2 Methods for Generating Meso- and Macropores in Zeolites

240



9 Textural Characterization of Mesoporous Zeolites

Al extraction



1



Si migration



Si

2



Si



3



Ultrastable Y



Figure 9.1 Schematic presentation of the formation of

mesopores. The grid represents the zeolite framework, the

black dots are framework aluminum atoms, the open circles

are aluminum atoms extracted from the framework, and the

dotted lines indicate the mesopores. (Adapted from [79].)



generation and the so-called symmetry index, that is, the ratio of the XRD (X-ray

diffraction) peak intensities ([111] + [241]/[350]) [100]. The symmetry index has

been proposed as an indicator of the extent of stacking faults inside the framework.

Dealumination is assumed to take place preferentially at these stacking faults.

The mesoporous mordenites thus obtained exhibit a 3D structure and have been

applied in industrial processes for the transalkylation of diisopropyl benzene and

benzene to yield cumene and (hydro)isomerization of alkanes.



9.2 Methods for Generating Meso- and Macropores in Zeolites



Dealumination leads, apart from mesopore generation, to an increase in the

framework Si/Al ratios, thereby enhancing hydrophobicity and (hydro)thermal

stability of the zeolites. One of the most well-known examples of mesoporous

zeolites obtained by dealumination is the family of USY and related materials,

which are invaluable industrial cracking catalysts. Meanwhile, the increased Si/Al

ratios also lead to a decrease in the acid site density and consequently enhance the

acidity of the acid sites, both exerting strong effects on the catalysis performance.

Besides acids, other chemicals can also be used to withdraw aluminum from

the framework; examples are SiCl4 [83, 101], (NH4 )2 SiF6 [83, 102], and EDTA

(ethylenediaminetetraacetic acid) [57, 103]. However, caution should be taken to

avoid severe framework collapse when the extraction of aluminum by EDTA or

(NH4 )2 SiF6 is faster than the migration of silicon species during the treatment [104].

9.2.1.2 Desilication

As with aluminum, framework silicon can also be selectively extracted from the

framework to generate mesopores. This has mostly been done by treating a zeolite

with a base, for example, NaOH, KOH, LiOH, NH4 OH, and Na2 CO3 , or a specific

acid like HF. While base treatment is usually used for the removal of amorphous

gel impurities from zeolite crystals, its potential for generating mesopores has

been long ignored. Dessau reported in 1992 that hollow crystals were obtained by

refluxing large ZSM-5 crystals in Na2 CO3 aqueous solution [60]. It was revealed that

highly selective dissolution of the interior of the crystals had occurred, while the

exterior surface remained relatively intact. This result hinted toward the inhibiting

role of aluminum during the base treatment and gave direct evidence of aluminum

zoning in large-crystal ZSM-5 synthesized in the presence of quaternary directing

agents. Suboti´ et al. further investigated the role of aluminum during the base

c

treatment of ZSM-5 [105, 106]. However, the evolution of the porous structure was

not investigated in detail. Ogura et al. reported the first explicit evidence of mesopore formation in ZSM-5 crystals by NaOH treatment [62]. Groen et al. reported the

detailed investigation of base treatment conditions for optimizing the mesopore

formation in a series of papers [11, 107–122]. They found that for ZSM-5 crystals

there appears to be an optimal window of Si/Al ratio in the parent zeolite, that is,

SiO2 /Al2 O3 = 50–100 (molar ratio) to achieve optimal mesoporosity with high mesopore surface areas up to 235 m2 g−1 , while still preserving the intrinsic crystalline

and acidic properties (Figure 9.2) [119]. While ZSM-5 is still the most intensively

investigated zeolite [39, 63–65, 123–128], desilication has recently also been applied

to mordenite [74, 112], beta [113, 129], and ZSM-12 [75]. More recently, hollow

nanoboxes of ZSM-5 and TS-1 were obtained by treating the zeolites with tetrapropylammonium hydroxide solution. A dissolution–recrystallization mechanism was

proposed for the formation of such unique nanoporous structures [130, 131].

Similar to dealumination, desilication inevitably modulates the framework Si/Al

ratios; however, in this case, it resulted in decreased Si/Al ratios. Moreover, some

extraframework aluminum species are often observed after the base treatment

[132]. Therefore, an additional acid treatment or ion-exchange step is needed to

remove these species for opening the micropores and mesopores.



241



242



9 Textural Characterization of Mesoporous Zeolites



Al prevents Si

extraction

Si/Al ≤ 15



NaOH



Limited

mesopore

formation



Aluminum

Silicon



Si/Al ∼ 25 – 50



NaOH



Si/Al ≥ 200



NaOH



Optimal Si/Al

range

Mesopores in the

range 5 – 20 nm



Excessive Si

dissolution

Large mesoand macropores



Figure 9.2 Simplified schematic representation of the influence of the Al content on the mechanism of pore formation

during the desilication treatment of MFI zeolites. (Reprinted

with permission from [119].)



9.2.1.3 Detitanation

For zeolite containing other metals in the framework, mesopores can also be generated by similar selective leaching method. Schmidt et al. reported that mesoporous

ETS-10 titanosilicate can be obtained by postsynthesis treatment with hydrogen

peroxide [67–69]. The materials exhibited increased external surface areas and

improved performances in the Beckmann rearrangement of cyclohexanone oxime

to ε-caprolactam.

The various postsynthesis treatment methods mentioned above have proved

effective for generating mesopores in zeolites. The intracrystalline diffusion pathlength is shortened by virtue of the mesopores. Improved diffusion [59, 74, 75] and

catalytic properties [39, 67, 112–114, 123–125, 133] have been generally observed.

However, it is still difficult to generate the additional pores in a controllable way

in terms of pore shape, pore volume, and connectivity. Moreover, it inevitably

involves the change of framework composition, and partial amorphization of the

crystalline structure. Therefore, in practical applications, it is sometimes difficult to disentangle the contribution of improved textural properties from that of

varied framework composition. Another effect that has received less attention is

the change of morphology and surface properties of the zeolite crystals by the

postsynthesis treatment [62, 97], which can exert strong effects on the adsorption

properties.



9.2 Methods for Generating Meso- and Macropores in Zeolites



9.2.2

Templating Method



A more straightforward method for generating mesopores is the templating

method, which has been extensively used for the synthesis of mesoporous materials

[16, 134, 135]. Different from the postsynthesis treatment method, templates are

employed during the zeolite crystallization and selectively removed afterward.

Therefore, the framework composition can be predetermined on the basis of the

synthesis gel. Various types of template materials have been employed. On the basis

of the structure and rigidity, they can be roughly categorized into hard template

and soft template [10, 15].

9.2.2.1 Hard Template

Hard templates here refer to materials with relatively rigid structure that are

not supposed to deform during the zeolite synthesis. In this respect, carbon

materials are ideal candidates for generating mesopores because of their inertness,

rigidity, robustness, diversity, and easy removal by combustion [8, 12, 15, 16].

Carbon was initially used for making nanosized zeolites by the confined growth

of zeolite crystals between the spaces of carbon particles [136–138]. Afterward,

it was revealed that, by controlling the growth of the zeolite crystals, the carbon

templates can be entrapped inside the zeolite crystals [41]. Mesoporous crystals

with intracrystalline mesopores can be synthesized by the subsequent removal of

carbon via combustion (Figure 9.3).

Carbon-templating routes have received wide attention and have developed

rapidly during the last few years. Different types of carbon materials, for example,

carbon black [40, 139–146], carbon nanofiber [40], carbon nanotube [42, 147], carbon

Pores created by

combustion of

carbon particles



Carbon particles

about 12 nm



O2



+ CO2



550 °C



About 1 µm zeolite

crystal grown in

pore system of carbon



Mesoporous zeolite

single crystal



Figure 9.3 Schematic presentation of the growth of zeolite

crystals around carbon particles. (Reprinted with permission

from [41].)



243



244



9 Textural Characterization of Mesoporous Zeolites



aerogel [43–45, 148–150], carbon by sugar decomposition [151, 152], ordered

mesoporous carbon [153, 154], and colloid-imprinted carbon (CIC) [155–158]

have been used for the synthesis of mesoporous zeolites with various framework

types, for example, MFI [40–45, 136–140, 142–144, 148–159], FAU [44, 138,

150], MEL [72, 140, 145, 159, 160], BEA [138, 159], MTW [140, 146, 161], CHA

[159], LTA [138], and AFI [159]. It is noted that using ordered nanoporous CIC

as the templates, Fan et al. have synthesized MFI single crystals with ordered

imprinted mesoporosity [158]. Recently, using the same concept, nanosized CaCO3

[46] and resorcinol-formaldehyde aerogels [162, 163] have also been employed

for synthesizing mesoporous zeolites. Similarly, zeolites featuring hierarchical

structures can been synthesized via macrotemplating using polystyrene beads [164,

165], resin beads [166, 167], polyurethane foam [168], and even biological materials

like bacteria [169], wood [170], and leaves [171, 172]. However, in most of the cases,

the products are formed by coating of the templates with zeolite synthesis solution,

and occur as polycrystalline ensembles of nanosized zeolites. Alternatively, such

macrostructure can also be formed by assembling preformed colloidal zeolite

precursor around the templates.

9.2.2.2 Soft Template

The concept of soft template is adapted from the micelle templating synthesis

of mesoporous materials [173, 174]. Ordered mesoporous materials have been

synthesized by assembling zeolite ‘‘seed,’’ with micelle templates. The ‘‘seed’’ can

be synthesized either from the precursor solution [175–191], or by decomposition of zeolites [192–195], the so-called top-down approach. Such materials, albeit

mostly do not exhibit a discernible zeolite phase, have shown enhanced thermal

and hydrothermal stability and enhanced catalytic activity compared to conventional mesoporous materials with amorphous pore walls [175, 176, 180–187, 189,

192–196]. In some diffusion-limited reactions, their catalytic properties are comparable and even surpass those of the microporous zeolite counterparts [175, 185,

186, 189, 193, 194, 197]. Attempts to generate mesopores using supramolecular

micelles during zeolite synthesis met little success and, in most of the cases, ended

up with a phase-segregated composite of zeolite crystals and mesoporous materials

with amorphous pore walls [198–200]. This stressed the importance of modulating

the interplay between the formation of the mesoporous structures and the pore

wall crystallization. Recently, this problem has been solved by an elegant approach

using novel templates of alkoxysilane-containing surfactant or polymer [34, 35, 38].

The presence of the siloxy group increased the interaction between the templates

and the pore wall, and helped in the retention of the mesoporous structure during

the pore wall crystallization (Figure 9.4). The resulting zeolites showed a highly

crystalline structure and uniform mesoporosity. Moreover, the mesopore size can

be controlled in a similar manner as for mesoporous materials, for example,

MCM-41, by using surfactant molecules with different chain lengths or polymers

with different molecular weights. In this sense, the mesoporous zeolite obtained by

this approach is a synergetic product, rather than a combination between zeolites

and mesoporous materials. Other templates, for example, starch [39, 201] and



9.2 Methods for Generating Meso- and Macropores in Zeolites



Si(OR)3



Si(OR)3



Si(OR)3

Zeolite

Nucleation



Si(OR)3



Si(OR)3



Silylated

polymer



Proto-Zeolite



Nucleated

zeolite–polymer

composite

Crystal

growth



Intracrystal

polymer network

formation



Figure 9.4 Conceptional approach to the synthesis of a

zeolite with intracrystal mesopores using a silylated polymer

as the template. (Reprinted with permission from [38].)



polydiallyldimethylammonium chloride [37], have also been used for the synthesis

of mesoporous zeolites.

Compared to the postsynthesis methods, the templating approach presents several advantages. First, mesopores can be generated without affecting the framework

composition, thus making it possible to investigate separately the effects of textural

and framework properties. Secondly, in principle, the pore volume, pore shape,

and connectivity can be tuned in a more controllable way by choosing proper

templates. However, the effects of these templates upon the zeolite composition

and phase purity of the final product should be taken into account, both during the

crystallization process as well as the template removal process.

9.2.3

Other Methods



Mesoporous zeolites can also be synthesized in the absence of templates. As mentioned before, in addition to the intracrystalline pores, mesopores can also arise

from the stacking of nanosized zeolites as intercrystalline pores. However, conventional colloidal or nanosized zeolites suffer from the difficulties of separation.

A solution to this is to synthesize assemblies of nanosized zeolites by controlled

nucleation and growth of the zeolite crystals [202, 203].

Alternatively, small zeolite particles can also be deposited onto mesoporous support [32, 33], thereby generating mesoporous composite materials. Gagae et al. [204]



245



246



9 Textural Characterization of Mesoporous Zeolites



and Stevens et al. [205] have obtained zeolite nanocrystals embedded in a mesomatrix by acidifying zeolite seed or precursor solutions. Recently, a new synthesis

approach, which combined the sol–gel transformation and the zeolite crystallization, was reported by Wang and coworkers [206–209]. By steam-assisted

crystallization of a previously aged precursor gel, mesoporous composite TUD-C

with integrated ZSM-5 nanocrystals was obtained [207]. Tetrapropylammonium

hydroxide, which was used as the template for the formation of the microporous

structure of zeolites, is believed to act also as a scaffolding agent for the mesopore

formation. Later on, using a similar concept, another composite material TUD-M

was synthesized by reversing the preparation steps, that is, a first hydrothermal

treatment of the precursor gel followed by the sol–gel transformation [208].

Summarizing this section, we can see that numerous methods are available

for generating meso- and macroporous zeolite materials. Each method features

its own advantages over the others, and they are complimentary to each other.

Another important issue concerning the different approaches for obtaining hierarchical zeolites, which is rarely addressed in the open literature, is the potential for

large-scale production and the corresponding costs. For traditional processes, such

as dealumination by steaming, engineers have created the means to scale-up from

laboratory grams to the tons necessary in industrial processes. For desilication,

the fast reaction time of 30 minutes could be problematic, resulting in inhomogeneously treated zeolites. The templating approaches have the disadvantage that

they bring additional costs with them, not only in raw chemicals as, for instance,

for the micelle route but also in the case of carbon tubes; an additional synthesis

step for obtaining the template is also needed. Despite the fact that increasing

the scale of a process generally leads to lower costs, upscaling and costs of zeolite

production are aspects that should not be overlooked.

Nevertheless, in order to arrive at a deeper understanding of the efficiency and

working mechanisms of these methods, as well as the impacts of these pores

on the zeolite performance in practical use, it is vital to have a clear picture of

the textural properties and structures of the hierarchical zeolites. In the following

part of this chapter, we elaborate on the different characterization techniques for

evaluating the textural porosity, with emphasis on some novel techniques (electron

tomography (ET), optical microscopy), as well as some well-known techniques that

have received less attention (thermoporometry, mercury porosimetry, etc.).



9.3

Characterization of Textural Properties of Mesoporous Zeolites

9.3.1

Gas Physisorption



Among the various techniques available for porous structure analysis, gas

physisorption is still the standard and mostly used technique [210]. This has been

due to the well-developed theory, as well as the easy operation and wide availability



9.3 Characterization of Textural Properties of Mesoporous Zeolites



of the experimental equipment. This technique accurately determines the amount

of gas adsorbed on a solid material at a certain temperature and pressure, and yields

important information on the porous structure (pore volume, specific surface

area, pore size distribution (PSD)) and the pore surface properties. Nitrogen and

argon are the mostly used adsorbates. Usually isotherms are measured at liquid

nitrogen temperature and pressures varying from vacuum to 1 bar. The presence

of pores of different dimensions can be discerned from the shape of the isotherms.

For microporous materials, like zeolites, Ar is advantageous over N2 because the

presence of a quadrupolar moment in N2 can result in enhanced interaction with

the heterogeneous surface of the zeolite framework, which results in difficulties

in accessing the pore sizes and pore shapes [211]. As the interpretation of the

recorded values usually relies on simplified models, the accuracy of the results is

strongly dependent on the validity of the assumptions inherent to the model [210].

For example, in reports on zeolites, the BET (Brunauer–Emmett–Teller) specific

surface area is often given [212]. However, since the micropore filling does not

fulfill the conditions for multilayer adsorption, which is the basis of BET theory,

the reported BET data do not represent a real physical surface area. In particular,

for materials with multiscale pores, the interference of mesopores on the

multilayer adsorption makes the interpretation more complicated. Nevertheless,

for comparative studies, the BET surface area can be used as a value proportional

to the volume adsorbed. For detailed description of gas physisorption analysis of

zeolites, we refer to [213]. Here, we only discuss some specific phenomena related

to zeolites featuring additional porosity.

Figure 9.5 shows typical nitrogen physisorption results of zeolite NaY and the

samples obtained after different postsynthesis treatments [214]. The descriptions of

the samples and the corresponding textural parameters derived from the isotherms

are listed in Table 9.1. For NaY, a type I isotherm was observed, which is

typical for chemisorption on nonporous materials, or physisorption on materials

with only microporosity. After postsynthesis modification, type IV isotherms

appeared, indicating the formation of mesopores. Moreover, different hysteresis

loops were observed for samples obtained by different treatments. Hysteresis

appearing in the multilayer range of physisorption isotherms is usually associated

with capillary condensation in mesopores. Although the factors that affect the

adsorption hysteresis loops are still not well understood, the shapes of hysteresis

loops have often been identified with specific pore geometries. The steep hysteresis

loop in sample HMVUSY (high-meso very ultra stable Y) indicates the presence

of close-to-cylindrical mesopores with relatively uniform PSD, while the flat and

wide hysteresis loop in XVUSY (eXtra very ultrastable Y) is more complex, and

can be attributed to the presence of slit pores or ink-bottle-shaped pores with small

openings, or due to the effect of a pore network, which is discussed in more detail

below.

For materials with multiscale pores, especially for microporous materials like

zeolites, the comparative analysis methods, like t plot [215], αs plot [213], and

θ plot [216] methods, are generally used for the estimation of pore volume and

specific surface area of pores of different dimensions by comparing the isotherms



247



9 Textural Characterization of Mesoporous Zeolites



0.8

HMVUSY + 0.15



0.7



XVUSY



0.6



NaY − 0.08



Vol ads (ml g−1)



USY



0.5

0.4

0.3

0.2

0



0.2



0.6



0.4



0.8



1



p /p 0



(a)

0.9

0.8



HMVUSY

XVUSY

USY

NaY



0.7

dV /dlog d



248



0.6

0.5

0.4

0.3

0.2

0.1

0

1



(b)



10



100



Pore diameter (nm)



Figure 9.5 Nitrogen physisorption

isotherms and the BJH pore size distribution curves calculated by the desorption

branches for several Y zeolites (from top to

bottom: HMVUSY, XVUSY, USY, and NaY).



For clarity, the isotherm of NaY has been

shifted 0.08 cm3 g−1 downward and the

isotherm of the HMVUSY has been shifted

upward by 0.15 cm3 g−1 . (Reprinted with

permission from [214].)



of samples under investigation with those of reference samples. For the t-plot

analysis, the multilayer film thickness of adsorbate (t-values) is determined on a

reference nonporous solid with similar surface properties to the samples under

investigation. Alternatively, standard reference plots can be used, for instance, the

Harkins and Jura equation for silica and alumina substrates:

t (nm) = 0.1



13.99

0.034 − log(p/p0 )



1/2



(9.1)



In a t-plot analysis, the volume adsorbed on the sample under investigation at

different pressures is plotted against t and the corresponding statistical average



9.3 Characterization of Textural Properties of Mesoporous Zeolites

Physical properties of NaY and the mesoporous

Ys: USY, XVUSY, and HMVUSYa.



Table 9.1



Sample



NaY

USY

XVUSY

HMVUSY



Si/Al bulk

(at/at)



Si/Al XPS

(at/at)b



a0 (nm)



% Yc



Vmicro

(cm3 g−1 )d



Vmeso

(cm3 g−1 )d



Sext

(m2 g−1 )e



2.6

2.6

39.3

5.0



2.8

1.1

71.3

1.4



2.469

2.450

2.423

2.427



100

87

72

71



0.34

0.26

0.28

0.15



0.05

0.11

0.25

0.47



8

63

120

146



Results from [214].

(ultrastable Y) was obtained by steaming treatment, XVUSY (eXtra very ultrastable Y) was

obtained by steaming twice and acid leaching, HMVUSY (high-meso very ultrastable Y) was obtained

by hydrothermal treatment.

b Si/Al ratios determined by XPS measurements.

c Relative crystallinity.

dV

micro and Vmeso : micropore volume and mesopore volume determined by t-plot analysis.

e Sum of external and mesopore surface area calculated from the t plot.

a USY



layer thickness is calculated from the standard isotherm obtained with a nonporous

reference solid. For a nonporous sample, a straight line through the origin is

expected. Deviation in shape of the t plot from linearity indicates the presence of

pores of a certain dimension. By this means, values like the micropore volume,

mesopore volume, macropore volume, and mesopore and external surface areas

can be calculated. In a similar manner to that of the t plot, the textural properties

can also be evaluated from the αs plot, wherein αs is defined by the ratios between

the volume adsorbed in the sample under investigation and the volume adsorbed

in the reference solid at a certain relative pressure [217]. For the assessment of

microporosity, the thickness of the multilayer is irrelevant and it has been suggested

to replace t by the ‘‘reduced’’ adsorption [210]. Nevertheless, both methods generally

give consistent results for micropore assessment.

Using the t-plot method, the effects of the postsynthesis treatment on the porous

structure was investigated for the zeolite Y samples shown in Figure 9.1. From

Table 9.1, one can see that upon steaming and acid-leaching treatments, the mesopore volume increases significantly at the expense of their micropore volume. This

has been attributed to the blocking of the micropores with extraframework species

generated during the hydrothermal treatment and/or the partial amorphization of

the framework.

From the isotherms, the PSD curves can be derived. For mesoporous materials,

the BJH (Barret–Joyner–Halenda) method [218], which is based on the Kelvin

equation for the hemispherical meniscus, is still the most frequently used method.

The adsorption process in mesopores often associates with capillary condensation.

Preferentially, the desorption branch of the isotherms is used for PSD analysis,

as it is closer to the thermodynamic equilibrium [219]. Figure 9.5b shows the

PSD curves calculated from the desorption branches of the isotherms using the



249



250



9 Textural Characterization of Mesoporous Zeolites



BJH method. For NaY, as expected, essentially no mesopores were observed. For

HMVUSY, a centered peak was observed around 10 nm, which correlated well with

the steep hysteresis observed in the isotherms. Peaks around 3–4 nm in the PSD

curves appeared for all the treated samples, which corresponded with the sudden

closing of the hysteresis loop in the isotherms at a partial pressure of 0.4–0.5. Such

a coincidence does not imply similarity among the porous structures of the treated

samples. Peaks in this range have often been observed with mesoporous materials,

especially zeolites with mesopores, and have been erroneously attributed to the

presence of true pores with sizes of 3–4 nm. It can actually be related to the nature

of the adsorptive rather than solely the nature of the adsorbent. This phenomenon

is often referred to as the cavitation effect, and has been discussed by Neimark and

coworkers [220, 221].

As depicted in Figure 9.6, for spherical pores with narrow necks smaller than

4 nm, cavitation of the pores always occurs at a partial pressure corresponding to

an apparent pore size of around 4 nm [222]. For nitrogen physisorption, pores with

sizes smaller than 4 nm show no hysteresis and exhibit reversible adsorption and

desorption isotherms. This is due to the instability of the hemispherical meniscus

during desorption in pores with a critical size of ∼4 nm, which is caused by the

increased chemical potential of the pore walls provoking spontaneous nucleation

of a bubble in the pore liquid. In ink-bottle pores as shown in Figure 9.6, the

pore emptying during desorption is delayed because the meniscus in the necks is

strongly curved and prevents evaporation. The tensile strength of the condensate

in the cavities has a maximum limit, which corresponds to a partial pressure of

2 nm



Vads (cm3 STP g−1)



Vads (cm3 STP g−1)



Vads (cm3 STP g−1)



4 nm



Vads (cm3 STP g−1)



6 nm



10 nm



0.0 0.2 0.4 0.6 0.8 1.0



p /p0



p /p0



1



10



100



Pore diameter (nm)



1



6 10



100



Pore diameter (nm)



0.0 0.2 0.4 0.6 0.8 1.0



dV/dlog d (cm3 g−1)



p /p0



dV/dlog d (cm3 g−1)



dV/dlog d (cm3 g−1)



0.0 0.2 0.4 0.6 0.8 1.0



p /p0

dV/dlog d (cm3 g−1)



0.0 0.2 0.4 0.6 0.8 1.0



1



4 10



100



Pore diameter (nm)



Figure 9.6 N2 adsorption and desorption isotherms at 77 K

(middle) and corresponding PSD (bottom) as derived from

BJH model of 10-nm cavities with entrance sizes between

2 and 10 nm. (Reprinted with permission from [222].)



1



4 10



100



Pore diameter (nm)