Bergaya F. Handbook of Clay Science

Подождите немного. Документ загружается.

zeolites catalysts have a limited micropore size, and are rapidly poisoned by metals in

the heavy oil fractions.

Since the ﬁrst announcement of the commercial availability of PILC (Vaughan

et al., 1979 ), their use in petroleum cracking alone exceeded that of other catalysts

(Adams, 1987). PILC also found applications in environmental protection and as

molecular sieves, selective adsorbent s (Karamanis et al., 1997), thermal insulators,

and electro chemical and optical devices (Mitchell, 1990).

Research into the synthesis of PILC builds on the pioneering work by Barrer and

McLeod (1955) who obtained microporous materials by replacing the interlayer

exchangeable cations in smectites with tetraalkylammonium ions (Barrer, 1986). The

synthesis of the ﬁrst organic pillared clay minerals with an interlayer distance of

0.5–0.6 nm and a permanent porosity was independently reported by Mortland and

Berkheiser (1976) and Shabtai et al. (1976). However, organic pillared clay minerals

of this type are thermally unstable at high temperatures (>250 1C) when interlayer

collapse occurs as in pristine (unmodiﬁed) clay minerals. However, below the de-

composition temperature organic pillared clays can usefully serve as molecular

sieves, selective adsorbents, and catalysts.

Clay minerals intercalated with inorgani c species retain their micro- (and meso-)

porosity after heating at X300 1C. Although polymeric (hydr)oxymetal cations were

used as pillaring agents for clay minerals since the late 1960s (Sawhney, 1968), the

paper by Brindley and Sempels (1977) is regarded as the ﬁrst account on the prep-

aration of an inorganic pillared clay mineral. The procedure consists of exchanging

the Na

ions in the pristine smectite with oligomeric (hydr) oxy aluminum cations

which are then converted by heat treatment into Al- oxides. By propping the smectite

layers permanently apart, these Al-oxide species act as ‘pillars’ in the interlayer

space.

Here we give the recent IUPAC

definition, followed by a brief summary of the

fundamental principles of pillaring. These include a description of the different clay

minerals used, the pillaring species, the properties of the PILC products, and possible

modes of bonding between pillar and clay mineral surface. The numerous applica-

tions of PILC are not descri bed here. For their industrial and environmental ap-

plications, see Chapters 10.2, 11.1 and 11.2.

7.5.3. IUPAC DEF INITION OF PILLARING AND PILLARED CLAY

MINERALS

Bergaya (1990) suggested that three criteria must be met for successful pillaring:

(i) a preliminary step of quasi-reversible intercalation of various species (charged

and/or unch arged, organic or inorganic) causing up to a 5-fold increase in basal

001

) spacing of the clay mineral by XRD; (ii) after heating at high temperatures,

IUPAC: International Union of Pure Applied Chemistry.

Chapter 12: Pillared Clays and Clay Minerals394

this spacing may decrease slightly but must not collapse, and its ﬁnal value must be

maintained; and (iii) the last and fundamental criterion is that the product must have

an accessible porosity, that is, the intercalated species must not ﬁll the entire in-

terlayer space as in chlorite. However, subsequent research into the pillaring of

different layered materials, and for different applications, led Bergaya (1994) to

modify the above conditions slightly. The three new criteria for pillaring are: (i)

intercalation, generally by exchanging the interlayer inorganic cations with cationic

pillars, increasing the d

001

spacing by at least 0.7 nm (yielding a structure with

properties resembling those of zeoli tes in their applications); (ii) a free height (which

is not always correctly evaluated by porosity measurements (Bergaya et al., 1993a ;

Bergaya, 1995); and (iii) the basal spacing and free height do not change when the

material is heated to at least 200 1C and in some cases up to 700–800 1C, under

anhydrous or hydrothermal atmosphere, or when the pH is varied (placed in acidic

or basic solutions). The latest techni cal report of the IUPAC (Schoonheydt et al.,

1999) gives a similar definiti on of pillared clays (enlarged to pillared layered solids):

‘‘y pillaring is a process by whi ch a layered compound is transformed in a thermally

stable micro- and/or mesoporous material with retention of the layer structure’’. The

report adds that ‘‘a pillaring agent is any compound which maintains the spacing

between adjacent layers upon removal of the solvent and which induces an exper-

imentally observable pore structure between the layers called ‘interlayer region’ ac-

cessible to molecules at least as large as N

’’. Thus, a ‘pillared material’ must fulﬁll at

least three criteria: (i) chemical and thermal stability (without furth er specifying the

chemical or temperature conditions); (ii) a certain layer ordering that enables at least

001

spacing to be determined (although a rational series of basal reﬂections is not

necessary); and (iii) accessibility of the inter layer space to molecules at least as large

as N

(thus, N

adsorption–desorption isotherms must be determined, and analyzed

via t- and a-plots; however, no order in the pillars and consequently in the pores is

required).

We should also add that different abbreviations were used to denote pillared

materials, in particular pillared clay minerals. The most widely used are PILC (for

‘pillared interLayered clays’) and PLS (for the more general ‘pillared layered struc-

tures’). CLS (for ‘cross-linked smectites’) was also used (Lahav et al., 1978;

Pinnavaia et al., 1985; Sterte and Shabtai, 1987) when the authors propose some

chemical bonding between the pillars and the clay mineral layers. Other abbrevi-

ations that occur in the literature are ‘ELS’ (for expanded layered struc tures) and

‘MELS’ (for molecularly engineered layered structures). The technical report of the

IUPAC (Schoonheydt et al., 1999) accepted (with some restrictions) ELS as denoting

pillared compounds, and MELS for pillared derivatives. The IUPAC report also

recommends the term pillared ‘LDH’ (for layered double hydroxides) rather than

pillared ‘anionic clays’ (see Chapter 13.1).

The number of layered materials, including clay minerals and LDH (Vaccari,

1998), that can be pillared, is very large. This chap ter is limited to the phyllosilicates.

The main factors inﬂuencing the intercalating/pillaring process are directly related to

7.5.3. IUPAC Definition of Pillaring and Pillared Clay Minerals 395

the nature of the clay mineral host, the pillar guest (intercalated species), and ex-

perimental conditions.

7.5.4. HOST CLAY MINERALS

The nature of the clay mineral host (or parent material) is very important since

this determines layer composition and charge as well as the site of isomorphous

substitution (in the octahedra l and/or tetrahedral sheet). The dimens ion and shape of

the clay mineral layer s and the different arrangements of particles and aggregates are

also inﬂuential. These arrangements are related to the initial concentration of the

aqueous clay dispersion. Single layers and particles can occur in very dilute disper-

sions, while aggregates are formed in concentrated suspensions. Natural swelling

clay minerals (montmorillonite, hectorite, beidellite, saponite) and synthetic

smectites (saponite and Laponite) were mostly used as hosts because the ﬁrst step

in pillaring is the intercalation of the pillaring agent by cation exchange. Vermiculite

(Suquet et al., 1991), phlogopite (del Rey-Perez-Caballer o and Poncelet, 2000), talc

(Urabe et al., 1991), F-tetrasilisic mica (Sakurai et al., 1990), rectorite, which is a

regularly interstratiﬁed pyrophyllite-beidellite (Brody and Johnson, 1990; Occelli,

1991; Guan and Pinnavaia, 1994) and any other interstratiﬁed clay mineral-smectite

(Jie et al., 1986; Sterte, 1990) also featured as mineral hosts.

Another important factor is the nature of the exchangeable cation, initially

present. Na

-exchanged smectites were used in the majority of cases because they

are dispersible in water, facilitating intercalation of the pillaring agents by cation

exchange. In some instances, the clay minerals are puriﬁed, or organic ally modiﬁed,

prior to intercalating the pillars.

Since impurities associated with naturally occurring clays may give rise to inho-

mogeneous particle distribution, and prior puriﬁcation of such clays is laborious and

time consuming, synthetic clay minerals of high purity were used as hosts

(Kloprogge, 1998). Then direct synthesis of pillared clay minerals from synthetic

layer silicat es is an alternative approach to obtaining PILC.

7.5.5. PILLARING SPECIES

Bergaya (1990) set out the different metal ions that were studied as pillaring

agents in the form of a ‘Periodic Table of elements’. An updated version of this is

shown in Table 7.5.1. The vacant squares in this table denote the number of elements

that have yet to be investigated. Although the hydrolysis behaviour of metal cations

(Baes and Mesmer, 1976) is generally understood, that of aluminium is more fully

known. For this reason , Al-pillared clay minerals were the ﬁrst to be prepared, and

extensively documented.

Chapter 12: Pillared Clays and Clay Minerals396

A. (Al

)

-Pillaring Agent

Hydrolysis Products

The composition of the Al-pillaring solutions strongly depends on the preparation

conditions. The degree of hydrolys is or basicity (OH

base

/Al

salt

ratio) is an important

factor controlling solut ion pH, and hence the nature of the Al species. The controlled

hydrolysis of Al salts gives rise to a number of species, such as [Al(H

]

mon-

omers an d different polymers. Even so, the [Al

(OH)

]

cation

denoted as ‘Al

’(Fig. 7.5.1) is assumed to be the pillaring species. Its pseudospher-

ical structure was deduced from X-ray crystallograph y using a well-crystallized Al

sulphate (Johansson, 1960, 1962). The geometrical (M

) structure was al-

ready dedu ced from XRD patterns by Keggin (1934) whose name is attached to this

structure. However, the ﬁrst ‘Keggin’ structure was ammonium 12-molybdophos-

phate, an anionic metal–oxygen cluster synthesized by Berzelius in 1826 (Ohler and

Bell, 2002).

Al-NMR spectroscopy was helpful in determining the Al environment in Al

Keggin ions both in solution and in the solid state (Akitt et al., 1972; Akitt and

Farthing, 1978; Pinnavaia et al., 1984). The structure of Al

was also analyzed by

other methods, including SAXS, Raman spectroscopy, and light scattering. SEM

photographs of Al

crystals show a tetrahedral symmetrical structure of the

tridecamers (Furrer et al., 1992 ). Using

Al-NMR spectroscopy, Parker and Kiricsi

(1995) reported a facile, reproducible method, and the optimal experimental con-

ditions, for the preparation of thermally stable Al Keggin cations. The decompo-

sition kinetics of Al

indicates the formation of ‘defective’ Al

dimers (Al

which become the main active pillaring intercalant after lengthy thermal ageing.

Under some particular conditions (NH

), Al Keggin cations can be stabilized by self-

polymerization leading to the formation of Al

(Vaughan, 1988b).

Table 7.5.1. The different elements that were used as pillars (underlined elements) or co-

pillars (and/or by post-exchange or by impregnation of the PILC) shown as the Periodic Table

Be B

Mg Al Si P

Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge

Y ZrNbMo RuRhPd

(i)

La Ta Os Pt

(i)

Ce Pr Nd

Notes: Al is surrounded by a bold square, because it is the ﬁrst and most studied element. Some elements

are only used for impregnation of PILC denoted by subscript (i).

7.5.5. Pillaring Species 397

Procedures for Obtaining Al

precursors can be obtained by three main methods. The widely used method is

hydrolysis of aqueous Al salt solutions (often 1 M) with a base (NaOH solution or

solid Na

). Al

is the dominant species in AlCl

solutions of 10

1

–10

3

M and

an Al/OH ratio of 2–2.2 (Bottero et al., 1980, 1982). Al salts with anions other than

chloride (Akitt and Farthing, 1981a), and M

as base where M is not Na

, was

also used (Akitt and Farthing, 1978). The second method involves dissolution of Al

metal (Akitt and Farthing, 1981b) in HCl or in aqueous AlCl

solution, which is

acidic because of the dissociation of water molecules induced by the high-polarizing

power of the small Al

cation. The second method has the advantage of excludi ng

interfering ions, and is preferred for the industrial preparation of commercial Al

chlorohydrates

(Jones, 1988). The third method involves electrolysis of AlCl

so-

lution, yielding pure Al

with an OH/Al ratio of 2.4 (Akitt and Farthing, 1981c).

Factors Inﬂuencing Al

Formation

Although the method of preparation is clearly of prime importance, many other

factors inﬂuence the formation of pure Al

in solution, such as the nature and initial

concentration of the reagents used, the degree of hydrolysis (OH/Al ratio), the rate

of adding the reactants, the temperature, and the time and temperature of ageing the

hydrolyzed solutions.

Fig. 7.5.1. The structure of Al

(Keggin cation).

Chlorhydrol (abbreviated as ACH) from Chemical Reheis Company. Locron L from Hoechst.

Chapter 12: Pillared Clays and Clay Minerals398

B. Other Pillaring Agents

The solution chemistry of other elements (M), like Zr, Cr, and Fe that are potentially

capable of acting as pillaring agents was reviewed in a special volume of Catalysis

Today, edited by Burch (1988). References related to other elements such as Be, Mg,

Ti, Nb, Ta, Mo, Ni, Cu, B, Si, and Bi pillaring specie s were mentioned in the review

by Bergaya (1990).

During the last 15 ye ars, a large volume of literature appeared on the above

elements as well as new ones like Ga, Sn, and V. Since Al and Ga have very similar

chemical properties, the Ga

ion appears to be isostructural with Al

(Bellaloui et

al., 1990; Bradley et al., 1990a, 1990b; Bradley and Kydd, 1991; Tang et al., 1993).

Sn was used as metal complex precursor (Palinko et al., 1993; Bodman et al., 2002).

When a vanadium precursor in the form of VOCl

is refluxed with a H

montmorillonite in benzene, the corresponding pillared clay mineral is obtained

(Choudary et al., 1990).

The theses by Molinard (1994) and Montarge

s (1997) also contain a bibliography

about Al and the above-mentioned elements.

There are also numerous patents from many countries, some of which are men-

tioned by Burch (1988) and Konta (2001). Since the patent lit erature is not easily

accessible, this is not referred to in this chapter.

C. Mixed Al– M and M– M’ Pillaring Agents

The intercalation/pillaring of clay minerals with solutions containing two different

cations was investigated since the 1980s. One of the pillaring cation is usually Al. The

second inorganic cation is then added in various molar fractions to improve the

thermal, adsorptive, and catalytic properties of the pillared products (Gil et al.,

2000a; Konta, 2001).

Incorporation of Zr, Ga, transition metal elements like Cr, Fe, Cu, Mo, Ru, and

lanthanides into Al-pillaring solutions was reported. Such mixed pillaring agents

were wide ly used over the past two decades as the following references indicate:

– mixed Al–Zr: Occelli (1986), Occelli and Finseth (1986), Moreno et al. (1999);

– mixed Al–Ga: Vieira Coelho and Ponce let (1990, 1991), Bradley et al. (1990a,

1990b, 1990c, 1992), Gonzalez et al. (1991, 1992), Bradley and Kydd (1991, 1993),

Tang et al. (1993, 1995), Bagshaw and Cooney (1995), Pesquera et al. (1995),

Hernando et al. (1996a,b), Benito et al. (1999);

– mixed Al–Ge: Lee et al. (2001);

– mixed Al–Cr: Carrado et al. (1986a, 1986b), Skoularikis et al. (1988), Zhao et al.

(1995), Storaro et al. (1997), Toranzo et al. (1997);

– mixed Al–Fe: Barrault et al. (1988, 1992), Lee et al. (1989), Bergaya and Barrault

(1990), Rightor et al. (1991), Bergaya et al. (1991, 1993b), Zhao et al. (1993a),

Bakas et al. (1994), Lenarda et al. (1994), Storaro et al. (1995, 1996), Ladavos et al.

(1996), Mandalia et al. (1998);

7.5.5. Pillaring Species 399

– mixed Al–Cu: Frini et al. (1997), Barrault et al. (1998);

– mixed Al–Mo: Gil and Montes (1997);

– mixed Al–Ru: Lenarda et al. (1994), Storaro et al. (1995);

– mixed Al–La: Sterte (1991), Trillo et al. (1991, 1993a, 1993b), Zhao et al. (1993b);

– mixed Al–Ce, Pires et al. (1998).

Sterte (1991) prepared mixed solutions of Al not only with La and with Ce, but

also with Pr a nd Nd. In contrast to La and Ce, the two last lanthanides fail to give a

pillared product. Moreover, Dominguez et al. (1998) compared the experimental

results with theoretical calculations using mixed solutions of Al–Ga, Al–La, and

Al–Ce.

The rapid growth in the use of mixed pillared solutions is due to demand for

improvement in speciﬁc properties. In particular, the heat stability of AlMPILC is

greater than that of the classical Al pillared clay minerals.

Sometimes, elements as Pt and Pd act not as co-pillars, but are specifically used

for impregnation of the PILC (Table 7.5.1 ). Moreover, a great number of MPILC is

often used as support to other combination of elements (see, for example, Fe and Cr

supported on TiPILC (Cheng et al., 1996).

The use of Al-free, mixed Fe–Cr (Akc-ay, 2004), Fe–Zr (Heylen and Vansant,

1997), and Cr–Zr (Toranzo et al., 1998) pillaring solutions was also reported. Again,

the aim here is to produce pillared clay miner als with improved properties for speciﬁc

applications.

D. Pillaring Agents with More Than Two Cations

The use of mixed pillaring solutions containing more than two metal cations was also

reported:

– for Al–La–Ce by Gonzalez et al. (1992), Mendioroz et al. (1993), Booij et al.

(1996a,b);

– for Cr–Fe–Zr by Jamis et al. (1995a, 1995b).

E. New Pillaring Agents

The literature also mentions the use of metal complexes as pillaring agents. Many of

these involve the chlorides of transition metals like niobium, tantalum, and molyb-

denum (Christiano et al., 1985; Christiano and Pinnavaia, 1986), and tin (Palinko

et al., 1993).

Pillared clay minerals obtained from inorgano-organo pillaring agents were also

been described. Coordination compounds, organometallic complexes, surfactants,

and/or polymers are the source of the organic carbon in the pillaring agents (Gil

et al., 2000a).

The early papers by Loeppert et al. (1979) and Pinnavaia (1983) on the prep-

aration of clay minerals pillared with organometallic complexes indicate that the

Chapter 12: Pillared Clays and Clay Minerals400

thermal stability is not high, because the C–C bonds decompose at about 400 1C. The

PILC can swell in water and the intercalated complexes remain exchangeable.

However, organometallic precursors based on Si complexes (Endo et al., 1980,

1981) are converted into silica pillars when heated at 500 1C. Tetraethoxysilane

(TEOS) is often used in Al–Si pillaring agents (Wada and Wada, 1980; Sterte and

Shabtai, 1987; Figueras, 1988; Zhao et al., 1992; Fetter et al., 1995; Gil et al., 1995).

Other organometallic precursors based on iron complexes (Yamanaka et al., 1984;

Doff et al., 198 8 ; Martin-Luengo et al., 1989; Maes and Vansant, 1995) are con-

verted into iron oxide pillars when heated at 350 1C. Heating clay minerals inter-

calated with organo-inorgano binuclear oxobridged Fe complexes (Dick and Weiss,

1998) gives rise to zeolite-like channels in the products that show interesting prop-

erties depending on the ligands used.

Direct intercalation of metal oxide sols (DIMOS) precursors in the presence of

silica and/or alumina or other aqueous oxide sols leads to basal spacings of more

than twice the thickness of the host clay mineral layer (see references in Pinnavaia,

1992; Cool and Vansant, 1998).

Some cases are more complex. For example, intercalated mixed organometal

complexes are converted into oxide sol particles on calcination (Yamanaka and

Takahama, 1993). This leads to pillars formed by clusters of SiO

particles covered

by TiO

on their surface. However, by intercalating the sols of the two oxides

simultaneously with an organic structure-directing agent, and burning the latter off,

mixed SiO

and TiO

pillars are obtained.

The intercalation of imogolite (a tubular alumino-silicate) into montmorillonite

(a layered alumino-silicate) gives rise to a tubular silicate-layered silicate (TSLS )

nanocomposite (Pinnavaia and Johnson, 1986).

Co-intercalation of Al (and/or La) pillaring species and a cationic surfactant as

precursors (Srinivasan and Fogler, 1990a, 1990b) pro duces a cheap organo-inorgano

clay mineral with better adsorp tive properties than classical activated carbon. Co-

intercalation of Al

cations and a non- ionic surfactant (Michot and Pinnavaia,

1991; Michot et al., 1992, 1993) gives a pillared clay with enhanced capacity for

adsorbing toxic pollutants.

The pillaring of acid-activated clays was investigated since the early 1990s

(Mokaya et al., 1991; Mokaya, 1992) coinciding with research into pillaring proc-

esses with mixed agents.

7.5.6. PILLARING METHODS

A. Pillaring in Dilute Dispersions

The classical pillaring method involves two successive steps. The ﬁrst is the inter-

calation of the pillaring agent. This is done by adding a dilute solution of the

precursor slowly to an already prepared dilute clay mineral dispersion (Fig. 7.5.2).

7.5.6. Pillaring Methods 401

The property of the intercalated clay mineral is dependent on the nature of the

metallic cation, hydrolysis conditions, and type of clay mineral. Ageing time, with or

without stirring, temperature, mode of washing (ﬁltration and/or by dialysis), drying

(at room temperature, or at low tempe rature in an oven, or by freeze-drying) are also

Suspension

(2%)

Pillaring solution

Suspension

(2%)

M Cl

NaOH

Pillaring methods in dilute dispersions

Ex-situ method In-situ method

Added

simultaneously

Pillaring methods in concentrated medium

Pillaring

solution

Pillaring

solution

Clay slurry

in dialysis

bag

Distilled

water

Clay powder

pillaring oligomer

powder

in dialysis bag

Fig. 7.5.2. Comparison of different pillaring methods.

Chapter 12: Pillared Clays and Clay Minerals402

important parameters ( Pinnavaia et al., 1984). The second step consists of heating

the intercalated clay mineral. Here, the heating temperature (in air or in a closed

reactor with ﬂowing gas) as well as the duration and rate of heating have a strong

inﬂuence on the properties of the resultant PILC.

For the Al Keggin cation, the charge (7+) in solution changes upon further

hydrolysis, and after intercalation and heating. The ﬁnal charge of the pillars is

variable, but appears to be about 4+. Since each of the parameters mentioned above

is inﬂuential, data comparison is difﬁcult. Different PILC are commonly character-

ized by their interlayer distance (by XRD), surface area, and porosity (see Section

7.5.7).

Variants of the classical method were described, but all these still consume large

amounts of liquid. One variant involves the in-situ hydrolysis of the exchangeable

cation in the interlayer space (Yamanaka and Brindley, 1978; Brindley and Kao,

1980). This method therefore requir es pre-exchange of the clay mineral with a suit-

able cation and does not yield a satisfactory product. Another variant involves

adding a dilute solution of both the metal salt and the base simultaneously to a dilute

clay dispersion (Fig. 7.5.2). Here, the pillaring species is presumed to form directly in

situ (in the clay dispersion). The rate of addition of the two solutions needs to be

carefully controlled so that the ﬁnal pH does not vary much with every added drop

of the two con stituents. This requires vigorous stirring of the clay suspension, which

must remain clear during the whole course of addition, showing no sign of precip-

itation. The advantage of this latter in-situ method is less time-consuming than the

former because the intercalant does not have to be prepared separately, and no pre-

exchange of the starting clay mineral is required.

During intercalation at ambient temperature, ultrasonic treatment during 20 min

of Ca

-montmorillonite with chlorhydrol leads to PILC with improved textural

properties and higher stability (Katdare et al., 1997, 2000 ).

Microwave irradiation can be used as an adequate rapid procedure to reduce the

time of intercalation required in traditional intercalation step (Fetter et al., 1996).

However, the whole pillars are not completely affected by this treatment (de Andres

et al., 1999).

B. Pillaring in Concentrated Medium

Although pillared clay minerals attracted much attention over the past 30 years,

particularly from industry, they were not yet used as commercial catalysts. A few

kilograms of commercial PILC per batch (N.T.U.A., Athens) were produced at a

small pilot scale level (Kaloidas et al., 1995). One reason is that the pillaring pro-

cedure, developed in the laboratory, is difﬁcult to extend to an industrial scale. This

method is time-consuming, needs a large volume of liquid and repeated separations

of huge amounts of clay dispersions by centrifugation or ﬁltration. To produce

pillared clay minerals at an industrial scale, the above procedure must be simpliﬁed.

In particu lar, the volume of all reagents needs to be greatly reduced. It is only in the

7.5.6. Pillaring Methods 403