surface methoxides, as observed from
differing apparent activation energies at low and high partial
pressures of xylene. These reactions were run in the absence of
isomerization and diffusion limitations, and the formation of
surface methoxides for all methylation reactions is consistent
with post-reaction titration experiments with water forming 1:1
methanol molecules per zeolite Al. Inhibited DME isotopic
exchange is observed for para-xylene methylation only, indicative
of coadsorbed xylene molecules blocking the methoxide species
from further reactions with DME. Observed secondary kinetic
isotope effects for benzene, toluene, and xylene methylation
studies over H-MFI show that the rate-determining step is the
adsorption, respectively. X-ray diffraction studies for para-,
80
ortho-, and meta-xylene isomers in silicalite have noted missing
symmetries indicative of aromatic occupation of sinusoidal
channels, in addition to distorting the MFI straight channels to
induce a monoclinic to orthorhombic transition at room
temperature for aromatic loadings greater than three molecules
per unit cell. The preferential occupation of straight channels
over sinusoidal channels has been noted to lead to significant
kinetic differences in para-xylene isomerization compared to
ortho- and meta-xylene isomers over H-ZSM-5, which were
mostly invariant with the addition of MgO and CaO on a 1:1
basis with Al, indicating that occupation effects play a larger role
81
than sterics in this system. The less than 2-fold enhanced
adsorption on H-SPP compared to H-MFI therefore, could
possibly be explained by enhanced access to less favored
adsorption sites with less effect of sterics from neighboring
molecules, allowing for the achievement of a saturated zeolite
surface at lower aromatic partial pressures.
3
same across these aromatics and involves a transition from sp
2
hybridization to sp hybridization of the surface methoxide in the
transition state, which is consistent with that observed for
43
ethylene methylation. These experiments explain the distinct
behavior of xylene methylation from benzene and toluene
methylation over microporous and mesoporous MFI on a
mechanistic basis.
The methylation of xylene on MFI is distinct from elementary-
step methylation reactions of olefins, benzene, and toluene in
that, in the absence of diffusion limitations, rate saturation is
achieved relative to both xylene and DME partial pressures. This
phenomenon is attributed to the saturation of the zeolite surface
with xylene molecules adsorbing on a surface methoxide, which is
not observed for C −C olefins and benzene and toluene under
ASSOCIATED CONTENT
■
*
S
Supporting Information
2
4
Thiele moduli for methylation reactions over H-ZSM-5 and H-
SPP, and rate and energetic consequences at low and high xylene
partial pressures over H-ZSM-5 and H-SPP. This material is
available free of charge via the Internet at http://pubs.acs.org.
nearly identical reaction conditions and chemical conversions.
The product selectivity in methanol-to-hydrocarbons (MTH)
conversion can be attributed to the relative propagation of olefin
and aromatic methylation and cracking cycles. We report that
aromatic methylation pathways for benzene and toluene are
identical to those previously reported for olefin methylation and
hence, a direct comparison of rate parameters for olefin and
aromatic alkylation pathways can be inferred. Chemical titration
and isotopic scrambling studies at steady-state methylation
conditions evidence the involvement of surface methoxides in
both olefin and aromatic methylation cycles. In contrast to the
AUTHOR INFORMATION
■
Corresponding Author
*
E-mail: abhan@umn.edu.
Notes
The authors declare no competing financial interest.
1
999
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
ACKNOWLEDGMENTS
Research Article
(
37) Van der Mynsbrugge, J.; Visur, M.; Olsbye, U.; Beato, P.; Bjorgen,
■
M.; Van Speybroeck, V.; Svelle, S. J. Catal. 2012, 292, 201−212.
38) Ahn, J. H.; Kolvenbach, R.; Al-Khattaf, S. S.; Jentys, A.; Lercher, J.
A. ACS Catal. 2013, 3, 817−825.
39) Arstad, B.; Kolboe, S.; Swang, O. J. Phys. Chem. B 2002, 106,
2722−12726.
40) Svelle, S.; Arstad, B.; Kolboe, S.; Swang, O. J. Phys. Chem. B 2003,
07, 9281−9289.
41) Svelle, S.; Kolboe, S.; Swang, O.; Olsbye, U. J. Phys. Chem. B 2005,
09, 12874−12878.
42) Vos, A.; Rozanska, X.; Schoonheydt, R.; van Santen, R.; Hutschka,
F.; Hafner, J. J. Am. Chem. Soc. 2001, 123, 2799−2809.
The authors acknowledge financial support from The Dow
Chemical Company and the National Science Foundation
(
(
CBET 1055846)
(
1
(
1
(
1
(
REFERENCES
■
(
(
1) Chang, C. D. Catal. Rev. Sci. Eng. 1983, 25, 1−118.
2) Chang, C. D.; Kuo, J. C. W.; Lang, W. H.; Jacob, S. M.; Wise, J. J.;
Silvestri, A. J. Ind. Eng. Chem. Process Des. Dev. 1978, 17, 255−260.
(
(
3) Chang, C. D.; Silvestri, A. J. J. Catal. 1977, 47, 249−259.
4) Song, W. G.; Marcus, D. M.; Fu, H.; Ehresmann, J. O.; Haw, J. F. J.
(
(
(
(
43) Hill, I. M.; Hashimi, S. A.; Bhan, A. J. Catal. 2012, 285, 115−123.
44) Hill, I. M.; Hashimi, S. A.; Bhan, A. J. Catal. 2012, 291, 155.
45) Hill, I. M.; Ng, Y. S.; Bhan, A. ACS Catal. 2012, 2, 1742−1748.
46) Zhang, X.; Liu, D.; Xu, D.; Asahina, S.; Cychosz, K. A.; Agrawal, K.
Am. Chem. Soc. 2002, 124, 3844−3845.
5) Olsbye, U.; Bjorgen, M.; Svelle, S.; Lillerud, K. P.; Kolboe, S. Catal.
Today 2005, 106, 108−111.
6) Svelle, S.; Joensen, F.; Nerlov, J.; Olsbye, U.; Lillerud, K.; Kolboe,
S.; Bjorgen, M. J. Am. Chem. Soc. 2006, 128, 14770−14771.
7) Marcus, D. M.; McLachlan, K. A.; Wildman, M. A.; Ehresmann, J.
O.; Kletnieks, P. W.; Haw, J. F. Angew. Chem., Int. Ed. 2006, 45, 3133−
136.
8) Lesthaeghe, D.; Van Speybroeck, V.; Marin, G. B.; Waroquier, M.
Ind. Eng. Chem. Res. 2007, 46, 8832−8838.
9) Lesthaeghe, D.; Van Speybroeck, V.; Marin, G. B.; Waroquier, M.
Angew. Chem., Int. Ed. 2006, 45, 1714−1719.
10) Lesthaeghe, D.; Van Speybroeck, V.; Marin, G. B.; Waroquier, M.
(
(
V.; Al Wahedi, Y.; Bhan, A.; Al Hashimi, S.; Terasaki, O.; Thommes, M.;
Tsapatsis, M. Science 2012, 336, 1684−1687.
(
(
47) Berger, R.; Perez-Ramirez, J.; Kapteijn, F.; Moulijn, J. Chem. Eng. J.
2
(
002, 90, 173−183.
3
(
48) Van Speybroeck, V.; Van der Mynsbrugge, J.; Vandichel, M.;
Hemelsoet, K.; Lesthaeghe, D.; Ghysels, A.; Marin, G. B.; Waroquier, M.
J. Am. Chem. Soc. 2011, 133, 888−899.
(
Chem. Soc. 2009, 131, 816−825.
(
(
(
(
(
(
49) Svelle, S.; Tuma, C.; Rozanska, X.; Kerber, T.; Sauer, J. J. Am.
(
50) Pope, C. G. J. Phys. Chem. 1984, 88, 6312−6313.
51) Pope, C. G. J. Phys. Chem. 1986, 90, 835−837.
52) Vinek, H.; Lercher, J. A. J. Mol. Catal. 1991, 64, 23−39.
53) Rabiu, S.; Al-Khattaf, S. Ind. Eng. Chem. Res. 2008, 47, 39−47.
54) Odedairo, T.; Balasamy, R. J.; Al-Khattaf, S. Ind. Eng. Chem. Res.
011, 50, 3169−3183.
55) Cejka, J.; Zilkova, N.; Wichterlova, B.; Eder-Mirth, G.; Lercher, J.
A. Zeolites 1996, 17, 265−271.
56) Armaroli, T.; Bevilacqua, M.; Trombetta, M.; Alejandre, A.;
Ramirez, J.; Busca, G. Appl. Catal., A 2001, 220, 181−190.
57) Ruthven, D. M. Adsorption 2007, 13, 225−230.
58) Ruthven, D. M.; Eic, M.; Richard, E. Zeolites 1991, 11, 647−653.
59) Ruthven, D. M.; Vidoni, A. Chem. Eng. Sci. 2012, 71, 1−4.
60) Jentys, A.; Tanaka, H.; Lercher, J. J. Phys. Chem. B 2005, 109,
Chem. Phys. Lett. 2006, 417, 309−315.
(
(
(
(
(
(
(
11) Dessau, R. M.; LaPierre, R. B. J. Catal. 1982, 78, 136−141.
12) Dessau, R. M. J. Catal. 1986, 99, 111−116.
13) Dessau, R. M. J. Catal. 1987, 103, 526−528.
14) Dahl, I. M.; Kolboe, S. J. Catal. 1996, 161, 304−309.
15) Dahl, I. M.; Kolboe, S. J. Catal. 1994, 149, 458−464.
16) Dahl, I. M.; Kolboe, S. Catal. Lett. 1993, 20, 329−336.
17) Cui, Z.; Liu, Q.; Ma, Z.; Bian, S.; Song, W. J. Catal. 2008, 258, 83−
2
(
(
8
(
6
(
6.
18) Cui, Z.; Liu, Q.; Song, W.; Wan, L. Angew. Chem., Int. Ed. 2006, 45,
512−6515.
19) Svelle, S.; Ronning, P. O.; Olsbye, U.; Kolboe, S. J. Catal. 2005,
(
(
(
(
2
(
(
34, 385−400.
20) Svelle, S.; Ronning, P. A.; Kolboe, S. J. Catal. 2004, 224, 115−123.
2
(
254−2261.
21) Haw, J. F.; Song, W. G.; Marcus, D. M.; Nicholas, J. B. Acc. Chem.
61) Gobin, O. C.; Reitmeier, S. J.; Jentys, A.; Lercher, J. A. Microporous
Res. 2003, 36, 317−326.
22) Haw, J. F.; Nicholas, J. B.; Song, W. G.; Deng, F.; Wang, Z. K.; Xu,
T.; Heneghan, C. S. J. Am. Chem. Soc. 2000, 122, 4763−4775.
23) McCann, D. M.; Lesthaeghe, D.; Kletnieks, P. W.; Guenther, D.
Mesoporous Mater. 2009, 125, 3−10.
(
(
Liu, S. J. Mol. Catal. A: Chem. 2002, 181, 41−55.
(
(
62) Pope, C. G. J. Chem. Soc., Faraday Trans. 1993, 89, 1139−1141.
63) Chen, W.; Tsai, T.; Jong, S.; Zhao, Q.; Tsai, C.; Wang, I.; Lee, H.;
(
R.; Hayman, M. J.; Van Speybroeck, V.; Waroquier, M.; Haw, J. F.
64) Xiao, J. R.; Wei, J. Chem. Eng. Sci. 1992, 47, 1143−1159.
65) Roquemalherbe, R.; Wendelbo, R.; Mifsud, A.; Corma, A. J. Phys.
Angew. Chem., Int. Ed. 2008, 47, 5179−5182.
(
(
24) Song, W. G.; Haw, J. F.; Nicholas, J. B.; Heneghan, C. S. J. Am.
Chem. 1995, 99, 14064−14071.
Chem. Soc. 2000, 122, 10726−10727.
25) Mikkelsen, O.; Ronning, P. O.; Kolboe, S. Microporous Mesoporous
Mater. 2000, 40, 95−113.
26) Teketel, S.; Olsbye, U.; Lillerud, K.; Beato, P.; Svelle, S.
Microporous Mesoporous Mater. 2010, 136, 33−41.
27) Wang, C.; Wang, Y.; Xie, Z.; Liu, Z. J. Phys. Chem. C 2009, 113,
584−4591.
28) Bjorgen, M.; Olsbye, U.; Petersen, D.; Kolboe, S. J. Catal. 2004,
21, 1−10.
29) Kolboe, S.; Svelle, S.; Arstad, B. J. Phys. Chem. A 2009, 113, 917−
23.
30) Lesthaeghe, D.; De Sterck, B.; Van Speybroeck, V.; Marin, G. B.;
(
(
(
66) Xiao, J. R.; Wei, J. Chem. Eng. Sci. 1992, 47, 1123−1141.
67) Svelle, S.; Visur, M.; Olsbye, U. Top. Catal. 2011, 54, 897−906.
68) Lesthaeghe, D.; Van der Mynsbrugge, J.; Vandichel, M.;
Waroquier, M.; Van Speybroeck, V. ChemCatChem 2011, 3, 208−212.
69) Yamazaki, H.; Shima, H.; Imai, H.; Yokoi, T.; Tatsumi, T.; Kondo,
(
(
(
(
4
(
2
(
9
(
J. N. Angew. Chem., Int. Ed. 2011, 50, 1853−1856.
(
(
(
2
(
70) Wang, W.; Hunger, M. Acc. Chem. Res. 2008, 41, 895−904.
71) Bosacek, V. J. Phys. Chem. 1993, 97, 10732−10737.
72) Boronat, M.; Martinez, C.; Corma, A. Phys. Chem. Chem. Phys.
011, 13, 2603−2612.
73) Zicovich-Wilson, C. M.; Viruela, P.; Corma, A. J. Phys. Chem.
995, 99, 13224−13231.
74) Cheung, P.; Bhan, A.; Sunley, G. J.; Law, D. J.; Iglesia, E. J. Catal.
1
(
Waroquier, M. Angew. Chem., Int. Ed. 2007, 46, 1311−1314.
(
(
(
31) Ilias, S.; Bhan, A. ACS Catal. 2013, 3, 18−31.
32) Ilias, S.; Bhan, A. J. Catal. 2012, 290, 186−192.
2007, 245, 110−123.
33) Bjorgen, M.; Olsbye, U.; Svelle, S.; Kolboe, S. Catal. Lett. 2004, 93,
(75) Brogaard, R. Y.; Weckhuysen, B. M.; Nørskov, J. K. J. Catal. 2013,
300, 235−241.
3
(
(
7−40.
34) Bjorgen, M.; Olsbye, U.; Kolboe, S. J. Catal. 2003, 215, 30−44.
(76) Mukti, R. R.; Jentys, A.; Lercher, J. A. J. Phys. Chem. C 2007, 111,
3973−3980.
35) Bjorgen, M.; Svelle, S.; Joensen, F.; Nerlov, J.; Kolboe, S.; Bonino,
F.; Palumbo, L.; Bordiga, S.; Olsbye, U. J. Catal. 2007, 249, 195−207.
(77) Takaishi, T.; Tsutsumi, K.; Chubachi, K.; Matsumoto, A. J. Chem.
Soc., Faraday Trans. 1998, 94, 601−608.
(
36) Arstad, B.; Kolboe, S. J. Am. Chem. Soc. 2001, 123, 8137−8138.
2
000
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001

ACS Catalysis
Research Article
(
78) Lee, C.; Chiang, A. J. Chem. Soc., Faraday Trans. 1996, 92, 3445−
3451.
(
(
(
79) Mentzen, B. F. Mater. Res. Bull. 1992, 27, 953−960.
80) Nair, S.; Tsapatsis, M. J. Phys. Chem. B 2000, 104, 8982−8988.
81) Li, Y.; Jun, H. Appl. Catal., A 1996, 142, 123−137.
2
001
dx.doi.org/10.1021/cs400377b | ACS Catal. 2013, 3, 1992−2001