7/31/2019 9781849732772-00253

1/31

Hierarchical zeolites: materials withimproved accessibility and enhancedcatalytic activity

D. P. Serrano,1, 2 J. Aguado3 and J. M. Escola3

DOI: 10.1039/9781849732772-00253

1 Introduction

Conventional zeolites are crystalline aluminosilicates formed by a network of

SiO4 and AlO4 tetrahedra linked by shared oxygen atoms, which are usually

obtained in the form of crystals/particles of micrometer size (W1mm).1 The

presence of aluminium atoms into their framework confers them acidic

properties (both Bro nsted and Lewis) although they can also incorporate

other heteroatoms such as Ti, V, Sn, etc., allowing their application ascatalyst in many different types of chemical reactions.2 However, their major

feature is likely the occurrence of micropores of molecular dimensions

(generally 0.40. 75 nm), which turn them into molecular sieves. This unique

property has been also called shape selectivity and allows the zeolites to

discriminate among different reactants, products or even transition states

according to their shape and size. This phenomenon has led to remarkable

selectivities exhibited by zeolites in a large number of reactions. This is the

case of toluene disproportionation, wherein enhanced yields and selectivities

towards the p-xylene isomer have been attained over the ZSM-5 zeolite dueto its improved diffusion through the channel systems with regard to the

m/o- isomers.3 In fact, many of the current successful applications of zeolites

are based on their molecular sieves character.

However, zeolites fail when dealing with bulky substrates, usually

exhibiting low activities, often even below the values obtained with

amorphous materials such as silica-alumina. This result is to be expected as

the bulky substrates cannot access the active sites located inside the zeolite

micropores due to small size of the latter. Thus, only those sites situated

over the external surface of the catalyst particles and crystals, or close to the

micropore openings, are accessible for large molecules, and these represent

usually a low share (o 5%) of the total content of active sites. A clear-cut

example of this circumstance can be found in the catalytic cracking of

polypropylene over ZSM-5 zeolite,4 wherein just 11% of conversion

was observed at 3751C, despite the high acid strength of this zeolite, while a

99% conversion was obtained over the mild acid strength Al-MCM-41

mesoporous material. On the other hand, even if the substrate can enter into

the zeolite micropores, the diffusion rate is usually too slow bringing about

the appearance of intracrystalline mass transfer constraints, which limit

meaningfully the performance of the zeolite catalyst.5

1Department of Chemical and Energy Technology, ESCET, Universidad Rey Juan Carlos,c/ Tulipan s/n, 28933, Mostoles, Madrid, Spain2IMDEA Energy Institute, c/Tulipan s/n, 28933, Mostoles, Madrid, Spain3Department of Chemical and Environmental Technology, ESCET, Universidad Rey Juan

Carlos, c/ Tulipan s/n, 28933, Mostoles, Madrid, Spain

Catalysis, 2011, 23, 253283 | 253

c The Royal Society of Chemistry 2011

7/31/2019 9781849732772-00253

2/31

Seemingly, the increasing need of processing bulky compounds put the

zeolites into a difficult situation, specially when ordered mesoporous

materials such as MCM-416 and SBA-157 were discovered in the early and

late 90s, respectively. These materials possess pore sizes whose dimension

may be tailored within the 1.5 30.0 nm range by a suitable choice of

synthesis conditions (mostly type of surfactant, temperature and aging

time). However, the huge number of publications devoted to these materialssoon realized their main limitations, mainly derived from the amorphous

nature of their walls. Accordingly, ordered mesoporous aluminosilicates do

not possess the high hydrothermal stability and strong acidity of zeolites.

Therefore, zeolites are still the preferred choice in numerous industrial

applications.

An alternative solution to deal with bulky substrates has been the syn-

thesis and application of nanozeolites (crystal size below 100 nm) since they

show a high share of external surface area whose active sites are accessible

to large molecules.

8

However, nanozeolites are usually produced in lowyields since they are very difficult to separate from the synthesis medium.9,10

Recently, several new strategies have appeared in order to improve the

properties of zeolites when processing bulky substrates. Delaminated

zeolites,11 large pore zeolites12,13 and hierarchical zeolites14,15 are examples

of these new strategies. Delaminated zeolites, such as ITQ-2, result from the

delamination of a layered zeolite precursor (e.g. MCM-22) giving rise to

thin zeolite sheets (B2.5 nm thick) having a huge external surface area

(Z700m2 g1).11 However, this method of synthesis is inherently bound to

the occurrence of a zeolite layered precursor. The preparation of large pore

zeolites, having channels formed by rings with more than 12 members, is

another strategy that has been pursued in the late years. One of the main

achievements in this line has been the discovery of ITQ-33, a silicogerma-

nate having circular pores of 18 member-rings interconnected by 10

member-ring channels and a crystallographic pore diameter of 1.22 nm.13

This means a substantial enhancement in pore size considering that

a conventional zeolite Y shows roughly 0.75 nm of pore dimension.

Nevertheless, the main disadvantage of this approach is the need of

employing Ge in the synthesis, as well as the addition of special structure

directing agents (SDA) which are usually expensive reagents.One of the most successful strategies for improving accessibility is likely

the case of hierarchical zeolites.14,15 These zeolites are characterized by the

presence of a bimodal pore size distribution, formed by both micropores

and mesopores. The microporous structure is the one inherent to the

classical zeolite topology while the secondary mesoporous structure can be

generated by a variety of specific synthetic procedures. The presence of a

secondary porosity in hierarchical zeolites, usually in the mesopore range, is

responsible for the improved mass transfer properties of these materials,

which have proved to be advantageous in numerous reactions. On the otherhand, the surface area associated to this secondary porosity implies the

presence of active sites that are not sterically hindered for interacting with

bulky molecules. In some cases the mesopore surface area is confused with

the external surface area, likely due to the application of the t-plot method

for the joint calculation of both parameters. Although the nature and

254 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

3/31

performance of the active sites located in both mesopore and external

surface area are probably quite similar, they are different concepts. External

surface refers to the surface area located in the outer part of the zeolite

particles, whereas mesopore surface corresponds with the surface area in the

mesopore walls. Therefore, the occurrence of high external surface area

does not mean the presence of mesopores in the material. Thus, nanozeolites

show high external surface area but many times mesopores are absent. Inthis regard, hierarchical zeolites cannot be regarded truly as nanozeolites

(although they usually consist of zeolitic nanounits as building blocks).

From a catalytic viewpoint, most part of the active sites in hierarchical

zeolites are placed inside both micropores and mesopores, the latter having

improved accessibility. Henceforth, the main preparation strategies of

hierarchical zeolites as well as their applications in different reactions are

commented.

2 Methods of preparation of hierarchical zeolitesHierarchical zeolites can be prepared by different procedures which show

distinct features. Although all of them lead to materials with bimodal pore

size distributions, the features and contribution of the generated secondary

mesoporosity depends heavily on the chosen procedure. Some of these

synthesis strategies present common aspects which allow them to be clas-

sified as follows:

Dealumination

Desilication

Hard templating by carbon materials

Hard templating by polymers

Incorporation of organosilanes

Other methods

2.1 Dealumination

Dealumination methods represent the most classical alternative for de-

veloping mesoporosity in zeolites. They comprise steaming at elevated

temperatures and acid leaching.16 Steaming at high temperature of zeolite Y

has been traditionally performed for its application as catalyst in fluidcatalytic cracking reactors (FCC) since it creates mesopores by removal of a

certain amount of the aluminium from the zeolite framework. This treat-

ment causes an enhancement of the hydrothermal stability of the zeolite as

well as an improvement of the diffusion of bulky molecules inside the zeolite

pores. The steaming treatment consists of first performing an ion exchange

of an alkali metal containing zeolite Y with an ammonium salt solution in

order to reduce the alkali content to 1025% of its starting value, followed

by washing to remove the excess salt. Subsequently, the zeolite is heated to

2006001C to enable the migration of sodium ions towards easilyexchangeable sites. Then, a second ion exchange is performed with an

ammonium salt solution to remove completely the sodium. Finally the

zeolite is heated under a steam atmosphere at a temperature between

6008001C.17 The non framework aluminium resulting from the steaming

treatment can be removed from the zeolite by subsequent washing with

Catalysis, 2011, 23, 253283 | 255

7/31/2019 9781849732772-00253

4/31

dilute acids. In addition, as a result of the steaming treatment, some

shrinkage of the zeolite framework of around 0.020.03 nm takes place.

Acid leaching also allows mesopores to be created by a quite similar

mechanism, i.e. through the extraction of some aluminium from the

framework positions. A typical procedure is the treatment of the zeolite in

an slurry with a solution of ethylenediaminetetraacetic acid (EDTA) under

reflux for 18 h.18

Afterwards, the zeolite is heated with an inert gas at 8001C.However, this method has the drawback of removing preferentially the

aluminium present over the external surface leading to non uniform

distributions within the crystals.19 A related method which employs a strong

acid treatment has been reported recently by Van Oers et al.20 They attained

hierarchical Beta from nanoparticles of this zeolite using a procedure

consisting in the ageing at 1401C of a zeolite Beta nanoparticle solution,

followed by stepwise cooling or quenching. Then, this mixture is strongly

acidified with concentrated HCl and subjected to a second hydrothermal

treatment at 1501

C for 72 h. The obtained mesopore size depended on theconditions employed in the cooling step, giving rise to roughly 10 nm pores

with slow cooling and 6.0 nm pores with quenching.

Nevertheless, recent experimental evidences have been found that ques-

tion markedly the goodness of the mesoporous structure created in zeolite Y

by steaming. 3-D TEM technique has allowed the nature of the obtained

mesopores to be studied, indicating that their size encompass a broad range

(2 50 nm).21 But the most important finding has been that the mesopores

consist chiefly in cavities connected by micropores instead of a network of

mesopore channels connecting the outer surface and the internal micro-

pores. This is nicely illustrated in Fig. 1, which shows the 2-D and 3-D TEM

micrographs of a USY zeolite obtained by steaming of a NH4Y sample.

This finding has important consequences since this mesopore structure is

not expected to cause significant enhanced diffusion rates inside the zeolite

crystals. Likewise, other works have been devoted specifically to elucidate

this fact by measuring intracrystalline diffusion rates in steamed USY

Fig. 1 TEM micrographs of USY zeolite: a) 2D-TEM image of a crystal, b) 3D TEMreconstruction of a crystal. (Reprinted with permission from ref. 21, Copyright Wiley-VCHVerlag GmbH & Co KGaA, 2001)

256 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

5/31

zeolites using the PFG NMR technique.22 Two probe molecules were used:

n-octane and 1,3,5 triisopropylbenzene, with molecular diameters smaller

and larger than the zeolite micropores, respectively. The main conclusion of

this work was that the intracrystalline diffusion inside zeolite USY is

practically not affected by the presence of the mesopores generated by

steaming, due to the absence of an interconnected network structure. This

result indicates that the classical procedure of steaming is not reallyappropriate for the preparation of a network of mesopores, highlighting the

importance of developing new methods of synthesis of hierarchical zeolites.

2.2 Desilication

The technique of desilication is based on the treatment of zeolites with a

base (usually sodium hydroxide) under relatively mild conditions

(To363 K). Mesopores are formed due to the preferential removal of silica

from the framework. This is a remarkable difference from the acid leachingtreatment wherein aluminium atoms are selectively extracted. Ogura et al.23

carried out the desilication of ZSM-5 zeolite with 0.2 M NaOH aqueous

solution at 353 K. After the treatment, the ZSM-5 sample retains its crys-

tallinity and the majority of their microporous structure, whereas it contains

additional mesopores. The morphology of the zeolite is meaningfully

altered by the alkaline treatment leading to the appearance of voids and

grooves over the zeolite surface. This is clearly appreciated in Fig. 2 which

shows the SEM micrograph of both the raw sample and the alkaline treated

ZSM-5 zeolite. These authors also observed that the acidity of the zeolite

was little affected after the alkaline treatment according to ammonia TPD

measurements.24 Moreover, it was concluded that the size of the pores

formed by the treatment with NaOH is around 1.8 nm, being indeed

supermicropores.25

A key parameter of desilication by alkaline treatment is the Si/Al atomic

ratio of the zeolite.26 Thus, a Si/Al atomic ratio in the range 20 50 has been

found as optimal in the case of ZSM-5 zeolite giving rise to intracrystalline

mesopores with a size around 10 nm. In contrast, for lower Si/Al atomic

ratios, the high aluminium concentration in the framework prevents the

removal of silicon leading to almost no mesopore formation. On the con-trary, if the Si/Al is very high (Si/Al W 200), too much silicon is extracted

creating large pores. This can be appreciated in Fig. 3, illustrating the

evolution of the porous structure of the material with the alkali treatment

for different Si/Al atomic ratios.

On the other hand, the presence of extraframework aluminium proved to

inhibit silicon extraction due to realumination during the alkaline treat-

ment. However, this problem could be solved by previous acid washing in

order to eliminate the extraframework aluminium species.27 An interesting

fact that highlights the importance in this method of the Si/Al atomic ratiois the existence of Al concentration gradients in large ZSM-5 crystals. Thus,

the external zeolite surface is often rich in Al species, while the concen-

tration of aluminium in the inner part of the crystals is much lower. In this

case, the application of the desilication procedure inevitably brings about

the generation of hollow zeolite architectures.28 Therefore, special synthetic

Catalysis, 2011, 23, 253283 | 257

7/31/2019 9781849732772-00253

6/31

methods which lead to more uniform Al concentration into the zeolites are

required if the desilication procedure is to be applied.

The alkali treatment can be applied using also organic bases (TPAOH and

TBAOH) instead of NaOH, which causes a higher extent of the aluminumleaching.29 In this way, the mesopore volume can be optimized by incorpor-

ating in the alkaline treatment both NaOH and quaternary ammonium cat-

ions,30 such as TPA or TBA . These cations act as pore growth moderators,

being necessary for the success of the method that they do not enter the zeolite

micropores (TMA does not work). The addition of these cations together

with the alkaline solution allows the mesopore surface area to be increased

without reducing the micropore volume (the ratio between these two

parameters has been called hierarchy factor). In addition, the mesopore size

decreased from 10.0 nm to 4.5 nm on augmenting the TPA

concentration.Based on FTIR measurements using CO as probe molecule, it has been

concluded that the acid strength of the Bro nsted sites of the zeolitic

materials did not change significantly after desilication.31 However, the

appearance of new strong Lewis acid sites have been observed assigned to

dislodged Al species. The enhanced accessibility of the Bro nsted acid sites

Fig. 2 SEM micrographs of the original (a) and alkaline treated ZSM-5 (b). (Reprinted withpermission from ref. 23, Copyright Japan Chemical Society)

258 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

7/31

in the desilicated zeolites has been proved by FTIR using substituted

alkylpyridines such as 2-6-lutidine (size of 0.67 nm) and 2,4,6-collidine (size

of 0.74 nm).32 Collidine is too bulky to enter the MFI zeolite micropores

while lutidine can probe certain amount of the zeolite sites (below 50%).

This fact allowed an accessibility index (ACI) to be established as the ratio

of the Bro nsted acid sites detected with the probe molecule with regard to

the total number of Bro nsted sites, measured using a fully accessible probe

molecule such as pyridine (size of 0.57 nm). This accessibility index showed

a remarkable increase with the mesopore surface area reaching maximum

values of 0.4 and 1 with collidine and lutidine, respectively, for a mesopore

surface area of 277 m2 g 1.

The desilication procedure has been extended to other zeolites differentfrom ZSM-5, such as mordenite33 and ZSM-12.34 Its main advantage is that

the procedure to be applied is quite simple and does not involve the use of

expensive reagents. However, one inherent limitation of the method is the

need of having a Si/Al atomic ratio within a given range. Thereby, its

applicability is lower than other reported procedures that do not show this

bound. In addition, desilication involves a partial destruction of the zeolite

structure what should be carefully controlled in order to avoid the

appearance of amorphous material or the generation of extraframework

aluminium species.

2.3 Hard templating by carbon materials

The outburst in the synthesis of mesoporous/hierarchical zeolites has been

largely marked by the work of Jacobsen and col.35 It was preceded by the

development of the confined synthesis method for the preparation of

Fig. 3 Evolution of the porous structure of ZSM-5 zeolite during desilication by alkalitreatment as a function of the Si/Al ratio. (Reprinted with permission from ref. 27, CopyrightWiley-VCH Verlag GmbH & Co KGaA, 2005)

Catalysis, 2011, 23, 253283 | 259

7/31/2019 9781849732772-00253

8/31

nanozeolites, performed also by this research group.36 The latter consists of

the synthesis of nanozeolites inside a mesoporous inert carbon matrix.

Subsequently, the carbon is burnt off releasing the nanozeolites with a

crystal size of around 20 nm that could be tailored by controlling the pore

size of the used inert matrix. The preparation of mesoporous zeolites could

be carried out by modifying slightly this method. Thus, if an excess of

zeolite gel is used in the synthesis, the zeolite grows over and around theparticles of the carbon matrix forming large zeolite crystals which en-

capsulates the inert carbon matrix. Once the zeolite crystal is formed, the

carbon matrix is eliminated by combustion in air giving rise to large zeolite

crystals containing mesopores roughly of the size of the carbon matrix

particles (see Fig. 4).

Seemingly, the dimension of the mesopores can be easily controlled by

adjusting the size of the starting carbon particles. In this sense, in the original

work of Jacobsen, carbon black pearls 2000, with a mean size of 12 nm, were

employed leading to a mesopore volume of 1.01 cm

3

g

1

and a pore sizewithin the range 5 50 nm. The dimensions of the zeolite particles so obtained

were relatively large (0.3 1.2 mm), being formed by agglomerates of smaller

crystals. However, these nanocrystals show their zeolite planes aligned

spanning throughout the entire agglomerate which is indicative of them being

organized as a single zeolite crystal. In addition, electron diffraction patterns

also indicate that the apparent agglomerates are actually single zeolite crys-

tals. Other carbon sources have been also used as matrixes such as multiwall

carbon nanotubes (MWNT) which led to narrower mesopore distributions

than with carbon black pearls.37 In addition, the use of carbon fibres also

allowed cylindrical mesopores to be obtained with low tortuosity.38

This method was further applied successfully to the synthesis of a rela-

tively large number of hierarchical zeolites, such as MFI,39 MEL,40 MTW,41

BEA and CHA.42 However, one of the limitations of the method is the

availability of adequate carbon templates to obtain the desired mesoporous

structures. Accordingly, a modification of the carbon template method has

been proposed also by Jacobsen et al.43 which allows at least partially this

drawback to be overcome. They devised a procedure consisting of using a

Fig. 4 Scheme of the synthesis of mesoporous zeolites by carbon templating according to thework of Jacobsen. (Reprinted with permission from ref. 35, Copyright American ChemicalSociety, 2000)

260 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

9/31

carbohydrate (sucrose) instead of a carbon template. The method comprises

the incipient wet impregnation of silica gel with a sucrose solution, followed

by its thermal decomposition under inert atmosphere (Ar). Thus, a carbon-

silica composite is formed. Subsequently, a base and the corresponding

zeolite template are added and the mixture is left crystallizing. Afterwards,

the carbon template is eliminated by combustion in air and a hierarchical

zeolite is formed. The resulting mesoporous structure can be tailored tosome extent by controlling the concentration of the sucrose solution. Fig. 5

shows TEM micrographs of one of the mesoporous zeolites so obtained.

It should be remarked that hierarchical zeolites obtained by the carbon

templated method of Jacobsen et al. contain interconnected micropores

and mesopores, so enhanced mass transfer rates are to be expected, unlike

it occurred with the hierarchical USY zeolites obtained by steaming. This

fact has been proved by 3D-TEM of a hierarchical zeolite synthesized

using carbon nanotubes as hard-templates. The 3D images showed that

the mesopores formed defined channels spanning throughout the zeolitecrystal.44

Carbon aerogels monoliths have been also used as templates for the

generation of mesoporosity in zeolites.45 In this case, the mechanism of

formation of mesopores is slightly different from those shown above

according to the Jacobsen method. Carbon aerogels were obtained from

resorcinol-formaldehyde gels after drying with CO2 under supercritical

conditions and pyrolysis under nitrogen atmosphere at 1323 K. The

resulting carbon aerogel possessed mesopores with 23 nm size. The for-

mation of the hierarchical zeolites takes place by the crystallization of the

ZSM-5 zeolite inside the mesopores of the carbon aerogel. The latter is

Fig. 5 TEM micrographs of hierarchical zeolites obtained using concentrated sucrosesolutions as precursors of the carbon template. (Reprinted with permission from ref. 43,Copyright American Chemical Society, 2007)

Catalysis, 2011, 23, 253283 | 261

7/31/2019 9781849732772-00253

10/31

subsequently removed by burning off under oxygen leaving available the

space corresponding to the carbon aerogel pore walls. The obtained

material is a ZSM-5 zeolite with a mesopore size of 11 nm. This method

presents the added advantage that the mesoporous zeolite has been

synthesized as a monolith.

Ordered mesoporous carbons, prepared by nanocasting, have been also

used successfully as templates for the synthesis of hierarchical zeolites.46,47

This is the case of CMK-3, an ordered mesoporous carbon attained by

nanoreplication of pure silica SBA-15. The hierarchical zeolites obtained

employing CMK-3 as template present mainly supermicropores or small

mesopores with a size around 2 nm. These are mesopores smaller than those

reported previously with other carbon sources. The textural properties of

the hierarchical zeolites can be tuned by changing the type of CMK-3

carbon used. It should be pointed out that the size of the mesopores in

CMK-3 carbon can be tailored using SBA-15 templates with different

mesopore size. A modification of this method consists of impregnatingdirectly the composite SBA-15/carbon or MCM-41/carbon with TPAOH

and left the mixture crystallizing hydrothermally under steam.47 After cal-

cination, a mesoporous ZSM-5 is formed. In this case, the size of the

mesopores is around 3.5 nm and 10 15 nm depending on whether MCM-

41 or SBA-15 are used as raw templates, respectively.

In summary, carbon-templating offer many possibilities for synthesizing

hierarchical zeolites due to the large availability of carbon materials that

can be employed. In contrast, the main disadvantage of these carbon-based

strategy is related to the need of burning the carbon template, which

represents a great amount of material being destroyed and having, in many

cases, a high cost. Moreover, the carbon combustion may generate high

temperatures that can damage the zeolite structure.

2.4 Hard templating by polymers

Closely related with the use of carbon based templates, is the application of

polymers as hard templates for the synthesis of hierarchical zeolites. In this

regard, the work of Xiao et al.48 deserves special mention. These authors

have employed mesoscale cationic polymers, like polydiallyldimethyl-

ammonium chloride, for the preparation of hierarchical Beta zeoliteby a one-step hydrothermal route. The size of the mesopores obtained

varies in the range 5 40 nm, which is in agreement with the molecular

dimension of the cationic polymer used. According to the authors, the

method can be extended to the synthesis of other zeolites different from Beta.

Polystyrene spheres have also been employed for obtaining hierarchical

zeolites with a microporous/macroporous pore structure.49 Likewise, the

use of nanospheres of poly(methyl methacrylate) (PMMA) has allowed

mesoporous microspheres of zeolite ZSM-5 to be prepared by a mechanism

involving self assembly of the aluminosilicate source, the PMMAnanospheres and TPAOH under basic conditions.50 Subsequently, hydro-

thermal crystallization takes place and the hierarchical zeolite is obtained

after calcination. Mesopores with about 13 nm diameter are achieved

with this method despite that the PMMA nanosphere size is far higher

(around 80 nm). Mesoporous zeolite A was also obtained using as templates

262 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

11/31

aerogels formed by sol-gel polymerization of resorcinol-formaldehyde.

The hierarchical zeolite A so obtained showed mesopores with a size of

roughly 15 nm. This method has the advantage that the nanoscale structure

of the polymer aerogel template can be tuned just by changing the resorcinol -

catalyst ratio, allowing some modification and control of the mesopore

size.51

2.5 Incorporation of organosilanes

One alternative initially attempted to obtain hierarchical zeolites was based

on putting together in the synthesis gel the structure-directing agent of the

zeolite and that of an ordered mesoporous solid in order to obtain a hybrid

material formed by an array of ordered mesoporous with crystalline zeolitic

walls. However, this approach has failed to get the desired product as it

usually led to a physical mixture of segregated phases.52 Pinnavia et al.53

succeeded in solving this problem by a two-step synthesis strategy involving

the previous formation of a precursor solution containing the zeolite seedsthat subsequently were assembled into the mesostructure by aggregation

around surfactant micelles. However, the materials so obtained lack of

zeolitic crystalline features, being wide-angle X-ray amorphous, which

makes difficult to establish their real crystalline character.

In order to avoid the phase separation of the mesoporous material

and the zeolite, Choi et al.54 designed a specific amphiphilic organosilane

((3-trimethoxysilyl) propyl/hexadecyldimethylammonium chloride) of

formula [(CH3O)3SiC3H6N(CH3)2C16H33]Cl. This amphiphilic organosi-

lane is a surfactant molecule that contains a hydrophobic alkyl chain, a

quaternary ammonium group (zeolite structure-directing agent) and a

hydrolysable trimethoxysilyl moiety. The latter interacts strongly with the

growing zeolite crystals due to the formation of covalent bonds with the silyl

moiety avoiding the separation of the phases. The silanization constitutes a

crucial step for the success of the method. The amount of amphiphilic

organosilane used is low (4% mol), being first added to a typical syn-

thesis composition of ZSM-5 zeolite, containing tetrapropylammonium

hydroxide (TPAOH). The evolution of the crystallinity of the products

obtained using this method with the synthesis time is shown in Fig. 6,

which depicts the low and wide-angle XRD patterns. At short synthesistimes (3 h), only a mesoporous phase with amorphous walls is obtained.

However, after 12 h, the appearance of zeolite MFI begins to be appreciated

in the high angle XRD pattern, and finally, after 2 days, the presence of a

mesoporous structure with zeolitic pore walls is appreciated.

The authors suggested a solution-based mechanism to explain the for-

mation of the mesoporous zeolite, involving the dissolution of the starting

mesophase, followed by the crystallization of the mesoporous zeolite crys-

tals from the dissolved species. The mesopore diameter can be tailored

within the range 2 20 nm by changing the length of the alkyl chain of theorganosilane or by modifying the hydrothermal conditions of the synthesis.

The mesoporosity created by this procedure is claimed to be rather uniform.

On the other hand, the mesopore size can be widened up to 24 nm by using

triblock copolymers (EO20PO70EO20) as pore expanding agents.55 This

method for the synthesis of hierarchical zeolites has allowed the preparation

Catalysis, 2011, 23, 253283 | 263

7/31/2019 9781849732772-00253

12/31

of MFI, LTA54,55 and sodalite56 materials. Regarding the acidity of the

samples, and according to infrared mass spectroscopy/temperature

programmed desorption measurements, hierarchical ZSM-5 synthesized by

this method contains Bro nsted acid sites of similar strength as those of

conventional ZSM-5 but in lower number.57

Different approachs have been also developed by other authors to syn-thesize hierarchical zeolites by perturbing the crystallization mechanism

through the addition of organosilanes. Thus, our research group5861

envisaged a procedure for the preparation of hierarchical zeolites by using

seed silanization agents. The method is based on the fact that during the early

stages of crystallization of MFI zeolites from clear synthesis solutions, the

precursors are nanounits with a particle size of 25 nm.6264 If the

agglomeration of these nanocrystals into bigger entities is partially hindered

by the presence of bulky organic substituents anchored over their external

surface, the obtained zeolite shows, after calcination, the occurrence of

mesopores whose dimension correspond to the voids occupied by the bulky

substituents during aggregation. This strategy comprises the following stages:

a) Precrystallization of the zeolite synthesis gel to promote the formation

of ZSM-5 protozeolitic nanounits, using tetrapropylammonium hydroxide

(TPAOH) as structure directing agent.

Fig. 6 XRD patterns of mesoporous MFI zeolite generated using amphiphilic organosilanealong different synthesis times. (Reprinted with permission from ref. 54, Copyright McMillan

Publishers Ltd., 2006)

264 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

13/31

b) Functionalization of the zeolite seeds by anchoring organosilanes on

their external surface. These organosilanes form a surface passivating layer

that avoids or reduces meaningfully the nanounit aggregation.

c) Crystallization to complete the zeolitization of the functionalized

protozeolitic units. This stage can be performed hydrothermally under

standard zeolite crystallization temperatures (e.g. 1701C).

d) Calcination to remove the silanization and the structure directingagents rendering accessible both micropores and mesopores in the hier-

archical zeolite.

As it can be appreciated in Fig 7, the hierarchical ZSM-5 so obtained is

made up by 200 400 nm aggregates formed by ultrasmall ZSM-5 units

whose size varies within the 5 10 nm range depending on the synthesis

conditions. Interestingly, the lattice fringes share their orientation among

the different nanounits which indicates that they are not really independent

nanocrystals but a significant degree of intergrowth exists between them.

This is an important aspect as it implies that the actual size of the crystallinedomains is quite larger than the size of the nanounits, which is expected to

improve the stability of these materials compared to nanozeolites formed

solely by independent nanocrystals.

On the other hand,27Al-MAS NMR analyses point out that most of

the aluminium atoms are incorporated inside the framework of the

as-synthesized zeolites. Additionally, 1D and 2D NMR analyses provided

evidences about the location of the TPA and silanization agent (pheny-

laminopropyltrimethoxysilane, PHAPTMS) moieties bearing out that they

Fig. 7 TEM images of hierarchical ZSM-5 obtained by crystallization of silanized seeds.(Reprinted with permission from ref. 8, Copyright Royal Society of Chemistry, 2008)

Catalysis, 2011, 23, 253283 | 265

7/31/2019 9781849732772-00253

14/31

are located inside the zeolite nanounits and grafted over their external

surface, respectively. This can be appreciated clearly in the schematic

diagram shown in Fig. 8.

The effects induced by the seed silanization treatment heavily depend

on the chosen synthesis variables, in particular upon the nature and

concentration of the seed silanization agent and on the precrystallization

stage. Different bulky organosilanes have been used as seed silanization

agents: octadecyltrimethoxysilane, isobutyltriethoxysilane, 3-aminopropyl-

trimethoxysilane and phenylaminopropyltrimethoxysilane.8 The former

(octadecyltrimethoxysilane) was only anchored onto the protozeolitic units

in a reduced extent (1.3 wt %), being not successful in developing meso-

porosity. In contrast, both isobutyltriethoxysilane and 3-aminopropyl-

trimethoxysilane led to materials having mesopores with a size of 8.0 nm

and an external surface area of around 200 m2 g1. But the most interesting

result has been achieved with phenylaminopropyltrimethoxysilane, leading

to materials having 2.0 3.0 nm mesopores, their walls being formed by the

smallest ZSM-5 nanounits (5 10 nm). Regardless of the organosilane used,

the crystallinity of all the samples has been proved by both XRD patterns,which indicate the absence of amorphous material, and FTIR, that shows

the appearance of the 550 cm 1 band, typical of the asymmetric stretching

mode of five membered rings present in ZSM-5 zeolite.65

The concentration of the seed silanization agent is also an important

variable. Thus, an optimum of 12 mol%, referred to the total silica content,

appears to exist with phenylaminopropyltrimethoxysilane as seed silaniza-

tion agent.61 On the other hand, the precrystallization step for the

formation of the nanozeolite precursors containing TPA occluded is

completely necessary, since in its absence the synthesis of an amorphousmaterial took place. In addition, the temperature used in the pre-

crystallization step influenced deeply the BET surface areas of the samples:

698m2 g 1 at 901C and 785m2 g1 at 401C. The latter value is specially

remarkable since it represents the highest reported BET surface area for a

hierarchical ZSM-5 zeolite.61

Fig. 8 Location of TPA and PHAPTMS moieties in hierarchical ZSM-5 obtained from sila-nized seeds. (Reprinted with permission from ref. 61, Copyright American Chemical Society, 2009)

266 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

15/31

The seed silanization approach can be considered as a general route for

synthesizing hierarchical zeolites since it is not bound to just ZSM-5 zeolite.

Hierarchical Beta,59 mordenite66 and TS-160 zeolites have been prepared by

the seed silanization procedure. In the case of hierarchical Beta, BET

surface areas as high as 857 m2 g 1 have been reported using phenylami-

nopropyltrimethoxysilane as seed silanization agent. A great proportion of

supermicropores (1.7 1.8 nm) were formed in addition to mesopores inhierarchical Beta zeolite.

Another method based on the use of organosilanes has been devised by

Pinnavaia et al.67 In this case, the procedure consists of employing a silane

functionalized polymer, as shown in the schematic diagram depicted in

Fig. 9. During the nucleation of the zeolitic entities, a silylated polymer

(e.g. silylated polyethyleneimine oxide) is grafted over their external surface

hindering the nanozeolite aggregation but not avoiding the zeolite crystal

formation. This leads to the generation of an intracrystalline polymer

network inside the zeolite structure. Finally, the polymer is removed bycalcination forming a zeolite with intracrystalline mesopores. Fig. 10 shows

the N2 adsorption isotherm at 77 K of hierarchical zeolite (MSU-MFI),

obtained using this approach, compared with a reference ZSM-5 sample

wherein the differences in the porosity between both samples are evident (see

the steep adsorption of MSU-MFI in the p/p0 range of 0.15 0.6). The

polymers employed had molecular weights within the range 600 25000

leading to mesopore sizes around 2.0 and 3.0 nm, respectively. Silylation has

Fig. 9 Mechanism of formation of a mesoporous zeolite using a silylated polymer. (Reprintedwith permission from ref. 67, Copyright Wiley-VCH Verlag GmbH & Co KGaA, 2006)

Catalysis, 2011, 23, 253283 | 267

7/31/2019 9781849732772-00253

16/31

probed to be essential for the success of the procedure since the use of a

non-silylated polyethyleneimine was not effective. In addition, the degree of

silylation is also important for the generation of mesopores. The procedure

is not limited to ZSM-5 zeolite but it has also been applied to the synthesis

of hierarchical zeolite Y.

These organosilanes based methods have the advantage of the possibility

of tailoring the mesopore size by choosing the adequate synthesis

conditions. This is particularly true for the seed silanization and theamphiphilic organosilane strategies. In addition, although organosilanes

may be expensive reagents, the employed amounts are relatively small and

their removal by calcination takes place at temperatures (300 5001C) low

enough for not damaging the structure of the hierarchical zeolite.

2.6 Other methods

A variety of other methods, that cannot be classified within any of the

previous categories, has been reported for the preparation of hierarchical

zeolites. Just as interesting examples, three of them are described in thissection. Zhang et al.68 employed bacteria (Bacillus subtilis) as templates in

the synthesis of microporous/macroporous MFI hierarchical zeolites. The

zeolite nanounits grow occupying the void spaces of the bacterial template,

so the formation of a microporous/macroporous framework occurs after

calcination. The final products are fibres formed by 0.5 mm channels and

Fig. 10 N2 adsorption isotherm at 77 K of a hierarchical ZSM-5 (MSU-MFI), prepared usinga silylated polymer, and of a conventional ZSM-5. (Reprinted with permission from ref. 67,

Copyright Wiley-VCH Verlag GmbH & Co KGaA, 2006)

268 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

17/31

100 nm thick walls of silicalite. Other curious method to prepare

hierarchical zeolites is the one based on the usage of the leaves of the plant

Equisetum arvense as template.69 The presence of silica in the plant pro-

motes zeolite crystallization leading to a microporous/mesoporous zeolite

(roughly 0.79 cm3 g1 of intracrystalline mesoporosity). On the other hand,

inorganic compounds have also been used as templates for the preparation

of hierarchical zeolites. In this regard, nanosized CaCO3 can act as templatefor the preparation of hierarchical silicalite-1. After the synthesis, the car-

bonate can be easily dissolved by acid treatment, developing the secondary

porosity (pore size within 50100 nm). In this procedure, the presence of

hydroxyl groups over the surface of the nanosized CaCO3 is essential to

interact with the silanol groups of the silica leading to the encapsulation of

the salt inside the zeolite crystal. Thus, if the hydroxyl groups are protected

by fatty acids leading to hydrophobic CaCO3, the synthesis is unsuccessful

yielding conventional zeolites instead of the hierarchical ones.70

3 Singular features of hierarchical zeolites

The presence of a bimodal micro/mesoporous structure provides hier-

archical zeolites with an improved accessibility to the active sites, which in

many cases influences positively on their catalytic activity compared to

conventional zeolites with micrometer crystal sizes. Hierarchical zeolites

possess a collection of singular properties that are commented henceforth

according to the following order:

Improved surface area.

Increase in mass transfer rates.

Resistance to deactivation.

High dispersion of active phases.

3.1 Improved surface area

For conventional zeolites, having crystal sizes in the micrometer range, the

proportion of external surface area is usually negligible, i.e. the BET surface

area corresponds almost completely with the surface area associated to the

micropores. However, in the case of hierarchical zeolites the presence of

mesoporosity implies that a great part of the surface area is related to thelatter, while a reduction is usually observed in the micropore surface area

compared with the standard zeolites. This is an essential aspect since the

mesopore surface area is not sterically hindered, as it occurs with the surface

area of the micropores, being capable of adsorbing and interacting with

bulky compounds.

Other interesting fact is that in most cases hierarchical zeolites have been

reported to present enhanced BET surface area, which also occurs for

nanozeolites. This fact can be interpreted as a result of the strongly

restricted adsorption that takes place within the zeolite micropores, whereasadsorption on the external/mesopore surface area is not so limited. Thus, in

the case of nanocrystalline ZSM-5, a good correlation has been found

between the enhancement of the BET surface area and the reduction in

the size of the nanocrystals.9 A similar trend is expected to be also valid

for hierarchical zeolites, with an increase of the BET surface area as the

Catalysis, 2011, 23, 253283 | 269

7/31/2019 9781849732772-00253

18/31

contribution of the secondary mesoporosity is more pronounced. Accord-

ingly, the BET surface area can be used as a parameter for comparison

among the different afore commented synthesis procedures of hierarchical

zeolites.

Fig. 11 illustrates the ranges, as well as the average values, of BET

surface areas reported for hierarchical ZSM-5 zeolite synthesized with the

most important methods previously discussed. It can be observed that thestrongest effect on the BET surface area is obtained with the seed silani-

zation strategy61 reaching values close to 800 m2 g1. Moreover, this

method allows the BET surface area to be adjusted in a wide range by

changing the synthesis conditions. Significant enhancements in the BET

surface area of ZSM-5 are also obtained with the use of silane-containing

polymers,67 desilication29 and amphiphilic organosilanes,54 although with

clearly lower values (up to around 600 m2 g1). This is a remarkable

achievement as the BET surface area of a conventional ZSM-5 is about

400m

2

g

1

. In contrast, lower BET surface areas are exhibited by thehierarchical ZSM-5 samples obtained using different types of carbon

materials as hard-templates. Thus, the use of carbon blacks71 just gives rise

at most to a BET surface area of 418m2 g 1, very close to the value

corresponding to a conventional ZSM-5. These results indicate that the

1 2 3 4 5 6 7 80

100

200

300

400

500

600

700

800

900

1000

Carbonnanotubes

Carbon

aerogels

Carbon

blacks

Mesostructured

carbons

Amphiphilic

organosilane

Desilication

Silane

polymer

Seed

silanization

BETsur

facearea(m2g-1)

Method of preparation

Fig. 11 Ranges of BET surface area values reported for hierarchical ZSM-5 samples preparedaccording to different synthesis methods (seed silanization,8,58,61 silane polymer,79

desilication,26,30 amphiphilic organosilanes,76,77 mesostructured carbons,46,47 carbonblacks,42,70 carbon aerogels45 and carbon nanotubes44).

270 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

19/31

degree of modification of the zeolite textural properties depends strongly on

the type of method employed for the generation of the hierarchical porosity.

Those methods based on the incorporation of organosilanes into the

synthesis gel appear to cause a stronger and more effective modification of

the textural properties of the zeolite in terms of surface area than the route

using carbon templates.

The improvement of the BET surface area, as well as the presence ofa high share of non-microporous surface area, opens the possibility for

the functionalization of hierarchical zeolites with different agents.

Thus, hierarchical ZSM-5 has been successfully functionalized with

3-aminopropylsilane, reaching the incorporation of 25% more organic

groups than in the case of using an ordered mesoporous SBA-15 support.

The organic-functionalized hierarchical zeolites show the added advantages

of their high hydrothermal stability and reusability. This has been proved in

the Pd-catalyzed Sonogashira cross coupling reaction72 of terminal alkynes

with chlorobenzenes under the presence of Na2CO3. In this reaction,the crystallinity of the hierarchical zeolite made possible their reuse since

conversion just decreased slightly (from 96 to 91%) after 5 reaction cycles.

In contrast, for SBA-15 the conversion values dropped from 84 to 38% due

to the lower hydrothermal stability of this material compared to the

mesoporous zeolite.

3.2 Increase in mass transfer rates

The existence of an interconnected network of mesopores and micropores is

expected to favour the intracrystalline mass transfer phenomena in hier-

archical zeolites. The mesopores allows a faster diffusion of the reacting

molecules towards the active sites leading to enhanced kinetics. In the same

way, the products may diffuse faster from the active sites reducing the extent

of secondary reactions which alters the obtained selectivity. This fact has

been nicely shown by Christensen et al.73 in the catalytic alkylation of

benzene with ethene. Fig. 12.a) shows the activity so obtained (TOF values)

versus the temperature, while Fig 12.b) illustrates the selectivity towards

ethylbenzene versus benzene conversion. These results indicate clearly

that the activity of the mesoporous zeolite is always higher than that of

the conventional ZSM-5 in the whole range of temperatures studied(583 643 K). Fig 12.b) evidences that selectivities towards ethylbenzene

attained over the hierarchical ZSM-5 are higher than the values obtained

over the conventional ZSM-5 catalyst. The occurrence of mass transport

constraints when using a conventional ZSM-5 zeolite with regard to the

hierarchical zeolite was concluded from the lower values of the activation

energies obtained with the former (59 vs 77 kJ/mol, respectively). Con-

sequently, this enhanced mass transport rates drives to higher benzene

conversions for the hierarchical zeolite as well as to larger selectivities

towards ethylbenzene. The latter effect has been ascribed to the shorterdiffusion path present in the hierarchical zeolite which suppresses the

secondary ethylation reaction.

Further analyses confirmed that the mass transport rates of both benzene

and ethylbenzene are diffusion limited over the conventional zeolite

since the value of the Thiele modulus was higher than 0.1, while over the

Catalysis, 2011, 23, 253283 | 271

7/31/2019 9781849732772-00253

20/31

hierarchical zeolite was much lower. In addition, experiments of diffusion

of isobutane point out that the effective diffusivity is three times higherover the hierarchical zeolite.74 These results have been obtained using

hierarchical zeolites prepared according to the carbon templating route.

Similar conclusions have been drawn by Groen et al.75 with hierarchical

zeolites synthesized using the desilication method. Thus, a two-order of

magnitude improvement has been denoted in the diffusion of neopentane

inside desilicated ZSM-5, due to the shorter diffusion path length and the

presence of an accessible network of mesopores.

3.3 Resistance to deactivationBesides the above described improvement of the intraparticle mass trans-

port, hierarchical zeolites show another key feature when compared with

conventional ones: increased catalyst lifetime due to their high resistance to

deactivation. The large size of the mesopores hinders the occurrence of

micropore blocking phenomena by coke, typically found in many catalytic

Fig. 12 Benzene alkylation with ethane over conventional and mesoporous ZSM-5 catalysts:a) TOF; b) selectivity towards ethylbenzene. (Reprinted with permission from ref. 73,Copyright American Chemical Society, 2003)

272 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

21/31

reactions when using conventional zeolites. In this regard, coke precursors

can diffuse out of the hierarchical zeolite easily due the presence of

mesopores, which avoids or delays their transformation into coke deposits,

slowing down the zeolite deactivation. This fact has been clearly shown by

Ryoo et al.76 for a hierarchical MFI zeolite, synthesized using an amphipilic

organosilane, in three different reactions: isomerisation of 1,2,4

trimethylbenzene, cumene cracking and esterification of benzyl alcohol withhexanoic acid. Fig. 13 illustrates the deactivation curves obtained in these

three reactions wherein the superb performance of the hierarchical

MFI zeolite over conventional MFI zeolite and ordered mesoporous

Al-MCM-41 can be appreciated.

In the isomerisation of 1,2,4-trimethylbenzene (Fig. 13.a), the conversion

obtained over conventional MFI catalyst drops quickly from 30 to 8% after

30 min, while for Al-MCM-41 falls from 13 to 5%. In contrast, the con-

version over hierarchical MFI slowly decreases from 25 to 17% after much

longer reaction times (180 min). In cumene cracking (Fig. 13.b), the con-version obtained over Al-MCM-41 is below 10% along the whole reaction

time (140 min). Hierarchical MFI and conventional MFI initially show al-

most complete conversion (B 95%) for this reaction but the evolution of

the activity with the time shows rather different trends between them.

Thus, hierarchical MFI keeps its high conversion (W 90%) while for the

conventional MFI conversion drops to around 50% after 140 min.

Likewise, as it can be observed in Fig. 13.c), the benzyl alcohol conversion

over conventional MFI and Al-MCM-41 drops after 5 reaction cycles from

22% and 81%, respectively, to almost zero. In contrast, over hierarchical

MFI only a slight decrease from 90 to 80% in conversion is observed after 5

reaction cycles.

The slow deactivation undergone by hierarchical zeolites has been mainly

ascribed to the presence of mesopores that favour a fast diffusion of the

coke precursors out of the zeolite, avoiding their accumulation inside the

micropores which would cause pore blocking phenomena. Likewise, the

difference in performance between the ordered mesoporous Al-MCM-41

materials and the hierarchical ZSM-5 zeolite (both of them possess

mesopores) has been assigned to the stronger acidity of the hierarchical

ZSM-5 as well as to the higher concentration of Al sites in the Al-MCM-41.The latter proposal takes into account that the Al sites are relatively close in

Al-MCM-41, which can promote the formation of polymeric coke

precursor species.

Another example of resistance to deactivation over hierarchical MFI has

been denoted in the methanol to hydrocarbon reaction (MTH).77 The

lifetime of the catalyst has been observed to increase more than three

times over the hierarchical ZSM-5 with regard to conventional ZSM-5.

This effect has been also related to the deposition of coke on the mesopores

of the hierarchical zeolite, while for the conventional sample it takes placepreferentially inside the zeolite micropores having a stronger deactivating

effect by pore blockage. This is nicely shown in Figs 14.a and 14.b, re-

spectively, wherein the coke content as well as the hydrocarbon production

obtained over conventional (ZST-12) and mesoporous ZSM-5 (OSD-5) are

shown.

Catalysis, 2011, 23, 253283 | 273

7/31/2019 9781849732772-00253

22/31

Fig.

13

Deactivationperformanceofhierarchical

MFIindifferentreactions:a)1

,2,4,

TMBisomerisation,

b)cumenecrackingandc)benzylalc

oholesterificationwith

hexanoicacid

.(Reprintedwithpermissionfromref.76

,CopyrightRoyalSocietyo

fChemistry,

2006)

274 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

23/31

It can be observed that the hydrocarbon production over the con-

ventional ZSM-5 catalyst drops to practically 0% after a time on stream of

40 h while over OSD-5 it is still around 50% after 130 h, indicating the

longer lifetime of the mesoporous ZSM-5. In addition, both Figs 14.a and

14.b provides information of the coke content, distinguishing between in-ternal coke (inside the micropores) and external coke (mainly within the

mesopores). It is clear that the internal coke is formed before and in a far

greater amount over the microporous ZSM-5 than over hierarchical ZSM-5,

wherein the coke is mostly external (W80%). The internal coke possesses a

stronger deactivation effect, which explains the fast decrease observed in the

conversion with the conventional ZSM-5.

3.4 High dispersion of active phases

The presence of mesoporosity in hierarchical zeolites offers great oppor-tunities for the preparation of bifunctional catalysts through the in-

corporation of other active phases, such as metals and metal oxides, in close

contact and with an improved interaction with the zeolite support. The

presence of mesopores is expected to lead towards better dispersions of the

active phase. This has been shown by Christensen et al.78 when impreg-

nating both mesoporous and conventional zeolites with Pt (B2 wt %), PtSn

alloy (B1 wt %) and a b- MoC2 carbide (B10 wt %). For the conventional

zeolites the metals are deposited mainly over the outer surface, frequently as

large particles according to EDS scan measurements associated to the TEMimages. In contrast, the same technique indicates that for the mesoporous

zeolite the metal nanocrystals are evenly placed within the mesopores.

Moreover, in this case, large metal particles are not observed. This result is

not only ascribed to the higher BET surface area and pore volume of the

hierarchical zeolite but to the occurrence of a much higher amount of lattice

Fig. 14 Evolution along the time on stream of the coke and the hydrocarbon production in themethanol conversion: a) microporous ZSM-5 (ZST-12) and b) mesoporous ZSM-5 (OSD-5).(Reprinted with permission from ref. 77, Copyright Elsevier, 2010)

Catalysis, 2011, 23, 253283 | 275

7/31/2019 9781849732772-00253

24/31

defect sites than in the case of conventional zeolites. These defect sites are

assumed to act as preferred deposition points for the active phase, leading to

improved dispersions. Additional examples of the benefits derived from

using hierarchical zeolites as supports of different active phases are de-

scribed in the next section dealing with the catalytic applications of these

materials.

4 Catalytic applications

The remarkable and singular properties shown by hierarchical zeolites have

brought about the potential catalytic applications of these materials in

numerous reactions, specially those wherein steric or diffusion limitations

are encountered. The next paragraphs review the literature works dealing

with the application of hierarchical zeolites in a variety of reactions, which

have been classified into three groups (oil refining and petrochemical re-

actions, fine chemicals reactions and environmental catalysis).

4.1 Oil refining and petrochemical reactions

A potentially interesting application of hierarchical zeolites is in petroleum

refining processes since conventional zeolite catalysts cannot refine about

20% of a petroleum barrel due to the steric hindrances posed by bulky

molecules. The application of hierarchical zeolites could diminish this

amount increasing meaningfully the profitability of the refining. In addition,

it is expected that the share of gasoline and light alkenes might also be

enhanced by the application of hierarchical zeolite catalysts.

This approach has been tested by Pinnavaia et al.79 that studied the

cracking of gas-oil over a mesoporous zeolite (MSU-MFI) and compared it

with a conventional ZSM-5. These authors observed increased conversions

over mesoporous MSU-MFI, accompanied by higher yields of gaseous

products (LPG), gasoline and light cycle oil (LCO) and lower amounts of

coke. In addition, much more light olefins were also detected.

Another petroleum-related application of hierarchical zeolites is the

desulfuration of gasoline and diesel fuels, which is a very important process

in order to comply with the target of reducing their sulphur content below

10 ppm.80 Thereby, 4,6, dimethyldibenzothiophene, a highly refractorysulphur-containing organic compound, has been hydrodesulfurized (HDS)

over Pt, Pd and Pt-Pd mesoporous ZSM-5 zeolite (total metal content of 0.5

wt %) leading to higher sulphur removal efficiency than metal/microporous

zeolite or metal/g-Al2O3. Fig. 15 shows the HDS conversion attained over

the different Pd containing catalysts: mesoporous Na(90%)/H (10%)-

ZSM-5 (MNZ-5), conventional Na-ZSM-5 (NZ-5) and g-Al2O3. It can be

appreciated the remarkable performance of Pd/mesoporous Na/H-ZSM-5

reaching 86% conversion with only 3% for conventional Pd/Na-ZSM-5 and

21% for Pd/g-Al2O3. This result has been ascribed to a proper combinationof acidity and mesoporosity in the hierarchical zeolite. In line with this

result, Pd/mesoporous Beta hydrodesulfurized 4,6 dimethyldibenzothio-

phene at 2501C under 62 bar of hydrogen better than Pd/Al-MCM-41 (51 vs

35%) due to the higher acidity of the zeolite.81 Additionally, Pd/mesopor-

ous Beta has shown to be more active in the hydrogenation, at 2501C and 40

276 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

25/31

bar of hydrogen, of the bulky aromatic pyrene than Pd/conventional Beta,

Pd/Al-MCM-41 and Pd/g-Al2O3. The difference in performance with the

Pd/conventional Beta lies in its larger mesopore volume, whereas regarding

the other two catalysts this result has been also related to its greater

acidity.82

Other reactions of interest are the aromatization and isomerisation of

1-hexene wherein hierarchical zeolites, obtained by desilication, showed

enhanced stability. Thus, after a time on stream of 14 h at 3501C, the

hierarchical zeolite prepared by desilication with 0.5 M NaOH shows

selectivity towards aromatics of 19.1% while the selectivity values obtained

with the conventional ZSM-5 drop to 5.1%.83 Likewise, in butene aroma-

tization at 3501C it has been found that after a time on stream of 34 h, the

conversion over hierarchical ZSM-5 remains rather stable at 99% while the

conversion over conventional HZSM-5 drops to 93%.84 This performance

has been ascribed to a lower deposition of coke inside the micropores,reducing the extent of micropore blocking.

Hierarchical Mo/HZSM-5 also shows enhanced selectivity to aromatics

due to a larger tolerance to coke in the catalytic dehydroaromatization

of methane.85 Hence, after 720 min of reaction at 1003 K, the methane

conversion was 11% with hierarchical Mo/HZSM-5 while the conversion

over conventional Mo/HZSM-5 dropped to 3.9%. In addition, the select-

ivity towards benzene was 68% over the hierarchical zeolite instead of 37%

over the conventional catalyst.

In the case of 1-butene isomerisation at 701C, an enhanced activity hasbeen found over H3PW12O40 supported on mesoporous silicalite-1 com-

pared with H3PW12O40 supported on a conventional silicalite-1.71 These

differences are specially remarkable since the initial conversion was 72% for

the former and less than 1% for the latter. Alkylation of benzene with

ethene over microporous ZSM-5 has also been improved over mesoporous

Fig. 15 HDS conversion of 4,6, dimethyldibenzothiophene obtained over different Pd-containingcatalysts. (Reprinted with permission from ref. 80, Copyright Wiley-VCH Verlag GmbH & CoKGaA, 2008)

Catalysis, 2011, 23, 253283 | 277

7/31/2019 9781849732772-00253

26/31

ZSM-5, as it was previously commented in point 3.b).73 Mesoporous

mordenite33 have also been tested for the same reaction at 438 K. This

catalyst brings about a 5-6 fold increased production of ethylbenzene

compared to conventional mordenite. The explanation to this behaviour has

been the presence of mesopores which reduces the deactivation of the

catalyst.

On the other hand, confocal fluorescence microscopy86

has been used tostudy the oligomerization of 4-methoxystyrene. The bimodal porous

structure of the mesoporous zeolite gives rise to the preferential formation

of dimeric carbocations due to the shortening of the micropore diffusion

path, precluding the appearance of higher oligomers. This result is clearly

bound up to the above indicated resistance to deactivation shown by

hierarchical zeolites.

4.2 Fine chemistry reactions

Hierarchical ZSM-5 has been tested in various reactions involving bulkymolecules with potential application in Fine Chemistry, like the protection

of benzaldehyde with pentaerythritol, condensation of benzaldehyde with

2-hydroxyacetophenone or the esterification of benzyl alcohol with

hexanoic acid.87 High activities were observed for these three reactions over

hierarchical ZSM-5. In addition, by means of experiments of dealumination

of the mesopore walls with tartaric acid, it was concluded that bulky mol-

ecules react at the Al sites placed over the mesopore walls.

Hierarchical ZSM-5 has been evaluated in the Friedel-Crafts acylation of

anisole at 1201C with either acetyl chloride or acetic anhydride as acylating

agent and compared with nanocrystalline HZSM-5 and Al-MCM-41.88

Superior conversions were achieved over the hierarchical ZSM-5 due to the

right combination of improved accessibility provided by the mesopores and

the high acidity and crystallinity of the zeolite.

Mesoporous sodalite has been also successfully applied in base catalyzed

reactions such as Knoevenagel condensation, Claisen-Schmidt conden-

sation and acetonyl acetone cyclization.56 This material shows enhanced

activity with regards to CsNaX and KAl-MCM-41 catalyst. Thus, in

Knoevenagel condensation of 4-isopropylbenzaldehyde with ethylcyanoa-

cetate at 353 K, the conversion obtained with K-mesoporous sodalite was78%, while KAl-MCM-41 and CsNaX gave 45 and 36% conversion values,

respectively. This result has been related to the basic sites present in the

mesopores of the hierarchical zeolite.

Pd exchanged mesoporous sodalite and NaA zeolite have been also

applied for different aryl coupling reactions (Suzuki, Heck and Sonoga-

shira) involving bulky substrates.89 These hierarchical catalysts have been

reported to show high activity and reusability avoiding the usual problem of

Pd leaching and agglomeration provided that the reactions are carried out

under air. In this regard, the mesoporous zeolite stabilizes the Pd2

specieseliminating the formation of unwanted agglomerates. Mesoporous MFI

zeolite show much higher activity (98%) than conventional ZSM-5 (3.9%),

Al-MCM-41 (25%) and ZSM-5 seed assembled mesoporous materials

(SAM, 64%) in the synthesis of jasminaldehyde (all of them were prepared

with Si/Al=20). In addition, the selectivity obtained with the mesoporous

278 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

27/31

zeolite was outstanding (98%), since the other materials presented values

below 80%. These results have been ascribed to the highly mesoporous

structure and strong acidity of the hierarchical MFI.54 These same authors

also tested the synthesis of vesidryl with the same catalysts obtaining again

superb results with the mesoporous MFI zeolite.

Other interesting application of hierarchical zeolites as catalysts is in

epoxidation reactions using Ti-containing hierarchical zeolites. Thus, in theepoxidation of cyclohexene39 with hydrogen peroxide, a yield of products

ten times higher has been detected over mesoporous TS-1 compared to

conventional TS-1. Likewise, hierarchical TS-1, obtained through the seed

silanization route, has been employed as catalyst in the epoxidation of

1-octene60 at 1001C with a bulky oxidant (TBHP, hindered to enter the

zeolite micropores), leading to much higher conversion (42% vs 5% of

conventional TS-1) with 100% epoxide selectivity and TBHP efficiency

higher than 90%. This result is specially relevant as it has opened the

possibility of using organic hydroperoxides as oxidants, instead of solelyhydrogen peroxide, in combination with TS-1 zeolite.

4.3 Environmental catalysis

Hierarchical zeolites have been also investigated in a number of reactions

within the field of environmental catalysis. Among them, the decomposition

of NO over hierarchical Cu-ZSM-11 and Cu-ZSM-5 deserves special

mention.90 In this case, the improved accessibility causes an enhanced

activity of the hierarchical zeolites because of the formation of dimeric and

oligomeric Cu species within the mesopores instead of the preferential

formation of monomeric Cu species over conventional Cu-ZSM-11 and

Cu-ZSM-5. In addition, hierarchical Cu-ZSM-11 was two-fold more active

than mesoporous Cu-ZSM-5 due to the occurrence of solely straight

microporous channels, wherein the active sites are preferentially located.

Likewise, desilicated Fe-ZSM-5 samples have shown enhanced N2O

decomposition activity.91 This has been assigned to the fully exchange of

iron in desilicated ZSM-5 samples due to its enhanced accessibility without

formation of iron oxides. In contrast, for large zeolite crystals, iron

exchange is diffusion controlled and leads to the deposition of inactive iron

species formed by hydrolysis.One less successful application of hierarchical zeolites has been the

catalytic pyrolysis of lignocellulose92 from beech wood at 5001C. Hier-

archical Beta zeolite yields less liquid bio-oil and more coke and char than

Al-MCM-41, leading to increased production of aromatics and PAH due to

the stronger acidity of their acid sites.

On the other hand, hierarchical zeolites have shown to be remarkable

catalysts for the cracking of polyolefins. The latter are bulky substrates

wherein an easy access to the acid sites leads towards higher activities. Thus,

a bulky polyolefin such as polypropylene has been cracked at 3601C overhierarchical Beta and ZSM-5 zeolites, obtained both by a seed silanization

procedure, being compared to the conventional catalysts.58 The improved

accessibility of these hierarchical catalysts for bulky polypropylene

molecules caused a four-fold greater conversion than conventional zeolites.

A similar result has been obtained in the catalytic cracking of low density

Catalysis, 2011, 23, 253283 | 279

7/31/2019 9781849732772-00253

28/31

polyethylene at 3401C using a plastic/catalyst mass ratio=100 with differ-ent hierarchical ZSM-5 samples prepared also by a seed silanization

method. In this case, while a TOF value of just 0.018 s1 is achieved over

conventional ZSM-5, the hierarchical ZSM-5 samples leads towards TOF

values far higher, within the 0.3840.901 range, depending on the combin-

ation of mesoporosity/acidity of the hierarchical sample.61

5 Concluding remarks

The field of synthesis and applications of hierarchical zeolites has been

extremely fruitful in the last decade. A wide range of synthesis strategies

have been developed that successfully allowed these zeolites to be prepared

with bimodal microporous/mesoporous structure. In addition, the number

of potential applications is growing every year, mainly as catalysts in a large

variety of reactions. In this respect, a bright future can be envisaged for

hierarchical zeolites, specially in transformations dealing with bulky sub-strates or suffering of strong deactivation by pore blockage.

The development of hierarchical zeolites has obliged to change the con-

ventional vision of zeolites as just pure microporous materials with shape

selectivity properties. At present, the occurrence of mesoporosity is an

added feature to a zeolitic material which markedly improves its properties:

enhanced textural properties, faster intraparticle transport, reduction of

steric and diffusion constraints, improved dispersion of active phases, better

resistance to deactivation and higher catalytic performance in many re-

actions. Interestingly, these improvements in many cases do not imply ne-cessarily a decrease in the selectivity exhibited by the zeolite.

The availability of a large variety of synthesis methods, which for sure

will be optimized and enlarged in the future, makes possible to tailor and

define the contribution and features of the mesoporosity present in hier-

archical zeolites. Consequently, the research in the field of hierarchical

zeolites is expected to grow, specially when dealing with the following goals:

Better control of the mesopore size and distribution.

Optimization of the ratio mesopore/micropore surface to achieve the

best catalyst performance.

Deeper characterization of the nature and strength of the acid sites

present in the mesopores.

Increase in the number of catalytic applications, with a strong focus on

reactions requiring bifunctional catalysts, prepared by incorporation of

different active phases to hierarchical zeolites.

References

1 E. M. Flanigen, Stud. Surf. Sci. Catal., 2001, 137, 11.

2 G. Bellusi and M. S. Rigutto, Stud. Surf. Sci. Catal., 2001, 137, 911.3 M. A. Uguina, D. P. Serrano, R. Van Grieken and S. Venes, Appl. Catal. A,

1993, 99(2), 97.

4 J. Aguado, J. L. Sotelo, D. P. Serrano, J. A. Calles and J. M. Escola, Energy &

Fuels, 1997, 11, 1225.

5 M. Hartmann, Angew. Chem. Int. Ed., 2004, 43, 5880.

280 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

29/31

6 C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck, Nature,

1992, 359, 710.

7 D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka and

G. D. Stucky, Science, 1998, 279, 548.

8 D. P. Serrano, J. Aguado, J. M. Escola, J. M. Rodriguez and A. Peral, J. Mat.

Chem., 2008, 18, 4210.

9 J. Aguado, D. P. Serrano, J. M. Escola and J. M. Rodriguez, Micropor.

Mesopor. Mater., 2004, 75(1-2), 41.

10 L. Tosheva and V. P. Valtchev, Chem. Mater., 2005, 17, 2494.

11 A. Corma, V. Fornes, S. B. Pergher, Th. L. M. Maesen and J. G. Buglass,

Nature, 1998, 396, 353.

12 K. G. Strohmaier and D. E. W. Vaughan, J. Am. Chem. Soc., 2003, 125, 16035.

13 A. Corma, M. J. Diaz-Caban as, J. L. Jorda , C. Martinez and M. Moliner,

Nature, 2006, 443, 842.

14 K. Egeblad, C. H. Christensen, M. Kustova and C. H. Christensen, Chem.

Mater., 2008, 20, 946.

15 J. Perez Ramirez, C. H. Christensen, K. Egeblad, C. H. Christensen and J. C.

Groen, Chem. Soc. Rev., 2008, 37, 2530.16 Fluid catalytic cracking V: Materials and technological innovations, ed. M. L.

Ocelli, P. OConnor, Elsevier, Amsterdam 2001.

17 R. Szostak, Stud. Surf. Sci. Catal., 2001, 137, 261.

18 G. T. Kerr, J. Phys. Chem., 1968, 72, 2594.

19 B. Sulikowski, J. Phys. Chem., 1993, 97, 1420.

20 C. J. Van Oers, W. J. J. Stevens, E. Bruijn, M. Mertens, O. I. Lebedev, G. Van

Tendeloo, V. Meynen and P. Cool, Micropor. Mesopor. Mater., 2009, 120, 29.

21 A. H. Janssen, A. J. Koster and K. P. de Jong, Angew. Chem. Int. Ed., 2001,

40(6), 1103.

22 P. Kortunov, S. Vasenkov, J. Karger, R. Valiullin, P. Gottschalk, M. F. Elia,M. Perez, M. Stocker, B. Drescher, G. McElhiney, C. Berger, R. Glaser and J.

Weitkamp, J. Am. Chem. Soc., 2005, 127, 13055.

23 M. Ogura, S. Shinomiya, J. Tateno, Y. Nara, E. Kikuchi and M. Matsukata,

Chem. Lett., 2000, 882.

24 M. Ogura, S. Shinomiya, J. Tateno, Y. Nora, M. Nomura, E. Kikuchi and M.

Matsukata, Appl. Catal. A, 2001, 219, 33.

25 T. Suzuki and T. Okuhara, Micropor. Mesopor. Mater., 2001, 43, 83.

26 J. C. Groen, J. C. Jansen, J. A. Moulijn and J. Perez Ramirez, J. Phys. Chem. B,

2004, 108, 13062.

27 J. C. Groen, L. A. A. Peffer, J. A. Moulijn and J. Perez Ramirez, Chem Eur. J.,2005, 11, 4983.

28 J. C. Groen, T. Bach, U. Ziese, A. M. Paulaime-van Donk, K. P. De Jong, J. A.

Moulijn and J. Perez Ramirez, J. Am. Chem. Soc., 2005, 127, 10792.

29 S. Abello, A. Bonilla and J. Perez-Ramirez, Appl. Catal A, 2009, 364(1-2), 191.

30 J. Perez Ramirez, D. Verboekend, A. Bonilla and S. Abello , Adv. Funct. Mater.,

2009, 19, 3972.

31 M. S. Holm, S. Svelle, F. Joenssen, P. Beato, C. H. Christensen, S. Bordiga and

M. Bjorgen, Appl. Catal. A, 2009, 356, 23.

32 F. T. Starzyk, I. Stan, S. Abello , A. Bonilla, K. Thomas, C. Fernandez, J. P.

Gilson and J. Perez Ramirez, J. Catal., 2009, 264, 11.33 J. C. Groen, T. Sano, J. A. Moulijn and J. Perez Ramirez, J. Catal., 2007, 251, 21.

34 X. Wei and P. G. Smirniotis, Micropor. Mesopor. Mater., 2006, 97, 97.

35 C. J. H. Jacobsen, C. Madsen, J. Houzvicka, J. Schmidt and A. Carlsson, J. Am.

Chem. Soc., 2000, 122, 7116.

36 C. Madsen and C. J. H. Jacobsen, Chem Commun., 1999, 673.

Catalysis, 2011, 23, 253283 | 281

7/31/2019 9781849732772-00253

30/31

37 I. Schmidt, A. Boisen, E. Gustavsson, K. Stahl, S. Pehrson, S. Dahl, A.

Carlsson and C. J. H. Jacobsen, Chem. Mater., 2001, 13, 4416.

38 A. H. Janssen, I. Scmidt, C. J. H. Jacobsen, A. J. Koster and K. P. De Jong,

Micropor. Mesopor. Mater., 2003, 65, 59.

39 I. Schmidt, A. Krogh, K. Wienberg, A. Carlsson, M. Brorson and C. J. H.

Jacobsen, Chem. Commun., 2000, 2157.

40 M. Y. Kustova, P. Hasselriis and C. H. Christensen, Catal. Lett., 2004, 96, 205.

41 X. Wei and P. G. Smirniotis, Micropor. Mesopor. Mater., 2005, 89, 170.

42 K. Egeblad, M. Kustova, S. K. Klitgaard, K. Zhu and C. H. Christensen,

Micropor. Mesopor. Mater., 2007, 101, 214.

43 M. Kustova, K. Egeblad, K. Zhu and C. H. Christensen, Chem. Mater., 2007,

19(12), 2915.

44 A. Boisen, I. Schmidt, A. Carlsson, S. Dahl, M. Brorson and C. J. H. Jacobsen,

Chem. Commun., 2003, 958.

45 Y. Tao, H. Kanoh and K. Kaneko, J. Am. Chem. Soc., 2003, 125, 6044.

46 Z. Yang, Y. Xia and R. Mokaya, Adv. Mater., 2004, 16(8), 727.

47 S. I. Cho, S. D. Choi, J. H. Kim and G. J. Kim, Adv. Funct. Mater., 2004

14, 49.48 F. S. Xiao, L. Wang, C. Yin, K. Lin, Y. Di, J. Li, R. Xu, D. S. Su, R. Schlogl, T.

Yokoi and T. Tatsumi, Angew. Chem. Int. Ed., 2006, 45, 3090.

49 B. T. Holland, L. Abrams and A. Stein, J. Am. Chem. Soc., 1999, 121, 4308.

50 J. J. Zhao, J. Zhou, Y. Chen, Q. J. He, M. L. Ruan, L. M. Guo, J. L. Shi and

H. R. Chen, J. Mat. Chem., 2009, 19(41), 7614.

51 Y. Tao, H. Kanoh and K. Kaneko, Langmuir, 2005, 21, 504.

52 A. Karlsson, M. Stocker and R. Schmidt, Micropor. Mesopor. Mater., 1999, 27,

181.

53 Y. Liu, W. Z. Zhang and T. J. Pinnavia, Angew. Chem. Int. Ed., 2001, 40, 1255.

54 M. Choi, H. S. Cho, R. Srivastava, C. Venkatesan, D. H. Choi and R. Ryoo,Nature Mater., 2006, 5, 718.

55 K. Choo, H. S. Cho, L. C. De Menorval and R. Ryoo, Chem. Mater., 2009, 21,

5664.

56 G. V. Shanbhag, M. Choi, J. Kim and R. Ryoo, J. Catal., 2009, 264, 88.

57 K. Suzuki, Y. Aoyagi, N. Katada, M. Choi, R. Ryoo and M. Niwa, Catal.

Today, 2008, 132, 38.

58 D. P. Serrano, J. Aguado, J. M. Escola, J. M. Rodriguez and A. Peral, Chem.

Mater., 2006, 18, 2462.

59 J. Aguado, D. P. Serrano and J. M. Rodriguez, Micropor. Mesopor. Mater.,

2008, 115, 504.60 D. P. Serrano, R. Sanz, P. Pizarro and I. Moreno, Chem. Commun., 2009, 11,

1407.

61 D. P. Serrano, J. Aguado, G. Morales, J. M. Rodriguez, A. Peral, M.

Thommes, J. D. Epping and B. F. Chmelka, Chem. Mater., 2009, 21, 641.

62 A. E. Persson, B. J. Schoeman, J. Sterte and J. E. Otterstedt, Zeolites, 1994, 14,

557.

63 B. J. Schoeman and O. Regev, Zeolites, 1996, 17, 447.

64 V. Nikolakis, E. Kokkoli, M. Tirrell, M. Tsapatsis and D. G. Vlachos, Chem.

Mater., 2000, 12, 845.

65 D. T. On and S. Kaliaguine, Angew. Chem. Int. Ed., 2001, 40(17), 3248.66 J. Aguado, D. P. Serrano, J. M. Escola and A. Peral, J. Anal. Appl. Pyrol., 2009,

85, 352.

67 H. Wang and T. J. Pinnavaia, Angew. Chem. Int. Ed., 2006, 45, 7603.

68 B. Zhang, S. A. Davis, N. H. Mendelson and S. Mann, Chem. Commun., 2000,

781.

282 | Catalysis, 2011, 23, 253283

7/31/2019 9781849732772-00253

31/31

69 V. Valtchev, M. Smaihi, A. C. Faust and L. Vidal, Angew. Chem. Int. Ed., 2003,

43, 2782.

70 H. Zhu, Z. Liu, Y. Wang, D. Kong, X. Yuan and Z. Xie, Chem. Mater., 2008,

20, 1134.

71 K. Zhu, J. Hu, X. She, J. Liu, Z. Nie, Y. Wang, C. H. F. Peden and J. H. Kwak,

J. Am. Chem. Soc., 2009, 131, 9715.

72 D. H. Lee, M. Choi, B. W. Yu and R. Ryoo, Chem. Commun., 2009, 74.

73 C. H. Christensen, K. Johansen, I. Schmidt and C. H. Christensen, J. Am.

Chem. Soc., 2003, 125, 13370.

74 C. H. Christensen, K. Johansen, E. Tornqvist, I. Schmidt, H. Topsoe and C. H.

Christensen, Catal. Today, 2007, 128, 117.

75 J. C. Groen, W. Zhu Browuer, S. T. Huynink, F. Kapteijn, J. A. Moulijn and J.

Perez Ramirez, J. Am. Chem. Soc., 2007, 129, 355.

76 R. Srivastava, M. Choi and R. Ryoo, Chem. Commun., 2006, 4489.

77 J. Kim, M. Choi and R. Ryoo, J. Catal., 2010, 269, 219.

78 C. H. Christensen, I. Schmidt, A. Carlsson, K. Johansen and K. Herbst, J. Am.

Chem. Soc., 2005, 127, 8098.

79 D. H. Park, S. S. Kim, H. Wang, T. J. Pinnavaia, M. C. Papapetrou, A. A.Lappas and K. S. Triantafyllidis, Angew. Chem. Int. Ed., 2009, 48, 7645.

80 Y. Sun and R. Prins, Angew. Chem. Int. Ed., 2008, 47, 8478.

81 T. Tang, C. Yin, L. Wang, Y. Ji and F. S. Xiao, J. Catal., 2008, 257, 125.

82 T. D. Tang, C. Y. Yin, L. F. Wang, Y. Y. Ji and F. S. Xiao, J. Catal., 2007,

249(1), 111.

83 Y. Li, S. Liu, Z. Zhang, S. Xie, X. Zhu and L. Xu, Appl. Catal. A, 2008, 338,

100.

84 Y. Song, X. Zhu, Y. Song, Q. Wang and L. Xu, Appl. Catal. A, 2006, 302, 69.

85 L. Su, L. Liu, J. Zhuang, H. Wang, Y. Li, W. Shen, Y. Xu and X. Bao, Catal.

Lett., 2003, 91(3-4), 155.86 M. H. F. Kox, E. Stavitski, J. C. Groen, J. Perez Ramirez, F. Kapteijn and

B. M. Weckhuysen, Chem. Eur. J., 2008, 14, 1718.

87 V. N. Shetti, J. Kim, R. Srivastava, M. Choi and R. Ryoo, J. Catal., 2008, 254,

296.

88 D. P. Serrano, R. A. Garcia and D. Otero, Appl. Catal. A, 2009, 359, 69.

89 M. Choi, D. H. Lee, K. Na, B. W. Yu and R. Ryoo, Angew. Chem. Int. Ed.,

2009, 48, 3673.

90 M. Y. Kustova, S. B. Rasmussen, A. L. Kustov and C. H. Christensen, Appl.

Catal. B, 2006, 67, 60.

91 I. Melian-Cabrera, S. Espinosa, J. C. Groen, B. v/d Lenden, F. Kapteijn andJ. A. Moulijn, J. Catal., 2006, 238, 250.

92 K. S. Triantafyllidis, E. F. Iliopoulou, E. V. Antonakou, A. A. Lappas, H.

Wang and T. J. Pinnavaia, Micropor. Mesopor. Mater., 2007, 99, 132.