Recent trends and fundamental insights

in the methanol-to-hydrocarbons process

Item Type Article

Authors Yarulina, Irina; Chowdhury, Abhishek Dutta; Meirer, Florian;

Weckhuysen, Bert M.; Gascon, Jorge

Citation Yarulina I, Chowdhury AD, Meirer F, Weckhuysen BM, Gascon J

(2018) Recent trends and fundamental insights in the methanol-
to-hydrocarbons process. Nature Catalysis 1: 398–411. Available:
http://dx.doi.org/10.1038/s41929-018-0078-5.

Eprint version Post-print

DOI 10.1038/s41929-018-0078-5

Publisher Springer Nature

Journal Nature Catalysis

Rights The final publication is available at Springer via http://

dx.doi.org/10.1038/s41929-018-0078-5

Download date 16/11/2021 13:57:56

Link to Item http://hdl.handle.net/10754/628298

 Recent Trends and Fundamental Insights
in the Methanol-to-Hydrocarbons Process

Irina Yarulinaa,, Abhishek Dutta Chowdhuryb,, Florian Meirerb, Bert M. Weckhuysenb,* and Jorge
Gascona,*

a
King Abdullah University of Science and Technology, KAUST Catalysis Center, Advanced Catalytic
Materials, Thuwal 23955-6900, Saudi Arabia
b
Inorganic Chemistry and Catalysis, Debye Institute for Nanomaterials Science, Utrecht University,
Universiteitsweg 99, 3584 CG Utrecht, The Netherlands

These authors contributed equally to the work
*
Correspondence to: [email protected], [email protected]
Abstract

The production of high-demand chemical commodities, such as ethylene and propylene (methanol-to-
olefins), hydrocarbons (methanol-to-hydrocarbons), gasoline (methanol-to-gasoline) and aromatics
(methanol-to-aromatics) from methanol—obtainable from alternative feedstocks, such as carbon
dioxide, biomass, waste or natural gas through the intermediate formation of synthesis gas—has been
central to research in both academia and industry. Although discovered in the late 1970s, this catalytic
technology has only been industrially implemented over the last decade, with a number of large
commercial plants already operating in Asia. However, as it is the case for other technologies, industrial
maturity is not a synonym of full understanding. For this reason, research is still intense and a number
of important discoveries have been reported over the last few years. In this review, we summarize the
most recent advances in mechanistic understanding—including direct C-C bond formation during the
induction period and the promotional effect of zeolite topology and acidity on the alkene cycle—and
correlate these insights to practical aspects in terms of catalyst design and engineering.

2
 In 1977, scientists at Mobile published the first patent on the methanol-to-hydrocarbons process
(MTH)1 claiming “A lower alcohol and/or ether feed is selectively converted to a mixture of light olefins,
including ethylene and propylene, by catalytic contact of the feed, for example methanol or dimethyl ether,
at sub-atmospheric partial pressure, with certain crystalline aluminosilicate zeolite catalysts exemplified by
ZSM-5” (Fig. 1a). The possibility to obtain olefins (methanol-to-olefins, MTO), aromatics (methanol-to-
aromatics, MTA) and/or gasoline (methanol-to-gasoline, MTG) without relying on oil (methanol can be
produced from a wide range of carbon-containing sources, like biomass, waste, coal, natural gas or even
CO2) makes this family of processes a convenient alternative to the classical production routes of high-
demand chemicals and intermediates.
As it is the case for other grand old ladies of the petrochemical industry, such as Fluid Catalytic
Cracking (FCC),2 the MTH process, despite being already commercialized, still attracts a great deal of
attention from both industry and academia. Apart from the obvious economic driving force in improving the
process, from a scientific point of view, understanding (and eventually controlling) the complex reaction
mechanism behind such a complex overall stoichiometry remains a challenge. As we will elaborate below,
the process is dictated by a large set of elementary reactions and the full picture of the MTH reaction
mechanism is still elusive (Fig. 1b). On one hand, the chosen reaction conditions are important, as not all
reactions have the same kinetic order and energies of activation. On the other hand, the zeolite
microenvironment plays an even bigger role in defining product distribution and catalyst lifetime.3 MTH
serves therefore as an outstanding example of complexity in zeolite chemistry and heterogeneous
catalysis, as pinpointed in a number of classical4 and more recent reviews.5, 6, 7, 8, 9
In this review article, we highlight current challenges of MTH chemistry by critically analyzing the
existing literature to give an updated picture on the mechanism of reaction and to provide guidelines for
catalyst design. We believe that, despite a number of excellent reviews in this field, knowledge on
mechanistic aspects and structure-property relations in MTH has advanced tremendously over the last few
years and we expect this article to bridge the gap between those earlier reviews dedicated to mechanism
rationalization and catalyst engineering. To do so, we first summarize the recent efforts to unravel the
mechanism that leads to the formation of the first C-C bond (i.e. coupling of two methanol or dimethyl ether
molecules), a topic of intense debate over the last few decades. Formation of the direct C-C bond
continues with the formation of longer hydrocarbons inside the zeolite pores, this mechanism along with
the most recent spectroscopic evidences of its different steps are discussed in the second section. Finally,
with this mechanistic information in hand, we define a number of catalyst design rules that should help the
reader understand the effect of catalyst topology, acidity and reaction conditions on product distribution
and catalyst deactivation. The review is completed with our personal view on future challenges both from
the process and fundamental point of view.

The direct C-C bond.

3
 5, 6, 7, 8, 10
The MTH process involves a very complex reaction mechanism. As of now, more than 20
different proposals have been postulated in the literature.4 Surprisingly, the exact route for the formation
of the first carbon-carbon (C-C) bond during MTH was only very recently unveiled.11, 12, 13, 14, 15, 16, 17, 18 For
a long time, the scientific community assumed that the presence of traces of impurities (e.g. in the
methanol, catalyst, and/or carrier gas) was responsible for the formation of the direct C-C bond over any
direct mechanism (i.e. the coupling of two methanol molecules). This assumption was primarily attributed
to the lack of concrete experimental evidence in support of the direct mechanism. 19, 20 Moreover, the
feasibility of such direct coupling was anticipated to be low by theoreticians because of the high activation
energies and unstable reaction intermediates.21, 22
In 2006, Hunger and co-workers systematically
demonstrated that traces of organic impurities neither have any significant influence on product
distribution, nor do they control the formation of hydrocarbon pool (HCP, vide infra) species during MTH
reaction (Fig 1a).23 That means a direct mechanism may be operative, at least in the early stages of the
MTH reaction.16 Since then, other research groups (e.g. Kondo, Fan, Copéret, Lercher, Weckhuysen, and
Liu) have delivered both experimental and theoretical evidence in support of the existence of a direct
mechanism during the initial stages of MTH.11, 12, 13, 14, 15, 16, 17, 18, 23, 24, 25, 26, 27, 28, 29, 30 In essence, this reaction
can be sub-divided in two parts: direct mechanism and conventional HCP/Dual cycle mechanism, during
induction and autocatalytic period, respectively.
There are several excellent reviews available in the literature on this topic, which adequately describe
history and developments of direct mechanisms.4, 6 Although a consensus has currently been reached
regarding the existence of the direct mechanism, its actual nature/course of action is yet to be fully
established. In all direct mechanistic proposals, a surface methoxy species (SMS), formed upon adsorption
of methanol onto a Brønsted acid site is non-arguably the most experimentally verified intermediate (Fig.
2).13, 14, 18, 26 The ability of SMS to form a C-C bond at higher reaction temperature is a well-established
phenomenon during any zeolite-catalyzed hydrocarbon conversion process.18 The idea of a carbene-
insertion mechanism by SMS was first introduced by the group of Hunger, mostly based on NMR
spectroscopy and carbene-trapping experiments.18, 26, 27 The carbene character of SMS was proposed to
form through the polarization of the C-H bond of SMS by a neighboring adjacent oxygen (as depicted in
Fig. 2a-c).31, 32 The carbene/ylide nature of SMS was confirmed by a carbene trapping experiment with
cyclohexane as a probe molecule at ≥ 493K, where methylcyclohexane was formed through an insertion
reaction of carbene/ylide into the sp3 C-H bond of cyclohexane (Fig. 2a).18 Later, Kondo et al. also arrived
to the same conclusion with the help of IR spectroscopy: the C-C bond containing hydrocarbon species
originated from the coupling between carbene-like SMS and methanol/dimethyl ether (DME) (Fig. 2b).11, 12
Another convincing evidence in support of the existence of a carbene-type mechanism was provided
recently by Weckhuysen et al., by employing 2D magic angle spinning (MAS) solid-state NMR (both 1H-
13 13
C and C-13C) spectroscopy (Fig. 2c).13 The strong signals at 52.2 (13C) and 3.59 (1H) ppm were
assigned to surface adsorbed methanol, while the signals at 57.7 (13C) and 3.54 (1H) ppm were attributed
13
to SMS (red strip in Fig. 3a). Interestingly, a strong cross-peak between these C signals was also

4
observed, which was due to the close proximity of surface adsorbed methanol and SMS (blue strip in Fig.
3a and schematic illustrated in Fig. 2c). Such close proximity does not only reveal the ongoing reaction
between SMS and methanol (through the polarization of the C-H bond of SMS by a neighboring adjacent
oxygen atom), but also showcases the carbene/ylide-character of SMS.
Apart from the carbene mechanism, multiple analogues of oxonium and methane-formaldehyde-type
direct mechanistic proposals have also been shown in the recent literature. The conventional methane-
formaldehyde mechanism of the MTH reaction was originally postulated by Hutchings et al. in 1987.33 The
direct C-C bond coupling reaction between methane and formaldehyde, as a result of disproportionation
of methanol (through its hydride abstraction by SMS, as illustrated in Fig. 2d), was hypothesized to pose
an unrealistically high energy barrier.22 However, several recently postulated proposals aer conceptually
similar to this mechanism. The methoxymethyl cation mechanism proposed by Fan et al. was developed
from the original proposal of Hutchings et al., in order to make it energetically feasible.28, 29
Their
experimentally and theoretically verified proposal involves the formation of direct C-C bond via (i) a
methoxymethyl cation (CH3OCH2+ from SMS and DME, i.e. instead of direct formation of formaldehyde),
and (ii) its subsequent direct C-C coupling with another DME/methanol molecule to form CH3OCH2CH2OR
(R=H, CH3) (a precursor for olefins, as shown in Fig. 2e).28, 29 Similarly, the methyleneoxy mechanism was
13
recently reported by Liu et al. based on in-situ solid-state NMR spectroscopy by measuring C-methanol
14
conversion over H-ZSM-5 using a rotor reactor (Fig. 2f,g & Fig. 3b). Three major bands at 59.5, 69.0 and
80.0 ppm were generated almost instantaneously and were assigned to SMS, surface adsorbed DME and
trimethyl oxonium (TMO), respectively. The appearance of these three bands was accompanied by
formation of ethylene (detected by GC-MS) as well as higher hydrocarbons (chemical shifts of δ=20-
40ppm). Therefore, the direct C-C bond was proposed to be formed as a result of direct interaction
between the reactant (methanol/DME) and zeolite bound surface intermediates, SMS and TMO (Fig. 2f,g).
The important C-H bond activation step could be assisted by framework oxygen to form active
methyleneoxy species (R-O-CH2-H···zeolite), which eventually led to the simultaneous formation of both
olefin (Fig. 2h) and methane/formaldehyde (Fig. 2i). The unprecedented downfield response and relatively
broader nature of SMS at 69.0 ppm under in-situ reaction conditions (compared to δ=59.0 ppm under ex-
situ reaction conditions) was attributed to the strong interaction between the activated DME species and
active surface-zeolite catalyst (Fig. 2f, g). Although both methane and formaldehyde are proposed to be
involved in the catalytic cycle (via assisting the regeneration of SMS), their role towards the formation of
the direct C-C bond was not clarified.14
While numerous mechanistic reports have identified formaldehyde as a side-product, its fate/role during
the MTH reaction is still a subject of debate. Formation of formaldehyde on solid acid catalysts from
methanol is quite predictable, particularly at high temperatures. Interestingly, it could be easily associated
to the Koch-carbonylation mechanism of the MTH reaction (Fig. 2j), another direct mechanistic route
proposed independently by the groups of Lercher and Weckhuysen at the same time.13, 16 Lercher and co-
workers first proposed methyl acetate (CH3CO2CH3, derived via carbonylation of methanol/DME) as the

5
very first C-C bond-containing intermediate during the MTH reaction over H-ZSM-5.16 Spectroscopic
evidence for this proposal came from the Weckhuysen group by employing a combination of solid-state
NMR spectroscopy coupled with operando UV-visible diffuse reflectance spectroscopy (DRS) and on-line
mass spectrometry and it was shown that the Koch-carbonylation route occurred during MTH over H-
SAPO-34.16 In this case, the direct C-C bond containing zeolite-bound acetate species (Fig. 2j) is a Koch-
carbonylation product of SMS. Here, either methanol or formaldehyde, in principle, could act as a
carbonylating agent under MTH reaction conditions. Next, methyl acetate was formed after methoxylation
of surface-bound acetate species, which independently could initiate the formation of HCP species and
thus, olefins. In the carbonyl region of the NMR spectrum, clear cross-peaks to methyl carbon atoms were
observed at both long and short mixing times (Fig. 3c). The signal at 180.5 ppm shows only one cross-
peak, corresponding to zeolite-acetate species, at 22.5 ppm (Fig. 3c). Moreover, the signal at 178.5 ppm
13
has a clear cross-peak with a C signal at 55.1 ppm at longer mixing times only (i.e. methoxy-carbon is
13
not directly attached to the carbonyl-carbon). This C-methoxy signal correlates with a H signal at 3.82
ppm and has an additional very weak correlation signal with a methyl at 22.3 ppm ( 13C) at longer C-C
mixing times (Fig. 3c). This cross-peak pattern is a signature response from a methyl acetatemolecule.
Thus, all three responsible intermediates of this mechanism (SMS, zeolite-acetate and methyl acetate)
were spectroscopically identified by solid-state NMR spectroscopy. This Koch-carbonylation mechanism
of the MTH reaction has recently been theoretically verified by Pleassow and Studt. 30 Interestingly, the
simultaneous existence of (at least) two different direct C-C bond-forming routes (i.e. carbene and Koch-
carbonylation) were spectroscopically identified (Fig. 2c, j & Fig. 3a, c).
Another interesting recent report by Copéret et al. should be mentioned: the formation of the direct C-
C bond from DME alternatively proposed to be catalyzed by extra-framework aluminum atoms (Al) in acidic
zeolites (Fig. 2k).17 Herein, the C-C bond forming step initially involves generation of a transient Al-oxonium
species (i.e. formaldehydic oxygen coordinated to a Lewis acidic Al) through hydrogen abstraction from
an Al-methoxy species. Next, Al-oxonium species react with a methane molecule to yield the direct C-C
bond containing surface-ethanolic species (i.e. a precursor for ethylene). This contribution inevitably
sparks the controversy about the actual involvement of the zeolitic Lewis acid sites during the direct
mechanism of the MTH reaction.15, 17 However, such discussion is beyond the scope of the current review
and demands further research to understand the phenomenon at the molecular level.

Dual-cycle concept.
After a rather short induction period assigned to the formation of the direct C-C bond, the MTH process
continues with the steady-state formation of hydrocarbons in the so-called autocatalytic dual-cycle
concept.34 This notation is the result of merging the two mechanistic schemes developed in parallel for
ZSM-5 and SAPO-34.5 For cage-like zeolites (SAPO-34), able to accommodate large aromatics, light
olefins are believe to form via an indirect way through hydrocarbon pool species (Fig. 4). They can be
visualised as typical ship-in-a-bottle molecules, which can be formed in zeolites but cannot desorb. These

6
species are methylated yielding light olefins like ethylene and propylene by elimination reactions, thus
restoring the initial hydrocarbon pool species. For ZSM-5 and its analogues, formation of hydrocarbons
was rationalized as a result of consecutive methylation and cracking reactions already in the 1980s by
Dessau.35 The dual-cycle represents a compromise between both mechanisms (olefinic and aromatic
cycles) running in parallel (Fig. 4). Both cycles can be subcategorized further to elementary steps
described by six types of reactions: methylation and cracking of olefins, methylation and dealkylation of
aromatics, hydrogen transfer and cyclization, the latter two acting as the bridging step between the two
cycles.8 Using transient switching experiments 12
C/13C Svelle et al.36 showed that, for ZSM-5, all olefins
except ethylene are produced from the olefinic cycle. Ethylene evolution was linked to the presence of
lower methylbenzenes, suggesting that ethylene is mostly a product of the aromatic cycle.37 Sun et al.38
further contributed to this topic by performing seminal kinetic investigations at different conversion levels
in the presence of aromatic and olefinic co-feeds. They concluded that both cycles are active for ethylene
and propylene production, with the aromatic cycle giving similar selectivity for both olefins. The olefinic
cycle, on the other hand, was far more selective to propylene than ethylene, which implies that, if the two
cycles equally contribute to the product distribution, most ethylene will be formed in the aromatic cycle.39
The co-existence of the two cycles naturally renders them as competing.40 In this spirit, one can speculate
that a desired hydrocarbon range can be obtained by either stimulating or suppressing one over the over.
Such co-catalytic features of HCP species during the MTH reaction were later theoretically verified by van
Speybroeck et al,41 indirectly reinforcing the concept of hybrid organic-inorganic nature of a working MTH
catalyst, as originally proposed by Svelle et al. 36
UV-visible DRS is possibly the most utilized spectroscopic technique for the characterization and
identification of zeolite trapped organics. The biggest advantage of UV-visible DRS is its ability to
differentiate between carbocationic HCP species and their neutral counterparts. For instance, the
absorption band of any arenium HCP cation is lower in energy than any electronic transition of its neutral
counterpart. Moreover, UV-visible DRS provides insightful information regarding the zeolite framework
dependent formation of deactivating species during the course of reaction. In general, multiple bands at
around ≤295, 340-360, 385-410, 460-500 and ≥600 nm, are observed during a MTH reaction and were
typically attributed to neutral benzene/cyclopentadienyl species, dienylic carbocationic/methylbenzeniums
(up to three methyl groups), highly methylated areniums (specifically hexamethylbenzenium ions, HMB+),
trienylic and methylated polyarenium ions, respectively (Fig. 3d,e).25, 26, 42, 43, 44, 45, 46, 47 The formation and
characteristics of these bands are quite unique depending on the zeolite framework topology and acidity.
For instance, the specific role of alkyl cyclopentadienium ions over MFI zeolite (e.g. H-ZSM-5) was
identified by Jentoft and Wulfers using in-situ UV-visible DRS (Fig. 3d).46 Similarly, the Weckhuysen group
identified the nature of governing active and deactivating methylated aromatic species during the MTH
process over CHA zeolites (e.g. H-SAPO-34 and H-SSZ-13) employing a combination of operando UV-
13, 42, 47
visible DRS and online gas chromatography/mass spectrometry. Using a similar strategy involving
operando UV-visible DRS, the same group very recently monitored the formation of active HCP species

7
and the accumulation of coke molecules during both H-ZSM-5 and Mg-ZSM-5 catalysed MTH.48 Such
spatiotemporal UV-visible spectroscopic approach reveals the formation of a coke front at the beginning
of reactor bed, which travels towards the end until full deactivation. Mg modification resulted in slower
progression of the coke front and higher olefin selectivity. However, identification/assignment of zeolite-
trapped any organic reaction intermediates by UV-visible DRS in combination with theoretical calculations
is not always straightforward and, to some extent, confusing. For example, HMB+ and the 1-methylnaphthyl
cation display a similar absorption band at ~390 nm.41, 44 Unfortunately, both are probable intermediates
of the MTH reaction, while HMB+ is an active HCP species and the other is formed during the deactivation
period.41 This is the reason behind the enormous rise of utilization of solid-state NMR spectroscopy in
recent years for the accurate structural elucidation of trapped organics within zeolites (vide supra).13, 14, 17
The combination of solid-state NMR with UV-visible DRS has made significant advancement by the groups
of Hunger18, 25, 43 and Haw.49, 50, 51
Due to its capability to provide information at the molecular level, the utilization of solid-state NMR
spectroscopy is increasing gradually in the field of heterogeneous catalysis.13 Using 13C-enriched methanol
not only significantly increases NMR sensitivity, but also allows multi-dimensional solid-state NMR
correlation experiments to be performed in order to construct the accurate molecular structures along with
structural information of dominant carbenium ions involved in the HCP of the MTH-mechanism. 6, 18, 49 For
instance, the alkylated cyclopentadienium ion was detected as the primary HCP species over HMB+ within
H-ZSM-5 and H-SSZ-13 zeolites during MTH reaction, whereas hexa-/hepta-methylbenzenium cation is
widely acknowledged as the governing HCP species within SAPO-type molecular sieves (like SAPO-34,
6, 13, 24, 51, 52, 53, 54
DNL-6) and H-Beta (Fig. 3f). This observation was quite consistent with the UV-visible
13, 42, 47
DRS reports by the groups of Jentoft and Weckhuysen. These results demonstrate that the
formation of dominant carbenium HCP species is entirely dependent on the zeolite’s framework and
acidity.
In a nutshell, the scientific community now accepts that the direct mechanism exists during the early
stages of MTO reaction and that hydrocarbon pool species vary depending on zeolite framework, acidity
and reaction conditions. Numerous recent mechanistic reports provide solid experimental and
spectroscopic evidence and the simultaneous existence of multiple direct C-C bond forming routes during
the MTO reaction is quite likely.6
Based on this mechanistic knowledge, several strategies can be put forward to drive selectivity of the
process in different directions depending on the desired product. The first approach is related to the
manipulation of the organic counterpart, i.e. changing the concentration of olefin and/or aromatic species.
Obviously, the second approach deals with the inorganic element and can be achieved by making use of
catalyst engineering. Taking propylene and ethylene (and/or aromatics) yields as the measure of cycle
dominance, we will now analyse how certain factors selectively propagate one cycle over the other.

Towards improved selectivity in MTH.

8
 When talking about catalyst and process design, the induction period was disregarded as a possible
tool to manipulate both selectivity and lifetime because for a long time it was considered as a negligible
part of the entire MTH mechanistic picture. Recent investigations indicated that formaldehyde – formed
through disproportionation of methanol during direct C-C bond formation – causes catalyst deactivation
via interaction with aromatic molecules resulting in the formation of polycondensed aromatics. 55 Besides,
formaldehyde formation is accompanied with the production of CH4.56 It thus becomes clear that in order
to achieve better catalyst stability and selectivity, the induction period should be re-engineered to avoid
formaldehyde formation. This issue can be addressed at both the catalyst and reactor level. At the reactor
level, utilization of a continuous stirred tank reactor (CSTR) instead of a plug flow reactor (PFR) leads to
lower local methanol concentrations and lower oxygen containing species, thus mitigating catalyst
deactivation.57 Moreover, utilization of DME instead of methanol avoids the potential formation of
formaldehyde and results in higher methylation rates in comparison to methanol, substantially prolonging
catalyst lifetime.56 At the catalyst level, it was proposed that addition of rare-earth oxides, such as Y2O3,
should selectively decompose formaldehyde thus preventing its further interaction with aromatic species.58
Utilization of DME (DME/water) instead of methanol seems to be the best solution, since apart from
mitigating catalyst deactivation it decreases the overall heat release (the reaction enthalpy of DME
dehydration is lower than that of methanol) making it more attractive from an industrial perspective.59

Taking advantage of autocatalysis.

Both olefins and aromatics are recognized as competitive co-catalysts, the excessive presence of a
certain product promotes the cycle from which it originates.38, 40 So the most straightforward strategy to
enhance the yield of olefins (propylene and butenes) or aromatics is to increase their concentration by co-
processing them with methanol. Several works report the effect of co-feeding a wide range of olefins and/or
aromatics with methanol on the product distribution, however, with the main aim to shed light on
mechanism details.60, 61
In a wide range of temperatures and at different methanol conversion levels, co-feeding of toluene
results in an increase of ethylene and methylbenzene concentration at the expense of propylene and
higher olefins as a result of the aromatic cycle propagation.40, 62, 63, 64 This strategy can be used to respond
to the constantly fluctuating market demands to produce a mixture of olefins with preassigned C2=/C3= ratio.
Olefins co-feeding turns out to be less straightforward and the propagation of the corresponding cycle
depends on other parameters. The JGC corporation reported significantly enhanced propylene yield (up
to 60-70%) when C4-C5 products were recycled with methanol feed at temperatures higher than 843K. 65,
66
Co-feeding a small amount of propylene at low conversion levels at 548-623 K results in a higher
contribution of the olefin cycle, confirmed by higher selectivity towards C3+ products,63 while at 723 K and
higher conversion levels there is no notable effect of olefins co-feed.40 In the first scenario, methylation
reactions are promoted over cracking, therefore any inclusion of short chain olefins like propylene results
in a dramatic increase of C3+ products. At high methanol conversion levels (>70%), the olefinic cycle

9
becomes more important,38 while further addition of olefins promotes both formation of higher olefins,
which with equal success are either cracked or aromatized, thus contributing to both cycles. This brings
another important conclusion – process parameters such as temperature and methanol space-time also
contribute to the promotion of one cycle over the other when bed effects come into play. Temperature is
another decisive parameter in determining the type of prevailing reactions. Higher temperatures promote
cracking of higher olefins rather than their cyclization, which can be perceived as promotion of the olefinic
cycle. Therefore, to maximize propylene production, the MTP process is carried out at temperatures higher
than 723 K.

Influence of zeolite topology on product selectivity.

The zeolite skeleton dictates its shape-selective properties, allowing the participation in the chemical
reaction of only those molecules able to fit inside the zeolite pores.67 This broad definition can be further
subcategorized into reactant-, product- and transition-state selectivity. Thus, being an inherent attribute of
zeolitic materials, shape selectivity can be effectively utilized to propagate one cycle over another via
product- and transition-state selectivity.68, 69
Small pore 8MR zeolites are composed of large cavities interconnected by narrow window openings
(Fig. 5a,b).70, 71 Window openings are limited by 8 atoms, big enough to let short-chain olefins pass through
but sufficiently small as to retain bigger molecules inside their cages, thus creating an ideal playground to
incubate the aromatic-based hydrocarbon pool.7 In such an architecture, cavity dimensions can influence
the nature of the aromatic intermediates (transition-state selectivity),72 while window dimensions impose
restrictions for the molecules formed inside (product selectivity). Thus, slight differences in both
dimensions can affect product distribution and propylene/ethylene ratio, however guaranteeing high
selectivity to short-chain olefins (up to 90%).73 Note that the hydrocarbon pool trapped inside also serves
as a scaffold for the formation of polycondensed species, eventually leading to catalyst deactivation. The
challenge for this type of zeolites is to achieve steady-state performance of the aromatic cycle by slowing
down coke formation, accomplished by utilization of silicoaluminophosphates with milder acidity.74 On the
other hand, steady-state performance with a constant coke content can be achieved by utilizing fluidized
bed reactors, which additionally provide the advantage of a better heat dissipation.7 Such a reactor design,
however, requires application of fluidizable catalysts, i.e. with specific particle size and density. CHA - the
most studied topology – is composed from cylinder-like cavities with big dimensions able to host aromatic
molecules up to pyrene (Fig. 5a,b).75 Such big dimensions ensure long lifetime in comparison to other 8MR
zeolites. Linked to the product distribution, the following conclusion can be drawn on the effect of cage
dimensions: the more spacious is the cage the higher is selectivity to propylene.72, 76, 77 Compared at the
same temperature (350 and 400oC), LEV with the smallest dimensions (7.5 x 6.5 Å) yields higher amount
of ethylene, while more spacious CHA (10.9 x 6.7 Å) and AEI (12.7 x 11.6 Å) – propylene.76 The product
distribution can be further linked to the confined aromatics, that is, steric limitations imposed by LEV favour
formation of methylbenzenes with limited amount of methyl groups which are responsible for ethylene

10
formation. The pear-shape of AEI is able to hold bulky aromatics, resulting in an unusually high selectivity
to butenes (propylene/ethylene/butane =2.8 / 1 / 1.1) at 400oC.78 Large cage dimensions do not guarantee
high selectivity to propylene/butenes if combined with very narrow pores hindering diffusion of the latter.79
An example is AFX (13.0 x 8.3 Å) with very small window-openings (3.4 x 3.6 Å) promoting rapid growth
of polycondensed species and therefore fast deactivation.77 Similarly, the more spacious cage of ERI
delivers higher selectivity to ethylene in comparison to CHA, which might be due to narrower window
openings of ERI (3.6 x 5.1 Å) in comparison to CHA (3.8 x 3.8 Å). 80
Medium pore 10-ring zeolites, on the other hand, are composed of straight and/or curved channels
(Fig. 5c-f). In such structures, the aromatic cycle is only able to operate in the more spacious channel
intersections.81, 82 Choosing zeolites with absence of those ensures discarding the aromatic cycle. One-
dimensional ZSM-22 with TON topology was the first zeolite for which suppression of the aromatic cycle
was postulated,83 the same observations latter being made for other 1D 10-ring zeolites (Fig. 5c,d).84, 85
The prevalence of the olefinic cycle combined with product shape selectivity results in the predominant
formation of C5+ products ranging from 50 up to 75 % depending on the conversion levels. The obtained
product mixture rather meets the requirements for gasoline (after hydrogenation step), with the formation
of propylene being fairly low. Selectivity to propylene can be further improved by optimizing acidic
properties of zeolites leading to up to 53 % of propylene selectivity at 450oC,86 while optimization of textural
properties dramatically prolongs catalyst lifetime.87 For zeolites with intersections, two cycles work in
parallel, and propagation of one cycle over the other can be achieved by other means (Fig. 5 e,f). 88
Logically, further increase of the number of T-atoms in window openings corresponding to 12-ring
zeolites allows the aromatic cycle to proceed inside the zeolite channels. Therefore, 1D SSZ-24 with
tubular channels comparable to the size of CHA cavities is almost exclusively selective to aromatics at low
conversion levels (≈90%) and >26 % at high conversion levels (Fig. 5 g,h).89 Besides, wider window
dimensions result in the formation of much more alkylated aromatics for zeolite beta and MOR (mainly
penta- and hexamethylbenzene) in comparison to 10-ring ZSM-5 (mainly BTX).90 Thus, in 8- and 12-ring
zeolites composed of large cavities and channels respectively, the aromatic cycle is preferred.
Nevertheless, being imprisoned by narrow 8-ring windows it selectively produces short chain olefins in one
case, and heavy aromatics with no such restrictions.

Impact of acidity on stability and product selectivity.

Concentration, location and strength of acid sites are the three most important parameters in defining
the overall acidity in zeolites. Concentration of Brønsted acid sites is primarily linked to the amount of Al
in the zeolite. A literature survey reveals a linear correlation between propylene selectivity and Si/Al ratio,
while a notable reduction in aromatics selectivity corresponds to the decrease of Al content in the zeolite
(established for MFI but also holding for other topologies).91, 92, 93, 94, 95, 96 Opposite trends strongly evidence
the competing nature of the two cycles and that higher acid site density propagates the aromatic cycle
(Fig. 6a-d). An increase of Al content enhances the chance of reactant molecules to interact with each

11
other, thus increasing the chance to form aromatics and ethylene – the product of the aromatic cycle.97
The observed trends also account for the improved selectivity to propylene for post-synthetically modified
catalysts with different elements as well demetalated zeolites. It has been extensively shown that
phosphatation and incorporation of alkaline-earth metals lead to a significant reduction of Brønsted acidity
and its effect can be regarded as an increase in the Si/Al ratio.98, 99 Yarulina et al. have shown that Ca
incorporation results in an almost 10-fold reduction of Brønsted acidity. As a consequence, hydride transfer
and cyclization reactions were suppressed, resulting in very low amounts of paraffins and aromatics and
maximizing propylene yield up to 53%.98 Apart from the development of the second type of porosity
improving diffusional properties of zeolites, desilication and dealumination of zeolites can be also viewed
as an instrument to dilute acid site density and inhibit secondary reactions.94
Zeolites with similar bulk properties can still exhibit substantially different catalytic behaviour, which is
a side effect of a heterogeneous distribution of Al within the crystal, also called Al zoning. For example, an
Al-rich rim is characteristic for ZSM-5 crystals synthesized using TPA+ as structure-directing agent.100 Al-
zoning obviously can be considered as a local enhancement of acid density, which, as explained above,
promotes the aromatic cycle. Hydrothermal synthesis conditions and the precursor composition mixture
can also promote different locations of Al within the crystal lattice. Here it should be pointed out that the
definition of Al zoning used in the following describes a gradient in the number of Al atoms per unit volume,
not a distribution gradient of Al over the available T sites per unit cell. Al zoning in zeolites has been
observed since the 1970s.101 Around that time and in the following years different distributions and effects
were observed and in 1993 Althoff and co-workers published a systematic study of the parameters
influencing Al zoning.102
Besides influencing desilication and dealumination, because strength and number of catalytically active
sites in zeolites affect the effective diffusion path of reactants, the presence and distribution of Al in the
zeolite framework also directly affects catalytic performance (Fig. 6). It was shown that several acid-
catalyzed reactions preferentially occur at the edges of large ZSM-5 crystals: in 2007, Roeffaers et al.
applied in-situ fluorescence microscopy to monitor catalytic, condensed-phase reactions in individual
zeolite crystals.103 Specifically, they utilized the (acid-catalyzed) formation of chromophores during self-
condensation of furfuryl alcohol to visualize the time dependent distribution of the fluorescent reaction
products. With their study they showed that the presence of sub-units in large, coffin-shaped ZSM-5
crystals is relevant to catalysis, and proposed that the interfaces between the components act as diffusion
barriers. The authors also observed unusual behavior of certain crystals, showing intense fluorescence
emission confined to the outer surface of specific facets of the ZSM-5 crystals and linked this unusually
intense emission to Al zoning. In the same year Kox et al. used in-situ UV-Vis micro-spectroscopy to study
the oligomerization of styrene occurring in the micropores of ZSM-5 zeolite crystals,104 linking reaction
kinetics to the diffusion and catalytic properties of straight and zigzag channels in large ZSM-5 zeolite
crystals. One year later Tzoulaki et al. studied the diffusion properties of large (aluminum-free) Silicalite-1
crystals using interference microscopy for monitoring time-dependent concentration profiles during

12
molecular uptake and release and found no evidence for a significant effect of internal transport
resistances (at the interface of sub-units) or surface barriers on mass transport in these crystals. 105 They
therefore concluded that an inhomogeneous Al distribution that is preferentially located close to the surface
and at the interfaces of the crystal segments enhances catalytic activity and coke formation and is
responsible for the observed diffusion barriers. In line with this hypothesis, later in 2008 Mores et al.
confirmed the heterogeneous distribution of coke as a function of reaction time and temperature in large
H-ZSM-5 and H-SAPO-34 crystals during the MTO reaction by a combination of in situ UV-visible and
confocal fluorescence microscopy.106 Three years later Weckhuysen et al.107 showed that with decreasing
Al concentration in large MFI crystals, coke formation decreases as well. Furthermore, a higher Brønsted
acid site density did not influence the type of coke species generated, but instead increased the rate of
formation of methyl-substituted aromatic species and the subsequent growth towards larger coke species.
One year later, in 2012, Chen et al. reviewed the effect of acid site density, acid site strength and the role
of coke formation on deactivation of SAPO zeolites74 and in 2015 Olsbye et al. reviewed the MTH process
inter alia discussing catalyst deactivation by coke formation.6
Catalyst deactivation by (hydro-) carbon residues is an important deactivation pathway in MTH.
However, as discussed, after the formation of the first C-C bond species hydrocarbons are formed
according to the autocatalytic dual-cycle concept, which implies that hydrocarbon species can be both
activating and deactivating species.6 Therefore it is interesting to obtain insights about the nature and
location of the first seeds of coke in zeolites. Intuitively one would expect the first coke species to form in
regions with a) highest accessibility and b) of highest activity – both of which can be influenced by Al
gradients in the catalyst. However, resolving the location of the first coke species formed at the length
scale of nanometers and relating it to (local) Al gradients remains extremely challenging. Recently, the first
use of atom probe tomography (APT) to investigate the 3-D distribution of elements in zeolites at the sub-
nm scale has been reported.108, 109 In 2016 Schmidt et al. used APT to study coke formation and its relation
13
to inhomogeneous Al distribution in large ZSM-5 crystals after MTH (using C-labelled methanol to
109
distinguish the formed coke species from contaminations). APT samples were taken from the (Al rich)
surface and the (Al poor) core of the ZSM-5 crystal showing a clear correlation of coke formation and
Brønsted acid site density. The authors reported the presence of carbon clusters (using the term cluster
as synonym for a group of closely positioned atoms), which revealed insights into the coke formation
mechanism, showing that the coke clusters form preferentially around areas with elevated Al content, even
13
in the Al-poor core of the crystal. The median size of the observed C clusters was around 36-69 carbon
atoms, which suggests that these clusters could contain several occluded aromatic species. Independently
of the location of the APT (i.e. taken from the surface or core of the large ZSM crystal), 13C clusters were
found in each sample, suggesting that in the beginning of the reaction methanol is present throughout the
crystal (Fig. 6). As the MTH reaction and catalyst deactivation progress these large clusters merge into
the coke rich regions observed e.g. in the surface of large ZSM crystals, i.e. in regions of elevated Al

13
concentrations. These observations were found to be in line with the previously suggested mechanism of
coke formation in ZSM-5 during MTH (references 6, 93-96, among others).
Variation in silicon sources was taken as a strategy by Wang et al.81 to obtain ZSM-5 zeolites with acid
sites located either in intersections or in straight and sinusoidal channels. Having the same Si/Al ratio,
zeolites with acid sites located in channels exhibited significantly higher selectivity to propylene and longer
lifetime. Such a location of Al inhibited the aromatic cycle which is operative in the more spacious
intersections and resulted in the similar effect as if utilization of 1D 10-ring zeolites.
Finally, the effect of acid strength was extensively compared over aluminosilicates and their SAPO
versions. The general observation is that C3=/C2= ratio and methanol conversion capacity over SAPOs is
higher suggesting lower formation rates of poly-condensed species.110 However any speculation on the
dominating cycle is rather dubious as it was studied over 8-ring zeolites where the aromatic cycle is
prevailing due to topology features. A similar comparison performed for 12-ring zeolites, reveals that while
for H-SSZ-24 (AFI) the product distribution is dominated by aromatics, H-SAPO-5 (AFI) displays high
selectivity to butenes and shows almost no aromatics. A combination of co-reaction experiments and
theoretical calculations reveals that the acid strength has a profound effect on the reactivity of co-catalytic
hydrocarbon pool species.89 For H-SSZ-24, benzene methylation is significantly higher than that for
propylene bringing the aromatic cycle into the dominating role - in sharp contrast to H-SAPO-5 exhibiting
similar rates for olefins and aromatics methylation.111

Host-guest interactions.
Throughout this review article, zeolite’s framework topology and acidity dependent formation of products
(both olefins and HCP species) during the MTH reaction have been highlighted. This feature could easily
be linked to the host-guest chemistry between inorganic zeolite and organic HCP species during the
reaction. More importantly, it also provides the necessary evidence in support of the hybrid organic-
36, 41
inorganic nature of working MTH catalyst (vide supra). During evaluation of HCP species during the
MTH process over three small-pore and eight-ring windows zeolites (CHA, DDR, LEV) by Goetze et al.,
generation of non-identical organics was noticed due to very small differences in cage size, shape, and
pore structure of the zeolite frameworks.112 By encompassing a combination of multivariate analysis of
operando UV-visible spectroscopy and online gas chromatography along with bulk chemical analysis of
the hydrocarbon deposits by GC/MS of extracted coke species and thermogravimetric analysis, it was
observed that CHA, DDR and LEV were preferentially formed alkylated aromatics and pyrene, 1-
methylnaphthalene, and methylated benzene and naphthalene, respectively. The molecular dimension of
these retained organic species (i.e. the guest molecule) is comparable to the dimensions of the respective
zeolite cages (i.e. the host molecule). As a result, lattice of all three zeolites were expanded during the
113
MTH process, as revealed by operando X-ray Diffraction (XRD). The expansion of the c-axis of the
CHA and LEV lattices (i.e. longitudinal direction of their cages) is about 0.9 % and 0.5 % during the MTH
process, which is more than in the direction of a- and b-axes (i.e. in the direction of the width of the cage).

14
Contrary, relatively more expansion in the direction of a- and b-axes of DDR lattice (i.e. 0.5 % compared
to 0.3 % along the c-axis) means it becomes wider during the MTH process due to coke deposition,
whereas both CHA and LEV lattices become longer. This observation provides further evidence for
distinctive host-guest chemistry between zeolite and retained organics during catalysis. To further
elucidate the concept of host-guest interactions during catalysis, the report from Liu et al. should be
mentioned. With the help of isotopic tracing and theoretical calculations, they showed that the cavity size
of different SAPO molecular bearing identical 8MR pore openings (i.e. SAPO-35, SAPO-34, and DNL-6)
control the molecular size as well as reactivity of the confined HCP-species, which eventually influences
72
the conversion and selectivity of olefins during the MTH reaction. Although such
confined/retained/trapped organics within the zeolite framework are believed to be rigid in nature, this is
not completely true. While elucidating the reaction mechanisms of zeolite catalyzed MTH (Fig. 3a,c) and
alkylation of aromatics, the same group adapted different solid-state NMR magnetization transfer
techniques, which were previously developed for spectral separation of biomolecules on the basis of their
mobility, in order to distinguish between mobile (i.e. molecule or group with fast tumbling or rotation) and
rigid (i.e. molecule physisorbed in/on zeolite) species (cf. both were trapped within zeolite). 13, 114 However,
this host-guest feature of zeolite catalysis has recently been developed and demands further research
activities to confirm its implications in the actual process. Nevertheless, it contributes to the fundamental
understandings of the zeolite-catalysed hydrocarbon conversion chemistry.

Summary and Outlook.

After more than 40 years, the methanol-to-hydrocarbons process remains as one of the most popular
topics of research within the zeolite catalysis community. The recent industrial implementation, with
several new commercial plants already running, has triggered even more attention from the academic and
industrial communities. In this line, the interest towards the formation of the direct C-C bond, which until
very recently was believed to be due to impurities, has risen over the last few years. The combined effort
by several research groups helped identify several feasible low-barrier direct coupling pathways for this
key mechanistic step, although feedstock impurities (in principle) could still shorten the induction period.115
Among them, formaldehyde formed through disproportionation of methanol was identified as a deactivating
species due to its high reactivity with aromatics and the undesired formation of poly-aromatic species. This
important discovery calls for new reactor concepts where reactant feeding should be performed carefully
to avoid high concentrations of methanol and the consequences thereof, including heat effects.
The next mechanistic step, the dual-cycle, is now better understood and can be viewed as the main
machine for controlling lifetime and selectivity to the desired products. The combination of NMR and UV-
visible DRS helped pinpoint not only active intermediates, but also deactivating ones, although there is a
fine line between them. It has been extensively shown that deactivating species can become reaction
intermediates if, for example, higher reaction temperatures are used. Moreover, their reactivity depends
on zeolite topology: channel/cage dimensions can cause spatial restrictions preventing certain

15
intermediates to participate in the mechanistic cycle. In 2012, Olsbye et al. posed the question whether
the two cycles – olefinic and aromatic – can be promoted one over another and whether one of them can
independently exist. The challenge has been partially addressed: recent catalytic results demonstrate that
it is possible to extend catalyst lifetime up to circa 1kg CH3OH/g zeolite while maintaining high propylene
selectivities (>50%). However, it should be further clarified if deactivating species in these cases are also
of aromatic nature.
Considering that for both cycles the active intermediates act as co-catalysts, reaction conditions
(temperature and methanol partial pressure), feed composition (i.e. the presence of olefinic or aromatic
co-feed), zeolite topology and acidity can be tuned to further enhance performance. Having this said, most
of the efforts over the last few years have been directed towards the selective formation of propylene (the
main product from the olefinic cycle). In a potential scenario of lack of ethylene, it would be desirable to
apply similar concepts towards the selective formation of this highly important chemical. How to achieve
this objective without compromising catalyst lifetime is however still an open question, as ethylene is the
main product of the aromatic cycle, its formation would in principle always be accompanied by a high rate
of formation of aromatics. Along the same line, considering potential fluctuations in the olefin market, in an
ideal scenario there should be a catalyst whose selectivity can easily be tuned by changing the reaction
conditions. Here the best solutions seem to go in the direction of co-feeding the required products to
promote the cycle of interest. This aspect again calls for additional efforts at the reactor level.
In summary, in our opinion, MTH is still a fascinating research topic with a number of scientific and
engineering challenges to be addressed in the near future. Surprisingly, the latter (engineering) challenges
have hardly been touched upon in the open literature, while these have been key in the success of the
currently most applied MTH technology, which relies on the fast deactivating SAPO-34 and on fluidized
bed technology. On a more fundamental note, we are looking forward to seeing the implications that the
recent discoveries in MTH chemistry may have on other high temperature hydrocarbon chemistries, such
as catalytic cracking and even direct methane activation, as well as bifunctional catalysis concepts, in
which MTH chemistry may play an instrumental role. Indeed, the combination of zeolites with other
functionalities opens new avenues for the direct conversion of CO2 or syngas to olefins, hydrocarbons and
aromatics.116, 117 In such a bifunctional systems, CO2 or CO are first hydrogenated to methanol which is
subsequently converted to hydrocarbons, the zeolite of choice being responsible for the second step. This
is a rather promising concept, yet there are a number of obstacles to overcome, such as overhydrogenation
of olefins to paraffins, high selectivity to CO and methane associated with metal functionality and how to
find a proper catalyst matching in terms of deactivation.

16
Figure 1. Milestones and mechanism development of the methanol-to-hydrocarbon (MTH) process. (a) MTH
chronology shows that forty years of combined efforts produced several mechanistic concepts for both initial
induction period and steady state period. The current simplified mechanistic picture is shown in (b). The simplified
scheme of the MTH mechanism reveals that though the aromatic cycle leads to the formation of light olefins, it also
acts as a scaffold for coke formation; therefore olefins formation is preferred over the olefins cycle.

17
Figure 2. Several proposed direct mechanistic routes during the early stages of the zeolite catalyzed
methanol-to-hydrocarbon (MTH) process. The proposed mechanism of the methanol conversion reaction between
surface-methoxy species (SMS) and (a) cyclohexane and (b) ethylene via the carbene-type intermediate. (c) The
spectroscopically verified intermediate consists of SMS and methanol, suggesting a possibility of C-H insertion of
methanol by carbene-like SMS species. Schematic illustration of (d) the conventional methane-formaldehyde
mechanism, and its modified analogue (e) the methoxymethyl cation mechanism. (f-i) The methyleneoxy mechanism
of the MTH reaction, involving either (f) the trimethyloxonium or (g) the SMS intermediate. The SMS involving route
initiates the formation of both (h) ethylene and (i) methane/formaldehyde simultaneously. Plausible routes for the
direct formation of the carbon-carbon bond during early stages of the MTH reaction through (j) Koch-type
carbonylation of SMS. (k) The methane-Al/oxonium mechanism for the conversion of methanol over alumina.

18
Figure 3. The spectroscopic signatures of crucial intermediates during the zeolite catalyzed methanol-to-
hydrocarbon (MTH) process. (a) Solid-state NMR spectra of methoxy in H-SAPO-34 after the MTH reaction for 30
minutes at 673 K. Zooms from 2D 13C-13C (blue strip) and 13C-1H (red strip) magic angle spinning (MAS) solid-state
NMR spectra with long mixing (150 ms) and cross-polarization (CP) contact time (500 ms), respectively, identifying
surface adduct between SMS and methanol (arrow in molecular structure indicates 13C-13C correlation). (b) In-situ
13C MAS solid-state NMR spectra measured during MTH reaction over H-ZSM-5 at 573K for 12 min. (c) Solid-state
NMR spectra of acetate species in H-SAPO-34 after the MTH reaction for 30 minutes at 673 K. Zooms from 2D 13C-
13C (blue strip) and 13C-1H (red strip) MAS solid-state NMR spectra with long mixing (150 ms) or CP contact time
(500 ms), respectively, indicating surface acetate and methyl acetate resonances. UV-visible diffuse reflectance
spectra of (d) H-ZSM-5 and (e) H-SAPO-34 being exposed to the MTH reaction under in-situ (573K) and operando
(673K) reaction condition, respectively. An absorption band at (d) 295 nm on H-ZSM-5 and (e) 387 nm on H-SAPO-
34 were assigned to a methylated cyclopentadienium and hexamethylbenzenium ions, respectively. ((f) 13C MAS
solid-state NMR spectra of retained organic species in H-SAPO-34 (upper spectrum, 573K, 15 min) and H-SSZ-13
(lower spectrum, 548K, 25 min) after continuous-flow of 13CH3OH for 15 min (*=spinning side bands, chemical shifts
on the molecular structures were theoretically calculated values).

19
Figure 4. Steady-state mechanism development of the methanol-to-hydrocarbon (MTH) process. (a) Dessau
and LaPierre described steady state kinetics of MTH reaction over ZSM-5 by a sequence of olefins methylation
followed by their cracking or aromatization to give either olefins or aromatics. (b) In contrast, Dahl and Kolboe
investigating MTH over SAPO-34 introduced the hydrocarbon pool concept – an active intermediate of aromatic
nature – through which light olefins are formed by dealkylation. (c) The dual-cycle postulates that there are two
competing cycles running in the zeolite channels governed by olefins and aromatics, both acting as co-catalysts for
MTH and being active hydrocarbon pool species.

20
Figure 5. Impact of topology on cycles propagation. (a) 8-ring zeolites are composed of large cavities with small
window openings; in such structures aromatics can be formed but cannot escape the cage (b). The aromatic cycle
is prevailing and such topologies are characterized by high selectivity to short-chain olefins. (c) 1D 10-ring zeolites
cannot accommodate aromatic molecules, therefore the aromatic cycle is suppressed in such type of topology. The
product distribution is characterized by a gasoline range of hydrocarbons. (e) The 3D 10-ring MFI structure is
described by two types of channels, straight and sinusoidal, forming relatively spacious intersections, which can host
the aromatic cycle. (f) Both the aromatic and olefinic cycles are running in zeolite channels and the product is
characterised by the presence of a wide range of hydrocarbons. (g,h) The channels of 12-ring zeolites are rather big
and can host aromatic hydrocarbons, therefore the aromatic cycle can run inside the channels of 1D AFI zeolite.

21
Figure 6. Impact of acidity on cycles propagation. (a, b) A literature survey (ref. 91-96) reveals (a) a noticeable
increase of propylene and (b) a reduction of aromatics formed along with an increase of the SiO2/Al2O3 ratio for
microporous ZSM-5 zeolites tested in the methanol-to-hydrocarbons reaction at 673 – 773 K and in the wide range
of WHSV. Opposite trends highlight the competing nature of aromatics and olefins. (c) A significant reduction of Al
in the zeolite framework leads to a decrease of BAS density, consequently propagating the olefinic cycle as a result
of a decreased amount of secondary reactions leading to olefins cyclization and aromatic formation. (d) An increase
of acid site density results in a higher probability for olefins to interact forming aromatic molecules. A reduction of
Brønsted acidity can be achieved by post-synthetic modification with alkaline-earth metals or demetalation
procedures (c, d) leading to similar trends.

22
Figure 7. Resolving the location of the first coke species formed during the methanol-to-hydrocarbon (MTH)
process using atom probe tomography (APT). APT was used to study coke formation and its relation to the
inhomogeneous Al distribution in large ZSM-5 crystals after MTH reaction.109 A cross section was prepared from a
large ZSM-5 crystal after MTH reaction using 13C-labelled methanol by focused ion beam (FIB) cutting (panels a and
b). For APT four samples (needles) were taken using FIB cutting from different regions of the crystal cross-section
considering Al zoning: high Al concentrations are indicated in blue in panel c also showing the position of the samples
investigated by APT. A typical needle is shown in panel d. APT then allowed mapping of the 3D distribution of Si, Al,
O, and 13C atoms within each needle (panel e). Statistical analysis of the distribution of the 13C atoms revealed the
presence of carbon clusters indicating the regions of first coke formation in the catalyst (panel f).

23
References.
1. Chang, C.D. & Lang, W.H. Process for manufacturing olefins. US4025576 A (1977).

2. Vogt, E.T.C. & Weckhuysen, B.M. Fluid catalytic cracking: recent developments on the grand old lady of
zeolite catalysis. Chem. Soc. Rev. 44, 7342-7370 (2015).

3. Mitchell, S., et al. Structural analysis of hierarchically organized zeolites. Nat. Comm. 6, 8633-8647 (2015).

4. Stöcker, M. Methanol-to-hydrocarbons: catalytic materials and their behavior. Microporous Mesoporous

Mater. 29, 3-48 (1999).A review covering the main achievements in catalyst design and mechanism
understanding of methanol-to-hydrocarbons process over the last century.

5. Olsbye, U., et al. Conversion of methanol to hydrocarbons: how zeolite cavity and pore size controls product
selectivity. Angew. Chem. Int. Ed. 51, 5810-5831 (2012).A methanol-to-hydrocarbons review dedicated
to mechanism understanding and describing main parameters affecting selectivty to hydrocarbons.

6. Olsbye, U., et al. The formation and degradation of active species during methanol conversion over
protonated zeotype catalysts. Chem. Soc. Rev. 44, 7155-7176 (2015).

7. Tian, P., Wei, Y., Ye, M. & Liu, Z. Methanol to Olefins (MTO): From Fundamentals to Commercialization.
ACS Catal. 5, 1922-1938 (2015).

8. Ilias, S. & Bhan, A. Mechanism of the Catalytic Conversion of Methanol to Hydrocarbons. ACS Catal. 3, 18-
31 (2013).

9. Van Speybroeck, V., et al. First principle chemical kinetics in zeolites: the methanol-to-olefin process as a
case study. Chem. Soc. Rev. 43, 7326-7357 (2014).

10. Schulz, H. About the Mechanism of Methanol Conversion on Zeolites. Catal. Lett., (2018).

11. Yamazaki, H., et al. Evidence for a "carbene-like" intermediate during the reaction of methoxy species with
light alkenes on H-ZSM-5. Angew. Chem. Int. Ed. 50, 1853-1856 (2011).

12. Yamazaki, H., et al. Direct production of propene from methoxy species and dimethyl ether over H-ZSM-5.
J. Phys. Chem. C 116, 24091-24097 (2012).

13. Chowdhury, A.D., et al. Initial Carbon-Carbon Bond Formation during the Early Stages of the Methanol-to-
Olefin Process Proven by Zeolite-Trapped Acetate and Methyl Acetate. Angew. Chem. Int. Ed. 55, 15840-
15845 (2016).An in-depth mechanistic investigation providing spectroscopic evidence in support of
the Koch-carbonylation mechanism of the methanol-to-hydrocarbon reaction.

14. Wu, X., et al. Direct Mechanism of the First Carbon-Carbon Bond Formation in the Methanol-to-
Hydrocarbons Process. Angew. Chem. Int. Ed. 56, 9039-9043 (2017).

15. Lercher, J.A. New Lewis Acid Catalyzed Pathway to Carbon–Carbon Bonds from Methanol. ACS Cent. Sci.
1, 350-351 (2015).

16. Liu, Y., et al. Formation Mechanism of the First Carbon-Carbon Bond and the First Olefin in the Methanol
Conversion into Hydrocarbons. Angew. Chem. Int. Ed. 55, 5723-5726 (2016).

17. Comas-Vives, A., Valla, M., Copéret, C. & Sautet, P. Cooperativity between Al Sites Promotes Hydrogen
Transfer and Carbon − Carbon Bond Formation upon Dimethyl Ether Activation on Alumina. ACS Cent. Sci.
1, 313-319 (2015).

18. Wang, W. & Hunger, M. Reactivity of surface alkoxy species on acidic zeolite catalysts. Acc. Chem. Res. 41,
895-904 (2008).An in-depth review providing versatile reactivity aspects of surface-methoxy species
during zeolite catalyzed hydrocarbon conversion.

24
19. Dahl, I.M. & Kolboe, S. On the reaction mechanism for propene formation in the MTO reaction over SAPO-
34. Catalysis Lett. 20, 329-336 (1993).This article illustrates the concept of hydrocarbon pool species
during zeolite catalyzed methanol-to-hydrocarbon process.

20. Song, W., Marcus, D.M., Fu, H., Ehresmann, J.O. & Haw, J.F. An oft-studied reaction that may never have
been: Direct catalytic conversion of methanol or dimethyl ether to hydrocarbons on the solid acids HZSM-5
or HSAPO-34. J. Am. Chem. Soc. 124, 3844-3845 (2002).

21. Lesthaeghe, D., Van Speybroeck, V., Marin, G.B. & Waroquier, M. What role do oxonium ions and oxonium
ylides play in the ZSM-5 catalysed methanol-to-olefin process? Chem. Phys. Lett. 417, 309-315 (2006).

22. Lesthaeghe, D., Van Speybroeck, V., Marin, G.B. & Waroquier, M. The Rise and Fall of Direct Mechanisms
in Methanol-to-Olefin Catalysis: An Overview of Theoretical Contributions. Ind. Eng. Chem. Res. 46, 8832-
8838 (2007).

23. Jiang, Y., et al. Effect of organic impurities on the hydrocarbon formation via the decomposition of surface
methoxy groups on acidic zeolite catalysts. J. Catal. 238, 21-27 (2006).

24. Dai, W., et al. Understanding the Early Stages of the Methanol-to-Olefin Conversion on H ‑ SAPO-34. ACS
Catal. 5, 317−326 (2014).

25. Dai, W., Wu, G., Li, L., Guan, N. & Hunger, M. Mechanisms of the deactivation of SAPO-34 materials with
different crystal sizes applied as MTO catalysts. ACS Catal. 3, 588-596 (2013).

26. Jiang, Y., Hunger, M. & Wang, W. On the Reactivity of Surface Methoxy Species in Acidic Zeolites. J. Am.
Chem. Soc. 128, 11679-11692 (2006).

27. Wang, W., Buchholz, A., Seiler, M. & Hunger, M. Evidence for an Initiation of the Methanol-to-Olefin Process
by Reactive Surface Methoxy Groups on Acidic Zeolite Catalysts. J. Am. Chem. Soc. 125, 15260-15267
(2003).

28. Li, J., et al. A route to form initial hydrocarbon pool species in methanol conversion to olefins over zeolites.
J. Catal. 317, 277-283 (2014).

29. Wei, Z., et al. Methane formation mechanism in the initial methanol-to-olefins process catalyzed by SAPO-
34. Catal. Sci. Tech. 6, 5526-5533 (2016).

30. Plessow, P.N. & Studt, F. Unraveling the Mechanism of the Initiation Reaction of the Methanol to Olefins
Process Using ab Initio and DFT Calculations. ACS Catal. 7, 7987-7994 (2017).

31. Kazansky, V. & Senchenya, I.N. Quantum chemical study of the electronic structure and geometry of surface
alkoxy groups as probable active intermediates of heterogeneous acidic catalysts: What are the adsorbed
carbenium ions? J. Catal. 119, 108-120 (1989).

32. Salehirad, F. & Anderson, M.W. Solid-State 13 C MAS NMR Study of Methanol-to-Hydrocarbon Chemistry
over H-SAPO-34. J. Catal. 314, 301-314 (1996).

33. Hutchings, G.J., Gottschalk, F., Hall, M.V.M.l. & Hunter, R. Hydrocarbon formation from methylating agents
over the zeolite catalyst ZSM-5. Comments on the mechanism of carbon?carbon bond and methane
formation. J. Chem. Soc. Faraday Trans. 83, 571-583 (1987).

34. Plessow, P.N. & Studt, F. Theoretical Insights into the Effect of the Framework on the Initiation Mechanism
of the MTO Process. Catal. Lett. 148, 1246-1253 (2018).

35. Dessau, R.M. & Lapierre, R.B. On the mechanism of methanol conversion to hydrocarbons over HZSM-5. J.
Catal. 78, 136-141 (1982).

36. Svelle, S., et al. Conversion of methanol into hydrocarbons over zeolite H-ZSM-5: Ethene formation is
mechanistically separated from the formation of higher alkenes. J. Am. Chem. Soc. 128, 14770-14771
(2006).

25
37. Bjørgen, M., et al. Conversion of methanol to hydrocarbons over zeolite H-ZSM-5: On the origin of the olefinic
species. J. Catal. 249, 195-207 (2007).

38. Sun, X., et al. On reaction pathways in the conversion of methanol to hydrocarbons on HZSM-5. J. Catal.
317, 185-197 (2014).

39. Wang, S., et al. Polymethylbenzene or Alkene Cycle? Theoretical Study on Their Contribution to the Process
of Methanol to Olefins over H-ZSM-5 Zeolite. J. Phys. Chem. C 119, 28482-28498 (2015).

40. Sun, X., et al. On the impact of co-feeding aromatics and olefins for the methanol-to-olefins reaction on
HZSM-5. J. Catal. 314, 21-31 (2014). Seminal kinetic investigations demonstrating the autocatalytic
nature of the mechanism and discussing the effect of the feed composition on the dominant reaction
pathways.

41. Van Speybroeck, V., et al. Mechanistic Studies on Chabazite-Type Methanol-to-Olefin Catalysts: Insights
from Time-Resolved UV/Vis Microspectroscopy Combined with Theoretical Simulations. ChemCatChem 5,
173-184 (2013).

42. Borodina, E., et al. Influence of the reaction temperature on the nature of the active and deactivating species
during methanol to olefins conversion over H-SSZ-13. ACS Catal. 5, 992-1003 (2015).

43. Dai, W., et al. Intermediates and Dominating Reaction Mechanism During the Early Period of the Methanol-
to-Olefin Conversion on SAPO-41. J. Phys. Chem. C 119, 2637−2645 (2015).

44. Hemelsoet, K., et al. Identification of intermediates in zeolite-catalyzed reactions by in situ UV/Vis
microspectroscopy and a complementary set of molecular simulations. Chem. Eur. J. 19, 16595-16606
(2013).

45. Qian, Q., et al. Single-particle spectroscopy of alcohol-to-olefins over SAPO-34 at different reaction stages:
Crystal accessibility and hydrocarbons reactivity. ChemCatChem 6, 772-783 (2014).

46. Wulfers, M.J. & Jentoft, F.C. The role of cyclopentadienium ions in methanol-to-hydrocarbons chemistry.
ACS Catal. 4, 3521-3532 (2014).

47. Borodina, E., et al. Influence of the Reaction Temperature on the Nature of the Active and Deactivating
Species During Methanol-to-Olefins Conversion over H-SAPO-34. ACS Catal. 7, 5268-5281 (2017).

48. Goetze, J. & Weckhuysen, B.M. Spatiotemporal coke formation over zeolite ZSM-5 during the methanol-to-
olefins process as studied with operando UV-vis spectroscopy: a comparison between H-ZSM-5 and Mg-
ZSM-5. Catal. Sci. Techol. 8, 1632-1644 (2018).

49. Haw, J.F., Song, W., Marcus, D.M. & Nicholas, J.B. The mechanism of methanol to hydrocarbon catalysis.
Acc. Chem. Res. 36, 317-326 (2003).

50. Xu, T., et al. Synthesis of a benzenium ion in a zeolite with use of a catalytic flow reactor. J. Am. Chem. Soc.
120, 4025-4026 (1998).

51. Haw, J.F., et al. Roles for cyclopentenyl cations in the synthesis of hydrocarbons from methanol on zeolite
catalyst HZSM-5. J. Am. Chem. Soc. 122, 4763-4775 (2000).

52. Li, J., et al. Observation of heptamethylbenzenium cation over SAPO-type molecular sieve DNL-6 under real
MTO conversion conditions. J. Am. Chem. Soc. 134, 836-839 (2012).

53. Xu, S., et al. Direct observation of cyclic carbenium ions and their role in the catalytic cycle of the methanol-
to-olefin reaction over chabazite zeolites. Angew. Chem. Int. Ed. 52, 11564-11568 (2013).

54. Song, W., Nicholas, J.B., Sassi, A. & Haw, J.F. Synthesis of the heptamethylbenzenium cation in zeolite: In
situ NMR and theory. Catal. Lett. 81, 49-53 (2002).

26
55. Bollini, P. & Bhan, A. Improving HSAPO-34 Methanol-to-Olefin Turnover Capacity by Seeding the
Hydrocarbon Pool. ChemPhysChem 19, 479-483 (2018).

56. Martinez-Espin, J.S., et al. New insights into catalyst deactivation and product distribution of zeolites in the
methanol-to-hydrocarbons (MTH) reaction with methanol and dimethyl ether feeds. Catal. Sci. Techol. 7,
2700-2716 (2017).

57. Muller, S., et al. Coke formation and deactivation pathways on H-ZSM-5 in the conversion of methanol to
olefins. J. Catal. 325, 48-59 (2015).

58. Hwang, A. & Bhan, A. Bifunctional Strategy Coupling Y2O3-Catalyzed Alkanal Decomposition with Methanol-
to-Olefins Catalysis for Enhanced Lifetime. ACS Catal. 7, 4417-4422 (2017).

59. Yarulina, I., Kapteijn, F. & Gascon, J. The importance of heat effects in the methanol to hydrocarbons reaction
over ZSM-5: on the role of mesoporosity on catalyst performance. Catal. Sci. Techol. 6, 5320-5325 (2016).

60. Mole, T., Whiteside, J.A. & Seddon, D. Aromatic co-catalysis of methanol conversion over zeolite catalysts.
J. Catal. 82, 261-266 (1983).

61. Wu, W.Z., Guo, W.Y., Xiao, W.D. & Luo, M. Dominant reaction pathway for methanol conversion to propene
over high silicon H-ZSM-5. Chem. Eng. Sci. 66, 4722-4732 (2011).

62. Ilias, S., Khare, R., Malek, A. & Bhan, A. A descriptor for the relative propagation of the aromatic- and olefin-
based cycles in methanol-to-hydrocarbons conversion on H-ZSM-5. J. Catal. 303, 135-140 (2013).

63. Ilias, S. & Bhan, A. Tuning the selectivity of methanol-to-hydrocarbons conversion on H-ZSM-5 by co-
processing olefin or aromatic compounds. J. Catal. 290, 186-192 (2012).

64. Khare, R. & Bhan, A. Mechanistic studies of methanol-to-hydrocarbons conversion on diffusion-free MFI
samples. J. Catal. 329, 218-228 (2015).

65. Ito, H., et al. Method for production of lower olefin. EP1955989 A1; 2008.

66. Chikamatsu, N., Funatsu, S., Ito, H., Oyama, K. & Yoshida, J. Propylene production process and propylene
production apparatus. EP2058290 A1; 2009.

67. Smit, B. & Maesen, T.L.M. Towards a molecular understanding of shape selectivity. Nature 451, 671-678
(2008).

68. Hereijgers, B.P.C., et al. Product shape selectivity dominates the Methanol-to-Olefins (MTO) reaction over
H-SAPO-34 catalysts. J. Catal. 264, 77-87 (2009).

69. Chen, D., Moljord, K., Fuglerud, T. & Holmen, A. The effect of crystal size of SAPO-34 on the selectivity and
deactivation of the MTO reaction. Microporous Mesoporous Mater. 29, 191-203 (1999).

70. Moliner, M., Martínez, C. & Corma, A. Synthesis Strategies for Preparing Useful Small Pore Zeolites and
Zeotypes for Gas Separations and Catalysis. Chem. Mater. 26, 246-258 (2014).

71. Zhong, J., et al. Increasing the selectivity to ethylene in the MTO reaction by enhancing diffusion limitation
in the shell layer of SAPO-34 catalyst. Chem. Comm., (2018).

72. Li, J., et al. Cavity Controls the Selectivity: Insights of Confinement Effects on MTO Reaction. ACS Catal. 5,
661-665 (2015).

73. Yarulina, I., et al. Methanol-to-olefins process over zeolite catalysts with DDR topology: effect of composition
and structural defects on catalytic performance. Catal. Sci. Techol. 6, 2663-2678 (2016).

74. Chen, D., Moljord, K. & Holmen, A. A methanol to olefins review: Diffusion, coke formation and deactivation
SAPO type catalysts. Microporous Mesoporous Mater. 164, 239-250 (2012).

27
75. Haw, J.F., Song, W.G., Marcus, D.M. & Nicholas, J.B. The mechanism of methanol to hydrocarbon catalysis.
Acc. Chem. Res. 36, 317-326 (2003).

76. Chen, J., et al. Spatial confinement effects of cage-type SAPO molecular sieves on product distribution and
coke formation in methanol-to-olefin reaction. Catal. Commun. 46, 36-40 (2014).

77. Bhawe, Y., et al. Effect of Cage Size on the Selective Conversion of Methanol to Light Olefins. ACS Catal.
2, 2490-2495 (2012).

78. Dusselier, M., Deimund, M.A., Schmidt, J.E. & Davis, M.E. Methanol-to-Olefins Catalysis with Hydrothermally
Treated Zeolite SSZ-39. ACS Catal. 5, 6078-6085 (2015).

79. Pinilla-Herrero, I., Olsbye, U., Márquez-Álvarez, C. & Sastre, E. Effect of framework topology of SAPO
catalysts on selectivity and deactivation profile in the methanol-to-olefins reaction. J. Catal. 352, 191-207
(2017).

80. Kang, J.H., et al. Further Studies on How the Nature of Zeolite Cavities That Are Bounded by Small Pores
Influences the Conversion of Methanol to Light Olefins. ChemPhysChem 19, 412-419 (2018).

81. Liang, T., et al. Conversion of methanol to olefins over H-ZSM-5 zeolite: reaction pathway is related to the
framework aluminum siting. ACS Catal. 6, 7311-7325 (2016).A comprehensive article showing the effect
of aluminium siting and location on selectivtiy and lifetime.

82. Bleken, F., et al. Conversion of methanol over 10-ring zeolites with differing volumes at channel intersections:
comparison of TNU-9, IM-5, ZSM-11 and ZSM-5. PCCP 13, 2539-2549 (2011).

83. Cui, Z.-M., Liu, Q., Song, W.-G. & Wan, L.-J. Insights into the mechanism of methanol-to-olefin conversion
at zeolites with systematically selected framework structures. Angew. Chem. Int. Ed. 45, 6512-6515 (2006).

84. Teketel, S., Svelle, S., Lillerud, K.-P. & Olsbye, U. Shape-selective conversion of methanol to hydrocarbons
over 10-ring unidirectional-channel acidic H-ZSM-22. ChemCatChem 1, 78-81 (2009).

85. Teketel, S., et al. Shape Selectivity in the Conversion of Methanol to Hydrocarbons: The Catalytic
Performance of One-Dimensional 10-Ring Zeolites: ZSM-22, ZSM-23, ZSM-48, and EU-1. ACS Catal. 2, 26-
37 (2012).This article discusses how slight changes in pore dimensions profoundly affect selectivity
to hydrocarbons establishing topology as a tool to control selectivity.

86. Jamil, A.K., et al. Selective Production of Propylene from Methanol Conversion over Nanosized ZSM-22
Zeolites. Ind. Eng. Chem. Res. 53, 19498-19505 (2014).

87. Molino, A., et al. Conversion of methanol to hydrocarbons over zeolite ZSM-23 (MTT): exceptional effects of
particle size on catalyst lifetime. Chem. Commun. 53, 6816-6819 (2017).

88. Ma, H., et al. Reaction mechanism for the conversion of methanol to olefins over H-ITQ-13 zeolite: a density
functional theory study. Catal. Sci. Techol. 8, 521-533 (2018).

89. Westgård Erichsen, M., Svelle, S. & Olsbye, U. The influence of catalyst acid strength on the methanol to
hydrocarbons (MTH) reaction. Catal. Today 215, 216-223 (2013).

90. Mikkelsen, Ø. & Kolboe, S. The conversion of methanol to hydrocarbons over zeolite H-beta. Microporous
Mesoporous Mater. 29, 173-184 (1999).

91. Abubakar, S.M., et al. Structural and Mechanistic Investigation of a Phosphate-Modified HZSM-5 Catalyst
for Methanol Conversion. Langmuir 22, 4846-4852 (2006).

92. Liu, J., et al. Methanol to propylene: Effect of phosphorus on a high silica HZSM-5 catalyst. Catal. Commun.
10, 1506-1509 (2009).

93. Hu, S., et al. Selective formation of propylene from methanol over high-silica nanosheets of MFI zeolite. Appl.
Catal. A 445, 215-220 (2012).

28
94. Mei, C., et al. Selective production of propylene from methanol: Mesoporosity development in high silica
HZSM-5. J. Catal. 258, 243-249 (2008).

95. Wen, M., et al. Monolithic metal-fiber@HZSM-5 core–shell catalysts for methanol-to-propylene. Microporous
Mesoporous Mater. 206, 8-16 (2015).

96. Wei, R., Li, C., Yang, C. & Shan, H. Effects of ammonium exchange and Si/Al ratio on the conversion of
methanol to propylene over a novel and large partical size ZSM-5. J. Nat. Gas Chem. 20, 261-265 (2011).

97. Khare, R., Liu, Z., Han, Y. & Bhan, A. A mechanistic basis for the effect of aluminum content on ethene
selectivity in methanol-to-hydrocarbons conversion on HZSM-5. J. Catal. 348, 300-305 (2017).

98. Yarulina, I., et al. Suppression of the Aromatic Cycle in Methanol-to-Olefins Reaction over ZSM-5 by Post-
Synthetic Modification Using Calcium. ChemCatChem 8, 3057-3063 (2016).

99. van der Bij, H.E. & Weckhuysen, B.M. Phosphorus promotion and poisoning in zeolite-based materials:
synthesis, characterisation and catalysis. Chem. Soc. Rev. 44, 7406-7428 (2015).

100. Danilina, N., Krumeich, F., Castelanelli, S.A. & van Bokhoven, J.A. Where Are the Active Sites in Zeolites?
Origin of Aluminum Zoning in ZSM-5. J. Phys. Chem. C 114, 6640-6645 (2010).

101. von Ballmoos, R. & Meier, W.M. Zoned aluminium distribution in synthetic zeolite ZSM-5. Nature 289, 782
(1981).

102. Althoff, R., Schulzdobrick, B., Schüth, F. & Unger, K. Controlling the spatial distribution of aluminum in ZSM-
5 crystals. Microporous Mater. 1, 207-218 (1993).

103. Roeffaers, M.B.J., et al. Space- and Time-Resolved Visualization of Acid Catalysis in ZSM-5 Crystals by
Fluorescence Microscopy. Angew. Chem. Int. Ed. 46, 1706-1709 (2007).

104. Kox, M.H.F., Stavitski, E. & Weckhuysen, B.M. Nonuniform Catalytic Behavior of Zeolite Crystals as
Revealed by In Situ Optical Microspectroscopy. Angew. Chem. Int. Ed. 46, 3652-3655 (2007).

105. Tzoulaki, D., Heinke, L., Schmidt, W., Wilczok, U. & Kärger, J. Exploring Crystal Morphology of Nanoporous
Hosts from Time-Dependent Guest Profiles. Angew. Chem. Int. Ed. 47, 3954-3957 (2008).

106. Mores, D., et al. Space- and Time-Resolved In-situ Spectroscopy on the Coke Formation in Molecular
Sieves: Methanol-to-Olefin Conversion over H-ZSM-5 and H-SAPO-34. Chem. Eur. J. 14, 11320-11327
(2008).

107. Mores, D., Kornatowski , J., Olsbye, U. & Weckhuysen, B.M. Coke Formation during the Methanol-to-Olefin
Conversion: In Situ Microspectroscopy on Individual H-ZSM-5 Crystals with Different Brønsted Acidity.
Chem. Eur. J. 17, 2874-2884 (2011).

108. Perea, D.E., et al. Determining the location and nearest neighbours of aluminium in zeolites with atom probe
tomography. Nat. Comm. 6, 7589 (2015).

109. Schmidt, J.E., et al. Coke Formation in a Zeolite Crystal During the Methanol-to-Hydrocarbons Reaction as
Studied with Atom Probe Tomography. Angew. Chem. Int. Ed. 55, 11173-11177 (2016).In this article, atom
probe tomography was used to spatially resolve the 3D compositional changes at the sub-nm length
scale in a partially deactivated single zeolite ZSM-5 crystal after the methanol-to-hydrocarbon
reaction.

110. Bleken, F., et al. The Effect of Acid Strength on the Conversion of Methanol to Olefins Over Acidic
Microporous Catalysts with the CHA Topology. Top. Catal. 52, 218-228 (2009).

111. Westgård Erichsen, M., et al. How zeolitic acid strength and composition alter the reactivity of alkenes and
aromatics towards methanol. J. Catal. 328, 186-196 (2015).

29
112. Goetze, J., et al. Insights into the activity and deactivation of the methanol-to-olefins process over different
small-pore zeolites as studied with operando UV–vis spectroscopy. ACS Catal. 7, 4033-4046 (2017).

113. Goetze, J., Yarulina, I., Gascon, J., Kapteijn, F. & Weckhuysen, B.M. Revealing Lattice Expansion of Small-
Pore Zeolite Catalysts during the Methanol-to-Olefins Process Using Combined Operando X-ray Diffraction
and UV–vis Spectroscopy. ACS Catal. 8, 2060-2070 (2018).

114. Chowdhury, A.D., et al. Electrophilic aromatic substitution over zeolites generates Wheland-type reaction
intermediates. Nat. Catal. 1, 23-31 (2018).

115. Vogt, C., Weckhuysen, B.M. & Ruiz-Martínez, J. Effect of Feedstock and Catalyst Impurities on the Methanol-
to-Olefin Reaction over H-SAPO-34. ChemCatChem 9, 183-194 (2017).

116. Gao, P., et al. Direct conversion of CO2 into liquid fuels with high selectivity over a bifunctional catalyst. Nat.
Chem. 9, 1019 (2017).

117. Jiao, F., et al. Selective conversion of syngas to light olefins. Science 351, 1065-1068 (2016).

30