GeolCarp_Vol53_No2_87

GEOLOGICA CARPATHICA, 53, 2, BRATISLAVA, APRIL 2002

87 — 92

SMECTITE REACTIVITY IN ALKALINE SOLUTIONS

SUSANA RAMIREZ

1,2*

, JAIME CUEVAS

, SABINE PETIT

, DOMINIQUE RIGHI

and ALAIN MEUNIER

Université de Poitiers, CNRS UMR 6532 HydrASA, 40, avenue du Recteur Pineau,

F-86022 Poitiers Cedex, France

Universidad Autónoma de Madrid, Facultad de Ciencias, Departamento de Química Agrícola,

Geología y Geoquímica, Campus Cantoblanco, E-28049 Madrid, Spain

(Manuscript received October 4, 2001; accepted in revised form December 13, 2001)

Abstract: Bentonite and alkaline solutions were reacted in batch experiments at 90 °C during 365 days to determine the
effect of the chemical composition and pH of the solution in stability of the smectite. Bentonite and concrete are consid-
ered for use as backfilling and sealing materials in a deep geological repository of high level radioactive waste. The pH
of the pore-water leached from concrete during degradation is above 11 for a long period of time. In these alkaline
conditions, reactions take place in smectite, affecting the properties of bentonite. The aim of this work is to identify
crystalchemical changes in smectite resulting from its interaction with synthetic cement pore water, taking into account
the temperature effect in order to consider the thermal impact originated by radioactive decay. The bentonite used comes
from the “La Serrata de Níjar” deposit (Almería, Spain). The bentonite comprised mainly high-charge montmorillonite.
Alkaline solutions are representative of cement pore water leached from concrete during the first stages of degradation.
Crystalchemistry characterization of smectite (< 0.5

m fraction of bentonite) was performed before and after the ex-

periments by differential thermal analysis, X ray diffraction (Hofmann & Klemen test, simplified alkylammonium method),
infrared spectroscopy and chemical analysis. The results suggest a preferential dissolution of some smectite layers, more
specifically octahedrally-charged layers.

Key words: radioactive waste repository, hydrothermal stability, alkaline solutions, smectite, bentonite.

Introduction

Bentonite  and  cement  will  be  used  as  engineered  barriers  in
the final disposal for the high level radioactive waste in crys-
talline rock. The cement’s degradation produces alkaline solu-
tions,  which  affect  the  smectite  crystalchemistry  and,  there-
fore, the properties of the bentonite. Because of the main role
of bentonite in the safety of the system, the effect of the alka-
line  conditions  on  the  smectite  stability  has  been  studied  for
several years. Some works focussed on the collapse of expand-
able smectite layers and, more specifically, on the formation of
illite or illite/smectite mixed-layers. Eberl et al. (1993) using
Wyoming bentonite observed the increase of the proportion of
illite layers up to 25 % after reacting for 9 months at 35 °C in
potassium  hydroxide  solution  (KOH  3  M).  Likewise,  a  ran-
dom mixed-layer containing non-expandable layers was also
obtained using a sodium hydroxide solution (NaOH 0.5 M) at
35 °C.  In  general,  the  formation  of  non-expandable  layers,
both in potassic and sodic solutions, depends on solution con-
centration  rather  than  on  temperature  and  reaction  duration.
According  to  Bauer  &  Berger  (1998)  and  Bauer  &  Velde
(1999), the smectite reacts with a potassic alkaline solution to
form an illite/smectite mixed-layer. The formation of mixed-
layer phases was seen as an intermediate step in a series of dis-
solution-precipitation processes. These authors propose three
steps in the alteration of Ceca and Ibeco smectites in potassi-
um hydroxide solution (KOH up to 4 M). Initially, there is dis-
solution of tetrahedral and octahedral sheets in smectite. Fur-
ther, the illite content increases up to 40 % at 35 °C and up to

90 % at 80 °C. Finally, new phases precipitate in the following
sequence: discrete mica, KI-zeolite (Barrer 1982), phillipsite,
K-feldspar and quartz. Recently, Rassineux et al. (2001) stud-
ied Wyoming bentonite reaction with pH 13.5 solutions at 35
and 60 °C up to 730 days. In spite of an apparent invariability
of the chemical composition and crystal structure of the smec-
tite layer, as determined by IR spectroscopy, a slight increase
of the expandability after octahedral charge neutralization was
observed. To explain this fact, the authors proposed the re-or-
ganization of the layer stacking during the reaction, so that the
number of charged tetrahedral sheets, adjacent to the same in-
terlayer space, increases.

According to these results it seems necessary to investigate

in detail the chemical and crystal structure of smectites which
have experienced the alkaline reaction. The objective of the
present work is to study the crystalchemistry of the smectite
in the Spanish bentonite after alteration in alkaline condi-
tions at 90 °C in order to consider the effects of (1) solutions
leached out from cement and (2) heat generated by the radio-
active decay.

Materials and methods

The bentonite used was extracted from the “Cortijo de

Archidona” deposit (“La Serrata de Níjar”, Almería, Spain),
which is the Spanish reference material for the high-level ra-
dioactive waste disposal. Powder X-ray diffraction (XRD) pat-
terns indicate that the bentonite is composed of 93 ± 3 % smec-

susana.ramirez-martin@hydrasa.univ-poitiers.fr

MECC ‘01

88 RAMIREZ, CUEVAS, PETIT, RIGHI and MEUNIER

tite, 2 ± 0.5 % quartz, 3 ± 1 % feldspars, 2 ± 0.2 % cristobalite,
1 ± 0.7 % calcite and 1.5 ± 0.1 % rhyodacitic volcanic glass
(average values from Cobeña et al. 1998; Linares et al. 1993).

Characterization by XRD was carried out using oriented

saturated and ethylene-glycol solvated samples. XRD

data were obtained with a PHILIPS 1729 diffractometer with a
Fe filtered Co tube operated at 40 mA and 40 kV. The diffrac-
tometer is associated with a Socabim DACO-MP recorder sys-
tem controlled by the DiffracAt software.

The IR spectra were obtained using a Nicolet 510 Fourier

transform IR spectrometer (FTIR) equipped with the OMNIC
software to measure the integrated intensity of the absorption
bands. Spectra were recorded in the 4000—400 cm

—1

range with

a resolution of 4 cm

—1

. Measure was made from dried KBr pel-

lets, which were prepared by mixing and pressing 1 mg of
sample with 150 mg KBr.

Major elements were determined by X-ray fluorescence

(XRF) using a PHILIPS PW-1404 X-ray Spectrometer with a
Sc/Mo tube and operated at 30 kV and 80 mA.

The DTA analyses were performed on 20 mg sample, using

a Netzsch STA 409 EP at a heating rate of 10 °C/min from 20 °C
to 1100 °C. Samples were loaded into platinum crucibles and
run in air atmosphere.

Experimental

Bentonite was altered in batch experiments carried out in

tightly-closed Teflon reactors during 365 days at 90 °C. The
solid/solution ratio was 1/3: 80 g of < 1 mm sieved bentonite
and 240 ml solution. These experiments are part of a series de-
tailed in Vigil et al. (2001) and Ramírez et al. (2001).

Alkaline solutions are representative of the cement pore-wa-

ter leached out during the first stages of concrete degradation.
According to Anderson et al. (1989), the pH of the cement
pore-water varies from 12.4 to 13.5, Na

and K

being the

main cations in the solution. The Na

and K

concentrations

range between 0.05 and 0.3 mol/dm

while the Ca

concentra-

tion is below 2.5

—3

mol/dm

. The chemical compositions

of alkaline working solutions used in the experiments are
shown in Table 1. Alkaline solutions were standardized by ti-
tration with a potassic phtalate solution (stock solutions) or
with a sulphuric acid solution (working solutions).

At the end of the reaction time, the reactor was cooled to

room temperature. Then, control of reactors weight was per-
formed. No significant weight loss was observed. The solids
were separated from the solution by centrifugation. The solids
were dried in a glove-box, in CO

free atmosphere and sieved

to < 1 mm. Mineralogy of bulk samples is reported in Ramírez
et al. (2001) and Vigil et al. (2001). According to semi-quanti-
tative XRD analyses and SEM observations, the main mineral-
ogical changes after the sodic-potassic alteration tests are the
formation of zeolite and dissolution of poorly crystallized min-
erals, especially feldspars and volcanic glass. Under SEM, no
dissolution features were observed on calcite crystals. To char-
acterize the crystalchemistry in both initial and treated sam-
ples, the < 0.5

m fraction was separated by centrifugation

and Ca

saturated. This fraction is characterized as nearly a

pure smectite by XRD and FTIR analyses. The structural for-

mulas were calculated from chemical analyses. For evaluating
amount and localization of the layer charge, the following
methods were used: (1) the simplified alkylammonium proce-
dure of Olis et al. (1990) using XRD and (2) cation exchange
capacity (CEC) evaluation using IR spectroscopy from NH

saturated samples (Petit et al. 1998, 1999). Evaluation of layer
charge was made before and after using the Hofmann & Kle-
men treatment (Hofmann & Klemen 1950), so octahedral
charge was evaluated separately. The lithium saturation was
performed on calcium saturated samples by one week dialysis
against LiCl 1 M solutions. In order to assure complete satura-
tion with Li

, LiCl dialysis solutions were renewed every 48

hours. Then, the samples were heated at 250 °C in Pt crucible
for 12 hours.

Results

XRD patterns of Ca

-saturated and ethylene-glycol solvat-

ed samples exhibit the basal reflections of a smectite with two
ethylene glycol layers in the interlayer space (Fig. 1): intense
peak at 16.97

—1

nm and higher order peaks at 8.39—8.50

—1

nm, 5.54—5.57

—1

nm and 3.34—3.35

—1

nm. The pres-

ence of small amounts of 10

—1

nm collapsed layers in a

random mixed-layer, in both initial and treated samples, was
deduced from the irrationality of the diffraction patterns and
from the broad d(001) reflection with a high low-angle shoul-
der (Moore & Reynolds 1989). The percentage of illite like
layers, interstratified with smectite layers, was quantified us-
ing the procedure described by Inoue et al. (1989). According

Table 1: Chemical composition of initial solutions.

solution

(mol/dm

)

(mol/dm

)

(mol/dm

)

223

× 10

–4

12.6

0.08

0.17

8.52

× 10

–4

13.2

0.17

0.33

4.07

× 10

–4

13.5

Fig. 1. XRD patterns of samples before and after the alteration ex-
periment. Basal reflection (in nm), saddle/001 peak intensity ratio
and percentage of illite like layers randomly interstratified with
the smectite layer.

sample

test

initial

saddle/001 peak ratio

0.17

0.27

0.28

0.17

Illite like layer (% )

test

initial

d 005

0.335
0.334
0.334

0.334

d 002

0.846
0.849
0.850

0.839

d 003

0.557
0.555
0.554

0.556

d 001

1.697

SMECTITE REACTIVITY IN ALKALINE SOLUTIONS 89

to this method, the saddle/001 peak intensity ratio is correlated
with the proportion of illite in a random mixed-layer I/S. The
calibration curve for the saddle/001 peak intensity ratio versus
the percentage of 10

—1

nm collapsed layers was constructed

from calculated XRD patterns using NEWMOD© (Reynolds
1985). The average (

) and maximum crystallite sizes (N

max

)

are defined by the parameters

= 4 and N

max

= 10, respective-

ly. These values were selected because they produce the best
fit with experimental XRD patterns. In the initial sample, 10 %
of 10

—1

nm collapsed layers are found interstratified with

90 % of 17

—1

nm expanded layers. After alteration by alka-

line treatments only slight changes in the saddle/001 peak in-
tensity ratio were observed, and attributed to a decrease of
crystallite size rather than change in the percentage of inter-
stratified illite-like layers.

After neutralization of the octahedral charge (Hofmann &

Klemen treatment), expandability decreases for all samples
(Fig. 2). However, a broad and poorly defined peak near 17.30

—1

nm indicates the presence of expandable layers and, con-

sequently, layers having a tetrahedral charge. The sample treat-
ed with the calcic solution (a experiment) shows a diffracto-
gram similar to that of the initial sample. The increase of the
17.30

—1

nm reflection intensity, as well as the shift of

d(002) reflections to higher angles, are evident after the sodic-
potassic treatment (b and c experiments). The d(002) reflection
position reveals a mixed-layer with expanded layers (8.57

—1

nm, second order) and collapsed layers (9.6

—1

nm).

That is indicative of the increase in the proportion of tetrahe-
drally-charged  expanded  layers.  From  simulated  diffracto-
grams  obtained  with  NEWMOD©,  about  40  %  of  expanded
layers  are  present  in  the  initial  sample  after  neutralization  of
the octahedral charge whereas about 50 % were found for the
treated samples in b and c experiments.

The XRD patterns of the initial and treated samples after in-

tercalation with C

are shown in Fig. 3A. Alkylammo-

nium ions in the interlayer space of the clay may adopt either a
monolayer (13.6

—1

nm), a double-layer (17.7

—1

nm) or a

pseudotrimolecular layer (21.7

—1

nm) complex depending

on the magnitude of the layer charge (Olis et al. 1990). The

d(001) position points out the formation of an alkylammoni-
um double-layer in both the initial and altered samples from a
experiment. A slight shift of peak position toward lower an-
gles is shown in the samples treated with sodic-potassic solu-
tion (b and c experiments) up to 17.97

—1

nm and

17.83

—1

nm, respectively. This slight shift would be in-

duced by a decrease of the coherent scattering domain size
(Moore & Reynolds 1989). Actually, delamination of smectite
clay particles is often observed when clays are treated with al-
kaline and sodic solutions (Lagaly 1981). For the initial sam-
ple, and after neutralization of octahedral charge (Hofmann &
Klemen treatment; Fig. 3B), the position of the d(001) reflec-
tion is slightly lower than 13.6

—1

nm. This would indicate

an interstratification between collapsed layers (9.6

—1

nm)

and monolayer alkylammonium complex (13.6

—1

nm).

Collapsed layers are assumed to have no tetrahedral charge,
whereas layers which accept intercalation of alkylammonium
cations must have a minimal tetrahedral charge. The position
of the d(001) reflection in the treated samples was not signifi-
cantly affected, which would indicate that no significant
change in the amount of tetrahedral charge was induced by the
treatments.

Recently, Petit et al. (1998, 1999) have proposed a method

for evaluation of the layer charge in NH

-clays by IR spectros-

Fig. 3. XRD patterns of the initial and treated samples after inter-
calation with the alkylammonium cation (nC = 12) before (A) and
after (B) the Hofmann & Klemen treatment.

Fig. 2. XRD patterns of the initial and treated samples after the
Hofmann & Klemen treatment.

a test

b test

test

initial

2θ

0.932 nm

0.938 nm

0.909 nm

0.913 nm

1.730 nm

test

initial

2θ

1.773 nm

1.777 nm

1.783 nm
1.797 nm

test

initial

2θ

1.334 nm

1.335 nm

1.344 nm

1.360 nm

90 RAMIREZ, CUEVAS, PETIT, RIGHI and MEUNIER

copy. In order to compare the samples on the same quantitative
basis, normalization of the spectra was performed using the Si-
O band at 1000 cm

—1

as an internal reference band. The

band  in  the  IR  spectra  of  the  initial  and  treated  samples  is
shown in Fig. 4. Measurements were made before and after the
Hofmann  &  Klemen  treatment  and,  consequently,  the  cation
exchange capacity (CEC) obtained is assumed to result from
permanent plus variable charges and tetrahedral plus variable
charges,  respectively.  The  difference  between  the  two  CEC
values corresponds to the octahedral charge (Table 2). In the
sample treated with calcic solution (a experiment) the percent-

age of octahedral charge is similar to the initial one, while af-
ter sodic-potassic solution treatments (b and c experiments)
the octahedral charge decreases from 98 % to 77 % of the total
charge.

Structural formulas (Table 3) were calculated from the

chemical analysis of the Ca

saturated < 0.5

m fraction. Ac-

cording to XRD data, the chemical analysis does not corre-
spond to a single homogeneous phase, but to heterogeneously
charged smectite layers. Nevertheless, the obtained data were
used to compare cationic distribution between the treated and
initial samples. After the experiment, the increase in Mg

as-

signed to the octahedral sheet is evidenced. The amount of oc-
tahedral Mg

cation increases with initial pH of experimental

solutions. After a experiment (calcic solution at pH 12.6), the
Mg

content is 0.92 per unit cell, close to the initial value,

while in sodic-potassic solution at pH 13.5 (c experiment) it
rises up to 1.04 per unit cell. Simultaneously, the amount of
octahedral Al

decreases from 2.68 in the initial sample to

2.55 in the c treated sample.

Fig. 5. Differential thermal analyses (DTA) of samples before and
after the alteration experiment.

Table 2: Normalized integrated intensity of the

band

(S NH

) in the IR spectra before and after the Hofmann & Klemen

treatment (HK) and octahedral charge associated in the initial and
treated samples.

S NH

(arbitrary units)

before HK

after HK

octahedral

charge (%)

initial

8.26

0.17

a experiment

7.12

0.23

b experiment

6.21

1.28

c experiment

4.67

1.06

Table 3: Structural formula before and after alteration experiment
calculated from the chemical analysis of the calcium saturated
< 0.5

m fraction.

structural formula (O

(OH)

anion basis)

initial

(Si

7.72

0.28

)

(Al

2.68

0.39

0.85

0.01

)

0.62

0.09

a experiment

(Si

7.77

0.23

)

(Al

2.68

0.38

0.92

0.01

)

0.55

0.07

b experiment

(Si

7.71

0.29

)

(Al

2.67

0.38

0.99

0.02

)

0.50

0.08

c experiment

(Si

7.75

0.25

)

(Al

2.55

0.40

1.04

0.01

)

0.55

0.18

Fig. 4.

band in the IR spectra of the initial and treated sam-

ples before (A) and after (B) the Hofmann & Klemen treatment.

The DTA patterns of the samples are shown in Fig. 5. At

low temperatures, the curves show a strong endothermic peak
at nearly 100 °C corresponding to the loss of pore water and
another endothermic peak at about 170 °C due to the elimina-
tion of interlayer water. The dehydroxylation peak at about
630 °C is characteristic of dioctahedral smectites (Mackenzie
1970). In addition, another endothermic peak is observed at
high temperature. As the initial pH of the experiment increas-
es, the temperature of this peak decreases: 859 °C in the initial
sample and 851, 847 and 837 °C after the 12.6, 13.2 and 13.5

1500

1450

1400

1350

0.4

0.5

0.6

0.7

initial

test

wave number (cm

-1

)

absor

ban

1440

1420

1400

1380

1360

0.34

0.36

0.38

0.40

0.42

test

initial

test

absor

ban

wave number (cm

-1

)

200

400

600

800

-16

-14

-12

-10

-8

-6

-4

-2

837 ºC

851 ºC

847 ºC

859 ºC

630 ºC

170 ºC

95 ºC

initial

a test

b test

c test

(

T (ºC)

SMECTITE REACTIVITY IN ALKALINE SOLUTIONS 91

Fig. 6. OH stretching region in the IR spectra of the initial and treat-
ed samples before (A) and after (B) the Hofmann & Klemen treat-
ment.

pH experiment, respectively. This peak is correlated with the
breakdown  of  the  clay-lattice  and  formation  of  amorphous
phases  from  the  dioctahedral  smectites  (Grim  &  Brandley
1940;  Mackenzie  1970).  The  decrease  of  the  temperature  of
lattice breakdown suggests that the smectite structure has been
weakened by the treatments.

The IR spectra of the treated samples do not show any sig-

nificant structural alteration with regard to the initial sample.
The OH stretching region before and after the Hofmann &
Klemen procedure is shown in Fig. 6. The band at 3630 cm

—1

attributed to the vibration of Al

OH in octahedral sheets is ob-

served. After Hofmann & Klemen treatment, the shift of this
band up to 3635 cm

—1

and the appearance of a vibration attrib-

uted to AlMgLiOH band at 3670 cm

—1

evidences the migration

of Li

from the interlayer into the previously vacant octahedral

positions (Madejová et al. 2000).

Discussion and conclusions

Changes in the chemical composition and crystalchemistry

characterization indicate contrasted behaviour according to the

chemistry of altering solutions. No changes were observed for
the sample treated with solution a. Consequently, the Spanish
smectite is considered to be stable in Ca(OH)

saturated solu-

tion at 90 °C. For samples treated with solution b and c, a rela-
tive increase of magnesium with reference to aluminum in the
octahedral sheet is observed. Milodowski et al. (1990) found
the same results after reaction of bentonite (Wyoming and Cal-
cigel) and cement pastes at ambient temperature. In order to
explain  this  fact,  these  authors  considered  the  preferential
leaching of aluminum. In the present experiment, aluminum is
actually released to the solution during smectite alteration and
consumed  for  crystallization  of  zeolite  (Ramírez  et  al.  2001;
Vigil  et  al.  2001),  of  the  average  chemical  composition
(Na

2.47±1.61

0.58±0.28

1.88±0.48

0.25±0.10

)(Al

5.06±0.17

0.07±0.03

0.05±0.02

)

10.24±0.36

·12 H

Changes observed on XRD patterns of ethylene glycol sol-

vated and alkylammonium saturated samples were attributed
to a reduction of the coherent scattering domain size. Preferen-
tial dissolution of some smectite layers would contribute to
that reduction. Actually, susceptibility to alteration may vary
according to the crystalchemistry of smectite layers. Righi et
al. (1998) have demonstrated that montmorillonitic smectites
are more susceptible to weathering in soils with alkaline pH
(

≈

9) than beidellitic smectites.
Changes in CEC measured by IR spectroscopy are not fully

understood.  According  to  the  results,  total  CEC  decreases
from the initial sample to the a, b and c treated samples. Prob-
ably the alkaline treatments have dramatically altered the sur-
face of clay particles and consequently, it is likely that large
changes  in  the  amount  of  variable  charge  have  occurred.  In
such a situation it is difficult to evaluate accurately the eventu-
al alteration of the permanent charge. If we assume that Hof-
mann & Klemen treatment does not alter the proportion of per-
manent  and  variable  charges,  the  difference  between  CEC
before and after the Hofmann & Klemen treatment allows an
evaluation of the octahedral charge. With reference to the total
charge,  relative  amount  of  octahedral  charge  decreases  with
the b and c treatments. This would suggest that the more octa-
hedrally-charged layers are preferentially altered by the treat-
ments.

The transformation of smectite to illite-like layers has been

frequently proposed as an alteration process in an alkaline me-
dium (Eberl et al. 1993; Bauer & Berger 1998; Bauer & Velde
1999).  The  illitization  process  is  generally  described  by  the
formation of high tetrahedrally-charged layers and, according
to Cuadros & Altaner (1998), is associated with the increase of
aluminum  and  decrease  of  magnesium  in  the  octahedral
sheets.  This  chemical  evolution  was  not  observed  in  the
present  work.  Actually,  IR  spectroscopy  does  not  show  any
structural transformation of the reacted smectite. Furthermore,
the  results  are  better  explained  by  preferential  dissolution  of
some smectite layers, more specifically octahedrally charged
layers.

Acknowledgments: The experimental tests were carrying out
in the laboratories of the Autonomous University of Madrid
(Spain) while the main analyses were done with the UMR
CNRS 6532 (Poitiers, France) facilities. The research was sup-
ported by the European Commission thanks to the Project No.

3750

3700

3650

3600

3550

3500

3450

0,4

0,5

0,6

0,7

0,8

initial

test

3630 cm

-1

wave number (cm

-1

)

3800

3750

3700

3650

3600

3550

3500

0,3

0,4

0,5

test

initial

test

3670 cm

-1

3635 cm

-1

wave number (cm

-1

)

92 RAMIREZ, CUEVAS, PETIT, RIGHI and MEUNIER

FI4W-CT96-0032 and to the Marie Curie individual fellow-
ship No. FIKS-CT-2000-50510. The authors thank R. Dohr-
mann and an anonymous reviewer for their valuable and con-
structive comments which improved this paper.

References

Andersson K., Allard B., Bengtsson M. & Magnusson B. 1989:

Chemical composition of cement pore solution. Cem. Concr.
Res. 19, 327—332.

Barrer R.M. 1982: Hydrothermal chemistry of zeolites. Academic

Press ed., New York, 1—360.

Bauer A. & Berger G. 1998: Kaolinite and smectite dissolution rate

in high molar KOH solutions at 35 °C and 80 °C. Appl.
Geochem. 33, 905—916.

Bauer A. & Velde B. 1999: Smectite transformation in high molar

KOH solutions. Clay Miner. 34, 259—273.

Cobeña J.C., Martín M., Ramírez S., Vigil de la Villa R., Cuevas J. &

Leguey  S.  1998:  Estudio  detallado  de  la  mineralogía  y  microe-
structura  de  una  bentonita  compactada  sometida  a  procesos  de
calentamiento e hidratación. Bol. Soc. Esp. Miner.  21—A, 66—67.

Cuadros J. & Altaner S.P. 1998: Characterization of mixed-layer il-

lite/smectite from bentonites using microscopic, chemical and
X-ray methods: constraints on the smectite-to-illite transforma-
tion mechanism. Amer. Mineralogist 83, 762—774.

Eberl D.D., Velde B. & McCormick T. 1993: Synthesis of illite-

smectite from smectite at earth surface temperatures and high
pH. Clay Miner. 28, 49—60.

Grim R.E. & Bradley W.F. 1940: Investigation of the effect of heat

on the clay minerals illite and montmorillonite. J. Am. Ceram.
Soc. 23, 242—248.

Hofmann U. & Klemen E. 1950: Lost of exchangeability of lithium

ions in bentonite on heating. A. Anorg. Allg. Chem. 262, 95—99.

Inoue A., Bouchet A., Velde B. & Meunier A. 1989: A convenient

technique for estimating smectite layer percentage in randomly
interstratified illite/smectite minerals. Clays and Clay Miner.
37, 227—234.

Lagaly G. 1981: Characterization of clays by organic compounds.

Clay Miner. 16, 1—21.

Linares J., Huertas J., Reyes E., Caballero E., Barahona E., Guardio-

la J.L., Yáñez J., Romero E., Delgado A., Rodríguez J. & Mar-
tín-Vivaldi M.T. 1993: Investigación de bentonitas como

materiales de sellado para almacenamiento de residuos radi-
activos de alta actividad. ENRESA. Pub. Téc. 01/93.

Mackenzie R.C. 1970: Simple phyllosilicates based on gibbsite- and

brucite-like sheets. In: Mackenzie R.C. (Ed.): Differential ther-
mal analysis. Volume 1: Fundamental Aspects. Academic
Press, London and New York, 497—537.

Madejová J., Bujdák J., Petit S. & Komadel P. 2000: Effects of

chemical composition and temperature of heating on the infra-
red spectra of Li-saturated dioctahedral smectites. (I) Mid-in-
frared region. Clay Miner. 35, 739—751.

Milodowski A.E., Hughes C.R., Kemp S.J. & Pearce J.M. 1990:

Characterisation of the bentonite alteration in reacted cement-
bentonite blocks from swelling-test experiments. British Geo-
logical Survey. Technical Report WG/90/39C.

Moore D.M. & Reynolds R.C. 1989: X-ray diffraction and the iden-

tification and analysis of clay minerals. Oxford University
Press, New York, 1—332.

Olis A.C., Malla P.B. & Douglas L.A. 1990: The rapid estimation of

the layer charges of 2:1 expanding clays from a single alkylam-
monium ion expansion. Clay Miner. 25, 39—50.

Petit S., Righi D., Madejová J. & Decarreau A. 1998: Layer charge

estimation of smectites using infrared spectroscopy. Clay Min-
er. 33, 579—591.

Petit S., Righi D., Madejová J. & Decarreau A. 1999: Interpretation

of the infrared NH

spectrum of the NH

-clays: application to

the evaluation of the layer charge. Clay Miner. 34, 543—549.

Ramírez S., Cuevas J., Vigil R. & Leguey S. 2001: Hydrothermal al-

teration of “La Serrata” bentonite (Almería, Spain) by alkaline
solutions. Appl. Clay Sci. Accepted.

Rassineux F., Griffault L., Meunier A., Berger G., Petit S., Vieil-

lard  P.,  Zellagui  R.  &  Munoz  M.  2001:  Expandibility-layer
stacking  relation  during  experimental  alteration  of  a  Wyo-
ming  bentonite  in  pH  13.5  solutions  at  35  and  60

C. Clay

Miner. 36, 197—210.

calculation of one-dimensional diffraction patterns of mixed-
layered clays. R.C. Reynolds, Jr., 8 Brook Dr., Hanover New
Hampshire.

Righi D., Terribile F. & Petit S. 1998: Pedogenic formation of high

beidellite in a vertisol of Sardinia (Italy). Clays and Clay Min-
er. 46, 167—177.

Vigil R., Cuevas J., Ramírez S. & Leguey S. 2001: Zeolite forma-

tion during the alkaline reaction of bentonite. Eur. J. Mineral.
13, 635—644.