Bergaya F. Handbook of Clay Science

Подождите немного. Документ загружается.

Hansbo, S., 1960. Consolidation of clay with special reference to inﬂuence of vertical sand

drains. Study in connection with full-scale investigations at Ska

-Edeby. Swedish Geotech-

nical Institute, Proceeding No 18.

Horseman, S. T., Harrington, J. F., 1997. Study of gas migration in MX-80 buffer bentonite.

National Environmental Research Council, British Geological Survey. Report WE/97/7.

Kato, H., Muroi M., Yamada, N., Ishida, H., Sato, H., 1995. Estimation of effective diffu-

sivity in compacted bentonite. In: Murakami, T., Ewing, R. C. (Eds). Scientiﬁc Basis for

Nuclear Waste Management XVIII. Materials Research Society Symposium Proceedings

353, 277–284.

Pusch, R., 1994. Waste disposal in rock. Developments in Geotechnical Engineering, 76.

Elsevier Amsterdam.

Pusch, R., Adey, R., 1999. Creep in buffer clay. SKB Technical Report TR-99-32. SKB,

Stockholm.

Pusch, R., Bo

rgesson, L., Erlstro

m, M., 1987. Alteration of isolating properties of dense

smectite clay in repository environment as exempliﬁed by seven pre-Quaternary clays. SKB

Technical Report TR 87-29. SKB, Stockholm.

Yong, R. N., Mohamed, A.M.O., 1992. A study of particle interaction energies in wetting of

unsaturated expansive clays. Canadian Geotechnical Journal 29, 1060–1070.

Chapter 6: Mechanical Properties of Clays and Clay Minerals260

Handbook of Clay Science

Edited by F. Bergaya, B.K.G. Theng and G. Lagaly

Developments in Clay Science, Vol. 1

261

Chapter 7

MODIFIED CLAYS AND CLAY MINERALS

F. BERGAYA

, B.K.G. THENG

AND G. LAGALY

CRMD, CNRS-Universite

d’Orle

ans, 1F-45071 Orleans Cedex 2, France

Landcare Research, Palmerst on North, New Zealand

Institut fu

r Anorganische Chemie, Universita

t Kiel, D-24118 Kiel, Germany

This composite chapter deals with the physical and chemical modiﬁcation of clays

and clay minerals in the broad sense. Acid activation and therm al treatment (heat-

ing), described in Chapters 7.1 and 7.2, have long been used to produce materials for

certain practical applications. The surface properties and reactivity of clay minerals

may also be modiﬁed by adsorption and intercalation of small and polymeric organic

species. Chapter 7.3 reviews the important topic of clay mineral–organic interac-

tions. Clay minerals have been implicat ed in the abiotic origins of life on earth

because of their ability to adsorb, protect, concentrate, and transform biomolecules.

On this basis, the possible role of clay minerals in life’s origins is included here

(Chapter 7.4). In relation to acid activation and thermal treatment, the pillaring of

clay minerals (Chapter 7.5) is a recent method of mo diﬁcation. As the term suggest s,

the process commonly involves intercalation of cati onic species acting as ‘pillars’ to

prop the mineral layers apart. Subsequent heating gives rise to a permanently porous

material, useful for organic catalysis and other environmental applications.

DOI: 10.1016/S1572-4352(05)01007-X

Handbook of Clay Science

Edited by F. Bergaya, B.K.G. Theng and G. Lagaly

Developments in Clay Science, Vol. 1

263

Chapter 7.1

ACID ACTIVATION OF CLAY MINERALS

P. KOMADEL AND J. MADEJOVA

Institute of Inorganic Chemistry, Slovak Academy of Sciences, SK-845 36 Bratislava,

Slovakia

One of the most common chemical modiﬁcations of clays, used for both industrial

and scientiﬁc purposes, is their acid activation. This consists of the treatment of clay

with a mineral acid solution, usually HCl or H

. The main task is to obtain

partly dissolved material of increased speciﬁc surface area, porosity and surface

acidity (Komadel, 2003). The manufactured materials are widely available, relatively

inexpensive solid sources of protons, effective in a number of industrially signiﬁcant

reactions and processes. Acid attack on clays also occurs naturally, e.g., in the

interaction of acid mine drainage with clay minerals (Gala

n et al., 1999; Dubı

kova

al., 2002). Mining waste containing sulphides is the most common and greatest

antropogenic source of acidity. Progressive oxidation leads to the production of

protons and sulphates in leaching waters, which generally also mobilize large

amount of metals by dissolution of minerals. These waters inﬂuence the composition

of surface waters but also have an impact on surrounding soils and terrestrial eco-

systems.

From the industrial point of view, the term ‘acid-activated clays’ was reserved

mainly for acid-treated bentonites. Bentonite has always had a multitude of markets

and acid-activated bentonite was a traditional product for many decades. It is usu-

ally a Ca

-bentonite that was treated with inorganic acids to replace divalent cal-

cium ions with monovalent hydrogen ions and to leach out ferric, ferrous,

aluminium and magnesium ions thus altering the layers of smectite and increasing

the speciﬁc surface area and porosity. This results in the production of bleaching

earths, clays suitable for a range of bleaching or decolourising applications, in which

they compet e against natural bleaching earths (Siddiqui, 1968 ; Kendall, 1996).

This chapter mainly reviews acid treatment of smectites, the dominant minerals in

bentonites. Other pro ducts include environmentally benign catalysts or their sup-

ports, which are used in various chemical reactions such as Friedel-Crafts alkylation

and acylation, dimerisation and polymerisation of unsatur ated hydrocarbons, etc.

(Adams, 1987; Brown, 1994; see also Chapter 10.2), or colour developers in car-

bonless copying papers (Fahn and Fenderl, 1983). Acid-treated clays pillared with

DOI: 10.1016/S1572-4352(05)01008-1

oxyhydroxyaluminium species are used to prepare clay-modiﬁed electrodes (Falaras

et al., 2000a), as adsorbents for oil clariﬁcation (Mokaya et al., 1993; Falaras et al.,

2000b; Pagano et al., 2001 ) and as catalysts (Mokaya and Jones, 1994 ; Bovey and

Jones, 1995; Bovey et al., 1996).

7.1.1. PROPERTIES AND AUTO-TRANSFORMATION OF

-EXCHANGED CLAY MINERALS

The acidity of acid-untreated smectites has two sources: (i) the compensating

cations; these may have a strong polarizing effect on coordinating water molecules,

most of which are in the interlayer spaces and may not be easily accessible; and (ii)

speciﬁc site s at the layer edges, where unsaturated ‘‘broken’’ bonds occur; these may

be compensated by OH groups formation, leading to Brønsted acid sites such as

Si–OH and also coordinately unsaturated Al

and Mg

easily formed at the

edges, behave as Lewis acid sites (Lambe rt and Poncelet, 1997).

The ﬁrst step in acid treatment is that the protons replace the exchangeable

cations and then they attack the layers (C

el and Komadel, 1994). The exchange

reaction is fast if there is good contact between acid and smectite, and the quantity of

available protons is sufﬁcient. The substitution rate is independent of the smectite if

the mineral contains only swelling layers. In contrast to smectites saturated with

metal cations, proton-saturated smectites are unstabl e. The layers are attacked by

surface and interlayer hydrated protons, even after drying the separated activated

smectite, similar to what occurs in solution. This process, known as ‘auto-transfor-

mation’, spontaneously changes H

-smectites to their (Al

,Fe

,Mg

) -forms

on ageing (Barshad and Foscolos, 1970). In aqueous dispersion at 90 1C the process

is completed within 4 days (Janek and Komadel, 1999).

To study the properties of H

-clays, maximal saturation by protons and stability

of the product are required. Various preparation methods were tested. The best

results are obtained by passing the clay suspension through a succession of H

–OH

–

–H

ion-exchange resins. H

-forms of o2 mm fractions of ben tonites with

varied Fe

contents are prepared by this method. Potentiometric titrations of pro-

ton-saturated ﬁne fractions of bentonites have been used to characterize the acid sites

at the smectite–water interface in dispersions. The titration curves have revealed that

the number of strong acid sites varies and accounts for 60–95% of the total acidity in

the freshly prepared H

-forms (Janek et al., 1997). Layer-charge distributions of all

samples are inhomogeneous. This distribution is changed after oxalate pretreatment

of the samples, due to the removal of readily soluble phases that might have blocked

exchange sites. After auto-transformation, the alkylammonium exchange method

(Lagaly, 1994) reveals inhomogeneous charge density distributions; the fraction of

layers of the highest charge decreases. Comparison of cation exchange capacity

(CEC) obtained from potentiometric curves and the CEC calculated from the mean

layer charge has conﬁrmed that the attack of protons occurs from particle edges.

Chapter 7.1: Acid Activation of Clay Minerals264

However, for several samples the structural attack may also occur from the inter-

layer space. Autotransformation of the H

-smectites also decreases the mean layer

charge. Protons preferentially attack the octahedral Mg

during the auto-trans-

formation. A number of strong acid sites decreases and the number of weak acid sites

increases on ageing.

The titration data obtained are used in a thermodynamic calculation of proton

afﬁnity distribution. Numerical solving of an integral adsorption equation reveals a

continuous distribution of proton interaction sites. Proton afﬁnit y distributions

clearly detected up to ﬁve different proton interaction sites in all the smectite–water

systems, within accessible experimental range of pHs between 2 and 12. The amoun t

of the strongest acid sites decreases on ageing, while the amount of all weaker acid

sites increases with the progress of autotransformation. The strongest acid sites are

connected with free protons present in the dispersion while the weaker acid sites are

connected with the titration of released structural Al

,Fe

,Mg

cations and/or

their hydrolysed species, and deprotonation of SiOH groups. These results indicate

the sources of acidity in acid activated bentonites (Janek and Komadel, 1993). Hy-

drated aluminium ions in freshly proton-saturated dispersions contribute to a group

of weak acid sites, which also include oligomeric hydroxoaluminium cations. The

amount of these sites increases during auto-transformation. The fres hly prepared

proton-saturated dispersions show low pH values and the particles interact by edge-

to-face contacts. This increases the viscosity in comparison with the sodium forms at

pH close to 7 (Janek and Lagaly, 2001). A kinetic study of proton-promoted dis-

solution of K

-montmorillonite in solutions with constant KCl concentrations, us-

ing both titration and batch equilibration experiments, shows that adsorption of H

and dissolution of Al

occur (Zysset and Schindler, 1996).

Different clay minerals are treated with HCl or NaOH at room temperature for 2

weeks to obtain information on variable surface charge. Both treatments lead to an

increase of variable surface charge, while the actual charge value increases and

decreases depending on the mineral and the treatment. The heterogeneity of charge-

generating surface groups is observed in natural minerals. Duri ng acid treatment, the

number of weakly acidic surface functional groups increases, while the number of

groups of stronger acidic character decreases. Dissolut ion of Al

prevails over Si

in acid treatments. Illite and kaolinite are most resistant to acid attack. The surface

areas of the minerals computed from both water and nitrogen adsorpt ion isotherms

increase with acid and alkali treatments. The effects of acid and alkali attacks are

controlled by the individual character of the minerals (Jozefaciuk, 2002; Jozefaciuk

and Bowanko, 2002).

7.1.2. METHODS OF INVESTIGATION

The methods being employed to charact erize acid-activated silicates include

chemical analysis, XRD, Mo

ssbauer, Fourier transform infrared (FTIR), and

7.1.2. Methods of Investigation 265

MAS-NMR spectroscopies; Scanning electron microscopy, transmission electron

microscopy (TEM) and high resolution transmission electron microscopy

(HRTEM); acidity, surface area and pore size measurements, etc. Usually a com-

bination of several methods is needed for sufﬁcient characterization of the materials

obtained (C

el and Komadel, 1994 ; Vicente et al., 1994; Breen et al., 1995b;

Komadel et al., 19 96b; 1995a; 1995b, 1996b; Gates et al., 2002).

Chemical analysis of solid and/or liquid reaction products, infrared spectroscopy

(IR) and MAS-NMR spectroscopies are very sensitive to the nature and content of

the octahedral atoms and thus, also to the changes that occur in different stages of

acid attack (Breen et al., 1 995a, 1995b).

IR spectroscopy has been used for identiﬁcation of the main mineral and the

admixtures in the ﬁne fractions of bentonites (Farmer, 1974; Madejova

et al., 1992,

1995 Russell and Fraser, 1994; Madejova

and Komadel, 2001). It is also a routine

characterization technique for acid-treated clays, very sensitive to modiﬁcations of

the clay structure resulting from acid treatment. As protons penetrate into the clay

layers and attack the OH groups, the resulting dehydroxylation connected with

successive release of the octahedral atoms can be readily followed by changes in the

characteristic absorption bands attributed to vibrations of OH groups and/or oc -

tahedral cations. Comparative IR studies of acid-treated smectites (Madejova

et al.,

1998) as well as saponites, sepiolite and palygorskite (Vicente et al., 1996a) have been

published.

Changes in the IR spectra of SWy-1 montmorillonite (with quartz admixture)

after treatment with 6 M HCl (Figure 7.1.1) demonstrate the alteration of the chem-

ical bonds in the structure.

The spectrum of the untreated sample shows an intensive band at 1048 cm

–1

attributed to the Si–O stretching vibrations, and bands at 524 and 466 cm

–1

assigned

to Al–O–Si (octahedral Al) and Si–O–Si bending vibrations, respectively. Three

peaks in the hydroxyl bending region at 917 cm

–1

for Al

OH, 886 cm

–1

for AlFeOH,

and 850 cm

–1

for AlMgOH reﬂect that octahedral Al

is partially replaced by Fe

and Mg

. The doublet at 799 and 779 cm

–1

indicates the presence of quartz im-

purity in the sample, which has been conﬁrmed by XRD.

Si MAS-NMR shows

that 12% of total Si

is associated with quartz (Tka

et al., 1994; Madejova

et al.,

1998).

After 8 h of HCl treatment, no signiﬁcant changes are seen in the IR spectra of

SWy-1 montmorillonite except for a slight upward shift of the Si–O stretching band

(Dn ¼ 6cm

–1

) together with a weak decrease in the intensities of the OH and

Al–O–Si bending bands. This reﬂects the partially diminishing content of octahedral

cations due to acid attack.

As the treatment time progresses, a more pronounced decrease is observed in the

intensities of the bands related to octahedral cations. Changes in the Si

environ-

ment with prolonged acid treatment are reﬂected in both the position and the shape

of the Si–O stretching band. In addition to the tetrahedral Si–O band near

1058 cm

–1

, the IR spectrum of the sample treated for 12 h shows a pronounced

Chapter 7.1: Acid Activation of Clay Minerals266

absorption near 1100 cm

–1

. This absorption band is assigned to Si–O vibrations of

amorphous silica with a three-dimensional framework formed during acid treatment.

The spectrum of the sample treated for 18 h conﬁrms a high degree of structural

decomposition. The Si–O band of amorphous silica at 1103 cm

–1

dominates the Si–O

stretching region and only inﬂections related to OH bending bands are present in the

950–800 cm

–1

region. Anothe r characteristic band of amorphous silica near 800 cm

–1

increases in intensit y with treatment time and gradually overlaps the doublet of

quartz. The band near 524 cm

–1

, which is most sensitive to the presence of residual

in the octahedral sheets, decreases stepwise in intensity with increasing layer

decomposition.

After 30 h of treatment, the small band near 524 cm

–1

in the spectrum of the 18 h

treated sample, is changed to an inﬂection. This reﬂects a very high, but yet incom-

plete dissol ution of montmorillonite in 6 M HCl (Madejova

et al., 1998).

7.1.3. ACID DISSOLUTION OF SMECTITES

Treatment of clays with strong inorganic acids is frequently called ‘‘acid disso-

lution’’ or ‘‘acid activation’’ of clays. Depending on the extent of acid activation, the

resulting solid product also contains unaltered layers and amorphous three-dimen-

sional cross-linked silica, while the ambient acid solution contains ions according to

the chemical composition of the smectite and acid used.

1200 1000 800 600

30 h

18 h

12 h

8 h

0 h

Al-O-Si

524

917

886

850

799

1103

1048

Absorbance

Wavenumbers / cm

-1

Fig. 7.1.1. IR spectra of SWy-1 montmorillonite treated with 6 M HCl at 95 1C for 0, 8, 12, 18

and 30 h. Q is quartz. (From Madejova

et al., 1998.)

7.1.3. Acid Dissolution of Smectites 267

Early acid-dissolution studi es of dioctahedral smectites in HCl by Osthaus (1954,

1956), based on solution analysis, indicated faster dissolution of octahedral than

tetrahedral sheets. Assays of solid reaction products employing advanced spec-

troscopic techniques provided experimental evidence that acid treatments dissolve

central atoms from the tetrahedral and octahedral sheets at similar rates (Luca and

MacLachlan, 1992; Tka

et al., 1994).

Luca and MacLachlan (1992) studied the dissolution in 10% HCl of two non-

tronites from Garﬁeld and Hohen–Hagen, by Mo

ssbauer spectroscopy. They ﬁt the

spectra either with two octahedral Fe

doublets or with two octahedral and one

tetrahedral Fe

doublet. Isomer shift and quadrupole splitting values obtained

from the two-doublet models correspond to octahedral Fe

and not to tetrahedral

as Osthaus (1954) has suggested. W hen a tetrahedral Fe

doublet is included

in the model used to ﬁt the Mo

ssbauer spectra of acid-treated Garﬁeld nontronite

samples, a slight increase in the intensity of that doublet occurs with increasing

dissolution. This is much lower than Osthaus (1954) indicated. No trend in the

intensity of the tetrahedral Fe

doublet is observed for acid-treated Hohen-Hagen

nontronite samples. Therefore, acid treatment appears to remove octahedral and

tetrahedral Fe

from the nontronite structure at about the same rate. Mo

ssbauer

and IR spectroscopies, and XRD indicate that the undissolved part of nontronite

remaining after acid treatment is structurally similar to the untreated nontronite.

Al and

Si MAS NMR spectroscopic studies on removal of tetrahedral and oc-

tahedral Al

from montmorillonite by 6 M HCl, yield very similar conclusions

(Tka

et al., 1994). The rates of dissolution of tetrahedral and octahedral Al

are

also comparable for montmorillonite. Three different types of structural units have

been identiﬁed in acid-treated samples, including (SiO)

SiOH units, remaining as a

result of poor-ordering of the framework without the possibility of cross-linking.

The extent of the dissolution reaction depends on both clay mineral type and

reaction conditions, such as the acid/ clay ratio, acid concentration, time and tem-

perature of the reaction. The composition of the clay layers substantially affects their

stability against acid attack; trioctahedral layers dissolve much faster than their

dioctahedral counterparts (Vicente et al., 1994, 1995b; Breen et al., 1995a; Komadel

et al., 1996b). Higher substitutions of M g

and/or Fe

for Al

in dioctahedral

smectites increase their dissolution rate in acids (Breen et al., 1995b). For 15 di-

octahedral smectites, there is a good correlation of the Mg

and Fe

contents with

the half time of dissolution in 6 M HCl at 96 1C(Nova

k and C

el, 1978). Effects of

smectite type, acid concentration and temperature on the ha lf time of dissolution in

0.2 L HCl/g smectite acid/clay ratio in closed systems (no substances being add ed or

removed) are summarized in Table 7.1.1.

The rate of dissolution of various atoms, obtained from chemical analysis of the

liquid reaction products, indicates the presence of different phases in bentonite.

Readily soluble octahedral and tetrahedral constituents, and ‘‘insoluble’’ portions of

constituent atoms can be calculated from the dissolution curves. This provides in-

formation on the distribution of atoms in the sample (C

el and Komadel, 1994).

Chapter 7.1: Acid Activation of Clay Minerals268

Readily soluble portions include exchangeable cations and easily soluble admixtures

such as goethite (Komadel et al., 1993) and calcite (Komadel et al., 1996b). The most

common ‘‘insoluble’’ phases found in the ﬁne fractions of bentonites include kao-

linite, quartz, anatase and volcanic glass. Halloysite is the most decomposed mineral

after treatment in sulphuric acid of different concentrations, followed by montmo-

rillonite, pyrophyllite and kaolinite (Kato et al., 1966). Low octahedral substitution

is one reason for the observed low dissolution rate of pyrophyllite compared with

montmorillonite. The other reason is the presence of collapsed, non-swelling inter-

layers in pyrophyllite.

A series of reduced-charge montmorillonites is prepared via Li

ﬁxation at ele-

vated temperatures (Hofmann–Klemen effect) to explore how the expandability of

the interlayer spaces inﬂuences the extent of dissolution. As the negative layer charge

of these samples decreases substantially, pyrophyllite-like features appear in both IR

(Madejova

et al., 1996) and

Si MAS NMR (Gates et al., 2000 ) spectra. The in-

creased content of non-swelling interlayer spaces is conﬁrmed by both XRD and

HRTEM images (Komadel et al., 1996a). The dissolution of reduced-charge mont-

morillonites in HCl indicates that pyrophyllite -like layers surrounded by non-swell-

ing interlayer spaces dissolve much more slowly than montm orillonite layers of

similar chemical composition located between swelling interlayer spaces (Figure 7.1.2).

This clearly shows that protons attack the layers not only from the particle edges but

also from the swollen interlayer spaces. Non-swelling illite and kaolinite are also

more resistant to HCl attack than montmorillonite or vermiculite (Jozefaciuk and

Bowanko, 2002).

IR spectroscopy in both middle and near regions was applied for the investigation

of acid dissolution of reduced-charge montmorillonites. Reduction of the layer

Table 7.1.1. Effects of smectite type, acid concentration and temperature on half time of

dissolution in 0.2L HC1/g smectite in closed systems

Smectite HC1 (M) T(1C) t

0.5

(h)

Effect of smectite type

Nontronite 6.0 95 0.16

Mg-rich montmorillonite 6.0 95 6.2

Al-rich montmorillonite 6.0 95 8.0

Effect of acid concentration

Hectorite 0.25 20 4.6

Hectorite 0.50 20 2.6

Hectorite 1.00 20 1.7

Effect of temperature

-beidellite 6.0 50 12.0

-beidellite 6.0 60 6.0

7.1.3. Acid Dissolution of Smectites 269

charge connected with development of non-swelling interlayer spaces substantially

decreases the extent of the dissolution of SAz-1 montmorillonite in HCl. New bands

at 3744 and 7314 cm

–1

are assigned to the vibrations of Si–OH groups formed by acid

treatment. These bands are indicative of the extent of acid attack, even when no

differences are observed in the 1300–400 cm

–1

spectral region, commonly used to

follow this process. Both the IR spectra and solut ion analysis reveal that non-swell-

ing interlayers develop on heating, causing a marked reduction in dissolution rate.

The results obtained conﬁ rm that acid attack of the smectite structure occurs at both

interlayer surfaces and edges. However, if the accessibility of the layers to protons is

low, due to non-swelling interlayer spaces, the dissolution is reduced and takes place

mainly from the particle edges (Pa

lkova

et al., 2003).

An in situ observation by AFM shows that the dissolution of hectorite and non-

tronite particles in acid solutions occurs inward from the edges; the basal surfaces

appear to be unreactive during the timescale of the experiments. The hectorite (0 1 0)

faces appear to dissolve about six times more slowly than the lath ends, usually

involving ‘‘broken’’ bonds at the edge surfaces. The edges visibly dissolve on all

sides, and appear to roughen somewhat. On the other hand, the (0 1 0), (1 1 0), and

1 0) faces on nontronite particles are exceptionally stable. Thus, dissolution fronts

originating at broken edges or defects would quickly become ﬁxed along these faces,

after which no more dissolution is observable. These observations can be explained

10% S

100% S

10% S

1.0

0.8

0.6

0.4

0.2

0.0

20 30

Undissolved portion

Dissolution time (h)

1300 1100 900 700 500

419

pyrophylite

1121

Absorbance

Wavenumbers (cm

-1

)

Fig. 7.1.2. Ca

-saturated 100% smectite and a reduced-charge smectite with about 10%

swelling interlayers. Left: dissolution of Al

in 6 M HCl at 95 1C; Right: pyrophyllite-like

features in the IR spectra. From Komadel et al. (1996a).

Chapter 7.1: Acid Activation of Clay Minerals270