my?

 

 

 

 

 

 

.. $..n.ul.;.p.
Em? _ .

o—a

 

3; E5

"+5313

MINIMUM")IIIIUHHHIWIHUHUlHlflllllHllUNl

301766 7043

 

LIBRARY

Mlchlgan State
University

 

 

 

This is to certify that the

othesis entitled ~
HETEROSTRUCTURED CLAYS WITH
REGULARLY ALTERNATING INTERLAYERS OF
ORGANIC AND INORGANIC EXCHANGE CATIONS

presented by

Nouter Laurens IJdo

has been accepted towards fulfillment
of the requirements for

Ph . D 0 degree in Chemistry

%flii flaw—flu"
T omas Pinnavaia

Major professor

 

Date December 16, 1998

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINE return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1M clam-9.14

 

HETEROSTRUCTURED CLAYS
WITH
REGULARLY ALTERNATIN G INTERLAYERS
OF
ORGANIC AND INORGANIC EXCHANGE CATIONS

By

Wouter Laurens IJdo

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1998

ABSTRACT

HETEROSTRUCTURED CLAYS
WITH
REGULARLY ALTERNATIN G INTERLAYERS
OF
ORGANIC AND INORGANIC EXCHANGE CATIONS

By

Wouter Laurens Udo

Few investigations have been reported on mixed organic and inorganic exchanged
forms of smectite clays. Such mixed ion clay systems may have interesting materials
properties since they can potentially combine hydrophilic- and hydrophobic- clay
interlayers into a single phase material. This work contains the first examples of the
synthesis of mixed ion heterostructured fluorohectorite clays that have regularly
alternating organic and inorganic exchanged interlayers. The potential of such an
interlayer arrangement becomes apparent when (hydrophobic) organic modified clay
galleries need to be dispersed in water.

Three complementary strategies to the synthesis of mixed ion heterostructured
clays are described. Pathway I involves the partial replacement of hydrated metal
exchange ions by the direct addition of onium ions to a clay suspension. Route 11
embodies the reverse approach where a homoionic organic exchanged clay is treated with

a concentrated NaCl solution to produce an organic and inorganic mixed ion

heterostructured clay product. Equal molar quantities of the two homoionic end member
clays are reacted in water to produce a mixed ion heterostructured clay in approach H1.

The mixed ion heterostructure formation process was probed with two series of
onium ions in an attempt to elucidate this spontaneous segregation behavior of the
intercalates. The onium ions used were of the type CnH(2n+1)N(CmI-I(2m+l ))3+ in the series
(n=-4-22 & m=4) and (n=16 & m=1-5). The onium ion alkyl tail length determines the
extent of metal ion replacement in a partial ion exchange reaction. Quantitative onium ion
loading is obtained when the onium ion alkyl tail exceeds 10 carbon units. Segregation of
cationic species into distinct alternating galleries is directed by the onium ion head group.

Large onium ion head groups (propyl- or larger, m 2 3) move the center of onium
charge away from the silicate layers and preclude commingling of cationic species within
one interlayer. This forces the smaller hydrated metal ions to take positions on the other
side of the organic exchanged interlayers and results in heterostructure formation. The
smaller methyl- and ethyl- onium ion head groups, m=l or 2, effectively neutralize the
clay layer charge and do not produce heterostructured clays. Instead, they initiate the
formation of phase segregated homoionic parent end member clays.

However, heterostructure formation is not limited to mixed ion clay systems with
half a cation exchange equivalent of onium ions. Complete heterostructure conversion
also occurs for mixed ion clays that have an overall onium ion fraction between 0.35 and
0.50. Also, intercalation experiments established the relationship between clay layer
charge location and 1:1 C1¢5H33NBu3+ and Na+ mixed ion heterostructure formation

Finally, a model is proposed that describes the processes and requirements needed

when mixed ion heterostructured clays are synthesized.

To my parents, brother and sister.

iv

ACKNOWLEDGMENTS

I would like to thank my advisor, Prof. Thomas J. Pinnavaia for his support and
guidance over the years. I am very grateful for the help and encouragement he gave me
throughout my graduate studies. I am, however, most thankful for his generous advice on
so many non-chemistry related issues. His understanding and enlightening views have
shown me the way.

In addition, I was lucky to have such nice group members. I would especially like
to thank my friends Dr. Steve Bagshaw, Dr. Peter Tanev, Dr. Louis Mercier, Dr. Kathy
Severin, Dr. Jarrod Massam and Dr. Tarek Abdel-Fattah. Also, I like to thank my
chemistry buddy and friend Beth Gardner for the great times I have had in the lab over the
years.

I am grateful to Prof. E.F. Westrum Jr. who convinced me to pursue a graduate
career, to Prof. E.H.P. Cordfunke who taught me all the basics so well and to Prof. Pieter
Krijgsman who showed me what chemistry is all about.

I have to thank the entire Wijnands family, who have always encouraged and
supported me. I am grateful to my father, who introduced me to chemistry and showed me
the significance of it, and to my brother and sister, who have always believed in me.

My deepest gratitude goes to my mother, who taught me to look at the important
Commercial aspects of chemistry. She has always supported me and she understands me

so well. And yes, it was another adventure.

 

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................................... ix
LIST OF FIGURES .......................................................................................................... x
CHAPTER 1 INTRODUCTION 1
1.1 Clays and clay structure ........................................................................... l
1.2 Smectite clay interlayer properties and cation exchange ......................... 5
1.3 Organo clays ............................................................................................. 7
1.4 Natural clays ........................................................................................... 10
1.5 Synthetic clays ........................................................................................ 12
1.6.1 Clay applications, an overview .............................................................. 14
1.6.2 Polymer - clay nanocomposite ............................................................... 15
1.6.3 Liquid crystal - clay composites ............................................................. 19
1.6.4 Clay as adsorbent material ..................................................................... 20
1.6.5 Clays as catalysts .................................................................................... 22
1.7 Pillared clays .......................................................................................... 22
1.8 Mixed ion clays ...................................................................................... 24
1.9 Research objectives ................................................................................ 27
l . 10 References .............................................................................................. 3 1
CHAPTER 2 SYNTHESIS OF LAYERED SILICATE HETEROSTRUCTURES

2.1
2.2
2.3
2.4
2.5
2.6
2.7

WITH REGULARLY ALTERNATING ORGANIC AND
INORGANIC GALLERIES: PRECURSORS TO PILLARED

RECTORITE-LIKE INTERCALATES ................................................. 36
Abstract .................................................................................................. 36
Introduction ............................................................................................ 37
Mixed ion heterostructure formation ...................................................... 38
Synthesis of a pillared rectorite-like analog ........................................... 44
A provisional mixed ion heterostructure formation modal .................... 50
Experimental .......................................................................................... 50
References .............................................................................................. 54
vi

CHAPTER 3

3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9

CHAPTER 4

4.1
4.2
4.3
4.4

4.5

4.6
4.7

4.8
4.9

CHAPTER 5

5.1
5.2
5.3
5.4
5.5

5.6

CORRELATION BETWEEN ONIU M ION GEOMETRY AND
THE STAGING BEHAVIOR OF ORGANIC AND INORGANIC
GALLERY CATIONS IN LAYERED SILICATE

HETEROSTRUCTURES ...................................................................... 55
Abstract .................................................................................................. 55
Introduction ............................................................................................ 57
Experimental .......................................................................................... 60
Preferred mixed ion heterostructure formation pathway ........................ 62
Onium ion chain length effect on heterostructure formation ................. 63
Head group size effect ............................................................................ 72
A heterostructure formation model according to nanolayer tactoids ..... 75
Concluding remarks ............................................................................... 78
References .............................................................................................. 79
GALLERY STACKING ORDER OF FLUOROHECTORITES

INTERLAYERED WITH ORGANIC AND INORGANIC
EXCHANGE CATIONS: EVIDENCE FOR THE FORMATION

OF HETEROSTRUCTURED SOLID SOLUTIONS ............................ 81
Abstract .................................................................................................. 81
Introduction ............................................................................................ 83
Experimental .......................................................................................... 85
Heterostructure formation at a half an exchange equivalent of

CMHBNBu; ions, fQ = 0.5 ....................................................................... 87

Mixed ion clays formed when homoionic Na‘-FH is exchanged with
less than half a CEC equivalent of CwHBPBu,+ ions, 0.35 S fQ 5 0.590

Metal ion valency and 1:1 mixed ion heterostructure formation ......... 103
Preservation of the distinctive gallery properties associated with the

interlayers of mixed ion heterostructured materials ............................. 106
Concluding remarks ............................................................................. 110
References ............................................................................................ 1 12

THE RELATIONSHIP BETWEEN 2:1 CLAY LAYER CHARGE

LOCATION AND THE 1:1 CmeNBuz,+ AND Na‘ MIXED ION

HETEROSTRUCTURE FORMATION PROCESS ............................ 1 13
Abstract ................................................................................................ 113
Introduction .......................................................................................... 1 15
Experimental ........................................................................................ 119
Synthesis of a lithium-fluorohectorite, Lim-[Mg,_86Lim]Si,O,oF, ........ 123
Synthesis of fluorohectorite-like clays with some charge originating

in the tetrahedral sites ........................................................................... 130
Synthesis of a fluorosaponite clay, Nam-[Mg6](Si,Alm)OzoF, ............. 139

vii

5.7

5.8

5.9

CHAPTER 6

6.1
6.2
6.3
6.4
6.5

A materials application for a 1:1 CHI-1,,NBu,’ and Li+ mixed ion
heterostructured clay; the adsorption of 2.4-dichlorophenol from

water ..................................................................................................... 146
Concluding remarks on CmHnNBu; and N a+ mixed ion

heterostructures .................................................................................... 150
References ............................................................................................ 152
A MODEL FOR THE FORMATION OF LAYERED SILICATE

HETEROSTRUCTURES WITH REGULARLY ALTERNATING

ORGANIC AND INORGANIC CI‘ION GALLERIES ....................... 154
Abstract ................................................................................................ 154
Introduction .......................................................................................... 155
The mixed ion heterostructure formation model .................................. 158
Verification of the mixed ion heterostructured clay formation model. 159
Conclusions .......................................................................................... 160

viii

Table 1.1

Table 3.1

Table 3.2

Table 3.3

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 5.1

Table 5.2

LIST OF TABLES

Idealized structural formulas for some general 2:1 layer silicates. .......... 4

Comparison of X-ray basal spacings for Q+- and mixed ion Na+,
Q+- fluorohectorites, where Q+ is CnH(2n+1)NBu; ................................ 66

TGA determination of the exchange ion compositions for 1:1 Na+,
Qt fluorohectorites, where Q+ is CnH(2n+1)NBu; ................................ 67

Surfactant head group size and X-ray basal spacings ((1001) for
homoionic Q+ fluorohectorite (Q+-FH) and 1:1 mixed Na+, Q+
fluorohectorite heterostructures (Na+, Q*-FHHS), where Q+ is
C16H33NR3“ ........................................................................................... 74

Comparison of XRD data of the mixed ion products shown in
Figure 4.2. The heterostructured clays are formed when a Na+-FH
clay suspension is exchanged with Q+ onium ions, where Q" is
C16H33PBU3+ ........................................................................................... 92

TGA determination of the C16H33PBu3+ exchange ion compositions
(fQ) for mixed ion heterostructured solid solutions produced when
Na+-FH is exchanged with a x % CEC equivalent of Q+, where Q+ is
C15H33PB113+. .......................................................................................... 95

Elemental analyses (ICP) of the exchange ion compositions for Na+
and Q+ mixed ion heterostructured solid solutions produced when
Na+-FH is exchanged with Q", where Q+ is C16H33PBu3+. .................... 95

Comparison of the X-ray basal spacings for homoionic A+-FH clay
and A“, Q+ -FH mixed ion products, where Q+ is Cur,H33NBu;;+ and
AM is Li+, Na“, Ca2+, Ba“, Al3+ or Ce“. ............................................. 104

2QSi MAS-NMR data of the aluminum substituted fluorohectorite-
like clays shown in Figure 5.5 and 5.6. Both clays were produced by
the UP flux method from a 1000 °C melt. ........................................... 136

29Si MAS-NMR data of the fluorosaponite clay shown in Figure
5.9. ........................................................................................................ 143

ix

LIST OF FIGURES

Figure 1.1 The ideal smectite clay structure. The clay layer is build from an
octahedral sheet which is sandwiched between tetrahedral sheets.
Negative charge develops on the clay layer when ions of lower
valency are substituted into the clay structure. Electrical neutrality is
maintained by cationic interlayer MM species. ........................................ 2

Figure 1.2 Organo clay interlayer structures. Intercalated onium ions form a
monolayer (A) coverage at low clay charge densities or small alkyl
chain lengths. Onium ions are grouped in a bilayer (B) when there is
not enough space available to pack all the ions in a single layer
arrangement. A paraffin like onium ion packing (C) occurs at even
higher charge densities. The alkyl chains may have kinks through
gauche conformers in all interlayer structure types .................................. 8

Figure 1.3 Clay nanocomposites. Structure (A) is a conventional clay - polymer
composite. Phase (B) is an intercalated composite. The clay
interlayers are intercalated with polymer, but the basal spacing is
fixed and does not change with composition. Exfoliated composite
(C) has individual clay layers dispersed throughout the whole
polymer system. As such, the distance between single layers is
related to the clay loading ....................................................................... 17

Figure 1.4 Dispersions of heterostructured clay and organo clay in water. The
inorganic gallery of a mixed ion heterostructure (A) can swell with
water to exfoliation and create a dispersion of single organo clay
entities. In contrast, organo clay (B) can barely breakdown into
smaller particles ...................................................................................... 29

Figure 2.1 Three synthetic pathways to regularly ordered C16H33PBu3+lNa+-
fluorohectorite heterostructures (denoted C): Pathway I is the
quantitative ion exchange of homoionic Na+-FI-I (denoted A) with a
stoichiometric quantity of surfactant cations. Pathway II is the
exchange of organo cations in homoionic C16H33PBu3+-FH
(denoted B) with excess NaCl. Pathway III is the spontaneous
formation of the heterostructure by ion redistribution reaction of the
homoionic parents. ................................................................................. 39

I:igure 2.2 XRD patterns (Cu-KG) of fluorohectorite derivatives: (A)
homoionic Na+-FH, (B) homoionic C16H33PBu3+-FH and (C) the
mixed ion heterostructures C16H33PBu3+lNa+-FHHS prepared by

Figure 2.3

Figure 2.4

_ Figure 2.5

Figure 3.1

Figure 3.2

Figure 3.3

Figure 3.4

pathways I, II and III The weak shoulder at 40 A in pattern B is due
to the presence of residue heterostructure formed in the exchange

reaction of Li+-FH with [C15H33PBu3+]Br. .........................................

TGA results of Na+-FH (stripes), C16H33PBU3+-FH (solid) and

C16H33PBU3+INa+-FHHS (stripe-dot). .................................................

Selective ion exchange reactions of fluorohectorite heterostructures.

XRD patterns of heterostructures formed by the selective ion

exchange reactions illustrated in figure 2.4. ..........................................

Three possible structures formed when the exchange sites of
smectite clays are occupied with two different cationic species,
namely, alkylammonium ions and dehydrated sodium ions.
Structure (A) denotes the two phase segregated homoionic parent
end member materials. Product (B) is the mixed ion homostructure
with both species intercalated within one gallery, while (C) is the
mixed ion heterostructure with the two cations segregated into two
interlayers. The stacking pattern of the organic and inorganic

galleries may be regular or random. ......................................................

XRD patterns (Cu-Kc) of intercalates formed by reaction of a Na”:

FH suspension and 0.5 equivalents of [C,,H(2,,+1)NBu3]+ onium
ions. The surfactant alkyl chain length was varied in two-carbon
units over the range n=4-22 while the tri-n-butyl head group was

kept constant ..........................................................................................

TGA curves for the mixed intercalates depicted in figure 3.2 with
n=4-20. Three separate areas corresponding to surfactant content are
indicated. The weight loss below 200°C corresponds to the release
of gallery water. The decomposition of the C22 surfactant occurs in
stages and is, for simplicity, not shown. Decomposition of the
organic onium is indicated by the weight loss in the region 200 —

550 °C ....................................................................................................

Dependence of XRD basal spacings on the alkyl chain length n for
homoionic CnH(2n+1)NBu3+- fluorohectorites (circles) and mixed ion
N a+, CnH(2n+1)NBU3+-fluorohectorite heterostructures (squares). The

dashed line corresponds to the calculated basal spacings for mixed
ion homostructures in which the two cationic species co-occupy the

interlayer galleries. ................................................................................

xi

.42

.43

46

.47

.59

.64

.66

.70

Figure 3.5 XRD patterns (Cu-Kn) of materials obtained when a Na+-FH
suspension is treated with half an equivalent of
[C16H33N(C,,,H(2,,n.,1))3]+ surfactant. The head group size was
increased in one-carbon units from m=l-5 to form an expanding
series that included Me, Et, n-Pr, n-Bu and n-Pe groups. ...................... 73

Figure 3.6 Summary of the reaction pathway leading to heterostructured,
homostructured, and phase segregated mixed ion intercalates

formed by replacing 50% of the exchange cations in Nat
fluorohectorite with CnH(2n+1)N(CmH(2m+1))3+ cations (n 2 10; m =
1 - 5). ...................................................................................................... 77

Figure 4.1 Time dependence of the XRD patterns (Cu-Kg) of air dried products
formed when a Na+-FH clay suspension is exchanged with half a
cation exchange equivalent of C151I33NBu3+ ions. A 40 A mixed ion
heterostructure is already the dominant phase within 10 minutes.
Heterostructure formation is accompanied by the production of
C16H33NBu3+-FH organo clay (diffraction peak at 27 A marked *).
The unreacted Na+-FH exhibits a diffraction peak at 12.4 A marked
H. The two homoionic end member clays are fully converted to the
heterostructured clay within 24 hours reaction time. ............................. 89

Figure 4.2 XRD patterns (Cu-Kg) of mixed ion products formed when 50, 45,
40 and 35 percent of the exchange cations of Na+-FH clay
suspension are replaced by C16I-133PBu3+ ions (fQ = 0.50, 0.45, 0.40
and 0.35, respectively). All products were formed after a reaction
time of at least 48 hours at ambient temperature before they were
washed and air dried for analysis ........................................................... 91

Figure 4.3 TGA curves of the products shown in Figure 4.2. The air dried
materials were formed by replacing 35-50% of the exchange ions in
a Na‘i-FH clay suspension with CH5H33PBu3+ ions. Desorption of
physisorbed gallery water occurs below 200 °C. The weight loss
between 400 and 700 °C corresponds to the decomposition of
interlayer onium ions. ............................................................................. 93

Figure 4.4 Model structures of CH3H33PBu3+ and Na“ mixed ion clays that
depend on onium ion composition fq. All cationic species in
structure (A) are segregated into separate galleries and stacked in a
regularly alternating fashion when the fraction of intercalated onium
ions is half, fQ is 0.50. Two models describe heterostructured solid
solutions that arise when less than half a fraction of onium ions are
exchanged into the fluorohectorite clay, 0.35 S fQ < 0.50. However,
both the organic and inorganic interlayers of these mixed ion

xii

heterostructured solid solutions are still regularly interstratified.
Model 1, structure (B), assumes an uneven distribution of cationic
species within successive galleries. Structure (B) has a balanced
quantity of organic and inorganic galleries although the distinctive
galleries are unequally populated. Model 11 assumes a partial mixing
of cationic species. Structure (C) represents this situation where
some metal ions are commingled within the organo gallery. Thus,
the C15H33PBU3+ intercalated interlayer has thus a certain Na” ion
accommodation capacity. ..................................................................... 101

Figure 4.5 XRD patterns (Cu-Kc) of products formed when homoionic Li‘“,
Na+, Ca2+, Ba“, Al3+ and Ce3+ fluorohectorite clays are exchanged
with half a cation exchange equivalent of C15H33NBu3+ ions each, f0
= 0.50. All clay slurries were reacted for at least 48 hours at ambient
temperature before the materials were washed, isolated and air
dried ...................................................................................................... 105

Figure 4.6 XRD patterns (Cu-Kn) of homoionic air dried Na+-fluorohectorite
(A) and homoionic air dried A113” pillared fluorohectorite clay (B).
The inorganic interlayer of the air dried 1:1 C16H33NBu3+ and Na+
mixed ion fluorohectorite heterostructure (C) is selectively pillared
with A113,7+ ions in air dried heterostructured derivative (D). The
gallery height expands with an extra 6—7 A upon pillaring of both
the homoionic and mixed ion materials. .............................................. 108

Figure 4.7 XRD patterns (Cu-Kn) of materials obtained when an organic
interlayer is swollen with epoxy (Epon 828) precursor. The gallery
height of homoionic C16H33NBu3+-FH clay (A) expands by 4 A
when the epoxy precursor penetrates the organic interlayer to give
epoxy swollen material (B). A slightly larger 7 A interlayer swelling
occurs when the organic gallery of a 1:1 Na+ and C1¢5H33NBu3+
mixed ion heterostructure (C) is swollen with the epoxy precursor
to give heterostructured derivative (D) ................................................ 109

Figure 5.1 XRD patterns (Cu-Kc) of air dried fluorohectorite films produced
by the UP flux method (the silicate impurity is marked * in the
XRD pattern). The clays were derived from melts which were held
for two hours at 855 and 1000 °C prior to cooling. The XRD of an
air dried commercial Corning fluorohectorite is included for
comparison. .......................................................................................... 124

Figure 5.2 298i MAS-NMR spectra of the clays shown in Figure 5.1. The 29Si

chemical shift for all clay samples is around -95 ppm relative to
TMS ...................................................................................................... 125

xiii

Figure 5.3 XRD patterns (Cu-Kn) of C15H33NBU3+ exchanged fluorohectorite
clays. (A) Homoionic onium ion exchanged organo clays of the
fluorohectorites synthesized from melts at 855 and 1000 °C. (B) Air
dried 1:1 mixed ion heterostructured products obtained when the
855 and 1000 °C Li+-fluoroclays were exchanged with a half an
exchange equivalent of C151133NBu3+ ions ........................................... 128

Figure 5.4 XRD patterns (Cu-Kc) of aluminum substituted fluorohectorite—like
clay films produced by the LiF flux method (the silicate impurity is
marked * in the XRD pattern). The air dried clays were produced
from 1000 °C melts. Reaction mixtures were formulated to have
2.5% (lower pattern) and 7.5% (upper pattern) tetrahedral aluminum
sites per OzoFa formula unit (Si7_gAlo,2 and SiyaAlrm respectively)...... 131

Figure 5.5 27A1 and ”Si MAS-NMR of an aluminum substituted
fluorohectorite-like clay produced by the LiF flux method from a
1000 °C melt. The reaction mixture was formulated to have 2.5%
tetrahedral aluminum sites per SigOzoF4 formula unit
(Si7,gAIo,2020F4). The 29Si resonance at -95.0 ppm is due to Si(0Al)
while the -90.3 ppm peak is ascribed to Si(lAl) groups. The Si/Al
ratio was calculated to be 20.5 from the relative intensities of the
29Si peaks, which corresponds to a Si7.63Alo,3-; tetrahedral sheet
composition. The 27A] resonance at 67.9 ppm indicates Al in
tetrahedral sites. The asterisks in the 27Al NMR spectra signifies
side spinning bands. Octahedral A1 with a resonance around 0 ppm
is not observed ...................................................................................... 133

Figure 5.6 27A1 and 29Si MAS-NMR of an aluminum substituted
fluorohectorite-like clay produced by the LiF flux method from a
1000 °C melt. The reaction mixture was formulated to have 7.5%
tetrahedral aluminum sites per SIstoF4 formula unit
(Si-,aAlofiOzoFa). The 29Si resonance at -94.5 ppm is due to Si(0A1)
while the -89.9 ppm peak is ascribed to Si(1Al) groups. The Si/Al
ratio was calculated to be 9.0 from the relative intensities of the 29Si
peaks, which corresponds to a Si7,2oAlo_go tetrahedral sheet
composition. The 27Al resonance at 66.8 ppm indicates Al in
tetrahedral sites. The asterisks in the 27A1 NMR spectra signifies
side spinning bands. Octahedral A] with a resonance around 0 ppm
is not observed ...................................................................................... 134

Figure 5.7 XRD patterns (Cu-Kn) for: (A) Completely C16H33NBu3“
exchanged air dried homoionic fluorohectorite-like clays which
have 2.5% and 7.5% aluminum substituted tetrahedral sites; (B)
Products formed when half an exchange equivalent of C15H33NBu3+
onium ions are intercalated into both the 2.5% and 7.5% aluminum

xiv

Figure 5.8

Figure 5.9

Figure 5.10

Figure 5.11

Figure 5.12

containing clays. The C15H33NBu3+ and Li+ exchange cations are

present in these air dried clays in a 1:1 ratio ........................................

XRD pattern (Cu-Kc) of synthetic air dried fluorosaponite. The clay
was produced by a NaF flux method at 1000 °C. The Na+ clay was
first converted to the Li+ exchanged form, washed free of salts and

impurities prior to being analyzed ........................................................

27A1 and 29Si MAS-NMR of a fluorosaponite clay produced by the
NaF flux method from a 1000 °C melt. The reaction mixture was
formulated to produce a Na, 14-[Mg6]Si6_36A11,14020F4 clay. The 27A1
resonance at 67.7 ppm indicates Al in tetrahedral sites. The asterisks
in the 27Al NMR spectra signifies spinning side bands. Octahedral
A1 with a resonance around 0 ppm is not observed. The 29Si
resonances at -93.9 ppm is due to Si(0Al) while the -89.1 and -84.2
ppm peaks are ascribed to silicon atoms in Si(1Al) and Si(2Al)
sites. Si(3Al) sites are not observed. The Si/Al ratio was calculated
to be 4.31 from the relative intensities of these three peaks, which

corresponds to a Si6,49A11,51020F4 tetrahedral sheet composition .........

XRD patterns (Cu-K0,) for: (A) Li+ exchanged fluorosaponite, (B)
Homoionic C16H33NBu3+ exchanged fluorosaponite clay, (C)
Products formed when half an exchange equivalent of C1¢5H33NBu3+
onium ions are intercalated Li+-FS clay. All clays were air dried

before being analde ...........................................................................

Adsorption isotherms (25 °C) for the removal of 2,4-dichlorophenol
from a vigorously agitated solution in water by homoionic
C16H33NBu3+ exchanged fluorohectorite (circles) and a 1:1
C16H33NBu3+ and Na+ mixed ion heterostructured fluorohectorite

clay (squares). The clay suspensions were equilibrated for five days ..

Change of the 2,4—dichlorophenol equilibrium concentration with
time when the solute is adsorbed under static conditions from water
onto homoionic C16H33NBu3+ exchanged fluorohectorite (circles)
and 1:1 C1¢5H33NBu3+ and Na+ mixed ion heterostructured
fluorohectorite clay (squares). Both air dried clays were added to

standing 2,4-dichlorophenol solutions (25 °C) ....................................

XV

138

141

142

145

148

149

CHAPTER 1

Introduction.

1.1 Clays and clay structure.

Clays are the most abundant minerals found at the earth’s surface. Not
surprisingly, one of the earliest applications of chemical processing was concerned with
the manufacturing of pottery using clay as the raw material. Clay minerals are (hydrous)
layered silicates, and usually have a particle size smaller than 2 pm. These phyllosilicates
are assembled from two basic components, a tetrahedral- and an octahedral sheet. The
tetrahedral sheet is composed from SiOa tetrahedra which are linked together in a
hexagonal arrangement through sharing of three basal oxygen atoms. The fourth apical
oxygen atom is connected to the octahedral sheet. The octahedral sheet is build from edge
shared octahedra, in a structure similar to brucite, Mg(OH)2 or gibbsite, Al(OH)3.
Condensation of the two sheet types defines a plane that contains all apical oxygen atoms
from the SiO.: tetrahedra and one unconnected octahedral OH group. One hydroxyl group

is positioned in a cavity formed by a hexagonal-like opening in the tetrahedral sheet [1].
Figure 1.1 illustrates this layer construction and represents the structure of a smectite clay.
Kaolinite, [A14]Si4010(OH)8, has a layered structure that joins one tetrahedral- and

one octahedral sheet to yield a 1:1 layer silicate. It is the major component in China clay.
The OCtahedral sheet of kaolinite is classified as dioctahedral as only two out of three

0mallsbdral sites are filled with aluminum ions. The serpentine group minerals

   

A

‘
._, §.-“';' 2...
h- A

WAKE?
V

    
 

 

Figure 1.1 The ideal smectite clay structure. The clay layer is build from an
°°tahedral sheet which is sandwiched between tetrahedral sheets. Negative charge
develops on the clay layer when ions of lower valency are substituted into the clay
Stmcmre. Electrical neutrality is maintained by cationic interlayer Mn+ species.

have a similar 1:1 sheet arrangement as kaolinite, but their chemical composition is
different. In the serpentines, all octahedral layer sites are filled with trioctahedral
magnesium, [Mg6]Si4010(OH)g.

A 2:1 assembly is formed when two tetrahedral sheets are condensed on an
octahedral sheet. Pyrophyllite, [A14]Si3020(OH)4 and talc, [Mg6]SigOzo(OH)4 are the
simplest di- and trioctahedral forms of 2:1 layered silicates. The individual layers of these
clays are bonded together by weak van der Waals forces that are operating between the
neutral layers themselves. Thus, stacks of layers are agglomerated into particles. The
weak cohesion between the layers is responsible for the soft nature of these clays.

Negative charge develops on 2:1 silicate layers when lower valency ions are
substituted into the clay structure. Positively charged interlayer ions compensate this
excess anionic layer charge in order to maintain electrical neutrality. These interlayer
(gallery) cations may be hydrated, inducing interlayer expansion, and swelling of the clay
[2]. Hence, interlayer access allows for the replacement of gallery cations by other
cationic species through cation exchange processes.

Smectite clays are defined as 2:1 layered silicates that swell with water [3]. The
layer charge of (natural) smectites varies roughly between 0.4 and 1.2 e' per Ozo(OH)4

unit. Montmorillonite, Mx-[Alangx]Si3020(OH)4, is a common dioctahedral smectite
with a layer charge arising mainly in the octahedral sheet (octahedral aluminum cations
are Partially replaced by magnesium). Beidellite, My-[A14]Sig-yAlyOzo(OH)4, is a
dioCtahedral clay too, but the charge on the layer is primarily located on the tetrahedral
Sheets (by the partial substitution of silicon by tetrahedral aluminum). Strictly, one can

define a montmorillonite-beidellite series with general formula: M(x+y)'[AI4-ngx]SIg-

yAl,020(OH)4. An analogous trioctahedral smectite clay series ranges from octahedrally
charged hectorite, Mx-[Mg6.,Lix]SigOzo(OH)4, to tetrahedrally charged saponite,
My-[Mg6]Sig-yAly020(OH)4. Other smectite clay minerals belonging to one of these series
may have octahedral substitutions by metals such as Fe2+'3+, V3+, Crz", Mn”, Co“, Ni”,
Cu“, or Zn“. Even octahedral vacancies can be a source of negative layer charge, as is
the case with stevensite. Tetrahedral charged sites arise mainly by substitution of Si4+
with Al3+ and rarely from replacement with cations like Fe3+ or B“.

Table 1.1 Idealized structural formulas for some general 2:] layer silicates.

 

 

 

 

 

 

Mmcral Layer ' Dioctahedral Trioctahedral
group charge
Pyrophmi‘e - [A14]Si802o(0H)4 [MgGISi8020(OH)4
Talc Pyrophillite Talc
Mx' [Al4-ngXISi8020(OH)4 Mx'IMg6-xLix18i8020(OH)4
Montmorillonite Hectorite
Smectite 0,3 - 1.2 e'
My'IA14JSi8-yAly020(0H)4 My'IMgtSJSis-yA1y020(OH)4
Beidellite Saponite
vermiculite 1 .4 - 1 .8 e' My-x’[Mg6-xAlx]SiS-yAIyOZHOH)‘:
Vermicullte ,
Mica 2.0 e' K2'1A14lSi6A12920(0H)4 Kz'IMgdSieAledeHh
Muscovrte Phlogoplte

 

 

 

 

 

 

Vermiculites are yet higher charged trioctahedral clays, about 1.4 to 1.8 e' per
020(0H)4 unit. The vermiculite water swelling capability is, however, subordinate to
smectite clays. Also, cation exchange reactions in vermiculites prove to be more difficult
When compared to smectites. Micas are the highest charged 2:1 silicate minerals. Ion
eXChange in natural micas is for all purposes practically impossible. Table 1.1

summarizes ideal formulas for some important 2:1 layer silicates [1-3].

1.2 Smectite clay interlayer properties and cation exchange.

Clay layer charge density controls gallery cation population. Both charge density
and charge location affect clay interlayer chemistry. The swelling properties of smectites
are a consequence of the counterbalance between attractive and repulsive forces between
layers [2]. The electrostatic interaction between the negative layers and the interlayer
cations is the main component of layer attraction. Hydration of interlayer cations and the
repulsion between the gallery cations themselves dominate the repulsive terms. The
amount of interlayer water of a clay suspension depends on the layer charge density as
well as on the type of interlayer species. For instance, a Li+ or Na+ exchanged
montrnorillonite suspension which is high in clay content swells with water to form a gel.
However, (osmotic) swelling [4] can be so extensive in more dilute suspensions, that the
active attraction between the layers is insignificant, and a dispersion of exfoliated (single)
clay layers is formed. The distance between platelets is thus practically related to the clay
content of the water suspension. In contrast, Ca2+-montmorillonite does not form an

exfoliated state but can only form water swollen phases [5].

Large monovalent cations, potassium in particular, fit well within the ditrigonal
cavity of the clay layer [2]. This firm fit binds the clay layers together. Removal of these
(dehydrated) cations is extremely difficult at high clay charge densities, as is the case with
micas and vermiculites. The lower charged smectites allow for hydration of even

pOtaSSium, despite the enclosed cation location, but exchange reactions are sluggish.

The location of negative charge on the clay layer has an influence on the interlayer

Cation position. Tetrahedrally charged sites seem to favor an inner-sphere complex

structure, where the gallery cation is localized within the ditrigonal cavity [6,7]. This
particular location may withhold or block this gallery cation from participating in
exchange reactions. Charge arising in the octahedral sheet of 2:1 layers is neutralized by a
hydrated complex in which the gallery cations reside at a position in the center of the
interlayers. These center cations are easily exchanged for other cationic species by means
of mass action.

Smectite clays may have both Brtinsted acid sites (proton donor) as well as Lewis
acid sites (electron pair acceptor, aprotonic). A source of Bronsted acidity arises from the
interlayer exchange metal ions [8]. Cations within the clay gallery are more acidic than in
aqueous solution. It is the strong polarized field by the gallery cation that causes the water
molecules in the hydration shell around the metal ion to be more acidic. Metal exchanged
clays can thus function as solid acids [9]. However, this Brdnsted acidity of the interlayer
cations diminishes as the water content of the clay increases. Lewis acid sites can arise
from the presence of tetrahedral aluminum [10]. The connection between a pair of silicon
and aluminum neighbors in a Si-O-Al linkage can be broken, by protonation for instance,
rendering a silanol group and an aluminum in three coordination. Obviously, this three
coordinated aluminum is a Lewis acid, and with the presence of water yields a Bronsted
site.

Polar molecules other than water can also penetrate the clay gallery. These
molecules can directly coordinate with the exchange ions or hydrogen bond with gallery
ion hydration water or perhaps even hydrogen bond with the clay surface oxygens. For
instance, ethylene glycol can expand the interlayer space of smectites, and is widely used

as a swelling agent in the identification procedure of clay minerals [11].

1.3 Organo clays.

The inorganic exchange ions of a clay can also be replaced by organic cations
such as ammonium, phosphonium or sulphonium ions [12,13]. Even protonated proteins
can be exchanged onto a clay. Binding of onium ions generates a hydrophobic surface on
the clay layers. These clays, called organo clays, are organophilic in nature. The interlayer
structure of these organo clays depends on the size of the intercalated onium ions as well
as on the clay charge density [14]. Figure 1.2 illustrates some different orientation
assemblies possible for alkylammonium ion exchanged organo clays. At low clay charge
densities, the alkyl chains adopt an orientation parallel to the clay surface and form a
lateral monolayer (A). At slightly higher clay charge densities, the space becomes
insufficient to pack all ions within one single layer, and a lateral bilayer forms instead
(B). At even higher clay charge densities, the alkylammonium ions radiate away from the
clay surface and adopt a paraffin like structure, Figure 1.2 (C). The packing arrangement
of the alkylammonium ions is not necessarily rigid but is thought to exhibit some
disordered liquid like behavior due to the presence of some gauche conformations in the
alkyl chain [15,16].

Once bound to the clay, onium ion displacement by hydrated inorganic ions
normally is a difficult process [2]. The van der Waals interaction between the alkyl chains
of the intercalated onium ions provide stability against desorption. Interestingly, the same
process allows for adsorption of onium ions in excess of the clay layer cation exchange
capacity. Charge neutrality is maintained by the simultaneous intercalation of anionic
species when excess onium ions invade the clay gallery. These types of intersalated clays

can thus act as anion exchangers. Of course, the excess onium ions can be desorbed by

 

 

 

 

 

   

[ -

 

 

 

 

 

 

(A) Monolayer (B) Bilayer

 

 

 

 

 

 

 

 

 

(C) Paraffin arrangement

Figure 1.2 Organo clay interlayer structures. Intercalated onium ions form a
monolayer (A) coverage at low clay charge densities or small alkyl chain lengths. Onium
ions are grouped in a bilayer (B) when there is not enough space available to pack all the
ions in a single layer arrangement. A paraffin like onium ion packing (C) occurs at even
higher charge densities. The alkyl chains may have kinks through gauche conformers in
all interlayer structure types.

washing of the clay, but the alkylammonium ions will withstand desorption better with
increasing chain lengths. Thus, one cation exchange capacity equivalent of these larger
onium ions is generally enough to convert a clay into an organo clay completely.

An interesting case arises when the organic cation to be exchanged onto the clay is
too large in footprint size with respect to the area available per exchange site to form a
monolayer or bilayer. Such cations may form a paraffin structure with the alkyl chains
inclined to the clay surface to accommodate all cations within the interlayer. When this is
not possible, for instance, because of steric reasons, the exchange will not proceed beyond
complete interlayer surface coverage, trapping some of the initial exchange ions [17].

Organic solvent molecules can penetrate the interlayers of organo clays [12,18].
The adsorption of organic liquid will increase the interlayer distance between the layers,
thus swelling the organo clay. In fact, the onium ions reorganize their alkyl chain
arrangement upon this swelling. The alkyl chain length of onium ion exchanged organo
clays needs to exceed a particular length for swelling to occur effectively.
Alkylammonium exchanged smectite clays with alkyl chain lengths up to ten carbon units
show minimal organophilic behavior. Longer onium ion modified organo clays swell and
disperse well in polar organic liquids. These systems can even gel an organic liquid at
relatively low organo clay content [12,18].

The organophilic properties of organo clays are also determined by the amount of
clay surface covered by the onium ions [12]. Incomplete clay surface coverage by onium
ions enhances the adsorption of a polar organic liquid. The uncoated silicate surface not
occupied by onium ions is first covered with polar solvent molecules before gelation of

the liquid occurs through swelling of the organo clay. For instance, octadecylammonium

exchanged montrnorillonite clay swells well in nitrobenzene [12]. However, the same
clay hardly swells in less polar toluene. Covering of the uncoated clay surface with polar
additives can improve the organophilic performance of organo clays in non-polar
solvents. A few percent of alcohol added to toluene assists the swelling behavior of
octadecylammonium montrnorillonite greatly. Solvation of an organo clay is thus a
function of compatibility with the solvent molecules. Swelling is improved when the clay
surface loses its inorganic character and becomes increasingly more comparable to the

organic liquid.

1.4 Natural clays.

Natural clays are products formed by the decomposition of igneous rocks.
Hydrothermal alteration and weathering of rocks are the main natural processes of clay
formation [3]. Hence, natural clays are poorly defined minerals. The chemical
composition of a clay mineral is not fixed. Even samples taken from slightly different
spots within one geological deposit can be very different from one another. The variations
include among others, differences in layer composition, charge density heterogeneity,
impurity content and different crystallite sizes.

The variation in layer charge composition is perhaps the most significant
characteristic of natural clays [19]. Irregularities in layer composition suggests
heterogeneity in layer charge density. Thus, all the silicate layers of a smectite clay do not
need to bear an identical layer charge. In fact, clays containing identical layers are very
rare. The gallery cation population varies from gallery to gallery in these heterogeneous

smectites. Clearly, this variation in interlayer cation density falls within limits and has a

10

certain distribution. The clay layers themselves may thus be homogeneously charged, but
may also be asymmetrically charged. Another important possibility is that the layer itself
has an irregular charge composition in the direction parallel to the layer.

The gallery cation density of homogeneously charged clays can be derived from a
cation exchange capacity measurement or from elemental analysis. For heterogeneously
charged clays, the elemental analysis or cation exchange capacity only gives average
values of layer charge densities. Information over the charge density distribution of
smectite clays can be obtained by the alkylammonium method [19].

The alkylammonium method employs the fact that alkylammonium ions adopt a
specific orientation within the clay gallery depending on onium size and layer charge
density [14]. Monolayers, bilayers and pseudotrimolecular interlayer structures each have
specific basal spacings. Transitions in interlayer structure and thus also in the clay basal
spacing arise when the chain length of a series of intercalated onium ions steadily
increases. For instance, the transition between a monolayer and a bilayer will be sharp
and well-defined for a clay with a homogeneous charge distribution. A heterogeneous
clay composed from only two layer types will show a staged transition between these two
layer forms. The step between the two layer forms arises from non-integral 001 reflections
[20,21]. This basal spacing has a value in-between the spacings obtained for the
monolayer and bilayer forms. The exact spacing is a function of the relative fractions of
the interlayer types present and, of course, on the degree of randomness in the layer
stacking. A clay that is composed from a multitude of differently charged layers will have
a broad transition range of non-integral basal spacings that fall between the two layer

arrangements.

11

1.5 Synthetic clays.

Synthetic clays have several advantages over natural clays. These include: 1. The
charge location on the layers of synthetic clays can be chosen. Octahedrally or
tetrahedrally charged clays can be synthesized. Even clays with layer charge arising in
both sheets can be prepared. 2. Selection of a desired charge density. 3. Synthetic clays
may be homogeneous in charge distribution. 4. The chemical composition of synthetic
clays can be controlled (e.g. dioctahedral or trioctahedral, type of ion substitution). 5. A
certain degree of control over the clay particle size. Small clay particles with high surface
areas are often desired for catalytic application. Nanocomposites with good barrier
properties benefit from clays with high aspect ratios [22,23]. However, natural clays are
often abundant, easy to produce (open pit mines) and cheap.

Hydrothermal methods are commonly used to synthesize smectite [24]. The
reaction conditions can vary widely, ranging from ambient temperature and pressures to
very high temperatures (>1500 °C) and pressures (55 kbar). However, moderate
temperatures (100 °C - 500 °C) and modest pressures up to a few kbars are most
commonly used. Starting materials are often freshly prepared stoichiometric gels
formulated to the desired clay composition. These synthetic methods have been used to
produce montmorillonites, beidellites, hectorites and saponites.

Noteworthy is the recent development of a non-hydrotherrnal synthetic method
that produces saponites suitable for catalytic applications [25]. The synthesis is based on
the decomposition of urea at 90°C and is excecuted at ambient pressure. The procedure
allows for some control over the clay surface area. In addition, large amounts of clay can

be produced relatively fast.

12

Clays can also be produced by a solid state reaction of metal oxides in the
presence of a fluoride source. The first synthetic clay produced by this method was a
fluorrnica. Reports on the synthesis date back as far as 100 years ago. Research efforts on
the synthesis of fluoromicas were increased in Germany during World War II, as
synthetic micas appeared to be a good substitute for the natural sheet micas used as
dielectric insulators in electronic equipment [26]. Synthetic mica research programs were
initiated in the United States after World War 11 since certain sheet like micas were
classified as strategic - critical. Reports of the results achieved on synthetic mica projects
were published by the Bureau of Mines [26-28]. The synthetic procedures involved
melting of a standard batch with composition K28iF5 , 6MgO , A1203 , SSi02 at 1340 °C.
On slowly cooling of the melt, crystals of fluor-phlogopite (K2Mg6A128i5020F4) were
formed. Selection of different batch materials and reaction conditions were probed to
better control the nature of the crystals obtained. Water swelling fluor—clays were
discovered during these exploratory experiments.

A high charge density fluorohectorite, Li1,14-[Mg4,g6Li1,pdSigOzoFlt can be
synthesized through a solid state reaction of SiOz and MgO in a LiF flux [29,30]. The
procedure involves melting of the (reagent grade) chemicals in a platinum container at
850°C. The composition of the fluorohectorite obtained is fixed. Changes in the reaction
mixture stoichiometry do not translate in fluorohectorites with different charge densities.
As such, the procedure is particularly well-suited for the production of fluorohectorite on
a laboratory scale. Fluorohectorite has also been produced commercially by Corning [31].
Their process consists of melting the starting materials into a glass. Later, the glass is

cooled to allow for crystallization of fluorohectorite. The fluorohectorites obtained by

13

both methods are well crystallized, have a homogeneous layer charge distribution and

have high layer aspect ratios.

1.6.1 Clay applications, an overview.

The estimated market value of all clays mined in the US. in 1997 was about $1.75
billion [32,33]. A large amount of this annual clay production is used for well-established
(low tech) applications [34]. Kaolinite is mainly used as a filler and coating material in
paper. Common clays, which contain a substantial amount of non-water swelling
materials (e.g. kaolinite, chlorite, illite, quartz) are used in the manufacturing of ceramics
such as tiles and bricks. Water swelling bentonite (industrial montrnorillonite) has diverse
low tech applications ranging from absorbent materials and insecticide dispersants to
drilling muds [18].

Advanced high tech uses of bentonite include applications such as clay liners and
clay walls in chemical waste disposal sites [35]. Bentonite may also be used as ion
entrapment material around canisters that contain nuclear waste [36]. Organo
functionalized bentonite is used, for example, in paints [18], polymer - clay
nanocomposites [37-39] and liquid crystal - clay composites [40]. Organo clays are also

excellent adsorbent materials for the removal of organic pollutants from water [41,42].

14

1.6.2 Polymer - clay nanocomposites.

Mixing of a continuous polymer phase (matrix) with a discontinuous phase (filler)
constitutes a composite. Clays have been used as filler material in polymeric materials for
a long time [34]. Production of these types of composites is dominated by the savings
achieved on the more costly polymer matrix. More advanced composites use fibers (e. g.
carbon, glass) as filler material. These fibers reinforce the polymer matrix, producing a
stiff and lightweight composite with improved mechanical properties. Orientation of the
fibers during the composite fabrication process can produce a composite with anisotropic
properties.

Technically, polymers reinforced with filler material in the nanometer length scale
are classified as nanocomposites. Clays fall within this length scale. Fulfilling this size
requirement, smectite clays also have an additional interlayer functionality. Gallery
chemistry can be put to use to regulate the degree of clay dispersion within thepolymer!
The concept compares well with the gelling behavior displayed when an inorganic ion
exchanged clay swells with water. The degree of dispersion of these clay layers is related
to the concentration of clay layers in water. A low clay content results in a complete
dispersion of layers. These water delaminated clay layers form an exfoliated phase.
Higher clay contents diminishes the distance between the individual clay platelets leading
to a water intercalated state. Of course, the dispersion of a clay in an organic polymer
matrix resembles the solvation of an organo clay by an organic liquid. Dispersion
problems encountered in the nanocomposite synthesis therefore include compatibility

issues.

15

Analogous to clay - water swelling behavior, there exist three types of clay -
polymer composites as shown in Figure 1.3. A conventional composite (A) consists of
clay aggregates distributed throughout the polymer. The clay interlayers are not
intercalated with polymer. On the other hand, intercalated nanocomposite (B), also has
clay aggregates dispersed throughout the polymer matrix, but the clay galleries are
intercalated with polymer. However, the amount of polymer penetrated into these
interlayers is fixed and is therefore independent of the clay / polymer concentration.
Exfoliated nanocomposites (C) have clay layers that are completely dispersed within an
polymer matrix. As such, there remain no clay aggregates. The aspect ratio of the separate
layers in these exfoliated nanocomposites is clearly larger than the aspect ratio of clay
particles in conventional nano scale composites. Also, the average distance between
exfoliated clay layers is a direct function of clay/polymer concentration.

Exfoliated clay - polymer nanocomposites may dramatically improve materials
properties. For instance, an exfoliated clay - nylon-6 composite containing only a few
weight percent clay, shows a significant improvement in tensile strength, impact strength
and heat distortion temperature [43]. This hybrid technology was used in the commercial
production of injection molded timing-belt covers made from exfoliated clay - nylon-6
nanocomposite.

The preparation of exfoliated nanocomposites is quite challenging as it involves
the homogeneous dispersion of an inorganic phase into an organic matrix. In essence, a
driving force has to be provided that leads to clay layer exfoliation. Several factors
oppose this exfoliation process. These include: electrostatic interactions between the clay

layers and the gallery cations, van der Waals interactions between the alkyl chains of

16

 

 

 

a r

I a

 

 

 

(A) Conventional Composite (B) Intercalated Nanocomposite

 

 

(C) Exfoliated N anocomposite

Figure 1.3 Clay nanocomposites. Structure (A) is a conventional clay - polymer
composite. Phase (B) is an intercalated composite. The clay interlayers are intercalated
with polymer, but the basal spacing is fixed and does not change with composition.
Exfoliated composite (C) has individual clay layers dispersed throughout the whole
polymer system. As such, the distance between single layers is related to the clay loading.

17

gallery onium ions and, for instance, the already polymerized material surrounding the
clay layers that prevents further movement of the clay layers. Several methods leading to
clay exfoliation have been developed based on the organophilic character of an
organoclay.

In the case of nylon-6 nanocomposites, protonated tit-amino acid (12-amino lauric
acid) exchanged montrnorillonite clay is first swollen with e-caprolactam [37]. Then, at
elevated temperature, the amino acid catalyzed ring opening polymerization of
caprolactam within the clay galleries provides the driving force for layer exfoliation.

Diamine cured epoxy nanocomposites are based on the catalytic activity of
protonated primary amine exchanged organo clays [44,45]. These protonated amines
catalyze the cross linking process of the epoxide with the diarnine curing agent. Hence,
the curing rate within the organo clay galleries is higher than outside the clay tactoids
[44]. The driving force for layer exfoliation is therewith provided by the curing activated
swelling of the clay interlayers. Tuning of the polymerization rate within clay interlayers
relative to the outside rate seems a key to a successful clay layer exfoliation process, and
is employed with other polymer systems as well. For instance, clay nanocomposites of
tetraethylorthosilicate (TEOS) cross linked siloxane catalyzed by a tin catalyst seem to
use this approach [46].

Today, clay - nanocomposites have been reported for nylon [37], epoxy [38],
polyimide [47], acrylonitrile [48], polyether [49], polysiloxane [46] and polypropylene
[50]. Exfoliated nanocomposite materials offer many other desirable characteristics next

to improvement of mechanical properties. These materials properties include: enhanced

18

gas barrier properties, improved resistance to solvents and application as fire retardant

material.

1.6.3 Liquid crystal - clay composites.

Nematic liquid crystal clay composites are a new development in organo clay
composite chemistry [40]. These type of composites allow for a durable selective
switching to occur between the nematic and the isotropic phases of a liquid crystal. This
phase switching occurs by changing the frequency of an applied electric field. The
composite cells have an electric-optical effect based on light scattering. In addition, these
cells show an optical memory effect.

The design of such devices involves the dispersion of a modified organo clay in a
liquid crystal. Such dispersions have to be stable against phase separation, but also
against sedimentation. However, exfoliation of the organo clay in the liquid crystal is not
a requirement. The liquid crystal - clay suspension is sandwiched between glass plates
which are equipped with a thin transparent conductive film. This conductive film allows
for the application of an electric field across the composite cell. When the device is
switched to the nematic phase by the electric field, both the organic liquid crystal
molecules and the clay platelets spontaneously order themselves parallel to the applied
field. The cell is now in a light transparent state. The clay layers remain in this parallel
orientation even when the electric field is turned off. The liquid crystal molecules
preserve their ordered structure along the immovable clay platelets and the cell stays
transparent. The clay layers conserve their orientational order in the absence of an electric

field and generate the memory effect.

19

Application of an electric field with another frequency can disrupt the long range
order by switching the liquid crystal to the isotropic state. This transition induces a
random alignment of the clay particles throughout the suspension. The cell is now in a
light scattering state. When the field is turned off, the random orientation of the clay
particles retains the random structure of the liquid crystal molecules and the cell stays in
the light scattering state. The pulse duration of the electric field determines the degree of
orientational order of clay particles, thus controlling the number of aligned nematic
domains in the cell. The light scattering efficiency of these cells can therefore be tuned to
desired levels. Light controlling glass could be a promising application for these type of

clay liquid-crystal composites.

1.6.4 Clay as adsorbent material.

Smectite clays are used to minimize leachate flow from chemical waste disposal
sites [35]. A clay liner on top of a waste site prevents the entry of water into the landfill.
Surrounding clay - soil cutoff walls minimize leachate flow from the site to the
environment. Besides decreasing the hydraulic conductivity, the clay barriers can also
participate in the sorption of polar solutes and, for instance, neutralize leachate acidity. In
addition, the cation exchange capacity of the clay barrier can preferentially bind heavy
metal ions onto the clay and so remove them from the leachate flow. Furthermore,
migration of non polar organic compounds can be minimized by the incorporation of
suitable sorbent materials in the barrier.

Organo clays are organophilic and can be effective sorbent materials for organic

groundwater contaminants [42,51]. The organic interlayer functions as a partitioning

20

medium to remove water dissolved non polar pollutants. This solubilizing property
compares well to the action of an organic solvent when one extracts organics from water.
However, the type of organic exchange cation used in organo clays dictates the specific
organic solubilizing capability. The hydrophobicity of an organo clay is regulated by the
onium ion alkyl chain length and the molecular arrangement of the intercalated cations
[41]. Evidently, the clay becomes more hydrophobic when the alkyl chain length
increases. Non polar species adsorb well on these type of hydrophobic clays. However,
when the polarity of the pollutant increases, the sorptive performance of these organo
clays diminishes. Clays exchanged with smaller onium ions can do a better job with these
type of contaminants.

Smaller ions such as tetramethylammonium ions position themselves on the clay
surface as discrete entities. Of course, this renders the clay less hydrophobic. In this case,
the interlayer can not function as a partitioning medium. However, the uncovered free
surface of the clay layer can now also interact with the sorbate species [41]. The
combination of hydrophobic and hydrophilic areas within the clay interlayers enhances
the sorptive performance when removing more polar organic species from water.

In addition to the adsorbent - adsorbate interactions, adsorbate - solvent
interactions also influence the adsorption process. When compared to water solvated
contaminants, a decreased adsorption onto hydrophobic organo clay arises when non
polar pollutants are dissolved in organic liquids.

Soils can be polluted with aromatic wastes produced by coal conversion plants or
oil based industries but also by leaks from underground gasoline storage tanks. Organo

clays can be used in the cleanup treatment of these type of contaminated soils. They are

21

effective sorbents in the removal of BTEX compounds (benzene, toluene, ethylbenzene
and o—xylene) which are the most toxic aromatic constituents of gasoline. Mixing of the
organo clay with the contaminated soil prevents leaching of the contaminants into the

groundwater system [52].

1.6.5 Clays as catalysts.

Clay minerals have been used as catalysts in a variety of chemical reactions. For
instance, acid treated clays were used in the catalytic cracking of petroleum before being
replaced by more efficient zeolites [53]. Today, clays are still used as cracking catalyst
support in the oil refining industry.

Most clay catalyzed reactions make use of the Lewis and Bronsted acidity of clays
[9]. Clays can function as solid acids in heterogeneous catalysis. In addition, catalytic
organic reactions can take place within the interlayer space. The catalytic activity can thus
be confined to a two-dimensional reaction space.

Some industrial processes use clay based catalysts but clay catalysts are mainly
used in laboratory synthesis. Examples of clay catalyzed reactions include processes such

as: isomerization, polymerization, hydrolysis, epoxidation and Diels-Alder reactions [9].

1.7 Pillared clays.

Clays can assume microporous character (with pores < 20 A) when certain large
cations are exchanged into the interlayers [54,55]. These exchange ions can hold the clay
layers apart permanently and can thus act as pillars. The pillaring cationic species have to

be strong enough to prevent collapse of the supported structure. In addition, the pillars

22

have to be spaced sufficiently far apart from each other in order to create free space for
molecular gallery guests [56]. Thus, the created interlarnellar voids depend on the density
and size (width and height) of the pillaring species.

The desired pillar separation 1S regulated by the clay charge density and the
valency of the employed gallery exchange species. Molecular selectivity in terms of
molecular sieving can therefore be tuned to specific adsorbates. This pillaring and shape
selectivity was first demonstrated by N(Me)4+ and N(Et)4+ exchanged montmorillonite
clays [57,58]. These modified clays intercalated many gasses and non polar organics,
whereas the parent Na” exchanged clays did not exhibit this behavior. These shape
selective separation capabilities may be altered by using short chain primary (di)amines or
other exchange cations such as Co(en)33+.

Organic pillared microporous clays are not very stable. Heating of these materials
causes a collapse of the pillared structure through decomposition of the pillaring agents.
Materials pillared with metal oxide clusters are more robust. These type of pillars are
formed by the intercalation of bulky inorganic polynuclear cationic species, followed by
calcination [54-56]. Microporous metal oxide pillared clays are thermally stable, may
have high surface areas and have pore sizes comparable to zeolites. Furthermore, the
metal oxide pillar itself can add additional catalytic activity to the microstructure. In
addition, some particular clays may enhance this activity as is the case with alumina
pillared derivatives of fluoro clays [59]. The structural fluorine provides a reactive center
for fluorine hydrolysis and renders an catalytic enhancing effect in these clays.

Polynuclear ionic species, such as the Keggin ion for instance,

A11304(OH)24(H20)127+, are commonly used as pillaring agents. Intercalation of the

23

polycation through ion exchange and subsequent thermal dehydration / dehydroxylation
converts the polynuclear complex into a metal oxide interlayer particle [56]. The clay
charge is balanced by protons after calcination. The clay layers are separated from each
other by the pillar, creating an 8-9 A free vertical span relative to the layers. Likewise.
other comparable metal oxide pillared clays have been prepared with gallium-, iron-,
chromium-, zirconium- and titanium- oxide pillars. The thermal stability of all these
pillared clays is, of course, limited by the stability of the clay layers.

Alumina pillared derivatives of rectorite possess high thermal and hydrothermal
stability in comparison to other alumina pillared clays [60]. Rectorite is a relatively rare
mixed layer 2:1 silicate mineral with a regular alternating stacking of high charge non-
swellable mica type galleries and low charge swellable smectite type galleries [61].
Intercalation chemistry only occurs in the smectite like interlayer. This increased thermal

stability has been ascribed to the presence of the rigid non-intercalated mica-type layers.

1.8 Mixed ion clays.

When the selective sorption properties of N(Me)4+ pillared montmorillonite clays
were first explored by Barrer and Brummer [62] and later by McBride and Mortland [63],
attempts were made to modify the lateral distance between the pillaring species through a
partial ion exchange of Na+ clay by N(Me)4+ cations. The idea was that if both cationic
species would mix within one interlayer, the free distance between the organic pillars
could be regulated by simply exchanging a desired amount of pillars into the system [64].
Such a structure would look like a mixed ion homostructure [65]. This structure class has

only one interlayer type but multiple cationic species are homogeneously mixed within

24

the interlayer regions. However, mixing of Na+ and N(Me)4+ cations within a gallery did
not occur. Instead, when organic ions entered an interlarnellar region, inorganic ions were
forced out of that particular gallery, prohibiting the formation of a mixed ion
homostructure. Thus, the partial exchanged material had both cationic species segregated
into different interlayers. These different interlayer types were found to stack at random.
Hence, this clay material can be classified as a randomly stacked mixed ion
heterostructure.

A mixed ion heterostructure has at least two different exchange ions segregated
into different interlayers [65]. Of course, the clay layers themselves are still of one kind.
Stacking of two unlike cation-occupied interlayers, for instance, can occur without
apparent order. This random interstratification can present itself at any given ratio
between both cationic species. Alternatively, a periodic arrangement of interlayers occurs
in regularly stacked mixed ion clay heterostructures. This regular alternation imposes
constraints to the possible exchange cation ratios. Regular interstratification of galleries is
expected to occur for exchange ion ratios of 50%, 33%, 25% etc., when only two cationic
gallery species are present.

Such regularly interstratified mixed ion clay heterostructures are reminiscent of
the staged structures encountered in graphite intercalation compounds [66]. A stage-n
graphite material has a repeat arrangement of n graphite layers which are followed by a
single layer of intercalated interlayer ions. However, staging in graphite is a reversible
process. A stage-n graphite structure can be reversibly altered into a stage-(n11) structure.

So far, clays have not shown this extraordinary intercalation behavior.

25

Méring and Glaeser [67,68] first proposed demixing of Ca2+ and Na“ ions in
certain montmorillonites. Arguments were made on basis of water adsorption isotherms.
Domains of demixed exchange ions better explained the observed isotherms than regions
of randomly mixed ions. However, the question whether cation demixed regions occur
within interlayer domains or in alternate galleries remained unanswered. Later
investigations on the system revealed a (de)mixing dependence on the Na+ to Ca2+ ratio
[69]. Random mixing of ions was observed at a Na+ ion content percentage larger than
50%. Consequently, a mixed ion homostructure is formed at these compositions. A partial
demixed distribution of ions occurred between 15% and 50% Na“ exchange. However,
the adsorption experiment could not differentiate between demixing by domains or by
alternate interlayers. At lower Na+ contents, the interlayer clay complex was mainly Ca2+
saturated while the external surfaces were predominantly exchanged with Na+ ions.

Staging-like behavior has been encountered in a peculiar synthetic fluoro mica
[70], Lil_23-[Mg5_36Clo,64]Si3020F4. The selective gallery intercalation process originates by
migration of interlayer lithium into vacant octahedral clay layer sites. It thus resembles
the well known Hofmann-Klemen effect [71] which leads to a reduction of clay charge
upon this lithium ion fixation. Normally, this migration effect does not initiate differences
in intercalation properties among galleries. However, this particular clay does form two
different interlayer types on lithium migration. Interlayer differentiation arises when an
interlayer donates all its exchange ions to fill all vacant octahedral sites in both
surrounding clay layers. Neighboring interlayers retain all their cations and neutralize
excess negative layer charge. Staging-like behavior is generated through lithium fixation

in octahedral sites but is, however, irreversible.

26

Interesting intercalation products were encountered when controlled amounts of
long chain quaternary alkylammonium ions were intercalated in taeniolite [72],
Liz-[MgaLi2]SigOzoF4, a synthetic fluoro mica. At all exchange concentrations of onium
ions employed, organo clay was the dominant product, although significant amounts of
mixed ion heterostructure and mixed ion homostructure were formed. The mixed ion

heterostructure was present in both the regularly as well as randomly stacked forms.

1.9 Research objectives.

Intercalation of diverse ionic species in smectite clays does apparently not
necessarily have to lead to the formation of segregated parent end member clays. There is
evidence, as discussed above, that the presence of different ionic species in a clayrmay
possibly lead to other intercalated phases. A mixed ion homostructure is formed when
cations commingle within interlayers. Mixed ion heterostructures can be generated upon
cation demixing. In some mixed ion clay systems, however, it is not clear whether this
demixing occurs by domains or by alternate interlayers. In addition, clays with alternating
organic and inorganic interlayers have never been produced without the presence of other
impurity phases. The formation of end members and mixed ion homostructures seems, till
now, always to be associated with heterostructure formation.

In the present work, novel but simple processes will be described concerning the
synthesis of heterostructured clays [73]. These heterostructured clays have regularly
alternating organic and inorganic ion exchanged galleries. In this sense, these
heterostructured materials combine the properties associated with the end member clays

into one material. Thus, for instance, this particular clay intercalation chemistry associates

27

“and.

clay exfoliation in water with exfoliation in organic media. This combinatory exfoliation
effect allows for the creation of unprecedented clay dispersions. The significance of this
is best illustrated by Figure 1.4. The inorganic interlayer of a heterostructure will swell
when brought into contact with water. Eventually, this water swelling results in an
exfoliation of layer ensembles. The exfoliated entity consists of two clay layers and an
organic interlayer. In other words, the exfoliated heterostructure suspension consists of
very well dispersed organo clay! Organo clay by itself is often too hydrophobic to be
wetted by water. Dispersion of an organo clay in water can thus be problematic.
Normally, organo clay aggregates can only fractionate into smaller particles when brought
into water. But, this wetting of an organo clay does require vigorous processing. Water
exfoliation of a heterostructured clay is, in comparison, an excellent organo clay delivery
system.

Thus, the fundamental intercalation chemistry regarding heterostructure formation
will be elucidated in this work. Factors that govern the formation process of mixed ion
heterostructures will be identified. These include both the influence of the onium ion
geometry and the clay layer charge location. For this, special series of surfactants have to
be prepared. Clays with specific charge density locations have to be synthesized. This, in
turn, requires the development of a synthetic process that allows for control over the
substitutions made within the clay structure. Heterostructure formation parameters can

then be established with these specially tailored clays and surfactants.

28

 

  

Organo clay
aggregates

Figure 1.4 Dispersions of heterostructured clay and organo clay in water. The
inorganic gallery of a mixed ion heterostructure (A) can swell with water to exfoliation

and create a dispersion of single organo clay entities. In contrast, organo clay (B) can
barely breakdown into smaller particles.

29

What effect does the fraction of intercalated exchange ions have on the stacking
order of mixed ion heterostructures? Can we only interstratify organic and inorganic
galleries in a 1:1 ratio, or are 2:1 and 1:2 structures, for instance, also possible? And if so,
is the gallery stacking regular, or random?

Once heterostructure formation parameters are demonstrated in an ideal clay
model, other less perfect natural clays can be investigated. In the end, this will complete
the mixed ion heterostructure formation picture. Only then can general guidelines be
formulated towards the formation of mixed ion heterostructures in other layered materials

capable of comparable intercalation chemistry.

30

1.10 References.

[1]

[2]

[3]

[4]
[5]

[6]
[7]
[8]
[9]
[10]
[11]
[12]

[13]

[14]
[15]
[16]

[17]

SW. Bailey (editor), Hydrous Phyllosilicates (exclusive of micas); Reviews in
Mineralogy 19, (1988)

A.C.D. Newman (editor), Chemistry of Clays and Clay Minerals, Mineralogical
Society 6, (1987)

E. Nemecz, Clay Minerals, Akademiai Kiado, Hungarian Academy of Science,
Budapest (1981)

FT. Madsen, M. Muller-Vonmoos, Appl. Clay Sci. 4, 143 (1989)

R. Kjellander, S. Marcelja, R.M. Pashley, J .P. Quirck, J. Phys. Chem. 92, 6489
(1988)

J. Greathouse, G. Sposito, J. Phys. Chem. B 102, 2406 (1998) .

F.C. Chang, N.T. Skipper, G. Sposito, Langmuir 14, 1201 (1998)

K.V. Raman, M.M. Mortland, Clays Clay Miner. 16, 393 (1968)

J .M. Adams, Appl. Clay Sci. 2, 309 (1987)

J .J . Fripiat, A. Leonard, J .B. Uyterhoeven, J. Phys. Chem. Ithaca 69, 3274 (1965)
RS. Dyal, S.B. Hendricks, Soil Sci. 62, 421 (1950)

J .W. Jordan, J. Phys. Colloid Chem. Ithaca 53, 294 (1949)

A. Weiss, Organic Geochemistry, (Englinton, Murphy; editors) Ch. 31, 737
(1969)

G. Lagaly, Solid State Ionics 22, 43 (1986)

RA. Vaia, R.K. Teukolsky, E.P. Giannelis, Chem. Mater. 6, 1017 (1994)

E. Hackett, E. Manias, E.P. Giannelis, J. Chem. Phys. 108, 7410 (1998)

A. Weiss, Angew. Chem. Int. Ed. 2, 134 (1963)

31

[18]
[19]
[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

TR. Jones, Clay Miner. 18, 399 (1983)

G. Lagaly, A. Weiss, Proc. Int. Clay Confi, Mexico 157 (1975)

IL. McAtee, Natl. Akad Sci-Natl. Research Counsel. 566, 308 (1958)

G.W. Brindley, G. Brown, Crystal Structure of Clay Minerals and their X -ray
Identification, Mineralogical Society 5, (1980)

T. Lan, P.D. Kaviratna, T.J. Pinnavaia, Chem. Mater. 6, 573 (1994)

LE. Nielsen, J. Macromol. Sci. (CHEM) A 1(5), 929 (1967)

N. Giiven, Hydrous Phyllosilicates (exclusive of micas); Reviews in Mineralogy
l9, Ch.13 (1988)

R.J. Vogels, Non-Hydrothermally Synthesized Trioctahedral Smectites, Ph.D.
Thesis, Utrecht University, Netherlands (1996)

RA. Hatch, R.A. Humphrey, W. Eitel, J .E. Comeforo, U.S. Bureau of Mines,
Reports of Investigations 5337 (1957)

RC. Johnson, H.R. Shell, U.S. Bureau of Mines, Reports of Investigations 6235
(1963)

HR. Shell, K.H. Ivey, U.S. Bureau of Mines, Fluorine Micas, Bulletin 647 (1969)

R.M. Barter, D.L. Jones, J. Chem. Soc. (A) 1532 (1970)

, R.M. Barrer, D.L. Jones, J. Chem. Soc. (A) 503 (1971)

G.H. Beal, D.G. Grossman, S.N. Hoda, K.R. Kubinski, Corning Glass Works,
U.S. Patents 4239519 and 4297139 (1980)

Mineral Commodity Summaries, U.S. Geological Survey (1997)

R.L. Virta, Clays [online] Available http://www.minerals.er.usgs.gov/minerals/

February 3, (1998)

32

[34]

[35]

[36]

[37]

[38]
[39]

[40]

[41]
[42]
[43]
[44]

[45]

[46]
[471

[48]

[49]

[50]

Kirk-0thmer Encyclopedia of Chemical Technology, 4'” ed. 6, 405 (1993)

B. Shawney (editor), CMS Workshop Lectures, Organic Pollutants in the
Environment, The Clay Minerals Society, Boulder CO 8, (1996)

CD. Hollister, S. Nadis, Scientific. Am. 1, 60 (1998)

A. Usuki, Y. Kojima, M. Kawasumi, A. Okada, Y. Fukushima, T. Karauchi, O.
Kamigaito, J. Mater. Res. 8, 1179 (1993)

T. Lan, T.J. Pinnavaia, Chem. Mater. 6, 2216 (1994)

PB. Messersmith, E.P. Giannelis, Chem. Mater. 5, 1064 (1993)

M. Kawasumi, A. Usuki, A. Okada, T. Karauchi, Mol. Cryst. Liq. Cryst. 281, 91
(1996)

M.M. Mortland, S. Shaobai, S.A. Boyd, Clays Clay Miner. 34, 581 (1986)

SA. Boyd, J .-F. Lee, M.M. Mortland, Nature 333, 345 (1988)

A. Okada, A. Usuki, Mater. Sci. Eng. 3, 109 (1995)

T. Lan, P.D. Kaviratna, T.J. Pinnavaia, Chem. Mater. 7, 2144 (1995)

T.J. Pinnavaia, T. Lan, Z. Wang, H. Shi, P.D. Kaviratna, A.C.S. Symp. Ser. 622,
250(1996)

S.D. Burnside, E.P. Giannelis, Chem. Mater. 7, 1597 (1995)

T. Lan, P.D. Kaviratna, T.J. Pinnavaia, Chem. Mater. 6, 573 (1994)

Y. Kojima, K. Fukumori, A. Usuki, A. Okada, T. Karauchi, J. Mater. Sci. Lett. 12,
889(1993)

M.S. Wang, T.J. Pinnavaia, Chem. Mater. 6, 468 (1994)

N. Hasegawa, M. Kawasumi, M. Kato, A. Usuki, A. Okada, J. Appl. Polym. Sci.

67, 87 (1998)

33

[51]

[52]
[53]
[54]
[55]

[56]

[57]
[58]
[59]
[60]
[61]
[62]
[63]

[64]

[65]

[66]

[67]

168]

GR. Rakhhandehroo, R.B. Wallace, S.A. Boyd, T.C. Voice, Soil Sci. Soc. Am. J.
62, 5 (1998)

S. Gitipour, M.T. Bowers, A. Bodosci, J. Colloid Interface Sci. 196, 191 (1997)
WP. Hettinger Jr., Appl. Clay Sci. 5, 455 (1991)

K. Ohtsuka, Chem. Mater. 9, 2039 (1997)

A. Corma, Chem. Rev. 97, 2373 (1997)

RT. Tanev, J .-R. Butruille, T.J. Pinnavaia, Chemistry of Advanced Materials,
(L. V. Interante, H.J. Hampden-Smith; editors) Ch. 8 (1998)

R.M. Barter, D.M. MacLeod, Trans. Faraday Soc. 51, 1290 (1955)

R.M. Barrer, Pure & Appl. Chem. 61, 1903 (1989)

J .-R. Butruille, T.J. Pinnavaia, J. Catal. 139, 664 (1993)

J. Guan, E. Min, Proc. 9‘“ Int. Congr. Catal. 104 (1990)

G. Brown, A. Weir, Proc. Int. Clay Conf., Stockholm 1, 27 (1963)

R.M. Barret, K. Brummer, Trans. Faraday Soc. 59, 959 (1962)

MB. McBride, M.M. Mortland, Clay Miner. 10, 357 (1975)

R.M. Barrer, Zeolites and Clay Minerals as Sorbents and Molecular S ieves,
Academic Press (1978)

W.L. IJdo, T.J. Pinnavaia, J. Solid State Chem. 139, 281 (1998)

SA. Solin, Annual Review of Materials Science (Kaufmann, Giordmaine,
Wachtman; editors) 27, 89 (1997)

J. Méring, R. Glaeser, Clay Miner. Bull. 2, 188 (1954)

J. Méring, R. Glaeser, Bull. Sop. Franc. Miner. Crist. 77, 519 (1954)

34

[69]

[70]
[71]
[72]

[73]

DH. Fink, F.S. Nakayama, B.L. McNeal, Soil Sci. Soc. Amer. Proc. 35, 552
(1971)

K. Urabe, I. Kenmoku, Y. Izumi, J. Phys. Chem. Solids 57, 1037 (1996)

U. Hofmann, R. Klemen, Z. Anorg. Chemie 262, 95 ( 1950)

K. Tamura, H. Nakazawa, Clays Clay Miner. 44, 501 (1996)

W.L. IJdo, T.Lee, T.J. Pinnavaia, Adv. Mater. 8, 79 (1996)

35

CHAPTER 2
Synthesis of Layered Silicate Heterostructures with Regularly
Alternating Organic and Inorganic Galleries :

Precursors to Pillared Rectorite-like Intercalates.

2.1 Abstract.

Smectite clay heterostructures with regularly interstratified organic and inorganic
exchange cations have been prepared for the first time by three reaction pathways. One
route used a thermodynamically favored cation exchange reaction between homoionic
sodium fluorohectorite and alkylphosphonium cations. In a second pathway, onium ions
in the clay were replaced by sodium ions. The third pathway reassembled homoionic
sodium and alkylphosphonium parent end members into the regularly interstratified
heterostructure. Differences in solvation properties for the inorganic and organic
exchange cations allowed the heterostructure to be transformed by ion exchange into a

rectorite—like alumina pillared analog.

36

2.2 Introduction.

Layered alumino-silicates, in particular the 2:1 mica-type structures, possess
diverse intercalation properties that have found a wide variety of materials applications
including chemical catalysis [1,2] and polymer nanocomposites [3-5]. Most of these 2:1
silicates are uniformly intercalated by molecules or cations. All galleries are thus
equivalent in these homoionic clays. A spontaneous segregation of diverse guest species
into separate but (regularly) stacked galleries by a process comparable to staged
heterostructures in graphite is rare [6,7]. Only a few clay minerals have a mixed layered
structure and may as such be heteroionic. Illite / smectite and synthetic mica /
montrnorillonite clays, for example, have galleries that are distinguishable. The galleries
in these clays differ in height and guest species, but the galleries are not regularly ordered
in the layer stacking direction. The only known 2:1 layered silicate with a discernible
regular arrangement of distinguishable galleries is rectorite.

Rectorite is a relatively rare mixed layer 2:1 silicate mineral with a regular
interstratification of high charge non-swellable mica type galleries and low charge

swellable smectite type galleries [8]. Intercalation of [A113O4(OH)24(H20)12]7+ ions into

the smectite galleries of such clay minerals, followed by calcination, yields microporous
alumina pillared clays. Alumina pillared derivatives of rectorite, possess high thermal and
hydrothermal stability in comparison to other alumina pillared clays [9]. This increase in
stability has been ascribed to the presence of the rigid non-intercalated mica-type layers
[9-11]. The effect of layer thickness on thermal stability has also been demonstrated for a
randomly interstratified pillared synthetic mica-montmorillonite clay [12]. Thus, the

benefits of rectorite compared to other smectites are substantial.

37

Smectite clays with a heterostructure composition analogous to rectorite might
afford similar materials properties. However, such smectite structures are unknown. In
the present work we report the first examples of regularly interstratified smectite clay
heterostructures.

Smectite clays intercalated with organo cations, such as alkylammonium ions,
form gallery assemblies with specific orientations in the layer stacking direction
depending on alkyl chain length and layer charge density [13,14]. Once bound to the clay,
onium ion displacement by hydrated inorganic ions normally is a difficult process. In the

case of fluorohectorite, however, onium ion redistribution is favored.

2.3 Mixed ion heterostructure formation.

Owing to the thermodynamic stability of mixed ion fluorohectorite
heterostructures (FHHS), especially when the gallery cations are Na+ and C16H33PBu3+,
FHHS materials may be formed in quantitative yield by three processes as shown in
Figure 2.1. Pathway I simply consists of replacing half of the sodium in Nat-FH by
surfactant cations. This reaction is quantitative with respect to alkylphosphonium ion
uptake. Complementary to I is pathway 11, where now the intercalated surfactant cation is
partially replaced by sodium when the organoclay end member is exposed to an aqueous
NaCl solution. Since the equilibrium lies in favor of the organo cation clay, a large excess
of Na+ ions is needed to drive the reaction by mass action. Approach III mixes equivalent

molar amounts of the parent end members, Na+-FH and C16H33PBu3+-FH, and produces

-.
y

spontaneously the same heterostructure.

38

Figure 2.1 Three synthetic pathways to regularly ordered C16H33PBU3+lNa+-
fluorohectorite heterostructures (denoted C): Pathway I is the quantitative ion exchange
of homoionic Na+-FH (denoted A) with a stoichiometric quantity of surfactant cations.
Pathway II is the exchange of organo cations in homoionic C16H33PBu3t-FH (denoted B)
with excess NaCl. Pathway III is the spontaneous formation of the heterostructure by ion
redistribution reaction of the homoionic parents.

39

 

OMDHODmeHvu—OS
as 852

I

5452 84:2 8:: e2

 

 

 

<9,

 

Humanism N:

 

 

 

H: San

 

 

1
6,55 Aosaez 8,5 «2

I

 

34.2

40

Figure 2.2 shows the XRD patterns for the starting materials Nat-FH and
C15H33PBu3t-FH and the three mixed ion C16H33PBu3+lNa+-FI-II-IS products produced
by pathways 1, II and 111 respectively. All heterostructure products show ordered d(()()1)
reflections corresponding to a basal spacing around 40 A, consistent with regular
alternating stacking of a 12.4 A layer from Nat-PH and a 28 A C16H33PBU3+-FH layer
with an inclined paraffin-like orientation of alkylphosphonium ions. The observation of
multiple 001 reflections at 20 A (dooz) and 13.3 A ((1003) along with the absence of non-
integral Bragg reflections, demonstrates regular rather than random interstratification of
the layers. Heterostructure formation is favored thermodynamically, as indicated by the
fact that parent end member reflections are not observed for the product formed by
pathway HI.

Other evidence in support of C16H33PBu3+lNa+-FHHS formation is provided by
the therrno gravimetric analysis or TGA curve for a sample formed by pathway 11 by ion
exchange of C16H33PBu3+-FH with NaCl, Figure 2.3. A weight loss consistent with a
50:50 mixture of parent end member galleries is indicated. This result, together with the
X-ray results is indicative of a regular stacking of organic and inorganic exchanged
fluorohectorite layers. As can be seen from the relative XRD intensities given in Figure
2.2, the heterostructures derived from pathways I and III show somewhat sharper
reflections than the product of pathway 11, which may indicate a difference in scattering
domain size. Peak broadening may also be the result of both turbostratic stacking of the

layers and limited layer stacking domain size during the formation process.

41

 

C
kwmthwhy III
JLApatlg/hy II

l

40

I
20A 1313 A

 

    

patgtiay I

l
28
40 I
' 14IA
B
1

T— ' l ' '7 ' l
2 4 6 8
Degrees (20)

 

 

Relative Intensity

 

 

 

0

Figure 2.2 XRD patterns (Cu-Kn) of fluorohectorite derivatives: (A) homoionic Nat-
FH, (B) homoionic C16H33PBu3t-FH and (C) the mixed ion heterostructures

C16H33PBu3+INa+-FHHS prepared by pathways 1, II and III. The weak shoulder at 40 A
in pattern B is due to the presence of residue heterostructure formed in the exchange
reaction of Li+-FH with [C15H33PBu3tlBr.

42

 

 

+
C1 6H3 3PBu3 /Na -FHHS

 

 

 

l k ________ \
901 *
5 1 Na+-FH
g 8541
‘3‘ :
:15.» 80-1
g ; C16H33PBu3+
75-:
7o{
65I'H'Tl "l""l "'l"
O 100 200 300 400 500 600
T [°Cl
Figure 2.3 TGA results of Nat-PH (stripes),

C16H33PBU3+/Na+-FHHS (stripe-dot).

43

C16H33PBU3t-FH (solid) and

The thermodynamic stability of C16H33PBu3+INat-fluorohectorite

heterostructures is best illustrated by the spontaneous nature of reaction III. The end
members spontaneously rearrange to form the interstratified heterostructure. Notice also,
that for pathway I the amount of surfactant used in the reaction is the stoichiometric
amount needed to form a 1:1 heterostructure. In fact, further conversion of the
heterostructure to the homoionic surfactant exchanged clay is relatively difficult owing to
the stability of the C16H33PBu3+INa+-FHHS, which is greater than its pure parent end
members. The regularly interstratified structure shows no tendency to segregate into
homoionic phases even upon storage in water. Once a heterostructure is formed,
repetitive exchange with surfactant is required to obtain the homoionic organo clay end
member. The XRD pattern (B) in Figure 2.2 for C16H33PBu3+-FH shows, in addition to
the first and second order reflections of homoionic clay at 28 A and 14 A, a weak (C1001)
reflection near 40 A due to the interstratified heterostructure. Despite the mass action of
excess surfactant, a trace of the heterostructure remains in the reaction product. Owing to
heterostructure stability it is possible to replace half of the surfactant cations from the
alkylphosphonium modified fluorohectorite in an easy one step process by ion exchange

with excess NaCl as denoted in scheme II.

2.4 Synthesis of a pillared rectorite-like analog.

Differences in the solvation properties of the organic and inorganic cations of the
heterostructure allows for the possibility of selective ion exchange without loss of regular
stacking order. Cations in either of the two different heterostructure galleries can be

selectively replaced by either organic or inorganic cations by simple ion exchange,

44

regardless of the intercalated species. We illustrate this principle with several examples of
new heterostructure derivatives synthesized in this fashion. Figure 2.4 illustrates an ion
exchange scheme followed to obtain a regular interstratified
[A11304(OH)24(H20)12]7+/K+ pillared rectorite-like analog. XRD patterns of the derived
hetero compounds are shown in Figure 2.5. Hydrated sodium in C16H33PBu3+/Na+-
FHHS was selectively replaced by potassium. The XRD pattern of the air-dried product
clearly shows a decrease in d-spacing for heterostructure (D), which is the result of partial
dehydration of gallery potassium ions. The disappearance of the second order ((1002)
reflection indicates that the layer sandwiched between the organic and inorganic galleries
has been further displaced from a symmetrical position between the two galleries, as
expected for potassium cations that are tightly locked in partially hydrated gallery
positions. The absence of additional reflections for other new phases confirms the
retention of a heterostructure with segregated galleries. Once the partially hydrated K+
cations are immobilized in mica-like galleries the alkylphosphonium surfactant may then
be exchanged without side reaction. The hydrophobic surfactant in C16H33PBu3tht-
FHHS was successfully replaced by tetrabutylammonium cations of lower
hydrophobicity. The TBA+/K+-FHHS was isolated as a pure phase only after several
washes, as indicated by the XRD pattern in Figure 2.5 (B). As the surfactant is replaced

by TBA+ cations, dooj peak broadening occurs. As represented in Figure 2.4 (E), the

TBA+ cations adopt an "on-edge" orientation in the galleries. Stuffing the galleries with

TBA+ causes layer stacking distortion, but gallery segregation is maintained.

45

 

 

 

 

 

 

 

 

 

TBAT D
EtOH

 

 

 

 

 

 

 

 

_fl-
Na Na Na Na Na $0,; EtOH : I]
M 23.4 28 A

 

 

 

 

 

 

 

 

 

 

 

Na Na Na Na Na 22? («1,9 61,9 29A

 

 

 

 

 

 

Figure 2.4 Selective ion exchange reactions of fluorohectorite heterostructures.

46

 

 

29A

 

 

K 11A| ON

I 28A

I37}.

Relative Intensity

 

I40A

 

 

 

 

2 4 6 i; 10
Degrees (20)

Figure 2.5 XRD patterns of heterostructures formed by the selective ion exchange
reactions illustrated in figure 2.4.

Rectorite-like analogs (F) and (G) in Figure 2.4 are obtained when the organic
TBA+ cations of reduced hydrophobicity are replaced with inorganic cations.
Tetrabutylammonium cations in TBA+IK+-FHHS were completely exchanged by sodium
using an excess of NaCl to drive the reaction through mass action. The (9001) spacings of
23.4 A for Nat/K+—FIIIiS, Figure 5 (F), suggests a regular stacking of Na+- and K‘-
layers. As no phase segregated Nat-PH is observed, we conclude that the potassium filled
galleries are mica-like. As the layer sandwiched between the Na+ and K+ galleries has
moved more to the center of symmetry of the unit cell in the layer stacking direction, an
increase in intensity for the second order reflections is expected and observed at 11.7 A.
The observed diffraction pattern deviates from that of rectorite because all the silicate
layers in Nat/IG-FHHS are of symmetrical charge whereas the 2:1 silicate layers in
rectorite are asymmetrical in layer charge and gallery cation population density. However,
this Na+/K+-fluorohectorite heterostructure is rectorite-like because the Nat galleries are
hydrated and exchangeable by other intercalants, whereas the K+ galleries are not as
easily accessible for exchange. Upon calcination of the air-dried Nat/K+-FHHS at 623 K,
the Na+ and K+ galleries completely dehydrate. The first order reflection of the Nat/Kt-
FHHS, Figure 5 (G), is now virtually absent and only an 11 A peak is observed. These
observations are consistent with a regularly alternating unit cell as shown in Figure 4 (G).
As diffraction is mainly due to the layers themselves, and to a lesser extend to the
exchange ions, only the 11 A reflection can be observed.

Heterostructure (G) was pillared with [A113O4(OH)24(H20)12]7+ cations in water

based on the assumption that the Na”r exchanged layers will rehydrate and exchange at a

48

faster rate than the K+ intercalated mica-like galleries. The XRD pattern of product (H)

indicates a successful pillaring as no first order reflection for pure A1137t-FH occurs at 19
A. Reflections at 29 A, 14.5 A and 9.7 A indicate formation of A1137+IK+-FHHS. Since

the K+ galleries do partially rehydrate during the exchange process, special care during
the pillaring process was needed in controlling the A1137+ reaction stoichiometry and the
reaction time in order to avoid formation of homoionic A1137+-FH. Alumina pillared
rectorite is thermally and hydrotherrnally exceptionally stable. These properties may not
be expected for A1137+lK+-FHHS after calcination since fluorohectorite defluorinates and
hydrolyzes at high temperature [15]. Also the charge density of the layers of
fluorohectorite is located in the octahedral sheet, whereas rectorite has asymmetrical
charge density in the tetrahedral sheets, which might also contribute to the stabilization of
the alumina pillared product. As expected, calcination at 673 K of our A1137+IK+-FHHS
rectorite-like product resulted in a partial collapse of the layers. However, the nitrogen
adsorption and desorption isotherms of the calcined product does show a pore size of
around 8 A consistent with a alumina pillared material. Partial collapse of the material
during calcination at 673 K is confirmed by the rather low observed BET surface area of
78 mZ-g'l. Although our alumina pillared fluorohectorite heterostructure lacks the thermal
stability of a pillared rectorite, the structural analogies to rectorite are quite evident.
Future studies of heterostructure formation and pillaring based on 2:1 clays such as

beidellite and vermiculite may afford more stable pillared rectorite-like analogs.

49

2.5 A provisional mixed ion heterostructure formation model.

Finally, we suggest a tentative explanation of the driving force for heterostructure
formation. The clay charge density, together with steric restrictions of the surfactant
cations, most likely limits the approach of the organo cations towards the silicate layer
surface, thereby inducing an unfavorable charge separation. To minimize the destabilizing
effect of charge separation in the organo cation galleries, the layer charge is compensated
and better neutralized by locating the smaller inorganic cations in the next adjacent
gallery. The high clay layer charge of the fluorohectorite restricts the organo cations in the
parent end member into an electrostatically unfavorable configuration. Consequently,
heterostructure formation lowers the restrictions imposed on the surfactant cations,
leading to a thermodynamically more stable intercalate. The formation of these
heterostructures and the demonstrated selective gallery ion exchange possibilities with
retention of gallery segregation is a fundamental advance in the intercalation chemistry of
clays and should lead to important new materials applications. Also, heterostructure
formation is not limited to alkylphosphonium ion surfactants. We have prepared FHHS
materials using alkylammonium ions via pathway 1, and the latter structures will be

described in future studies.

2.6 Experimental.
Starting materials: Synthetic high charge-density lithium-fluorohectorite,
Lil.12[Mg4,33Li1.12]Si3020F4, (Corning, Inc.), designated Li+-FH, was converted into the

sodium form, Nat-FH, by ion exchange with a twofold excess of an aqueous 0.1 mol-L'l

NaCl solution, washed free of excess salt and air-dried. Phosphonium modified

50

fluorohectorite was obtained from Li+-FH by a repetitive bulk exchange with a twofold
excess of 0.01 mol-L'l surfactant C16H33PBu3Br (Lancaster) in a 50:50 (v/v) water /

ethanol mixture. The procedure was repeated until a pure phase was indicated by X-ray

diffraction. The organoclay was washed first with ethanol, then water, and air-dried.

Heterostructure formation: Heterostructures, (HS), were produced by three
general methods. In pathway I the heterostructure was produced from Nat-PH and
C16H33PBu3Br by stirring a 1.0 : 0.5 molar ratio of clay : surfactant in deionized water at
ambient temperature for at least 48 h. Typically, an 0.003 mol-L‘1 aqueous surfactant
concentration was used. Excess salt was washed away with water before air-drying the
product.

Heterostructures obtained from route 11 were made by reversing the positions of

the exchange ions in route I. Surfactant was replaced from C16H33PBu3+-FH by bulk

mass action of sodium in a 2.5 mol-L'1 NaCl aqueous solution. About 0.30 g homoionic
C16H33PBu3t-FH was suspended in 40 mL deionized water. NaCl was then added until a
2.5 mol-L‘1 solution was obtained. The reaction mixture was stirred for 6 h at room
temperature before the suspension was washed free of excess salt. Since the driving force
for heterostructure formation was provided by the mass action of excess sodium salt, the
formation of heterostructure was closely monitored by XRD in order to establish the
optimum 6 h and prevent further conversion of the heterostructure to Na+-FH. After being

washed with water the clay was air-dried.

51

Method III mixed equivalent molar parts of Nat-FH and C15H33PBugt-FH in
deionized water. A heteroionic structure formation in this case was provided by the
spontaneous cation redistribution reaction of the homoionic parent end members. About
0.5 g air-dried clay was stirred in 100 mL water for at least 48 h at room temperature
before the C15H33PBu3+INa+-FHHS was isolated by centrifuging. The material was

washed with water and air—dried.

Cation exchange reactions of C16H33PBu3+lNa+-FHHS: Sodium ions in
C16H33PBu3+INa+-FHHS were exchanged for potassium by stirring the HS in a 0.01
moltL'1 KCl aqueous solution with four times the exchange capacity value of the HS in

KC] for at least 12 h at ambient temperature. This procedure was repeated three times to
ensure complete exchange and the final product was washed and air-dried. The
alkylphosphonium surfactant in C16H33PBU3+fl(+-FHHS was exchanged for
tetrabutylammonium (TBA) by stining the HS in ethanol with a ten-fold 0.02 moloL‘1
excess of the bromine salt of TBA for 12 h. The procedure was repeated, again at room
temperature, but a 50:50 (v/v) water / ethanol mixture was used as the solvent. The
product was washed until no change was observed in the XRD pattern. The final product

was stored in aqueous suspension, because the air-dried product hardly re-suspends in
water. Tetrabutylammonium in TBAt/Kt-FHHS was replaced with sodium by exchange
with a 0.01 mol-L'l NaCl solution in 25 fold excess. The product was washed and air-

dried. The resulting Na+/K+-FHHS was then calcined at 623 K for 4 h. An A113 pillared

exchanged A1137+/Kt-FHHS was prepared by resuspending the calcined Na+/K+-FHHS

52

in an aqueous solution containing one equivalent of [A113O4(OH)24(H20)12]7+ cations

prepared through slow base hydrolysis of Al3+ cations [16]. This pillared product was

then washed and calcined at 673 K for 6h.

Product characterization: X-ray diffraction (XRD) patterns were recorded on a
Rigaku rotaflex diffractometer equipped with 1/6_ divergence and Seller slits and Ni-
filtered Cu K,l X-ray radiation. All samples (except for the C16H33PBu3+/I(+-FHHS )
were analyzed as preferentially oriented films to enhance the intensity of the 9(001)
reflections. Data was collected with 001° 20 interval between 1 20 and 20 20 using a
scanning speed of 1° 20 per minute. Thermogravimetric analysis (TGA) was carried out
on a Cahn 121 TG analyzer. Typically, the sample size was about 50 mg and the heating
rate was 5 K per minute up to 873 K. Nitrogen adsorption and desorption isotherms were
measured at 77 K on a Coulter Omnisorp 360CX Sorptometer using standard continuous

adsorption procedures. The sample weight was about 100 mg after it was heated at 373 K

and 10'6 torr before measurements were made.

53

2.7 References.

[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]

[14]

[15]

[16]

F. Figueras, Catal. Rev. -Sci. Eng. 30, 457 (1988)

M. L. Occelli, Surf. Sci. Catal. 35, 101 (1988)

Y. Fukushima, S. Inagalti, J. Incl. Phenom. 5, 473 (1987)

V. Mehrotora, E. P. Giannelis, Mater. Res. Soc. Proc. 171, 39 (1990)
T. Lan, T. J. Pinnavaia, Chem. Mater. 6, 2216 (1994)

S. A. Solin, Adv. Chem. Phys. 49, 455 (1982)

M. S. Dresselhaus, G. Dresselhaus, Adv. Phys. 30, 139 (1981)

G. Brown, A. Weir, Proc. Int. Clay Conf., Stockholm 1, 27 (1963)

J. Guan, E. Min, Proc. 9’” Int. Congr. Catal. 104 (1990)

M. L. Occelli, Stud. Sun“. Sci. Catal. 63, 287 (1991)

J. F. Brody, J. W. Johnson, Proc. Mater. Res. Soc. 65 (1990)

K. Urabe, N. Kouno, H. Sakurai, Y. Izumi, Adv. Mater. 3, 558 (1991)
G. Lagaly, Solid State Ionics. 22, 43 (1986)

A. Justo, C. Maqueda, F. L. Perez-Rodruguez, G. Lagaly, Clay Miner. 22, 319

(1987)

I J .-R. Butruille, L. J. Michot, O. Barres, T. J. Pinnavaia, J. Catal. 139, 664 (1993)

S. Yamanaka, G. W. Brindley, Clays Clay Miner. 27, 119 (1979)

54

CHAPTER 3
Correlation between Onium Ion Geometry and the Staging Behavior of
Organic and Inorganic Gallery Cations in Layered Silicate

Heterostructures.

3.1 Abstract.

Heterostructured fluorohectorite clays with regularly alternating interlayers of
inorganic and organic exchange cations have been prepared through ion exchange
reaction of the sodium clay in aqueous suspension with half an equivalent of
alkylammonium ions. In order to elucidate this staging - like behavior of the intercalates,

the exchange process was examined for a series of onium ion surfactants of the type,
[CnH(2n+1)N(CmH(2m+]))3]+. One set of onium ions varied the alkyl chain length (n=4 -

22), while maintaining a constant onium head group size (m=4). A second set varied the
head group size (m=l - 5), while the chain length remained fixed (n=16). Two

fundamental factors, namely, alkyl chain length and head group size determined staging
behavior. The surfactant alkyl chain length determined the extent of Na+ replacement by
onium ions. Relatively little Na+ exchange (~10%) occurred for the short chain onium
ions with n=4, m=4 and n=6, m=4. The replacement of Na+ by onium ions became more

favorable as the alkyl chain length was increased to n=8 or 10. For very hydrophobic

surfactants with n212, onium ion uptake was essentially quantitative, affording 1:1 mixed

Na+- onium ion intercalates. The “footprint” or area covered by the onium ion head group

55

on the interlayer surface controlled heterostructure formation for the 1:1 intercalates. For
m23 the head group footprint matched the clay surface charge density and precluded the
mixing of organic and inorganic exchange cations within the same gallery. The
segregation of the hydrophobic organic cations and the hydrophilic inorganic cations on
the internal and external surfaces of two-nanolayer tactoids during the ion exchange
process was proposed as the pathway for staging behavior in the heterostructured
intercalates. The smaller footprints of trimethyl and triethyl onium ion head groups
(m=I,2) allowed for co-mingling of the exchange ions on single-nanolayer tactoids and to
the formation of phase - segregated organic and sodium ion clays upon stacking of the

tactoids.

56

3.2 Introduction.

The diverse intercalation chemistry of smectite clays (e.g. montrnorillonite) has
led to several important materials applications. For instance, the incorporation of large
polycations between the negatively charged 2:1 aluminosilicate layers gives rise to
microporous pillared structures that are useful in catalytic applications [1]. Other
interesting applications taking advantage of this rich intercalation chemistry include
polymer nanocomposites [2] and chemical sensors [3]. Organoclays are formed when the
hydrated exchangeable clay cations are replaced by long chain alkylammonium ions. The
specific properties of an organoclay depend on the particular intercalated surfactant, as
well as on the nature of the layered material itself. Important applications of these
organoclays stem from their hydrophobic properties [4]. Organoclays are used industrially
as rheological control agents in paints, greases, and oil-well drilling fluids [5] and as
adsorbents for removing organic pollutants from aqueous solutions [6].

Surprisingly few investigations have been reported on mixed organic-inorganic
ion exchanged forms of smectite clays. Mixed ion clays have the potential of combining
the hydrophilic properties of the native inorganic clay with the hydrophobic qualities of
an organo clay. A fundamental understanding of the structure and polar nature of mixed
ion clays may lead to materials applications. For example, the ability to mediate the
surface polarity of naturally occurring clays by partial ion exchange with onium ion
surfactants might be advantageous in trapping pollutant plumes in subterranean
environments.

As shown in Figure 3.1, several products may form when the inorganic cations of

a smectite clay are partially replaced by organic cations. Phase segregation can occur,

57

yielding starting material and newly formed organo clay (A). As indicated in Figures 3.1
(B) and (C), mixing of the two ions at exchange sites can lead to the formation of a single
homostructured phase or to two types of heterostructured phases. The mixed ion
homostructure (B) has all the interlayers occupied by both the metal ions and the onium
ions, whereas the heterostructures (C) have two types of segregated galleries, one
inorganic and one organic cation interlayer. The two galleries of the heterostructure may
be stacked either in a disordered, interstratified manner or in a regularly alternating
fashion. The regular stacking of the distinguishable organic and inorganic galleries is
reminiscent of staging behavior in graphite [7].

In general, homo- and heterostructured mixed ion intercalates of smectite clays are
relatively uncommon. Méring and Glaeser [8] first proposed the segregation of Na+ and
Ca2+ ions in the galleries of montrnorillonite, but there was no evidence for regular
heterostructure formation. A homostructured phase containing inorganic cations and a
minor fraction of short chain alkylammonium cations has been reported for partially
exchanged montmorillonites [9,10]. However, a disordered heterostructure formed when
comparable amounts of organic and inorganic cations occupied the exchange sites. A
mixed ion homostructure also was observed when the organic cation was
hexadecyltrimethylarnmonium [11]. Rectorite is an example of a clay mineral with staged

galleries [12]. However, only one of the interlayers is readily accessible for ion exchange.
Also, the clay - related potassium niobate K4Nb6017-nH20 has distinguishable interlayer

regions, with both galleries showing intercalation chemistry [13].

58

 

ll

pal

the
110i

and Na Na Na Na Na

 

 

 

Na Na Na Na Na

 

B. Mixed ion homostructure C. Mixed ion heterostructure
(Stacking may be regular or random)

Figure 3.1 Three possible structures formed when the exchange sites of smectite clays are
occupied with two different cationic species, namely, alkylammonium ions and
dehydrated sodium ions. Structure (A) denotes the two phase segregated homoionic
parent end member materials. Product (B) is the mixed ion homostructure with both
species intercalated within one gallery, while (C) is the mixed ion heterostructure with
the two cations segregated into two interlayers. The stacking pattern of the organic and
inorganic galleries may be regular or random.

59

Staging-like behavior has also been achieved for a swelling fluoromica through an
irreversible process that fixes small interlayer cations like lithium in vacant octahedral
positions in the silicate layers [14]. Mixed inorganic-organic onium ion heterostructures
have been recently reported for a swelling fluoromica [15], but the exchange process used
to form these compositions also gave mixtures of regularly and randomly interstratified
heterostructures as co-products.

We recently reported three complementary pathways for the formation of
regularly alternating mixed inorganic - onium ion fluorohectorite heterostructures [16].
We also demonstrated different intercalation properties for the staged interlayer regions
and were able to synthesize a derivative that compared favorably to the rare mineral
rectorite. In the present work we have investigated the intercalation process that leads to
the formation of fluorohectorite heterostructures. In particular, we have elucidated the

role of the onium ion geometry in the synthesis of regularly ordered heterostructures.

3.3 Experimental.

Clay preparation: A synthetic, high charge density lithium-fluorohectorite,
Lil_12[Mg4,33Li1.12]Si3020F4, (Corning, Inc.), was converted to the sodium form by
exchange with 0.1 M NaCl. Homoionic alkylammonium ion derivatives were obtained
from Li+—FH by ion exchange reactions with the bromine salts of the desired onium ions.

A two-fold excess of surfactant cations was utilized to insure complete exchange. All
organoclays were washed with a 50/50 (v/v) water/ethanol mixture until no change in the

X-ray diffraction pattern was observed for the air dried materials.

60

Surfactant synthesis: Two series of quaternary ammonium bromide surfactants

with varying chain lengths and head group sizes were synthesized from the corresponding
l-alkylbromides and trialkylamines, (Aldrich) [17,18]. For the C16H33N(CmH(2m+1~,)3Br
series, the ionic head group size m was varied from 1 to 5, while the surfactant alkyl chain
length was kept constant at sixteen carbon units. The CnH(2n+1)NBu3Br surfactant series

changed the alkyl chain length n in two-carbon unit intervals from 4 to 22 while keeping
the tri-n-butyl head group fixed. In all preparations a slight excess of trialkylamine was
used. The reactants were dissolved in ethanol and refluxed for three days before the
ethanol was removed by rotary evaporation. The products were recovered in either
solidified or oil form, depending on the surfactant chain length. The solids were
recrystalized twice from ethylacetate/pentane, whereas the oils were washed several times

with pentane.

Heterostructure preparation: All heterostructured clays were prepared in a

simple one-step procedure by the direct addition of a half an exchange equivalent of
surfactant cations to a Na+ clay suspension. In a typical preparation 0.30 g sodium-
fluorohectorite clay was exfoliated in about 100 mL of deionized water prior to surfactant
addition. The suspension was stirred after surfactant addition for at least 48h at ambient
temperature or at 80°C, in the case of the sparingly water soluble tri-n-pentyl terminated

surfactant. All the heterostructures were washed free of excess salt and air dried.

61

Product characterization: X-ray diffraction, XRD, patterns were recorded on a
Rigaku rotaflex diffractometer equipped with Ni-filtered Cu-Ka radiation. All of the clay

samples were analyzed as preferentially ordered films. A Cahn 121 TG
therrnogravimetric analyzer was used to determine the surfactant content of the clay

samples. The samples were heated at 5 °C per minute up to 800 °C.

3.4 Preferred mixed ion heterostructure formation pathway .

Previously, we developed three different synthetic strategies [16] to produce

staged 1:1 Na+ : onium ion fluorohectorite heterostructures. Figure 2.1 (chaper 2)

illustrates these three pathways for the case where the onium ion is C16H33PBu3+. The

mixed ion heterostructures presented in this paper contain alkylammonium ions instead of
alkylphosphonium ions. Pathway I, which represents the pathway used in the present
work, embodies the direct addition of a quantity of surfactant cations to obtain a reaction
mixture in which the overall ratio of onium to sodium ions was 0.5 : 1.0. Pathway II is the
reverse approach wherein a homoionic organoclay is treated with a concentrated NaCl
solution to produce the heterostructured products. This procedure proved to be more
susceptible to the creation of heterostructures with layer stacking defects, as judged by the
broadening of the 001 XRD reflections. Pathway III in Figure 2.1 involves the reaction of
equal molar quantities of the two end member homoionic clays in a water suspension.
Pathway III is more cumbersome than pathway I because it involves the preparation of

two homoionic clays. However, it does lead to the formation of the same heterostructured

62

products as the route I pathway. For these reasons pathway I was preferred over the other

two pathways.

3.5 Onium ion chain length effect on heterostructure formation.
Our clay heterostructures are defined as intercalates that contain equal amounts of
onium and sodium ions staged into regularly alternating galleries. The unit cell

dimensions along the layer stacking direction is expected to correspond to the sum of the
basal spacings for the Na+-FH and the organoclay end members. The effect of ammonium

chain length on heterostructure formation was studied first. The alkyl chain length of a

‘[CnH(2n+1)NBu3]+ surfactant was increased in two-carbon atom units (n=4-20) while

keeping the tri-n-butyl head group fixed. The XRD profiles for the 1:1 NaJr
alkylammonium reaction products are shown in Figure 3.2. The basal spacings for both
the homoionic organoclay end members and the 1:1 Na+- ammonium reaction products

are given in Table 3.1.
The short chain onium ions with n=4 and 6 formed little or no heterostructured
product, in part, because the sodium competed more favorably for exchange sites of the

clay than the onium ions. Consequently, the XRD patterns of the products contain
primarily the 12.4 and 6.2 A reflections characteristic of Na+-FH. For n=8 and 10, the
onium ions begin to displace increasingly more sodium from the exchange sites and some
heterostructured product, along with unexchanged Na+-FH is identified in the XRD

pattern.

63

 

 

 

 

 

 

 

 

WERE;

 

 

 

 

I

-
-
uni
q
—1
cl
-

1 I U I l 1

10 14 18
Degrees (20)

N
0\

Figure 3.2 XRD patterns (Cu-1(a) of intercalates formed by reaction of a Na+-FH
suspension and 0.5 equivalents of [C,,H(2,,+1)NBu3]+ onium ions. The surfactant alkyl
chain length was varied in two-carbon units over the range n=4-22 while the tri-n-butyl
head group was kept constant.

Table 3.1 Comparison of X-ray basal spacings for Q+- and mixed ion Na+, Q+-

fluorohectorites, where Q+ is CnH(2n+1)NBu;.

 

 

 

 

 

 

 

 

 

 

 

Gag???“ Q+-FH [A] Na*,Q+-FHHS [A] Difference [A]
22 29.7 43.7 14.0
20 28.6 41.7 13.1
18 28.4 40.4 12.0
16 26.9 39.4 12.5
14 25.9 38.2 12.3
12 25.1 37.8 12.7
10 23.0 35.3 12.3
8 22.1 34.0 1 1.9
6 15.8 - -
4 15.7 - -

 

 

 

 

 

 

At n 2 12, the replacement of sodium by onium ions is quantitative and well
developed 1:1 staged heterostructures are formed. For chain lengths in the range n=8-22,

the observed difference in basal spacings between the organoclay and heterostructured
products remain fairly constant and very near the expected 12.4 A spacing of Na+-FH.

The above interpretation of the XRD results is supported by thermogravimetric

analysis (TGA) of the reaction products (see Figure 3.3). The calculated and observed

TGA weight loss for the products obtained from the [C,,H(2n,t1)NBu3]+ surfactant series is

summarized in Table 3.2 for the products obtained when Na+-FH is exposed to half an

exchange equivalent of surfactant cations.

65

 

 

 

 

. s a
5 .2 i i 1
- iv =
: ‘xuu uqé:3q:_.‘;,\ E
10 — : ‘ :
9° j ; \ c4, c6 (9%)
a -1 ' “‘--1__ C (10%)
2 « 5   5 8 (12%)
E 15 '1 : \ :
.99 . : \ :
9 2 : . ‘~'...
3 20 - : : C10 (36%)
1 i 1 C12 (45%)
25 _~ : ' C14 (47%)
- 5 5 C16 (48%)
j ; . E18 (50%)
- ' ' 20 52%
30 j—III I I l I III II I I I I I I I l I I ( C)

0, 100 200 300 400 500 600
Temperature [°C]

Figure 3.3 TGA curves for the mixed intercalates depicted in figure 3.2 with n=4-20.
Three separate areas corresponding to surfactant content are indicated. The weight loss
below 200°C corresponds to the release of gallery water. The decomposition of the C22
surfactant occurs in stages and is, for simplicity, not shown. Decomposition of the
organic onium is indicated by the weight loss in the region 200 - 550 °C.

66

Table 3.2 TGA determination of the exchange ion compositions for 1:] Na+, Qi-

fluorohectorites, where Q+ is CnH(2n+1)NBu;.

 

 

 

 

 

 

 

 

 

 

 

Carbon chain Expected organic Observed organic Q :Na. who in
. a . j, the mixed 1011
length, 11 weight loss , wt%. weight loss , wt%
product
20 22.0 22.9 52:48
18 20.9 21.0 50:50
16 19.9 19.3 48:52
14 18.8 17.8 47:53
12 17.6 16.0 45:55
10 16.5 12.0 36:64
8 15.3 3.7 12:88
6 14.0 2.5 9:91
4 12.8 2.5 10:90

 

 

 

 

9 Expected weight loss for a water-free (dehydrated) 1:1 Na‘”:Q+ mixed ion clay.

9 Observed weight loss due to burning of Q+.

On the basis of the observed weight loss in the region 125-550 °C, about 20
percent of the supplied onium ions are exchanged into the clay galleries when n=4 and 6.

However, quantitative uptake of the surfactant cations and staged heterostructure

formation occurs when C12 or longer chain surfactants are used in the exchange process.

The onium ions with C8 and C10 chain lengths define a transitional region corresponding

to 12-36% onium exchange. In this transitional field there is an onset on heterostructure

formation as evidenced by the development of heterostructured related diffraction peaks

(see figure 3.2).

67

 

Based on these results; the extent to which staged heterostructures are formed
depends on the onium content. We refer to the relationship between staged

heterostructure formation and onium ion stoichiometry as the surfactant chain ejfect. The
C4 and C6 onium ions are too hydrophilic to quantitatively displace sodium from

exchange sites. No heterostructured intercalate can be formed under these equilibrium
conditions because there is little or no onium binding. The onium ion simply does not
compete effectively with sodium in the exchange process. Most of the aqueous onium

ions are thus lost upon washing and drying of the products. Increasing the chain length to

C3 leads to some heterostructure formation but the onium ion exchanged product is still

in equilibrium with a substantial amount of Na+-FH. The shoulder on the Na+-FH peak

near 13 A corresponds to a mixed ion homostructured material that has both cationic

species mixed within the same interlayers. This same type of product also is evident in

the C6 exchange system.
The surfactant chain effect starts to dominate at C10 and is fully operative for the
C12 onium ion as a result of the longer surfac1ant alkyl chains. A mixed ion

homostructure is absent in the C10 case, even though the equilibrium has shifted further in
favor of mixed cation intercalates due to the increased onium content. Increasing the
onium ion chain length beyond C10 results in a practically quantitative surfactant uptake,
as evidenced by TGA. However, these C12 and longer onium ions differ somewhat in the

fidelity of alternating onium and sodium ion intercalation, as judged by variations in the

68

widths of the 001 reflections. The highest ordered air - dried heterostructure products with
the smallest peak widths are obtained for the C14 and C16 surfactant chain lengths.

When the surfactant is added to the aqueous clay suspension, all exchange sites
are available for reaction. XRD studies of the reaction products as a function of time
indicates that some fully exchanged organoclay, as well as the mixed ion heterostmcture,
is initially formed at the initial stages of surfactant addition. This homoionic organoclay
phase reacts further with the Na+-FH component on aging of the reaction slun'y. Thus, the
actual exchange pathway involves a combination of pathways I and III of Figure 2.1.
Surfactants with longer tails, as in the case for n > 18, slow down this onium ion
reshuffling process between homoionic end member structures, because of increased van
der Waals interactions between surfactant tails and a decreased onium ion water
solubility. Hence, longer equilibration times are needed for the surfactants with longest
chain lengths.

Plots of the observed d-spacing as a function of the alkyl chain length for both the
homoionic '[CnH(2n+1)NBu3]+ organoclays (circles) and the staged 1:1
[C,,H(2n,t1)NBu3,]+/Na+ heterostructures (squares) are shown in Figure 3.4. The dashed

line in the figure is the spacing expected for a mixed ion homostructure in which the
onium and sodium ions co-occupy every gallery. The surfactant alkyl chain in such a
homostructure was assumed to have a 90° angle with respect to the clay layer surface.

The basal spacing was calculated by the relationship: 15.7 A + n-1.265 A, where 15.7 A is
the thickness of the clay layer (9.6 A) plus the van der Waals thickness of a NBu3 head

group (6.1 A) and where n is the number of carbon units in the alkyl chain [19].

69

 

d-spacing [A]

 

 

 

 

20 l ' T r l T l ' l ' I ' l
10 12 14 16 18 20 22
Alkyl chain length

Figure 3.4 Dependence of XRD basal spacings on the alkyl chain length n for homoionic
CnH(2n+1)NBu3+- fluorohectorites (circles) and mixed ion Na+, CnH(2n+1)NBu3+-
fluorohectorite heterostructures (squares). The dashed line corresponds to the calculated

basal spacings for mixed ion homostructures in which the two cationic species co-occupy
the interlayer galleries.

70

The slopes of the lines defined by the observed spacings for both the homoionic
organo clays and the mixed ion heterostructures, in Figure 3.4, are in good agreement
with each other. The surfactant alkyl chains in both structures each have inclined angles
of 33° and 36° respectively, with respect to the layer surface.

A best fit of the line for the homoionic organoclays gives an intercept at 18.0 A. In

comparison, an intercept at 30.3 A is obtained for a best line drawn through the basal
spacings for the staged heterostructured materials (d-spacings of the C8 and C10

transitional structures excluded). The 12.3 A difference in intercept for the two lines is
the value expected for a heterostructure in which the onium and sodium ions are staged in
regularly alternating galleries. The line for the mixed ion heterostructure and the expected
line for the mixed ion homostructure intersect around 43 A. At this particular basal
spacing no distinction can be made between heterostructure and a homostructure.
However, the observed spacing for every other point precludes a mixed ion

homostructure.

In summary, the tail length of the onium ion surfactant determines the extent of
sodium ion replacement upon the addition of half an equivalent of the surfactant. Small
onium ions are too hydrophilic to displace sodium quantitatively. A quantitative onium
loading is obtained only when the surfactant becomes more hydrophobic at surfactant

chain lengths of n212.

71

3.6 Head group size effect.

In a second set of intercalation experiments we investigated the effect of the
surfactant head group size on heterostructure formation. The [CnHaml)N(C,,,H(2,,,,t1))3]+
head group size m was increased one carbon unit at a time from Me, Et, n-Pr, n-Bu to n-
Pe, while the chain length was fixed at n=16. This chain length ensures a quantitative

onium ion uptake on the partial displacement of the sodium ions. Diffraction profiles for

the intercalation products are given in Figure 3.5. In contrast to the staged

Cut,H33NBu3+/Na+ heterostructure, phase segregated organo clay and sodium clay

products were obtained when Na+-FH was equilibrated with half an exchange equivalent
of trimethyl terminated quaternary ammonium surfactant (m=I). The first order (27 .3A)

and second order (13.7A) peaks for C15H33NMe3+-FH and the 12.4 A Na+-FH peak are

well developed. Thus, the small size of the surfactant head group causes the two cationic
species to segregate into discrete galleries during the exchange reaction. The nature of the
reaction products becomes more complex when a triethyl terminated surfactant (m=2) is
used. All possible products, homoionic parent end members and a mixed ion
heterostructured intercalate are found as products. The XRD shows predominantly the

segregated 27.2 A C16H33NEt3+-FH and 12.4 A Na+-FH phases, but also a small amount

of a heterostructured phase, as evidenced by the diffraction shoulder around 39 A. The
cationic species are thus evidently segregated into distinct interlayers. When the head
group size is increased to n-Pr, n-Bu and n-Pe (m23), the equilibrium clearly shifts

exclusively towards a staged mixed-ion heterostructure.

72

 

n-Pe

 

n-Bu

 

n-Pr

Et

 

Relative Intensity

Me

I'I'ITI'IWF'F'I'I

2 6 10 14 18
Degrees (20)

 

 

 

 

Figure 3.5 XRD patterns (Cu-Kg) of materials obtained when a Na+-FH suspension is

treated with half an equivalent of [C16H33N(C,,,H(2m+1))3]+ surfactant. The head group
size was increased in one-carbon units from m=l-5 to form an expanding series that
included Me, Et, n-Pr, n-Bu and n-Pe groups.

73

Clearly, the size of the surfactant head group is a key factor in determining

heterostructure formation. When the head group size is large relative to the charge density
of the layered host, there is insufficient space to accommodate a hydrated Na+ ion

between groups. We refer to the relationship between head group size and layer charge

density as the head group footprint efl'ect. The fluorohectorite layer charge density is
calculated to be one electron per 84 A2 based on an orthogonal
L11_12[Mg4,33L11,12]Si3020F4 unit cell [20]. The surfactant ionic head group areas, as

estimated based on the van der Waals radii, are given in Table 3.3. This table also
includes the basal spacings for all five homoionic organoclays and the d-spacings for the

three 1:1 mixed ion heterostructures.

Table 3.3 Surfactant head group size and X-ray basal spacings ((1001) for homoionic Q+
fluorohectorite (Q+-FH) and 1:1 mixed Na+, Q+ fluorohectorite heterostructures (N81, Q+-

FHHS), where Q+ is C16H33NR3+.

 

 

 

 

 

 

 

Surfactant head
Surfactant head . Q+-FH, Nai, Qi-FHHS,
group Slzea
gourp. R 2 door [A] (1001 [Al
[A 1

Me ‘ 34 27.3 -

Et 56 27.2 -
n-Pr 78 27.0 39.5
n-Bu 101 26.9 39.4
n-Pe 123 26.4 38.9

 

 

 

 

9- For comparison, the charge density of fluorohectorite is 1 electron per 84 A2.

74

 

The surface area per unit of charge is matched by the area of the head group (i.e.,
the “footprint”) only for the tri-n-propyl and larger surfactant head group sizes (i.e., m23).
The larger tri-n-butyl and tri-n-pentyl surfactants can fold back upon themselves, thereby
covering the basal surface area. A completely hydrophobic clay surface thus arises for
m23. The lack of free space between onium ions precludes the co-occupancy of both
cationic species within the same gallery. The hydrated sodium ions are now forced to take

position on the hydrophilic side of the clay layer.

3.7 A heterostructure formation model according to nanolayer tactoids.
Previously reported neutron scattering and NMR studies have shown that alkali
metal exchange forms of smectite clays in stirred aqueous suspension form tactoids

containing 1-3 nanolayers [21-23]. The number of nanolayers per tactoid depends on the
nature of the exchange cation and the shear rate [23]. The formation of staged Na+- onium

ion heterostructures can be understood in terms of an equilibrium between tactoids

containing one and two nanolayers. As illustrated in Figure 3.6, when half an equivalent
of a C,,H(2,,.,1)NR3+ onium ion contains an alkyl chain with n>10, the hydrophobic chain
effect will cause quantitative displacement of half the Na+ from the basal surfaces of both
types of tactoids. However, the position of the equilibrium between the two tactoid forms
will depend on the size of the NR3 head group. For R = propyl, butyl or pentyl, the head
group footprint effect causes preferential displacement of Na+ from the gallery region

defined by the two-layer tactoid. This two-layer structure optimizes the hydrophobic

interactions between onium ions while keeping the external surfaces hydrophilic for

75

solvation by water. Consequently, the tactoid equilibrium is shifted to the two-nanolayer
structure. Drying the tactoids leads to a regularly ordered heterostructure.
Conversely, for small head groups, R = methyl or ethyl, there is no head group
effect and the Na+ ions are randomly displaced from the basal surfaces. In order to
optimize the hydrophilic character of the basal surfaces, for solvation by water the tactoid
equilibrium is shifted toward the single nanolayer tactoid (cf., Figure 3.6). Drying the
suspension forms initially a hydrated homogeneous mixed ion intercalate. However,
further removal of the gallery water results in segregation of the organic and inorganic
ions within the galleries and the formation of layer stacking patterns characteristic of the

phase segregated end members.

Finally, we note that tactoids containing three nanolayers may be unstable in Na+-

. . + . .
fluorohectorite suspens1ons. We were unable to form 1:2 Na : onlum lon heterostructure.

Instead, a 1:1 staged heterostructure and an organo clay were formed.

76

 

Fig
and
Call

Na Na

 

 

 

 

 

 

 

Na Na
__> Na Na Na Na

3 Na _—
1 Na Na
iCnH(2n+I)N(CmH(2m+l))3+
l .
2N3

m=1, 2 "1:3, 4’ 5

’1 >10 Na Na

 

  

(Phase segregated)

 

(Homostructure)
(Heterostructure)

Figul‘e 3.6 Summary of the reaction pathway leading to heterostructured, homostructured,
and Phase segregated mixed ion intercalates formed by replacing 50% of the exchange
Cations in Na+-fluorohectorite with CnH(2n+1)N(CmH(2m+1))3+ cations (n 2 10; m = 1 - 5).

77

3.8 Concluding remarks.
We have elucidated the processes leading to the formation of staged

fluorohectorite heterostructure containing regularly alternating interlayers of inorganic
(Na+) and organic (alkylammonium) cations. Firstly, the alkyl chain of the onium ion
must be sufficiently long (n>10 carbon atoms) to quantitatively displace half of the Na+
ions through a hydrophobic binding effect. Secondly, the head group footprint must be
sufficiently large (m23 carbon atoms) to exclude binding of Na+ between onium ions.

These conditions favor the formation of bipolar two-nanolayer tactoids in aqueous

suspension with hydrophobic onium ions intercalated between the layers and hydrophilic

Na+ ions on the external surfaces. These dispersed bipolar tactoids should be ideally

suited for the adsorption of organic molecules from aqueous solution. Heterostructures

are formed when the bipolar tactoids are removed from suspension and dried.

78

3.9 References.

[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]

[13]

[14]
[15]
[16]
[17]
[18]
[19]

[20]

K. Ohtsuka, ChemMater. 9, 2039 (1997).

Y. Fukushima, S. Inagaki, J. Incl. Phenom. 5, 473 (1987).

U. Guth, S. Brosda. J. Schomburg, J. Appl. Clay Sci. 11, 229 (1996).

G. Lagaly, Solid State Ionics, 22, 43 (1986).

T. R. Jones, Clay Miner. 18, 399 (1983).

S. H. Xu, S. A. Boyd, Adv. Agron. 59, 25 (1997).

S. A. Solin, Adv. Chem. Phys. 49, 455 (1982).

R. Glaeser, J. Méring, Clay Miner. Bull. 2, 188, (1954).

R. M. Barrer, K. Brummer, Trans. Faraday Soc. 59, 959 (1962).

M. B. McBride, M. M. Mortland, Clay Miner. 10, 357 (1975).

S. Xu, S. A. Boyd, Soil Sci. Soc. Am. J. 58, 1382 (1994).

G. Brown, A. Weir, Proc. Int. Clay Conf. Stockholm 1, 27 (1963)

T. Nakato, D. Sakamoto, K. Kuroda, C. Kato, Bull. Chem. Soc. Jpn. 65, 322
(1992).

K. Urabe, I. Kenmoku, Y. Izumi, J. Phys. Chem. Solids 57, 1037 (1996).
K. Tamura, H. Nakazawa, Clays Clay Miner. 44, 501 (1996).

W. L. IJdo, T. Lee, T. J. Pinnavaia, Adv. Mater. 8, 83 (1996).

S. A. Buckingham, C. J. Garvey, G. G. Warr, J. Phys. Chem. 97, 10236 (1993).
G. G. Warr, T. N. Zemb, M. Drifford, J. Phys. Chem. 94, 3086 (1990).
L. Mercier, C. Detellier, Clays Clay Miner. 42, 71 (1994).

J. M. Barter, D. L. Jones, J. Chem. Soc. A 1531 (1970).

79

[21] D. J. Cebula, R. K. Thomas, J. Chem. Soc. Faraday I 76, 314 (1980).
[22] J. Fripiat, J. Cases, M. Francois, M. Letellier, J. Colloid Interface Sci. 89, 378
(1982).

[23] J. D. F. Ramsay, P. Lidner, J. Chem. Soc. Faraday Trans 89, 4207 (1993).

80

CHAPTER 4
Gallery Stacking Order of Fluorohectorites Interlayered with Organic
and Inorganic Exchange Cations: Evidence for the Formation of

Heterostructured Solid Solutions.

4.1 Abstract.

We report on the phases formed by the intercalation of onium (C16H33PBu3+) and
inorganic (Na+) ion mixtures in fluorohectorite clay with a layer charge density of 1.12 e'
per 020F4 unit cell. When the overall fraction of onium ions, fq, is in the range 0.35 S fQ S
0.50, a solid solution is formed in which the galleries regularly alternate in height at
values corresponding to the spatial segments of the organic and inorganic ion galleries at
ion composition fQ is 0.50. These thermodynamically stable heterostructured solid
solutions involve a partial mixing of Na+ ions in organic exchanged interlayers with a
retention of the distinctively heterostructured alternation of organic and inorganic
galleries.

Phase segregated Na+ clay and heterostructured solid solution clay are formed
when the Na" ion accommodation capacity of the organic interlayers is exceeded at onium
ion composition fQ is < 0.35. A two phase system is also produced when the overall
onium ion fraction fQ is > 0.5. The two phases at these ion compositions are a clay
heterostructure with 1:1 mixed Na+ and C115H33PBu3+ ion galleries and a phase segregated

C16H33PBu3+ exchanged organo clay. The alternating stacking order of organic and

81

inorganic galleries in mixed ion heterostructured clays is thus invariant with the overall
onium ion fraction f0.

Also, metal ion size and valency do not affect the heterostructure formation
process. Thus, an exfoliated inorganic ion exchanged clay precursor is not a requirement
for heterostructure formation. Hydrated gallery metal ions, when present, favor interlayer

positions on either side of a C16H33PBu3+ onium ion exchanged organic interlayer.

82

4.2 Introduction.

Previously, we have reported on fluorohectorite clay heterostructures which have
regularly stacked Na+ and C16H33PBu3+ ion exchanged galleries [1]. These mixed ion clay
structures are reminiscent of a staged graphite intercalation compound. For example,
assuming a rigid graphite layer model [2], one interlayer of intercalants is followed by n
sheets of graphite in a stage n structure [3,4]. A stage n graphite structure can easily be
modified into a stage (nil) arrangement. In other words, the layer stacking order n is
determined by the number of layers of host separating layers of guests.

Until now, we have only reported on the stacking order of Na+ and C16H33PBu3+
mixed ion clay structures where both cationic species are present in a 50:50 ratio. Here,
we investigate the gallery stacking order of mixed ion fluorohectorite clays in which the
total exchange ion composition ranges from completely organic, onium ion fraction f0 =
1.0, to completely inorganic, onium ion fraction fQ = 0.0.

A homoionic smectite clay normally has a fixed amount of charge balancing
interlayer cationic species. The number of exchange ions in a homoionic clay is fixed by
the clay cation exchange capacity and, of course, by the valancy of the interlayer cations.
Reversible staging behavior with the appearance of vacant interlayers, as seen in graphite,
is therefore implausible [3]. Only a peculiar irreversibly altered stage 2 like clay structure
has been observed so far. This modified clay has a discernible regular arrangement of
empty clay interlayers and Li+ exchanged galleries. The empty interlayers are formed by a
partial but irrevocable Li+ migration through heating of the pristine clay material. In this
process, every other Li+ gallery is depleted of all its Li“ ions to fill vacant octahedral sites

in both adjoining fluoro mica-like clay layers [5]. Further Li+ migration of the next Li+

83

interlayer is precluded since the hitherto created trioctahedral clay layer structure has no
additional vacant octahedral sites left.

Clay gallery differentiation may nonetheless be expected when various gallery
species are present. Our previously published 1:1 Na+ and C,,I'I(2,,,.1)N(C,,,H(2m,j))3+ mixed
ion heterostructures, with n 2 12 and 3 S m S 5, clearly demonstrate this gallery
dissimilarity [6]. Segregation of both cationic species creates distinctive galleries in these
heterostructures. Individual organic and inorganic exchanged interlayers are stacked in a
regular consecutive arrangement. The mixed ion heterostructure is thus made from
interstratified galleries.

A 50:50 percent Na+ and C16H33PBu3+ interlayer ion mixture results in the
development of two alternating galleries. Likewise, when the above gallery species are
present in a 66:33 ratio, a regularly 2:1 inorganic : organic gallery stacked structure may
be formed. Such an interstratified mixed ion product resembles a stage 3 graphite
intercalation compound. Of course, the formation of such a clay gallery arrangement is
entirely hypothetical. Production of a 1:1 mixed ion heterostructure instead is not
precluded. In this situation, the 1:1 heterostructured phase will be accompanied by
supplementary Na+ exchanged interlayers. The excess Na+ interlayers may be randomly
interstratified throughout the 1:1 heterostructure, but can also separate as a second
(homostructured end member) phase. On the other hand, a regular interstratification of
the supplementary Na” galleries and a 1:1 mixed ion heterostructure yields a 2:1
(inorganic : organic) mixed ion heterostructure.

Similarly, a 1:2 (inorganic : organic) mixed ion heterostructure may form with a

33:66 percent ratio of Na“ and C15H33PBu3” gallery ions. Thus, the gallery onium ion

84

fraction fQ creates the possibility of regular interstratification, but does not dictate which
staged like structure forms. Also, mixing of cationic species within an interlayer domain
may still occur for all heterostructures. Commingling of cationic species within the same
interlayer region appears when the segregation process is not complete. Furthermore,
incorporation of metal ions in an organic exchanged interlayer may result from cation
entrapment [7]. This entrapment process arises when exchange sites on the edges of metal
ion exchanged clay layers are occupied with alkylammonium ions and form a
hydrophobic barrier. Diffusion of hydrated interlayer metal ions out of the interlayer
region is then inhibited by the presence of this barrier and it may lead to cation
entrapment.

The gallery stacking order in mixed ion clays is clearly dependent on the relative
amounts of intercalated organic and inorganic exchange ions. The exchange ion fraction
is regulated by the direct addition of a desired amount of onium ions in a partial ion
exchange reaction of a Na+ fluorohectorite clay. In this paper we report how this gallery
stacking order changes as function of onium ion content. We also examined what effect
the interlayer metal ion type has on the heterostructure formation process. For this,
homoionic metal ion exchanged clays were prepared wherein the metal ion size and

valancy were varied prior to onium ion addition in a partial ion exchange reaction.

4.3 Experimental.
Clay preparation: Synthetic fluorohectorite, L11.12°[Mg4.33L11.121818020F4
(Corning, Inc.), denoted as Lii-FH, was converted to the Na“, Ca2+, Baz”, Al3+ and Ce”

forms by exchange with 0.1 M solutions of the corresponding chloride salts. Excess salts

85

were removed by washing of the clays with deionized water. The procedure was repeated
to ensure complete exchange. [C16H33P(C4H9)3+]Br was obtained from Lancaster and was
used without further purification. Related [Cid-133N(CmH(2m,l))3+]Br salts (with 3 S m S
5) were prepared as previously described [6].

Mixed cation fluorohectorite: Mixed cation derivatives of fluorohectorite were
prepared in a simple one-step/one-pot procedure by the direct addition of a desired
amount of onium ions to the Na+ clay suspension. The amount of [C 16H33PBu3+]Br
surfactant added corresponded to the desired fraction of the cation exchange capacity. The
cation exchange capacity (CEC) of fluorohectorite, as measured by the ammonia selective
electrode method [8], was 121 mqu 100g. In a typical mixed ion synthesis, 0.30 g of Na+
clay was first dispersed in about 100 ml deionized water and then a desired quantity of the
surfactant was slowly added. Reaction mixtures were stirred for at least 48 hours at
ambient temperature. All products were washed free of excess salt and air dried. A
similar procedure was used when the influence of the gallery metal ion size and valency
was investigated, except that for these products a half a CEC equivalent of C16H33NBu3+
ions was used.

Product characterization: XRD patterns were recorded on a Rigaku rotaflex
diffractometer equipped with Ni-filtered Cu-Ka radiation. All clay samples were analyzed
as ordered films. A Cahn 121 TG therrnogravimetric analyzer was used to determine the
onium ion content of the clays. The samples were heated at 5 °C per minute up to 800 °C
to determine the weight loss due to onium ion combustion. In addition, elemental analysis

was used to ascertain the onium ion to metal ion ratio.

86

4.4 Heterostructure formation at a half an exchange equivalent of C1,,H33NBu3+
onium ions, f0 = 0.5.

Onium ions with long alkyl chains have generally a higher affinity for clay
exchange sites than hydrated metal ions. Van der Waals interactions between the
surfactant alkyl chains drive this preferential onium ion uptake. Thus, a quantitative
uptake of onium ions occurs immediately when surfactant is directly added to a clay
suspension. Any amount of surfactant added therefore participates fully in our route I
(Figure 2.1) heterostructure formation approach. However, all clay exchange sites can
engage in onium ion uptake.

Figure 4.1 shows the XRD patterns for air dried products formed over time when
a half an exchange equivalent of C16H33NBU3+ ions are added to a Na+ fluorohectorite
suspension to obtain a product in which the fraction of onium ions fQ is 0.5. The 40 A
peak displayed in all time derived XRD profiles corresponds to the C16H33NBu3+ and Na“
mixed ion heterostructured intercalate. This spacing agrees well with the sum of the basal
spacings of C16H33NBu3+-FH (27 A) and Na+-FH clay (12.4 A). Homoionic end
members, C16H33NBu3+-FH and Nai-FH, are also clearly present in relatively large
amounts immediately after onium ion addition (10 minutes). However, the homoionic end
member products participate fully in heterostructure formation as their distinctive XRD
reflections disappear completely within 8 hours reaction time.

Fully C16H33NBu3+ exchanged organo clay is initially formed since all of the clay
exchange sites are involved in onium ion uptake. The organo clay phase is thus a kinetic
product that appears when the surfactant is directly added to the Na“ clay suspension.

Hence, the reaction mixture also contains unreacted Nai-FH clay. Onium ion

87

redistribution then shifts the homoionic end member clays to complete heterostructure
formation.

The gallery stacking order of the 1:1 Na+ and C16H33NBu3+ heterostructured
intercalate is evidently not influenced by cation redistribution processes. Kinetic factors,
such as onium ion redistribution, seem insignificant when the formation of 2:1 (inorganic
: organic) mixed ion heterostructures is attempted by means of direct surfactant addition.

pathway I.

88

 

 

 

 

 

 

  
  
 

 

 

 

 

 

 

40 A
.4
CO
«1: $13
>. c:
a (\l
5 A 24 hours
3
o 8 hours
.2 :1:
E 4 hours
é) * 96
* 2 hours
9(-
* 1 hour
10 min
' 1 ' 1 ' 1 TJl_"_l—'—T_

2 4 6 8 10 12 14
Degrees (20)

Figure 4.1 Time dependence of the XRD patterns (Cu-K0,) of air dried products formed
when a Na+-FH clay suspension is exchanged with half a cation exchange equivalent of
C16H33NBu3+ ions. A 40 A mixed ion heterostructure is already the dominant phase
within 10 minutes. Heterostructure formation is accompanied by the production of
C16H33NBu3+-FH organo clay (diffraction peak at 27 A marked *). The unreacted Na+-FH
exhibits a diffraction peak at 12.4 A marked **. The two homoionic end member clays
are fully converted to the heterostructured clay within 24 hours reaction time.

89

4.5 Mixed ion clays formed when homoionic Na*-FH is exchanged with less than
half a CEC equivalent of C16H33PBII3+ ions, 0.35 S fQ S 0.5.

A theoretical 2:1 mixed ion fluorohectorite heterostructure, made from two Na+
and one C1¢5H33PBu3+ exchanged galleries, is expected to have a 53 A repeat spacing
when the two interlayer types are regularly stacked (homoionic C15H33PBu3+-FH clay has
a basal spacing of 28 A while homoionic NaJ-FH clay has a 12.4 A repeat unit). Such a
structure is not observed when an one third exchange equivalent of C16H33PBu3+ ions is
added to a fluorohectorite clay suspension. Rather unexpectedly, the XRD pattern shows
only reflections characteristic of a 40 A 1:1 mixed ion heterostructured product! Even
phase separated Nai-FH clay is not observed. Figure 4.2 shows the XRD patterns for
products obtained at onium ion fractions fQ = 0.50, 0.45, 0.40 and 0.35. The XRD
patterns for all air dried materials are practically indistinguishable from each other. XRD
data for these Na+ and C16H33PBu3.+ mixed ion air dried products are compared in Table
4.1. The d-spacings of the third order diffraction peaks decrease only slightly with a
decreasing onium ion content. Surprisingly, the reduction in repeat distance of the
heterostructures is less than 0.6 A over the whole fQ composition range. The first order
diffraction peaks show an opposite trend but d(001) values derived from these peaks are
generally less accurate.

Domain sizes for the four mixed ion heterostructures with the different gallery ion
compositions were calculated for the 001 and 003 XRD diffraction peaks, Table 4.1. The
domain heights derived for both these peaks correspond to an uninterrupted stacking of

approximately three to four heterostructure repeat units.

90

 

k

 

 

 

 

 

 

Relative Intensity
p—h

 

 

 

 

 

Degrees (20)

Figure 4.2 XRD patterns (Cu-K0,) of mixed ion products formed when 50, 45, 40 and 35
percent of the exchange cations of Nai-FH clay suspension are replaced by C16H33PBu3"
ions (fQ = 0.50, 0.45, 0.40 and 0.35, respectively). All products were formed after a
reaction time of at least 48 hours at ambient temperature before they were washed and air
dried for analysis.

91

Table 4.1 Comparison of XRD data of the mixed ion products shown in Figure 4.2. The
heterostructured clays are formed when a Na+-FH clay suspension is exchanged with Q+

onium ions, where Q+ is C16H33PBu3+.

 

 

 

 

 

Onium ion PWHM Domain PWHM Domain
d(001) . d(003)

fraction [A] d(001) srze [A] d(oo3) size
fQ ‘ [20]l2 14001) [A]9 [29? I«(003) [A]9
0.50 40.2 0.676 1 16 13.36 0.532 148
0.45 40.5 0.607 130 13.27 0.501 157
0.40 40.9 0.629 125 13.20 0.484 163
0.35 41.4 0.648 121 13.17 0.444 177

 

 

 

 

 

 

 

 

9 The cation exchange capacity (CEC) of the fluorohectorite used is 121 [meq/lOOg].
8 Peak Width at Half Maximum in 20 of the (001) order diffraction peaks.
9 The mean domain size calculated according to the Scherrer equation, L = kA/Bcos(0), p

where k=0.89, 2:1.541871 A and [3 is the Peak Width at Half Max. in radians of 20.

These X-ray results imply that a single heterostructure is formed over the
composition range 0.35 S fQ S 0.50. Since the gallery cation compositions are quite
different for these mixed ion clays, the intercalates are best regarded as alloys or solid
solutions. The variable onium ion content is verified by therrnogravimetric analysis for
the reaction products, Figure 4.3.

The calculated and observed weight losses for the materials analyzed in Figure 4.3
are summarized in Table 4.2. The Q+ : Na" ratios in the mixed ion products were
calculated on basis of the weight loss due to air oxidation of the gallery onium ions. The

observed mixed ion ratios compare well with the expected fractions.

92

 

 

I I I I

p—A
O

N
O

llLlllllllllllljllllllllillll

Weight 1088 percent
17:

N
‘JI

I I Y I l I I I T I I I I I I U I I T I I I I I l

 

 

 

Figure 4.3 TGA curves of the products shown in Figure 4.2. The air dried materials were
formed by replacing 35-50% of the exchange ions in a Nai-FH clay suspension with
C15H33PBu3+ ions. Desorption of physisorbed gallery water occurs below 200 °C. The
weight loss between 400 and 700 °C corresponds to the decomposition of interlayer
onium ions.

93

Additional evidence in support of the onium ion composition was obtained by
elemental analysis on the same products, Table 4.3. Both analytical methods show a
somewhat lower than anticipated onium ion content for all mixed ion clay products.
However, both confirm that the clay intercalates have nearly the intended gallery ion
compositions. Also, every sequential mixed ion clay in this series differ by an equal
amount of onium ions.

A physical mixture of a regular 1:1 mixed ion heterostructure (50:50 C16H33PBu3+
and Na+ ion ratio, fQ is 0.50) and Na+-FH clay corresponding to an overall 35 % onium
ion content (fQ = 0.35) exhibits a clear Na+-FH XRD signature at 12.4 A. A phase
segregated Na+-FH clay is thus easily detectable by XRD. Therefore, we conclude that
none of the samples in Figure 4.2 contain phase separated N a+ clay.

In contrast, no evidence of a segregated Nai—FH phase is observed when an
exfoliated Nai-FH clay suspension is added to a 1:1 C15H33PBu3+ and Na+ mixed ion
heterostructured clay suspension at an overall onium ion ratio fQ = 0.35. This reaction
mixture was allowed to equilibrate for one minute before the product was centrifuged and
air dried for analysis. The XRD pattern (not shown) exhibited only diffraction peaks due
to heterostructured mixed ion intercalate, while the distinctive Nai-FH XRD reflections
were not observed. This result indicates that an onium ion - Na+ ion redistribution
reaction has occurred with retention of a mixed ion intercalate structurally analogous to a

1:1 mixed ion heterostructure.

94

Table 4.2 TGA determination of the C15H33PBu3+ exchange ion compositions (fQ) for
mixed ion heterostructured solid solutions produced when Na+-FH is exchanged with a x

% CEC equivalent of Q“, where Q+ is C15H33PBu3+.

 

 

 

 

 

 

Fraction of onium Expected organic Observed organic fQ for the mixed ion
ions Q+ addedg weight lossg, wt.% weight lossg, wt.% product
0.50 23.9 23.3 0.49
0.45 22.0 22.0 0.44
0.40 20.1 20.1 0.39
0.35 18.0 18.0 0.35

 

 

 

 

a The cation exchange capacity (CEC) of the fluorohectorite is 121 [meq/ 100g].
L’ Expected weight loss for a water free (dehydrated) N a1 and Q+ mixed ion clay.

9 Observed weight loss due to air oxidation of Q+.

Table 4.3 Elemental analyses (ICP) of the exchange ion compositions for Na+ and 0“
mixed ion heterostructured solid solutions produced when Na+-FH is exchanged with Q”,

where Q+ is C16H33PBu3+.

 

 

 

 

 

 

Fraction of onium fQ based on
Wt. % C Wt. % Si
ions Q+ addedi Wt. % Cl2
0.50 17.1 20.7 0.49
0.45 15.7 20.3 0.46
0.40 14.2 21.7 0.39
0.35 12.9 21.8 0.35

 

 

 

 

Q The cation exchange capacity (CEC) of the fluorohectorite is 121 [meg/100g].
9 Based on (0*. Nam.12-[MgraLirolsuozoFa

95

 

 

XRD is known to be effective in detecting a random interstratification of clay
layer stacking patterns. For example, mixed layer smectite - illite systems contain layer
stacking sequences that are not periodic [9]. Hence, such interstratified clay structures
exhibit irrational 001 diffraction harmonics. In contrast, the XRD patterns of the air dried
C16H33PBu3,+ and Na” mixed ion fluorohectorite heterostructures at compositions in the
range 0.35 S fQ S 0.50 exhibit rational diffraction patterns with nearly identical d-
spacings. Therefore, we conclude that the products at these compositions are
heterostructured solid solutions. The probability of random insertion of Na‘“ galleries into
a 1:1 heterostructure to form an interstratified intercalate is precluded by the diffraction
data. Moreover, a phase segregated homoionic Na+-FH clay phase is observed in the XRD
pattern when the overall onium ion content fQ is < 0.35. This indicates that phase
segregated homoionic Nai-FH is thermodynamically more stable than a randomly
interstratified arrangement of C16H33PBu3+ and Na“ galleries at this composition.
However, in the composition range 0.35 S fQ S 0.50, a regularly alternating sequence of
C16H33PBu3+ and Na” galleries is maintained, despite the variation in exchange ion
composition.

The structure of a heterostructured C16H33PBu3+ and Na“ mixed ion clay
synthesized with half an exchange equivalent onium ions, fQ = 0.50, is well understood
[1,6]. As shown in Figure 4.4 (A), equal concentrations of C16H33PBu3+ and Na” ions are
segregated into separated galleries in a regularly alternating stacking sequence. However,
the structure of the solid solutions formed at fQ = 0.45, 0.40 and 0.35 is more complex. At
least two alternative models could explain the observed XRD patterns for these

intercalates.

96

One explanation, Model 1, considers the gallery cation population in mixed ion
heterostructures. The total number of charge compensating gallery cations is regulated by
the clay layer charge, but the specific position of the charge balancing cations in the
galleries is somewhat adaptable [10,11]. Tetrahedrally charged clays have interlayer metal
ions associated with the ditrigonal cavity [10]. Clays with layer charge arising in the
octahedral sheet have hydrated metal ions positioned at the center of the interlayer [10].
The synthetic fluorohectorite clay used for our mixed ion materials has layer charge
arising in the octahedral sheet only. This particular charge location allows for effective
charge compensation by the gallery ions to occur on either side of the layer. A preference
for interlayer segregation can arise when one gallery (e.g., the onium ion gallery) is
precluded from accommodating hydrated metal ions. Previous investigations on the
nature of the mixed ion heterostructure formation process indicated that a large onium ion
head group, such as those in C16H33PBu3+, precludes commingling of unlike cationic
species within one interlayer [6]. As the relative population of Na+ ions increases at fQ <
0.50, the metal ions may prefer to occupy an inorganic than organic interlayer due to
more favored hydration effects. In other words, the cation population density on both
sides of a clay layer does not have to be the same, as illustrated in Figure 4.4 (B).
However, this uneven gallery distribution of exchange ions would lead to charge
separation in excess of 10 A, the thickness of a silicate layer. This is energetically
unfavorable. Consequently, we disfavor a heterostructured solid solution in which the
C15H33PBu3,+ and Na+ ions are segregated into regularly stacked but unequally populated

galleries.

97

Model II for the heterostructured solid solutions over the composition range 0.35
S fQ S 0.50 entails a partial mixing of cationic species within a gallery. Cation mixing
may occur for kinetic reasons in mixed ion clays when the cation segregation process is
not complete. For instance, the diffusion of hydrated metal ions out of a clay interlayer
can be impeded when onium ions with long alkyl chains migrate inwards. Some metal
ions are for that reason trapped within the clay interlayer when onium ions are
intercalated. Cation entrapment causes commingling of cationic species within organo
interlayers [7].

From our previous work [6], we know that the onium tri-butyl head group size
precludes commingling of cationic species within one interlayer. However, this occurs
when the onium ion composition fQ is 0.50. All exchange sites within an organic
interlayer are occupied with onium ions at this ion composition. The complete layer
surface is covered with organic material. Mixing of cationic species is therefore
prevented, and a mixed ion heterostructure as shown in Figure 4.4 (A) is formed.

However, not all exchange sites within an organic interlayer are occupied by
onium ions when a partial mixing of cationic species occurs at ion composition fQ is <
0.50. More importantly, the clay interlayer surface is not completely covered with
hydrophobic organic material. Metal ions can take position at these uncovered exchange
sites. However, these type of galleries are still predominantly organic in character. Such a
limited cation mixing allows for the formation of a mixed ion heterostructured solid
solution with regularly alternating organic and inorganic galleries at onium ion

compositions fQ is < 0.50. The structure of such a material is illustrated in Figure 4.4 (C).

98

A C16H33PBU3+ intercalated interlayer has thus a certain metal ion accommodation
capacity at onium ion compositions 0.35 S fQ S 0.50. This ion mixing within the
predominantly organic exchanged galleries is a thermodynamically driven process. This is
evident, since these heterostructured solid solution intercalates may also be obtained by
the spontaneous recombination reaction of the homoionic parent end member clays at
overall onium ion compositions 0.35 S fQ S 0.50.

In a C16H33PBu3+ organo fluorohectorite clay, the clay interlayer surfaces are
completely covered with a layer of onium ion head groups, while the alkyl tails on top are
inclined at a 33° angle with respect to the clay surface [6]. Because of these large onium
ion head groups, there are optimal van der Waals interactions between all onium ion alkyl
chains (head group chains and alkyl tail). Na+ ions that reside in C16H33PBU3+ exchanged
galleries do not notably influence the direction and orientation of all these onium ion
alkyl chains. Thus, relocation of the commingled metal ions from the predominantly
organic interlayers does not generate a further optimization of the van der Waals
interactions between the onium ion alkyl chains. Moreover, the presence of metal ions
next to sterically hindered onium ions may improve clay layer charge compensation. As
such, a partial but limited mixing of cationic species may lead to a thermodynamically
stable mixed ion heterostructured solid solution in the onium ion composition range 0.35
S fQ S 0.50.

The gallery height of an onium ion exchanged interlayer is normally dependent on
the population density of the intercalated onium ions. However, the packing of onium
ions is not rigid, but can exhibit some liquid like character [12]. The onium ion alkyl tails

in C16H33PBu3+ exchanged interlayers are oriented with a very modest inclination [6]. A

99

decrease in the number of onium ions within an organic gallery of a mixed ion
heterostructure may not result in a large decrease in gallery height. A small but gradual
decrease in the repeat distance along the layer stacking direction is detectable for the
mixed ion structures that contain progressively less onium ions. For instance, the d-
spacing decreases by 0.57 A upon reducing fQ from 0.50 to 0.35.

The Na+ ion accommodation capacity of the C.(5H33PBu3+ intercalated interlayers
of a heterostructured solid solution is exceeded when the onium ion composition fQ is <
0.35. As such, phase segregated homoionic Nai-FH clay phase is formed next to the
heterostructured solid solution phase.

Also, two segregated fluorohectorite clay phases are produced when the onium ion
composition fQ is > 0.50. The XRD pattern indicates a 1:1 C15H33PBU3+ and Na+ mixed
ion heterostructured material and homoionic C16H33PBu3+ exchanged organo clay. A
mixed ion heterostructure has evidently no extra onium ion accommodation capacity

when the onium ion composition fQ exceeds 0.50.

100

Figure 4.4 Model structures of C16H33PBu3+ and Na“ mixed ion clays that depend on
onium ion composition fQ. All cationic species in structure (A) are segregated into
separate galleries and stacked in a regularly alternating fashion when the fraction of
intercalated onium ions is half, fQ is 0.50. Two models describe heterostructured solid
solutions that arise when less than half a fraction of onium ions are exchanged into the
fluorohectorite clay, 0.35 S fQ < 0.50. However, both the organic and inorganic interlayers
of these mixed ion heterostructured solid solutions are still regularly interstratified. Model
1, structure (B), assumes an uneven distribution of cationic species within successive
galleries. Structure (B) has a balanced quantity of organic and inorganic galleries
although the distinctive galleries are unequally populated. Model 11 assumes a partial
mixing of cationic species. Structure (C) represents this situation where some metal ions
are commingled within the organo gallery. Thus, the C16H33PBu3+ intercalated interlayer
has thus a certain Na+ ion accommodation capacity.

101

A.

 

‘0 0 0 Q 0 g
[Na Na Na Na Na Na:

7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8 0 Q 0 0 07
LNa Na Na Na Na N;
rot-0.5
P0 0 0 07
NaNaNaNaNaNaNaNa
3 9 Q 9S
Na Na Na Na Na Na Na Na
rQ<o.s Model]

 

L ;

QNaQQNaE

L

Tia Na Na Na Na Na\
LQ Q Na Q Na 6
Na Na Na Na Nah;

; l
L \

fQ< 0.5 Model II

 

 

 

 

 

 

 

 

 

 

 

102

Clearly, C16H33PBu3+ and Na+ mixed ion fluorohectorite clay can form
heterostructures where both the cationic species are partially segregated into different
galleries, model II. The two unlike gallery types are regularly interstratified but can only
be stacked in a 1:1 alternating manner. This heterostructured stacking pattern is
independent of the gallery onium ion composition fQ. Thus, the formation of
heterostructures with 2:1 ordered Na+ and C16H33PBu3+ galleries is precluded when the
onium ion composition fQ is < 0.50. The production of a reverse structure, with 1:2
ordered Na+ and Cur,H33PBu3+ galleries, is also prohibited at onium ion compositions fQ >
0.50. Therefore, we were not able to form stage 3 like mixed ion fluorohectorite clay
structures. Moreover, the same effects were observed when n—Pr, n-Bu or n-Pe terminated
alkylammonium ions were used instead of the discussed tri(n-Bu) alkyl phosphonium

ions.

4.6 Metal ion valency and 1:1 mixed ion heterostructure formation.

Alkaline earth exchanged clays do not delaminate in water whereas the alkali Li+
or Na” metal exchanged clays do exfoliate in water [13]. We have investigated the effect
of metal ion valency and size on the heterostructure formation process. For this,
homoionic Li+, Nai, Ca2+, Ba2+, Al3+ and Ce3+ exchanged clays were treated with half an
exchange equivalent of C16H33NBU3+ ions, fQ = 0.50. The XRD profiles for the obtained
air dried reaction products are shown in Figure 4.5. These patterns show that all metal
exchanged clays do form heterostructured products. The stacking repeat distance for the

air dried mixed ion products and the homoionic metal clays are summarized in Table 4.4.

103

Table 4.4 Comparison of the X-ray basal spacings for homoionic A+-FH clay and A1, Q+

r11 mixed ion products, where Q“ is C161133NBu3+ and A‘” is Li“, Na“, Ca2+, Ba“, AP+

 

 

 

 

 

 

 

 

or Ce“.
Interlayer ion, AM-FH A“, Qi-FHHS Differencea
A“ 1A1 1A1 1A1
Li+ 12.4 39.4 27.0
Na+ 12.4 39.4 27.0
Ca2+ 15.0 41.7 26.7
13a2+ 12.5 39.3 26.8
A13+ 15.5 42.3 26.8
Ce3+ 15.4 42.1 26.7

 

 

 

 

9 The basal spacing of homoionic C1(5H33,NBu3+ exchanged fluorohectorite clay is 26.9 A.

A different amount of gallery water is associated with each particular metal clay.
Each homoionic metal ion exchanged clay has its own specific basal spacing. However,
all observed diffraction peaks for the mixed ion materials correspond to stacking repeat
distances that agree perfectly well with the sum of the basal spacing of organo clay and
the corresponding homoionic metal clay.

We conclude that metal ion size and charge do not direct the heterostructure
formation process. Demixing of cations with subsequent gallery differentiation is
evidently not initiated by the hydrated metal ions but rather by the onium ions [6]. Also,

an exfoliated clay precursor is not needed for 1:1 mixed ion heterostructure arrangement.

104

 

 

 

 

3'1 A 14.0 A

 

 

Relative Intensity

 

 

 

 

 

 

Degrees (20)

Figure 4.5 XRD patterns (Cu-K0) of products formed when homoionic Li“, Na”, Ca2+,
Baz”, A13+ and Ce3+ fluorohectorite clays are exchanged with half a cation exchange
equivalent of C16H33NB113+ ions each, fQ = 0.50. All clay slurries were reacted for at least
48 hours at ambient temperature before the materials were washed, isolated and air dried.

105

4.7 Preservation of the distinctive gallery properties associated with the interlayers
of mixed ion heterostructured materials.

Mixed ion clay assemblies with regularly stacked organic and inorganic galleries
are truly heterostructured materials as they involve the direct combination of both end
member clay phases into a one phase structure. Distinctive gallery properties associated
with each interlayer type are therefore preserved.

The inorganic interlayers of 1:1 Na“ and Cu5H33NBu3+ mixed ion fluorohectorite
heterostructures, fQ = 0.50, are capable of ion exchange. This exchange can be executed
with great selectivity. Intercalation of the A113;7+ Keggin ion in homoionic Na+-FH clay
expands the clay gallery height by an extra 6-7 A. A similar 6 A increase in gallery height
is observed when the inorganic interlayer of a 1:1 mixed ion heterostructure is pillared
with the A113,7+ pillaring agent, Figure 4.6. Exchange selectivity is validated as the organic
gallery remains unaltered. Moreover, the gallery stacking integrity of the mixed ion is not
affected by the selective ion exchange since no homoionic end member materials are
formed during the ion exchange.

The galleries of organic exchanged clays can be penetrated and expanded with
organic solvents or even polymers. Swelling of the organophilic gallery with organics
increases the interlayer distance. Epoxy precursor, (Epon 828, poly(bisphenol A-co-
epichlrohydrin), Shell), [13] swells homoionic C115H33NBu3+ exchanged clay by roughly 4
A. A 1:1 Na+ and C16H33NBu3+ mixed ion heterostructure is expanded with 7 A when
swollen with the Epon 828 epoxy precursor, Figure 4.7. The small disagreement in
interlayer expansion between both swollen states was caused by the homoionic

C16H33NBu3+ organoclay. This clay was not swollen to full capacity whereas the organo

106

gallery of the 1:1 mixed ion heterostructure was expanded to the maximum. Nevertheless,
the organophilic gallery of the mixed ion heterostructured material exhibits undiminished
organic interlayer characteristics.

Both the inorganic ion exchange- and the organic interlayer swelling- experiments
show that the individual gallery properties of 1:1 C16H33NBU3+ and Na+ mixed ion
heterostructures are comparable to those of the homoionic metal- and organo- clays.
Thus, the arrangement of distinct interlayers in a heterostructured manner does not affect
the performance associated with each gallery type. However, a heterostructured ordering
of unlike galleries has an extra added characteristic. The distinctive gallery properties are
also coupled and can cooperate next to each other. For instance, the dispersion of an
organo modified clay in water can be a difficult process. The wetability of such a clay
system is greatly improved when the organic interlayers are arranged in a heterostructured

form.

107

 

46

23

15.3 A
1114A

40
19.7 A

 

Relative Intensity

 

 

 

Degrees (20)

Figure 4.6 XRD patterns (Cu-Kc) of homoionic air dried Nai-fluorohectorite (A) and
homoionic air dried A1137+ pillared fluorohectorite clay (B). The inorganic interlayer of
the air dried 1:1 C16H33NBu31' and Na“ mixed ion fluorohectorite heterostructure (C) is
selectively pillared with A113“ ions in air dried heterostructured derivative (D). The
gallery height expands with an extra 67 A upon pillaring of both the homoionic and
mixed ion materials.

108

 

 

in

 

 

Relative Intensity

27.7 A :

Lo";
T A

 

 

 

 

 

I ' I ' I ' l '

2 4 6 8 10
Degrees (20)

Figure 4.7 XRD patterns (Cu-K1,) of materials obtained when an organic interlayer is
swollen with epoxy (Epon 828) precursor. The gallery height of homoionic C16H33NBu3+-
FH clay (A) expands by 4 A when the epoxy precursor penetrates the organic interlayer to
give epoxy swollen material (B). A slightly larger 7 A interlayer swelling occurs when the
organic gallery of a 1:1 Na+ and C15H33NB113+ mixed ion heterostructure (C) is swollen
with the epoxy precursor to give heterostructured derivative (D).

109

4.8 Concluding remarks.

Our MM and C15H33NBU3+ mixed ion heterostructured materials consist of
regularly interstratified organic and inorganic galleries. The consecutive arrangement of
distinctive galleries couple and preserve the characteristics associated with these
interlayer types. However, this heterostructured gallery order is independent of onium ion
composition fq. Both gallery types in heterostructured clays can only be stacked in a
continuously alternating manner. Thus, a 2:1 ratio of distinctive galleries does not
produce a corresponding 2:1 interstratified mixed ion heterostructure. Instead, a mixed
ion heterostructured solid solution material may form when the gallery onium ion ratio is
in the range 0.35 S fQ S 0.50. The mixed ion heterostructured solid solution phase is
accompanied by segregated homoionic parent end member clay when the onium ion
composition fQ is outside this range.

The formation of a mixed ion heterostructured solid solutions involves a
partial mixing of metal ions within onium ion intercalated galleries when the onium ion
composition fQ is in the range 0.35 S fQ S 0.50. Limited mixing of cationic species in
C16H33PBu3+ and Na+ mixed ion heterostructured solid solutions is a thermodynamically
driven process. As such, our heterostructured solid solution clay materials are stable
against decomposition into homoionic parent end member clays.

Metal ion size and charge do not influence the heterostructure formation process.
Ion demixing into separate alternating galleries is therefore completely dominated by the
onium ions. The regularly stacked organic and inorganic galleries are heterostructured

since they combine and conserve the properties of both end member clays into a one

110

phase material. As such, our heterostructured materials that accommodate both these

interlayer types are best described as interstratified rather than staged structures.

11]

4.9 References.

[1]
[2]

[3]
l4]

[5]
[6]
[7]
[8]
[9]

[10]
[11]
[12]

[13]

[14]

W.L. IJdo, T. Lee, T.J. Pinnavaia, Adv. Mater. 8, 79 (1996)

K. Okino, H. Touhara, Comprehensive Supramolecular Chemistry, Chapter 2;
Graphite and Fullerene Intercalation Compounds (Atwood, Davies, MacNicol,
Vo'gtle; editors) 7, 25 (1996)

SA. Solin, Adv. Chem. Phys. 49, 455 (1982)

SA. Solin, Annual Review of Materials Science (Kaufmann, Giordmaine,
Wachtman; editors) 27, 89 (1997)

K. Urabe, I. Kenmoku, Y. Izumi, J. Phys. Chem. Solids 57, 1037 (1996)

W.L. IJdo, T.J. Pinnavaia, J. Solid State Chem. 139, 281 (1998)

S. Xu, S.A. Boyd, Soil Sci. Soc. Am. J. 58, 1382 (1994)

E. Busenberg, C.V. Clemency, Clays Clay Miner. 21, 213 (1973)

RC. Reynolds, Jr., X—Ray Diffraction and the Identification and Analysis of Clay
Minerals, Oxford University Press (1989)

J. Greathouse, G. Sposito, J. Phys. Chem. B 102, 2406 (1998)

PC. Chang, N.T. Skipper, G. Sposito, Langmuir 14, 1201 (1998)

E. Hackett, E. Manias, E.P. Giannelis, J. Chem. Phys. 108, 7410 (1998)

R. Kjellander, S. Marcelja, R.M. Pashley, J.P. Quirk, J. Phys. Chem. 92, 6489
(1988)

T. Lan, T.J. Pinnavaia, Chem. Mater. 6, 2216 (1994)

112

CHAPTER 5
The Relationship Between 2:1 Clay Layer Charge Location and the 1:1

C1,,I-I33NBu3+ and Na‘ Mixed Ion Heterostructure Formation Process.

5.1 Abstract.

Heterostructured mixed ion clays may form when half an exchange equivalent of
C16H33NBu3+ ions are exchanged into, for instance, a homoionic Na+ fluorohectorite clay.
Such a 1:1 C15H33NBu3+ and Nai mixed ion heterostructured clay has both cationic
species segregated into separate galleries, but with the distinct interlayers arranged into a
regularly alternating fashion. Here we report on the relationship between the clay layer
charge location and the 1:1 C16H33NBu3+ and Nai mixed ion heterostructure formation
process.

Fluorinated clays were synthesized with specific charge density locations. Thus,
we were able to regulate the layer charge that arises from specific substitutions in the
octahedral and/or tetrahedral sheets of the 2:1 clay structure. Clays with contrasting layer
charge locations, such as fluorohectorite (octahedral charge only) and fluorosaponite
clays (tetrahedral charge only), were produced through a fluoride flux method. In
addition, two clays were produced in which the layer charge was carried on the octahedral
as well as tetrahedral sheets of the 2:1 clay structure.

Intercalation experiments established the relationship between clay layer charge
location and 1:1 C15H33NBu3+ and Na+ mixed ion heterostructure formation. A clay with

a layer charge arising in the central octahedral sheet is able to form a 1:1 C16H33NBu3+

113

 

and Na+ mixed ion heterostructure. The presence of charge derived from tetrahedral
substitutions in the 2:1 clay layer structure, however, inhibits heterostructure formation.
Instead, these latter clays form homostructured materials that have both C16H33NBu3+ and
N a“ cationic species mixed within the clay interlayer regions.

A 1:1 C15H33NBu3+ and Na+ mixed ion heterostructure combines hydrophilic and
hydrophobic interlayers into a single phase intercalate. This particular gallery
arrangement may be advantageous for the dispersion of C113H33NBu31+ galleries in water.
Adsorption isotherms (25 °C) show that the organic galleries of a homoionic
Cu5H33NBu3+ exchanged fluorohectorite clay and a heterostructured 1:1 C16H33NBu3+ and
Li+ mixed ion fluorohectorite have comparable capacities for the adsorption of 2,4-
dichlorophenol from water. However, the heterostructured clay out-performs the organo
clay when the adsorption of 2,4-dichlorophenol is followed in time. 2,4-Dichlorophenol is
adsorbed almost instantaneously from a static aqueous solution when air dried 1:1
C16H33,NBu3+ and Li+ mixed ion fluorohectorite heterostructure is added. In contrast, the
2,4-dichlorophenol concentration diminishes slowly upon the addition of air dried

homoionic C115H33NBu3+ exchanged fluorohectorite clay to such a standing solution.

114

 

5.2 Introduction.

Smectite clays have a substantial sorptive capacity for heavy metal ions [1]. The
binding of these inorganic environmental pollutants occurs through ion exchange. In
contrast, the sorption of non-polar organic contaminants on smectite clays is generally
restricted to the external clay surfaces only [2]. However, the sorptive potential towards
organic molecules can be improved when the clay interlayer is modified to allow for the
accommodation of non-polar organics [3,4]. The polar nature of a smectite clay interlayer
is altered when the hydrated gallery metal ions are exchanged by, for instance, long chain
alkylammonium ions. Thus, non ionic organic pollutants may readily penetrate the
organic clay interlayers of onium ion exchanged clays [5-7]. The organic galleries of
onium ion modified clays function as a partitioning medium for non-polar organic
solutes. Hence, organo clays can remove organic pollutants from waste water streams and
are, for example, commercially used in industrial waste water treatment systems [1].

The sorptive properties of organo clays can even be tailored to specific organic
contaminants [6]. Sorbate selectivity is generally achieved by the intercalation of specific
organic cations in the clay interlayer region. However, the adjustment of the interlayer ion
structure sometimes causes the organo clay to become too hydrophobic to be readily
wetted by water. Such modified clays are impractical for removing dissolved organic
contaminants from water.

We have previously reported on 1:1 C16H33NBu3+ and Na“ mixed ion
heterostructured fluorohectorite clays with regularly alternating organic and inorganic
exchanged cation galleries [8,9]. These structures incorporate both hydrophilic and

hydrophobic galleries within a single phase material. Obviously, these heterostructured

115

 

modified clays may readily be wetted by water through exfoliation of the inorganic
gallery, yet the water suspended heterostructured entities still carry the organophilic
character of the organic exchanged interlayer.

Three reaction pathways have been developed that lead to the self assembly of
mixed ion fluorohectorite heterostructures with regularly alternating organic and
inorganic galleries [8,9]. The simplest synthetic approach I to the preparation of 1:1 Na+
and C15H33NBu3+ mixed ion heterostructured clay is the partial ion exchange reaction of
the homoionic Na+ clay by C15H33NBU3+ onium ions. Route 11 embodies the reverse
approach were a homoionic clay is treated with a concentrated NaCl solution to produce a
mixed ion heterostructured intercalate. Equal molar quantities of the two homoionic
parent end member clays are reacted in water to produce a mixed ion heterostructure in
pathway III.

In previous work, we have also published on the relationship between the onium
ion geometry and the mixed ion heterostructure formation process [9]. The onium ion
head group size alone directs heterostructure formation in fluorohectorite clay. Mixed ion
heterostructures with regularly alternating interlayers of inorganic and organic exchange
cations form when C16H33N(CmH(2m+t))3+ (m=3, 4 or 5) onium ions are used in a partial
ion exchange reaction. Large onium ion head groups move the center of positive charge
away from the silicate layers and preclude commingling of cationic species within one
interlayer. This forces the smaller hydrated metal ions to take positions on the other side
of the organic exchanged interlayer, resulting in heterostructure formation. Smaller

trimethyl- and triethyl terminated onium ion head group sizes effectively neutralize the

116

 

clay layer charge and do not produce heterostructured clays. Instead, they initiate the
formation of phase segregated homoionic parent end member clays.

Intercalation experiments with a series of C,,H(2,,,t1,NBu3+ (n=4, 6 - 22) onium
ions revealed that the onium ion alkyl chain length determines the extent of metal ion
replacement in a partial ion exchange reaction. Quantitative onium ion loading is
obtained when the alkyl tail exceeds 10 carbon units. Thus, a structural modification of
the onium ion alkyl chain is not likely to influence the heterostructure formation process!

So far, we have only used a synthetic fluorohectorite clay for the synthesis of our
mixed ion heterostructured materials. It would be desirable to produce mixed ion
heterostructures with abundant natural clays. However, mixed ion heterostructures with
alternating galleries do not easily form when natural clays are used in a partial onium ion
exchange reaction. Often, clay structures are formed that have both cationic species
mixed within one interlayer instead.

In the present work, we report on the relationship between clay layer charge
location and mixed ion heterostructure formation. For this, we have synthesized water
swelling fluorinated clays with specific layer charge density locations. Thus, we were able
to regulate the metal ion compositions in the octahedral and tetrahedral sheets of synthetic
2:1 layered silicates. These model clays allow us to evaluate the significance of clay layer
charge location irregularities. This data is relevant when one attempts to make a
heterostructured material from a natural smectite clay.

Water swelling synthetic fluoroclays were first reported by the U.S. Bureau of
mines as part of a study on synthetic fluoromicas [10-12]. Later, Barrer and Jones [13-15]

investigated ion exchange reactions in synthetic water swelling fluorohectorites,

117

 

Li,,14-[Mg435Liu4]SigOan4. This high charge density clay has a layer charge arising from
Li+ for Mg2+ substitutions in the octahedral sheet. The laboratory synthesis of
fluorohectorite involves melting of the metal oxides in the presence of a fluoride source.
Corning, Inc. produced a comparable commercial fluorohectorite by a glass ceramic
method [16] and we have used this particular clay in our previous studies on mixed ion
heterostructure formation [8.9].

A taeniolite-like series with a varying 2:1 silicate layer charge that arises in the
octahedral sheet was reported by Kitajima and Takusagawa [17]. A NaxolMgw-
ijix]SigOzoF4 series with 0.6 S x S 2.0 was produced from the metal oxides and fluorides
in the melt at 1420 °C. The same investigators also reported on isomorphic tetrahedral
substitutions in taeniolite, Kz-[Mg(4.,x,Li(2-,)]S1(3-X)Z,020F4 with 0 S x S 2 [18]. Again, the
synthetic procedure involved melting of the metal oxides and fluorides in a sealed Pt
container at 1420 °C.

Reports on synthetic 2:1 layered fluorosilicates with a layer charge arising from
substitutions in the tetrahedral sheet only are more abundant [10-12]. However, most
synthetic procedures involve melting of oxides at high temperatures (~1400 °C). Lower
reaction temperatures (~1000 °C) were achieved when NaF was used as a flux. These flux
methods produced very high charge density sodium-4-micas, Na4[Mg6]Si4A14020F4
[19,20].

All of the fluorinated clays used in this work were prepared by a solid state
reaction of metal oxides and fluorides. Fluoride was added to aid the mineralization of
fluoroclay, but more importantly, the addition of alkali fluorides as a flux material

lowered the reaction temperature. All fluorinated clays were easily produced and in

118

relatively large amounts when compared to clays that are synthesized by hydrothermal
methods.

In addition, we report on the water dispersing properties of a 1:1 C16H33NBU3+ and
Li“ mixed ion fluorohectorite heterostructure and compare this property to the
performance of homoionic C16H33NBu3+ exchanged fluorohectorite clay. Also, we
explored the adsorption of 2,4—dichlorophenol from water by both types of modified

clays.

5.3 Experimental.

Synthesis of 2:1 clays with octahedral layer charge: Fluorohectorites were
synthesized from reagent grade LiF, MgO and 8102. LiF was used both as reagent and as
flux. Batches were formulated to contain 10 LiF, 4.86 MgO and 8 S102. The reagents
were mixed and transferred into a Pt crucible. A Pt lid covered the crucible during the
calcination process in order to minimize loss of the silicon component in the form of
gaseous SiF4. Also, the oven was placed in a well ventilated hood as some HF is
generated during the reaction. Batches were calcined in air at 855 or 1000 °C for 2 hours.
The molten glasses were then cooled to room temperature at 5°C/min. Melt products were
removed effortlessly from the crucible when the fluoro clays were first allowed to swell
with water. Water soluble lithium salts were removed by washing of the clay with
deionized water, while most insoluble and non-dispersible byproducts were removed by

sedimentation. Finally, the fluorohectorites were air dried.

119

 

Synthesis of 2:1 clays with tetrahedral and octahedral layer charge: The same
LiF melt procedure as described above was used when aluminum substituted
fluorohectorites were synthesized from the metal oxides. Reagent grade 'y-A1203 was used
as an alumina source. Batches were designed so that the clay products would contain
roughly 2.5% and 7.5% aluminum substituted tetrahedral sites per Si(g-x)Al,,OzoF4 clay
formula unit. The reaction mixtures were calcined at 1000°C for 2 hours before being
cooled to room temperature. The cooling rate was 5°C/min. The products were purified

by sedimentation, washed with water and air dried.

Synthesis of a fluorosaponite clay with a tetrahedral layer charge:
Fluorosaponites are produced when NaF instead of LiF is used in the solid state reaction
process. A decreased NaF amount was used when compared to the LiF quantity used in
the LiF flux method. In addition, some Mng was added at the expense of MgO. A typical
reaction contained 3.5 NaF, 4 MgO, 2 Mng, 6.86 SiOz and 0.57 A1203. Clays of good
crystalline quality were obtained when the reaction mixture was calcined at 1000°C for at
least 5 hours. However, volatilization of silicon and even the sodium component can be a
problem at these longer calcination periods. The room temperature cooled powder
product was suspended in a 2 M LiCl solution to exchange the Na+ interlayer ion for Li+
since it was found that only the Li+ exchanged fluorosaponite form was able to exfoliate
in water. Next, the clay suspension was washed with water, purified by sedimentation,

and air dried.

120

Formation of mixed ion clays: Mixed ion structures were prepared in a one step
procedure. Half a cation exchange capacity equivalent of onium ions were directly added
to a Li+—clay suspension in a partial ion exchange reaction. C16H33NBu3+ onium ions were
used to form the mixed ion clay structures and were synthesized as previously described
[9]. Typically, about 0.30 g Lii-clay was first dispersed in 100 ml deionized water before
surfactant was added. All partial ion exchange reaction mixtures were stirred at ambient

temperature for at least two days. Reaction products were washed free of excess salt and

air dried.

2,4-Dichlorophenol adsorption: Isotherrns (25 °C) were obtained for the
adsorption of 2,4-dichlorophenol from water onto homoionic C16H33NBu3+ exchanged
fluorohectorite and a 1:1 C16H33NBu3+ and Lii mixed ion heterostructured fluorohectorite
clay. The adsorption isotherms were constructed by weighing multiple 0.0500 mg
quantities of the air dried organo modified fluorohectorite clays into scalable
polypropylene bottles. Next, precise quantities of a standard 2,4-dichlorophenol solution
and water were added to the clay so that the amount of 2,4-dichlorophenol added was
equal to 1.0, 2.5, 5.0. 10, 15, 20, 25, 30, 35, 40, 45 or 50 umol per 100 ml total volume.
The bottles were agitated on a shaker table for at least five days and then centrifuged.
Next, residual 2,4-dichlorophenol was extracted into 5 ml hexane from 5 ml of the
Supernatant. The 2,4-dichlorophenol equilibrium concentration was then determined from

the optical density of 2,4-dichlorophenol at 284 or 292 nm using an IBM 9430 UV-VIS

Spectrophotometer.

121

In addition, the change in the 2,4-dichlorophenol equilibrium concentration was
also followed in time for a static (unstirred) adsorption procedure. Air dried powders of
homoionic C16H33NBu3+ exchanged fluorohectorite (0.1495 g) and air dried 1:1
C16H33NBu3+ and Li+ mixed ion heterostructured fluorohectorite clay (0.3155 g) were
each sprinkled onto the surface of a 500 ml standing solution of 2,4-dichlorophenol in
water (26.78 umol/ 100 ml). A 5 ml sample was taken every hour and the supernatant
solution was extracted with 5 ml hexane before being analyzed for 2,4-dichlorophenol
concentration.

Product characterization: X-ray diffraction, XRD, patterns were recorded on a
Rigaku rotaflex diffractometer equipped with Ni-filtered Cu-Ka radiation. All synthetic

clay samples were analyzed as powders and as preferentially ordered films.
The cation exchange capacity (CEC) of the fluoro clays were measured with the
ammonia selective electrode method [21]. A Cahn 121 TG thermogravimetric analyzer
was used to estimate the water content of the air dried Lii-clays and to determine the
NH,“ content of the NHf-clays used in the CEC measurements. All clay samples were
heated at 5 °C per minute up to 800 °C.
High resolution solid state MAS-NMR spectra were recorded on a Varian VXR
4005 spectrometer equipped with a 7 mm Varian CP mass probe. A sample spinning
frequency of 4 kHz was used for all NMR experiments. 29Si MAS—NMR experiments
Were run at 79.462 MHz using a pulse width of 4 us and a 870 s interval. 27Al MAS-
NMR experiments were performed at 104.228 MHz using a pulse width of 4.8 118 and a 2

S interval. Chemical shifts 8 are reported in ppm relative to Si(CH3)4 and AI(H20)63+.

122

5.4 Synthesis of lithium-fluorohectorite, Lil,14-[MgusLil,14]SistnF4.

High charge density fluorohectorite is readily produced by the reaction of MgO
and $10; in a LiF flux. Two distinct fluorohectorite clays were synthesized from melts
which were held at 855 or 1000 °C for two hours prior to cooling. The XRD patterns of
these two water-washed clay products are shown in Figure 5.1. A
Li),12-[Mg4_3gLi1.12]818020F4 fluorohectorite clay produced by Corning is included for
comparison in this figure. Evidently, crystalline layered fluoroclays are formed for both
synthetic reaction temperatures. However, the clay synthesized from the melt at 1000 °C
is accompanied by an impurity phase (d=6.71, 4.45, 3.03, 2.66, 2.22 A), presumably a
lithium silicate, while the fluorohectorite clay produced at 855 °C is almost free of this
impurity.

2981 MAS—NMR spectra of the two synthesized fluorohectorites are given in
Figure 5 .2 together with the NMR spectra of the commercial Corning fluorohectorite. All
three clays have a characteristic 29Si chemical shift positioned around -95 ppm. The
silicon environment is therefore the same in all fluoroclays. Tetrahedral silicon in a clay
layer resembles a silica Q3 site since one silicon is connected by bridging oxygens to three
other silicon neighbors. However, each silicon is also connected by one oxygen to the
octahedral sheet, unlike a silica Q3 site which is linked to a hydroxyl group. Our observed
chemical shifts therefore fall between the values normally found for silica Q3 (~ -100
ppm) and Q2 (~ -90 ppm) sites [22]. All three clays samples are free of amorphous SiOz
since no distinctive silica Q4 (~ ~110 ppm) sites are observed in the 2981 MAS-NMR

spectra [22].

123

 

 

 

 

 

 

 

12.2A
.2» JL LL :1: 1000 C if
E
T, 12.2A
.2.
*5 o
124uA

 

l Comingjk

IWUYIT'TIIITIUIUVVYITIIII

5 10 15 20 25 30
Degrees (20)

 

 

 

 

Figure 5.1 XRD patterns (Cu-Kn) of air dried fluorohectorite films produced by the LiF
flux method (the silicate impurity is marked * in the XRD pattern). The clays were
derived from melts which were held for two hours at 855 and 1000 °C prior to cooling.
The XRD of an air dried commercial Corning fluorohectorite is included for comparison.

124

 

29Si MAS-NMR

  

1000 °C

 

[I'llIIIIIIUIIIIIIIUIIIII'IUUII

-80 -90 -100 -110
Slppml

Figure 5.2 2981 MAS-NMR spectra of the clays shown in Figure 5.1. The 29Si chemical
shift for all clay samples is around -95 ppm relative to TMS.

125

 

Nevertheless, the sedimentation purification process did not remove all impurities
from the 1000 °C synthesized fluorohectorite clay. The presence of these impurities is
reflected in the values obtained for the clay cation exchange capacities as well as the
XRD patterns. The CEC values were determined to be 127 and 93 meq/ 100g respectively
for the 855 °C and 1000 °C synthesized clays. The Coming fluorohectorite was found to
have an ion exchange capacity of 121 meq/ 100g air dried clay. The fluorohectorite layer
charge densities are not accurate, especially for the 1000 °C sample, because of the
presence of the impurity mass.

The alkylammonium exchange provides a better understanding how the relative
layer charge density varies with the fluorohectorite synthesis temperature. This method
involves the intercalation of long chain alkylammonium ions into the clay. A clay with a
relatively higher layer charge density will have a larger basal spacing than a clay with a
lower layer charge density, due to a more vertical orientation of the alkyl chains in the
more populated galleries. Thus, the clay interlayer distance is a direct function of the
interlayer onium ion population and may therefore provide a measure of layer charge
density. This approach assumes that excess intersalated onium ion salts are removed from
the gallery by washing and that only the exchange ions needed to balance the clay layer
charge are returned in the galleries.

Different basal spacings were observed when C16H33NBu3+ onium ions were
intercalated into all fluorohectorites. The XRD profiles for the fully C16H33NBu3+
exchanged flux synthesized fluorinated clays are shown in Figure 5.3 (A). The organo
clay formed from the 855 °C melt product has a 30.5 A basal spacing, while the organo

clay obtained from the 1000 °C melt clay has a 28.8 A spacing. Homoionic C16H33NBu3+

126

exchanged Corning fluorohectorite shows a 26.9 A basal spacing [9]. These XRD results
indicate that the Coming fluorohectorite material must have a somewhat lower layer
charge density than both LiF flux derived fluorohectorite clay products since it has the
smallest interlayer separation of all three organo clays. Similarly, the 855 °C synthesized
fluorohectorite clay has a slightly higher layer charge density than the 1000 °C derived
clay product. Clearly, the cation exchange capacity obtained for the 1000 °C melt
products disagrees with these XRD results. The clay layer charge density is higher than
indicated by the CEC value.

Barrer and Jones [13] reported a clay unit cell composition of
Li1,14~[Mg4.36Li1_14]Si3020F4 for a fluorohectorite produced at 850 °C. Both our
fluorinated clays were not analyzed for elemental composition because of the presence of
impurities. We will therefore adopt this Li1_14-[Mg4,36Li1,pdSigOzoFa unit cell formula for
the 855 °C flux synthesized fluorohectorite clay as our synthetic method does not deviate
significantly from the procedure described by Barrer and Jones [13].

Clearly, the LiF flux method produces a high charge density fluoroclay. A
variation of the employed melt temperature only causes a minor change in layer charge
density. Based on the XRD behavior of alkylammonium exchanged products, we found
that a change in the metal oxide reaction mixture stoichiometry did not result in the
production of clays with very different layer charge densities. Hence, this flux method is
particularly well suited for the production of reproducible fluorohectorite compound clays

on a laboratory scale.

127

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

40.7A
.<
A B
29.4A «<—
v.
29 «t 3
:7, <1-
. J
s but 1900 C l 1000°C
o _s
2
E
a“; 30.3A 23°:
«it 532'
r~y°<
:2 a! .,
LA; 855C 855°C
1'1'1'1'1'1'1'1'1' 7'1'1'1'1'1'1'1'1'

 

 

 

 

 

 

26101418 26101418
Degrees (20) Degrees (20)

Figure 5.3 XRD patterns (Cu-Kn) of C1(5H33NBu3+ exchanged fluorohectorite clays. (A)
Homoionic onium ion exchanged organo clays of the fluorohectorites synthesized from
melts at 855 and 1000 °C. (B) Air dried 1:1 mixed ion heterostructured products obtained
when the 855 and 1000 °C Lii-fluoroclays were exchanged with a half an exchange
equivalent of C15H33,NBu3+ ions.

128

The flux reaction procedure does suffer from the formation of impurities when the
reaction temperature is executed at 1000 °C. However, we found that the amount of
impurities in the water washed fluorohectorite can be suppressed when some boric acid is
added to the reaction mixture prior to the melting of the reactants at 1000 °C. llB MAS-
NMR on such prepared clay samples does give a very modest 11B NMR signal. In
addition, only the characteristic -95 ppm 29Si resonance peak appears in the 29Si MAS-
NMR spectrum of these boron treated clays. It is unclear how much boron, if any, is
incorporated in the tetrahedral sites of the 2:1 clay layers. The purity of the boron treated
fluorohectorites was validated by XRD and cation exchange capacity measurements. A
fluorohectorite clay batch with about four weight % additional boric acid added and
synthesized at 1000 °C was measured to have an exchange capacity of 122 meq/ 100g
which compares well with the pure Corning fluorohectorite. Also, identical basal
spacings (29.6 A) were obtained when CunggNBug+ onium ions were intercalated into
Corning fluorohectorite and this boron treated fluorohectorite clay.

Figure 5.3 (B) shows the XRD patterns for air dried 1:1 mixed ion clay products
formed when half an exchange equivalent of C16H33NBu3+ onium ions are intercalated
into the 800 °C and 1000 °C flux synthesized fluorohectorite clays. The stacking repeat
distance of a 1:1 mixed ion heterostructure should agree with the sum of the basal
spacings of C16H33NBu3+ organoclay and Li+ exchanged fluoroclay. As expected, the air
dried mixed ion materials do form mixed ion heterostructured clays. The mixed ion
heterostructure derived from the 855 °C fluoroclay has repeat spacing (41.5 A) that is

only 1.2 A smaller than expected based on the sum of 30.5 A C16H33NBu3+ organo clay

129

and of 12.2 A Li+ clay. This discrepancy could be caused by small differences in

hydration for the heterostructured and the homoionic clays.

5.5 Synthesis of fluorohectorite-like clays with some charge originating in the
tetrahedral sites.

Partial substitution of silicon by aluminum in fluorohectorite clays was achieved
when some A1203 was mixed with the MgO and SiOz in a LiF flux. Such synthesized
clays have layer charge originating in both the octahedral and tetrahedral sheets. Hence,
Li+ is partially substituting for Mg2+ in the octahedral sites, whereas Al“ is ideally
incorporated in the tetrahedral sites. However, the melt has to be heated to 1000 °C in
order to substitute Al3+ into the tetrahedral sites of fluorohectorite. A lower processing
temperature does not lead to a partial substitution of silicon by aluminum as evidenced by
27Al and 2981 MAS-NMR. Two reaction mixtures were formulated to contain 2.5% and
7.5% aluminum substituted tetrahedral sites per 813020F4 clay formula unit. Figure 5.4
shows the XRD patterns of these two flux-synthesized clays.

Obviously, the 12.2 A clays contain the same impurity phase encountered in the
non Al substituted 1000 °C synthesized fluorohectorite; compare Figure 5 .1 and 5.4.

The presence of aluminum in the 2:1 clay layer and the location (tetrahedral or
octahedral sheet) of aluminum in these fluorohectorite-like clays is verified with 27Al and

2S'Si MAS-NMR. Both these MAS-NMR spectra’s are shown in Figure 5.5 and 5.6 for

respectively the 2.5% and 7.5% aluminum content batch compositions.

130

 

 

 

 

 

 

 

12.2A
.2.» °<’"
:2 «a,
.3 0' .5%
.5 J1 ’l‘lt J -
O.)
.2
E
d)
04
12.2A
o<
.. ,1
JL ‘5 2.5%ji
”PL ’1‘
,.,.

 

 

5 10 15 20 25 30
Degrees (20)

Figure 5.4 XRD patterns (Cu-KG) of aluminum substituted fluorohectorite-like clay films
PI‘Oduced by the LiF flux method (the silicate impurity is marked * in the XRD pattern).
The air dried clays were produced from 1000 °C melts. ReaCtion mixtures were
fOl‘l'nulated to have 2.5% (lower pattern) and 7.5% (upper pattern) tetrahedral aluminum
sites per 020E, formula unit (Si7,gAlo_2 and Si7,aAlo,6 respectively).

131

Both fluoroclays have 27A] MAS-NMR peaks positioned around 67 ppm which is
indicative of tetrahedral aluminum [22-24]. Octahedral aluminum with a peak centered
around 0 ppm is not observed [23,24]. Clearly, aluminum has a preference for the
tetrahedral sites in these flux synthesized clays. We speculate that the large excess of
lithium in the reaction mixture during the melt synthesis may outweigh aluminum in the
competition for octahedral 2:1 clay layer sites.

A variance in the environment around silicon sites is a result of the incorporation
of aluminum in the tetrahedral clay sheets. Aluminum can be distributed throughout the
tetrahedral clay sheets to form silicon sites with zero Si(0Al), one Si(lAl), two Si(2Al) or
three Si(3Al) nearest Al neighbors. The presence of tetrahedral aluminum in the
tetrahedral sites of our flux synthesized clay layers is confirmed with 29Si MAS-NMR
[24,25]. Isomorphous substitution of Al for Si in the tetrahedral clay sheets causes the
29Si NMR peak to shift to lower fields. Thus, the presence of distinct Si(nAl) groups
creates additional 2981 MAS-NMR resonances [22]. The -95 ppm 29s1 resonance is
therefore ascribed to Si(0Al) while the -90 ppm peak is due to Si(lAl) sites.

Clearly, the 29Si MAS-NMR spectra of our Al containing fluorohectorite-like
clays confirm the presence of aluminum in the tetrahedral clay sheets. The intensity of the
Si(lAl) resonance peak increases as expected when the Al content is raised from 2.5% to
7.5 % in the fluorohectorite reaction mixtures. However, silicon sites with more than one

Al neighbors, Si(2Al) and Si(3Al) sites, are not encountered in the 29Si MAS-NMR of

these two fluoroclay samples.

132

2°81 MAS-NMR 27Al MAS-NMR

 

 

[IUIIITITIIIIWWITIITI

-80 -90 -100 -110 150100 50 0 -50
Sippml 511mm]

 

Figure 5.5 27A1 and 29Si MAS-NMR of an aluminum substituted fluorohectorite-like clay
produced by the LiF flux method from a 1000 °C melt. The reaction mixture was
formulated to have 2.5% tetrahedral aluminum sites per 313020F4 formula unit
(Si7gAlopOzoF4). The 29Si resonance at -95.0 ppm is due to Si(0Al) while the ~90.3 ppm
Peak is ascribed to Si(lAl) groups. The Si/Al ratio was calculated to be 20.5 from the
FCIative intensities of the 29Si peaks, which corresponds to a Si7_63Alo,3-, tetrahedral sheet
Composition. The 27Al resonance at 67.9 ppm indicates Al in tetrahedral sites. The
asterisks in the 27A1 NMR spectra signifies side spinning bands. Octahedral A1 with a

resonance around 0 ppm is not observed.

133

29sr MAS-NMR 27Al MAS-NMR

°9
\D
so

 

 

I‘IthTIIIIIIIIIIITTI

-80 -90 -100 -110 150100 50 0 -50
Slppml 5lppm]

 

Figure 5.6 27A1 and 29Si MAS-NMR of an aluminum substituted fluorohectorite-like clay
produced by the LiF flux method from a 1000 °C melt. The reaction mixture was
formulated to have 7.5% tetrahedral aluminum sites per SigOzoFa formula unit
(SiuAlogOzoFa). The 2”Si resonance at -94.5 ppm is due to Si(0Al) while the -89.9 ppm
Peak is ascribed to Si(lAl) groups. The Si/Al ratio was calculated to be 9.0 from the
relative intensities of the 29Si peaks, which corresponds to a Si7,2oAlo_go tetrahedral sheet
cOmposition. The 27A1 resonance at 66.8 ppm indicates A] in tetrahedral sites. The
aSterisks in the 27Al NMR spectra signifies side spinning bands. Octahedral A] with a

resonance around 0 ppm is not observed.

134

Studies on the manner of Al distribution in the tetrahedral clay sheets have shown
that Lowensteins rule (avoidance of Al in adjacent tetrahedral sites) holds in 2:1 layered
silicates [22,24,25]. The Si/Al ratio, and thus the tetrahedral sheet composition, may then
be calculated for the two aluminum containing fluorohectorite-like clays from the
normalized 298i NMR Si(nAl) peak intensities (1,) [22]. The derived Si/Al ratios for these

two clays are listed in Table 5.1 and are calculated by:

Si/A1=23: I./i(a)h (v.1)

n=0 ":0

The 29Si MAS-NMR derived Si/Al ratios indicate that the A] content of both clays
are somewhat higher than was anticipated. We ascribe this increased Al content to the
partial volatilization of the silicon component in the form of SiFa during the flux reaction
process. The composition of the tetrahedral sheets of both 2:1 layered fluoroclays were
calculated from the Si/Al ratios on the basis of a Si(3-,)Al,OZOF4 unit cell and are given in
Table 5.1. The composition of the octahedral sheet is, however, not exactly known for
these two clays.

The presence of the small quantities of silicate impurities in both fluorinated clay
products renders elemental analysis obsolete in the determination of the 2:1 layer
compositions. However, cation exchange capacity measurements and the alkylammonium
intercalation method may provide an indication as to the magnitude of the total clay layer

charge densities.

135

Table 5.1 2981 MAS-NMR data of the aluminum substituted fluorohectorite-like clays

shown in Figure 5.5 and 5.6. Both clays were produced by the LiF flux method from a

 

 

 

1000 °C melt.
25Si
Al content Normalized Tetrahedral
resonance expected observed
prior to peak sheet
positions SizAl ratio SizAl ratio
melting3 intensity!2 composition
[PPm]
-95.029 0.854
2.5% 39.0 20.5 Si7_63Alo_37
-90.289 0.146
94.549 0.668
7.5% 12.3 9.0 Si7,zoAlo,go
-89.86’1 0.332

 

 

 

 

 

 

 

 

g The batches were formulated to have x% tetrahedral aluminum sites per Si(3-,)AlegoF4
formula unit.

9 The normalized intensities were derived by deconvolution of the 2981 resonance peaks
(peaks were fitted using Gaussian line shape analysis).

9 Silicon site without an Al neighbor, Si(0Al).

9 Silicon site with one A] neighbor, Si(lAl).

The measured ion exchange capacity of the two clay products decrease
respectively from 114 to 79 meq/ 100g when the aluminum content is raised from 2.5% to
7.5% in the fluoroclay synthesis mixtures. Several factors may cause this behavior. The
ammonium clays used in the ion exchange capacity measurements may have simply

contained different amounts of impurity phases. In addition, layer charge arising from
tetrahedral substituted sites may bind the interlayer exchange ion quite strongly, and
inhibits its ion exchange. As a result, the cation exchange capacity measured decreases

With an increased tetrahedral aluminum content.

136

However, intercalation of CungthBug+ ions into both aluminum substituted
fluorohectorite-like clays does not result in the formation of organo clays with very
different gallery heights, Figure 5.7 (A). A difference in the d-spacing for both these clays
is expected on the bases of the ion exchange capacity measurements. The
alkylammonium intercalation results do not support this expectation. The basal spacings
for the 2.5% and 7.5% air dried organo fluoroclays are 28.3 and 28.7 A respectively.
These organo clay interlayer heights are quite comparable and indicate that both clays
have similar layer charge densities, although the composition of the tetrahedral sheets
deviate significantly from each other as evidenced by MAS-NMR.

Figure 5.7 (B) shows the XRD patterns for the air dried 1:1 C16H33NBu3,+ and Li“
mixed ion clay products which are formed when half an exchange equivalent of
C16H33NBu3+ onium ions are intercalated into both the 2.5% and 7.5% aluminum
containing clays. A 1:1 mixed ion heterostructured clay has a stacking repeat order that
corresponds to the sum of the basal spacings of homoionic CmH33NBu3+ and Li“
exchanged fluoroclay. Such mixed ion heterostructured materials are not observed in the
XRD patterns for both mixed ion clays. Thus, the Al-substituted derivatives do not form
heterostructured intercalates.

Instead, 27 A and 14.4 A air dried products are formed for both clays. The 14.4 A
phase corresponds to a mixed ion material that has both cationic species intercalated
within the same gallery. The 5 A interlayer height (14.4 A basal spacing - 9.6 A 2:1 clay
layer thickness) points to a monolayer arrangement of onium ions. Inorganic ions co-
occupy the 5 A interlayer space since the homoionic C16H33NBu3+ organoclays have

much higher gallery heights, and have therefore more intercalated onium ions.

137

 

Relative Intensity

 

 

 

29.6 A

:1 Al%

 

 

JUL;

 

 

 

 

 

27.2 A

\ Al%

 

7.5%

 

<—27.2 A

<—14.4 A

<—13.5 A

[2.5%]

 

 

 

 

28.8A
K:
6! mi
3 <1-
LA <34 2.5%
2 6 10 14 18
Degrees (20)

I'I'I'I'I'I'ITI'I'

2 6 10 14 18
Degrees (20)

Figure 5.7 XRD patterns (Cu-Kc) for: (A) Completely C1(r,H33NBu3+ exchanged air dried
homoionic fluorohectorite-like clays which have 2.5% and 7.5% aluminum substituted
tetrahedral sites; (B) Products formed when half an exchange equivalent of C16H33NBu3,+
onium ions are intercalated into both the 2.5% and 7.5% aluminum containing clays. The
C16H33NBu3+ and Li+ exchange cations are present in these air dried clays in a 1:1 ratio.

The 27 A materials compare to fully C16H33NBu3+ exchanged clay but the
homoionic exchanged C16H33NBu3+ organo clays have slightly higher basal spacings. It is
therefore likely that some inorganic cations are trapped in the predominantly organic
exchanged interlayers of the 27 A products.

Thus, 1:1 C1(,H33NBu3+ and Na+ mixed ion heterostructured clays are not formed
when the 2:1 silicate layers carry some tetrahedral charge, in addition to octahedral layer
charge. However, both our synthetic fluoroclays used in these onium ion intercalation
experiments carry a significant amount of tetrahedral charge (Si-1,63Alog7 and SinoAlogo
respectively). For comparison, natural Arizona montrnorillonite has a smaller amount of
layer charge derived from tetrahedral substitutions, Nao,99-[A12,69Fe3+0,taMgLnan‘iom]

(Si8A10.04)020(0H)4-

5.6 Synthesis of a fluorosaponite clay, Nana-[Mgr](81,,“A11,14)OzoF4.

Fluorosaponites were produced with a method analogous to a procedure used in
the synthesis of sodium-4-micas [19,20]. Our synthetic saponite has a layer charge density
arising solely from tetrahedral substitutions in the 2:1 silicate layers. This is achieved
when NaF instead of LiF is used as flux in the solid state reaction process between the
metal oxides. Sodium does not occupy octahedral 2:1 layer sites but can only take
position in the interlayer region, which is opposite to lithium, which can found in both
locations.

A reaction mixture was formulated to produce a Na1,14-[Mg6]Si6g(,Alt.14020F4
saponite with a layer charge density comparable to the previously 855 °C synthesized

fluorohectorite, Lit,14-[Mg436Li1.14]SigOzoF4. The layer charge density locations in both

139

 

these clays are contrasting since the fluorohectorite clay has octahedral layer charge
whereas the fluorosaponite has tetrahedral layer charge.

The XRD pattern of the purified Li+ exchanged saponite product is shown in
Figure 5.8. The 12.2 A layered clay phase is free of crystalline impurity phases. This 2:1
fluoroclay has an octahedral sheet with a composition presumably similar to brucite. All
octahedral sites in the 2:1 structure are occupied by magnesium. The absence of
octahedral Al in the trioctahedral sheet is verified by 27Al MAS-NMR, Figure 5.9. The
67.7 ppm 27Al resonance is indicative of tetrahedral aluminum. Six coordinated
aluminum with a resonance peak positioned around 0 ppm is not observed in the
spectrum.

Three distinct silicon sites are observed in the ”Si MAS-NMR, Figure 5.9. These
different Si(nAl) sites confirm the presence of aluminum in the tetrahedral sheets of the
2:1 clay. The -93.9, -89.1 and -84.2 ppm 29Si resonances are attributed to Si(0Al), Si(lAl)
and Si(2Al) sites. Silicon with three aluminum neighbors, Si(3Al), is not observed for this
clay.

Normalized 29Si(nAl) NMR peak intensities are given in Table 5.2. The Si/Al
ratio is obtained from these intensities according to equation (VI) and calculated at 4.31.
This particular ratio corresponds to a Si6,4oA11,51 tetrahedral sheet composition. Again, the
clay layer charge density is somewhat higher than anticipated. We attribute this increased
aluminum content to evaporation of the silicon component during the synthesis process.
Formation of silicate impurities might also have lead to an increased incorporation of

aluminum in the tetrahedral sheet of the saponite clay.

140

 

12.2 A

 

 

 

 

 

5*

2

<1)

E

<1)

.2

*5

“<13 °<

0‘ o< «1: 3
A f :‘ I
5 10 15 20 25 30

Degrees (20)

Figure 5.8 XRD pattern (Cu-K0.) of synthetic air dried fluorosaponite. The clay was
produced by a NaF flux method at 1000 °C. The Na“ clay was first converted to the Li+
exchanged form, washed free of salts and impurities prior to being analyzed.

14]

”Si MAS-NMR ”Al MAS-NMR

v—tm [\
ow; 1\
00¢ ‘9

    

Experimental

 

econvoluted
[WWW [IIIIIIIIIIIIIIIIIIII

-70 -80 -90 -100-110 150 100 50 0 -50
filppm] €31me1

 

 

Figure 5.9 ”Al and ”Si MAS-NMR of a fluorosaponite clay produced by the NaF flux
method from a 1000 °C melt. The reaction mixture was formulated to produce a
Na],14-[Mg6]Si6,36A11,14020F4 clay. The 27Al resonance at 67.7 ppm indicates Al in
tetrahedral sites. The asterisks in the 27Al NMR spectra signifies spinning side bands.
Octahedral A1 with a resonance around 0 ppm is not observed. The 29Si resonances at -
93.9 ppm is due to Si(0Al) while the -89.1 and -84.2 ppm peaks are ascribed to silicon
atoms in Si(lAl) and Si(2Al) sites. Si(3Al) sites are not observed. The Si/Al ratio was
calculated to be 4.31 from the relative intensities of these three peaks, which corresponds
to a 816,49A1151020F4 tetrahedral sheet composition.

142

Normalized 29Si MAS-NMR Si(nAl) peak intensities may also be used as a
measure of the Si(nAl) site population in the tetrahedral sheets [22]. And, a random
distribution of aluminum throughout the tetrahedral sheet may be calculated for any Si/Al
ratio and compared to these experimental normalized peak intensities [22]. The Si(nAl)
population probability P, is given in Table 5.2 and was calculated under the restrictions

of Lowenstein’s rule for a random A] distribution model by:

P,, = -——6—— r" (l- r)3"' and 1h = Si/Al (V.Il)
n !(3 — n)!

Both the observed and calculated Si(nAl) population probabilities are in
agreement with each other for the random distribution of Si(nAl) sites throughout the
tetrahedral sheet, Table 5.2. The layer charge density is therefore uniformly distributed

within the fluorosaponite layers.

Table 5.2 2981 MAS-NMR data of the fluorosaponite clay shown in Figure 5.9.

 

 

 

 

 

 

Observed Si(nAl) Experimental Calculated
Silicon site,
peak position normalized peak population
Si(nAl)
[ppm] intensity‘ probability”, P,,
Si(0Al) -93.9 0.433 0.453
Si( 1A1) -89.1 0.438 0.41 1
Si(2Al) -84.2 0.129 0.124
Si(3Al) - - 0.012

 

 

 

 

9 The normalized intensities were derived by deconvolution of the 79Si resonance peaks.
(peaks were fitted using Gaussian line shape analysis).

'3 Calculated for a Si/Al ratio equal to the observed Si/Al value of 4.31.

143

 

Figure 5.10 shows the XRD patterns for (A) homoionic Li” and (B) homoionic
C16H33NBu3+ exchanged fluorosaponites, (FS), as well as (C) the air dried products
obtained when half an exchange equivalent of C115H33NBu3+ onium ions are exchanged
into the Lii-FS clay. A 1:1 mixed ion heterostructured product with a 41 A repeat unit
corresponding to the sum of the basal spacings of homoionic 28.9 A C16H33NBu3+
fluorosaponite and 12.2 A Lii-FS is not observed. Instead, two other mixed ion clay
phases are found next to some phase segregated 12.2 A homoionic Lii-FS. The 14.0 A
material is a homostructured clay that has both cationic species mixed within one gallery.
The 4.4 A interlayer height points to a monolayer arrangement of C16H33NBU3+ ions.
Obviously, Li” must co-occupy the gallery to maintain electrical neutrality.

The 26.9 A product has the onium ions organized in a paraffin arrangement but
does not quite have the same gallery height as the homoionic C15H33NBu3+ clay. Some
Li+ ions must be entrapped within the predominantly organic interlayer to reduce this
interlayer spacing.

Thus, a 1:1 C15H33NBu3+ and Nai mixed ion heterostructured clay is not formed
when the 2:1 silicate layers carry tetrahedral layer charge only. This is opposed to the
behavior of fluorohectorite clays that have octahedral layer charge only and do exhibit the

ability to form heterostructured materials.

144

 

N
.0)
\O
13.4A a»

 

 

 

LA 9.6A B

 

«t «z
o. N.
5!. 21
v
e-JLNX C
'5
t:
.8
E 29.1A
t1)
,3 14.4A
E“:
T)
a:

 

 

 

 

 

12.2 A
.JL 94A
2 6 10 14 18
Degrees (20)

Figure 5.10 XRD patterns (Cu-Kn) for: (A) Li+ exchanged fluorosaponite, (B)
Homoionic C16H33NBu3+ exchanged fluorosaponite clay, (C) Products formed when half
an exchange equivalent of C1(5H3,3NBu3+ onium ions are intercalated Li+-FS clay. All clays

were air dried before being analyzed.

145

5.7 A materials application for a 1:1 C1,;H33NBu3+ and Li” mixed ion
heterostructured clay; the adsorption of 2,4-dichlorophenol from water.

Organo clays have a large sorption capacity for non-polar organic solutes in
aqueous solution [1-7]. However, the organic modified clay is sometimes too
hydrophobic to be easily wetted by water. This inability to be dispersed in water may
become a problem in adsorption applications, where the organophilic galleries function as
partitioning medium for non-polar solutes dissolved in water. In contrast, our 1:1
C16H33NBu3+ and Li+ mixed ion heterostructured clay may be readily dispersed in water
because of the unique arrangement of the hydrophilic- and hydrophobic- galleries.

Adsorption isotherms were obtained for the removal of 2,4-dichlorophenol from
water by homoionic C15H33NBu3+ exchanged fluorohectorite (circles) and 1:1
C16H33NBu3+ and Li‘“ mixed ion heterostructured fluorohectorite clay (squares). Both
isotherms are shown in Figure 5.11. The amount of 2,4-dichlorophenol adsorbed on the
two clays is expressed in terms of mo] 2,4-dichlorophenol adsorbed per gram of onium
ions exchanged in the clays. Both clays have a type I isotherm which is characteristic of
strong adsorbate - adsorbent interactions [26]. The sorptive performance of both clays
compare well to each other, as expected, since the organic exchanged galleries do not
loose any organophilicity when arranged in a heterostructured manner.

The potential of heterostructured clay becomes apparent when the change of the
2,4-dichlorophenol equilibrium concentration is followed in time during the adsorption
process. Figure 5.12 illustrates this change in equilibrium concentration when 2,4-
dichlorophenol is adsorbed from water onto homoionic C16H33NBu3+ exchanged

fluorohectorite (circles) and 1:1 CI¢5H33NBu3+ and Na+ mixed ion heterostructured

146

fluorohectorite clay (squares). Both these air dried clays were added to standing (not
stirred) 2,4-dichlorophenol solutions in specific amounts so that both suspensions would
contain equal amounts of onium ion exchanged galleries.

The 2,4-dichlorophenol concentration reaches its final equilibrium concentration
almost instantaneously when the mixed ion heterostructured clay is added to the standing
2,4-dichlorophenol solution! In contrast, the equilibrium concentration only slowly
decreases over time when homoionic C16H33NBu3+ exchanged fluorohectorite clay is
added. The final equilibrium concentration is not even attained after 25 hours.

Clearly, the dispersion of the organic - inorganic heterostructured modified clay is
greatly facilitated by the presence of inorganic interlayers. The partitioning function of all
organic exchanged interlayers is fully operative in the water dispersed heterostructured
entities. Consequently, the 2,4-dichlorophenol concentration decreases sharply upon
addition of the air dried heterostructured clay.

The homoionic C16H33NBu3+ exchanged organo clay however, arranged in clay
aggregates, does suffer from a reduced organic interlayer access. As a result, the 2,4-
dichlorophenol concentration decreases through adsorption on the on the organo clay at a

significantly reduced rate.

147

 

Adsorbed
[mmol/gram onium ions]

 

 

 

 

O T I I I l I I I I l r I I I I I I I I

0 5 10 15 20

2,4-dichlorophenol equilibrium
concentration [umol/lOOml]

Figure 5.11 Adsorption isotherms (25 °C) for the removal of 2,4-dichlorophenol from a
agitated stirred solution in water by homoionic C115H33NBu3+ exchanged fluorohectorite
(circles) and a 1:1 C16H33NBu3+ and Na‘“ mixed ion heterostructured fluorohectorite clay
(squares). The clay suspensions were equilibrated for five days.

148

 

 

 

30

= .1

O :3 Cl Heterostructure

3% l o Organoclay

g ._. 25':

8 5 2o

C 8 . o

O ,_ oo

o 2 20.- 0 000

E C j o

.5 E .

E 3.- .

5-3 15-

:5 ‘ [:1

U‘ - 13 D D

m 1 DD EDD rm
10 I I I I I I I I I I I I I I 1 I I IT‘I I r I I

 

 

0 5 10 15 20 25
Time [hours]

Figure 5.12 Change of the 2,4-dichlorophenol equilibrium concentration with time when
the solute is adsorbed under static conditions from water onto homoionic C16H33NBu3+
exchanged fluorohectorite (circles) and 1:1 C15H33NBu3+ and Na+ mixed ion
heterostructured fluorohectorite clay (squares). Both air dried clays were added to
standing 2,4-dichlorophenol solutions (25 °C).

149

5.8 Concluding remarks on C15H33NBu3+ and N a+ mixed ion heterostructures.

Earlier work on mixed ion heterostructures showed the importance of the onium
ion head group size in relation to the clay layer charge density [9]. Here, we investigated
the relation between clay layer charge location and mixed ion heterostructure formation.
The layer charge position has a significant effect on the mixed ion heterostructure
formation process. 1:1 C|5H33NBu3+ and Na+ mixed ion heterostructures form with a
fluorohectorite clay that has charge arising in the central octahedral sheet only.

No demixing of the two gallery cation species into an alternating stacking of
interlayers occurs for fluorohectorite-like clays that carry some tetrahedral charge. Also,
no mixed ion clay heterostructures form when all the layer charge is derived from
tetrahedral substitutions in the 2:1 silicate layer structure. In both cases, organoclay and
homostructured clay phases are formed that have both cationic species mixed within
interlayers when tetrahedral charge is present on the synthetic fluoroclay layers.

A natural clay with a layer charge density similar to fluorohectorite, such
as an Arizona montrnorillonite for example, does not form a mixed ion heterostructured
clay when half an exchange equivalent of C16H33NBU3+ onium ions are exchanged into
the clay. This montrnorillonite clay forms a mixed ion homostructure instead and has both
gallery cationic species mixed within the interlayers. Elemental analysis of Arizona
montrnorillonite indicates that some layer charge arises from tetrahedral substitutions.
This, may explain why a natural clay like Arizona montrnorillonite does not form a 1:1
C1i5H33NBu3+ and Na” mixed ion heterostructure. In addition, natural clays may not have

an uniform charge density throughout the clay layers. A non-uniform charge distribution

150

may also contribute to the inability of Arizona montrnorillonite to form a mixed ion
heterostructure.

The advantage of combining the properties of a hydrophilic inorganic clay
(exfoliation in water) with those of an organo clay (pollutant adsorption) into a single
phase heterostructure has been illustrated for the first time through the adsorption of 2,4-
dichlorophenol in aqueous solution. The 1:] CI6H33NBu3‘ and Li’ mixed ion
heterostructured fluorohectorite clay has excellent water-dispersing properties in
comparison to the non-dispersable homoionic organo clay. In addition, the functionality
of the organic interlayers are improved when compared to an organo clay since the
heterostructured arrangement of interlayers also provides a better access to the organic

galleries which function as a partitioning medium.

151

 

 

References.

[1]

[2]
[3]
[4]
[5]
[6]

[7]
[8]
[9]

[10]

[11]

[12]
[13]
[14]
[15]

[16]

[17]

CMS Workshop Lectures, Vol. 8, Organic pollutants in the Environment, B.
Sawhney ed., the Clay Minerals Society (1996)

R.M. Barrer, Pure & Appl. Chem. 61, 1903 (1989)

SH. Xu, G.Y. Sheng, S.A. Boyd, Adv. Agron. 59, 25 (1997)

W.F. Jaynes, G.F. Vance, Soil Sci. Soc. Am. J. 60, 1742 (1996)

M.M. Mortland, S. Shaobai, S.A. Boyd, Clays Clay Miner. 34, 581 (1986)
S.A. Boyd, S. Shaobai, J .-F. Lee, M.M. Mortland, Clays Clay Miner. 36, 125
(1988)

W.F Jaynes, S.A. Boyd, Soil Sci. Soc. Am. J. 55, 43 (1991)

W.L. Udo, T. Lee, T.J. Pinnavaia, Adv. Mater. 8, 79 (1996)

W.L. Udo, T.J. Pinnavaia, J. Solid State Chem. 139, 281 (1998)

R.A. Hatch, R.A. Humphrey, W. Eitel, J .E. Comeforo, U. S. Bureau of Mines,
Reports of Investigations 5337 (1957)

RC. Johnson, H.R. Shell, U.S. Bureau of Mines, Reports of Investigations 6235
( 1963)

HR. Shell, K.H. Ivey, U.S. Bureau of Mines, Fluorine Micas, Bulletin 647 (1969)
R.M. Barrer, D.L. Jones, J. Chem. Soc. (A) 1532 (1970)

R.M. Barrer, D.L. Jones, J. Chem. Soc. (A) 503 (1971)

R.M. Barrer, D.L. Jones, J. Chem. Soc. (A) 2594 (1971)

G.H. Bea], D.G. Grossman, S.N. Hoda, KR. Kubinski, Corning Glass Works,
U.S. Patents 4,239,519 and 4,297,139 (1980)

K. Kitajima, F. Koyama, N. Takusagawa, Bull. Chem. Soc. Jpn. 58, 1325 (1985)

152

[18]
[19]
[20]
[21]

[22]

[23]

[24]

[25]

[26]

K. Kitajima, N. Takusagawa, Clay Miner. 25, 235 (1990)

W.J. Paulus, S. Komarneni, R. Roy, Nature 357, 571 (1992)

K.R. Franklin, E. Lee, J. Mater. Chem. 6, 109 (1996)

E. Busenberg, C.V. Clemency, Clays Clay Miner. 21, 213 (1973)

G. Engelhardt. High Resolution Solid-State NMR of Silicates and Zeolites, J.
Wiley&sons Ltd. (1987)

Woessner, D.E., Am. Miner., 74, 203, 1989

J. Sanz, J .M. Serratosa, J. Am. Chem. Soc. 106, 4790 (1984)

M. Lipsicas, R.H. Raythatha, T.J. Pinnavaia, I.D. Johnson, R.F. Giese, PM.
Costanzo, J .L. Robert, Nature 309, 604 (1984)

SJ. Gregg, K.S.W. Sing, Adsorption, Surface Area, and Porosity, Academic

Press, New York (1982)

153

CHAPTER 6
A Model for the formation of
Layered Silicate Heterostructures
with

Regularly alternating Organic and Inorganic Cation Galleries.

6.1 Abstract.

A model is presented that explains the formation of mixed ion heterostructured
clays with regularly alternating organic and inorganic galleries. Two criteria were defined
that constitute the requirements for heterostructure formation. Condition 1 defines layer
charge compensation conditions while condition 2 embodies the distribution of cationic

species in mixed ion clays. Mixed ion heterostructures are formed when both conditions

are fulfilled.

154

 

6.2 Introduction.

Mixed ion fluorohectorite heterostructures with regularly alternating organic and
inorganic cation galleries may be formed under certain circumstances. Several factors
govern the formation process. The formation conditions and boundaries of mixed ion
heterostructures were systematically investigated and described in detail in the previous
chapters.

The experimental results for the mixed ion heterostructure formation process
provide information upon which a heterostructure model may be built. This model is
developed as a general guideline for mixed ion heterostructure formation. It may be
useful when the synthesis of a heterostructured material is attempted based on the
intercalation chemistry of lamellar phases other than 2:1 layered silicates.

From the work on C15H33NBu3+ and Na+ mixed ion fluorohectorite clay

heterostructures we know that:

a. The two cationic species in a 1:1 C15H33NBu3+ and Na+ mixed ion heterostructure
are segregated into distinct clay interlayers. In addition, the two gallery types are regularly
stacked in the air dried product. A mixed ion heterostructure has a repeat unit which is
equal to the sum of the basal spacings of the homoionic onium- and metal ion —
exchanged clays. The mixed ion heterostructure repeat distance changes in proportion to

the gallery height of the homoionic organic- or inorganic- exchanged clays.

b. There are three synthetic pathways that produce mixed ion heterostructures. The

heterostructure is, however, a thermodynamically stable product. This is most obvious in

155

Pathway 111 In this approach, equal molar quantities of parent end member clays are
reacted in water to spontaneously produce a mixed ion heterostructured product. Pathway
I involves the partial replacement of hydrated metal ions by the direct addition of onium
ions to a clay suspension. Route II is the reverse approach where an organo clay is treated
with a concentrated NaCl solution to produce an organic and inorganic mixed ion

heterostructured clay.

c. The hydrated metal ions in mixed ion heterostructures can be of the type A“. For
instance, Na+ in a 1:1 C15H33NBu3+ and Na+ mixed ion heterostructure can be exchanged
through ion exchange by K+, or A113”. More importantly, An+ homoionic clay can be
directly converted into a 1:1 mixed ion heterostructure when half an equivalent of
C16H33NBu3+ ions are exchanged onto the clay, Pathway 1. Thus, an exfoliated clay is not
a requirement in a route I approach. However, a metal ion gallery can not be converted

into an organo gallery with a retention of the heterostructured arrangement.

(I. The organo interlayers in a 1:1 C16H33NBu3+ and Na” mixed ion heterostructure
are comparable with the galleries found in homoionic C16H33NBu3+ exchanged clay. Both
the Epon 828 epoxy swelling experiments and the 2,4-dichlorophenol adsorption
experiments indicate that the organo galleries are behaving the same in both organo

modified clays.

e. The onium ion alkyl chain length only determines the extent to which the metal

ions are replaced with onium ions in a partial ion exchange reaction, Pathway I.

156

Quantitative ion exchange is obtained for CnH(2n+1)NBu3+ onium ions with n_>_12. Metal

ion replacement is not complete when onium ions with shorter alkyl chains are used.

1'. Onium ion head group size controls mixed ion heterostructure formation.
Heterostructures are formed when the footprint of the onium ion matches the clay surface
charge density in a fluorohectorite clay; C1(5H33,N(C,,,H(2m+1))3+ with m23. Onium ion head
groups that do not cover the clay surface completely do not produce mixed ion
heterostructures; C1(5H33N(C,,,,I-I(2,,..,1,)3+ with mSZ. These smaller head group sizes allow

for the formation of phase segregated homoionic parent end member clays.

g. Mixed ion heterostructure formation is not limited to clay systems that contain
half an exchange equivalent of C15H33NBu3+ ions. A mixed ion heterostructured solid
solution is obtained when the overall onium ion fraction, fq, falls between 0.30 and 0.50.
The organic and inorganic galleries are still regularly stacked in a 1:1 alternating fashion
in these heterostructures, although there is seemingly an excess of metal ions in the clay
system. Furthermore, the C1¢5H33NBu3+ and Na+ mixed ion heterostructured solid solution
is accompanied by phase segregated homoionic Na+ clay when fQ drops below 0.30.
Homoionic C161-133NBu3+ exchanged clay and 1:1 C16H33NBu3+ and Na+ mixed ion
heterostructured clay are produced when the onium ion fraction fQ exceeds 0.50. Also, the

amount of organo clay formed increases with an increase in onium ion fraction fQ.

h. Mixed ion heterostructures were obtained only for 2:1 layered silicate clays that

had an octahedral layer charge only. The presence of layer charge derived from

157

tetrahedral substitutions in the 2:1 clay structure has an inhibiting effect on the
heterostructure formation process. These clays with tetrahedral layer charge formed
organo clays and mixed ion homostructures, wherein both cationic species are mixed in

the clay interlayers.

6.3 The mixed ion heterostructure formation model.

A mixed ion heterostructure formation entails communication between successive
interlayers. This prerequisite ensures the characteristic regular alternation of galleries.
This implies that the hydrated metal ions favor exchange positions on the opposite side of
an organic exchanged interlayer. This preference is initiated by steric restrictions of the
onium ions. The approach of the center of onium charge to the clay layer surface is
restricted by the size of the head group and this induces an unfavorable charge
compensation. Neutralization of layer charge is improved when the smaller hydrated
inorganic metal ions assume positions on the opposite side of the onium ion intercalated
gallery.

Segregation of cationic species is attained when the onium ion head groups cover
the entire clay surface in an organic interlayer. In other words, commingling of organic
and inorganic cations within an onium ion dominated gallery is precluded when the head
groups of the onium ions fully cover the silicate surface.

A heterostructured arrangement of the intercalated galleries improve clay layer
charge compensation. Notably, the clay layer charge in the homoionic onium ion

exchanged clay is not compensated for in an optimal manner. Consequently, a mixed ion

158

heterostructured clay is the thermodynamically more stable intercalate when both

homoionic parent end member clays are present in a clay suspension.

Thus, the formation conditions for a mixed ion heterostructured 2:1 clay are:

1. An onium ion head group must be sufficiently large to cover the entire basal surface of
the galleries and to spatially limit the approach of the onium ion center of positive
charge to the clay layer.

2. Commingling of cationic species within the organic interlayers of a heterostructured
clay is limited in part by differences in hydration. Up to 30 % of the organic cations in
a 1:1 heterostructure can be replaced by inorganic Na+ ions, giving rise to 1:1

heterostructured solid solutions over the composition range corresponding to 0. 35 SfQ

S 0.50 where fQ is the onium ion fraction in the mixed ion clay.

6.4 Verification of the mixed ion heterostructured clay formation model.

Attempts to synthesize a mixed ion heterostructure failed when half an exchange
equivalent of C16H33NEt3+ ions were intercalated in taeniolite, a synthetic clay of
composition Liz-[Mg4Li2]Si3020F4. These ethyl terminated onium ions should have been
large enough in size to cover the taeniolite layer surface completely. Apparently, these
onium ions compensate layer charge quite well. Requirement 1 is therefore not realized.
Again, taeniolite mixed ion heterostructures are formed when propyl- or larger onium ion

head group sizes are used, although they were always accompanied by the homoionic

159

parent end member clays. With such onium ions, both heterostructure formation
conditions are satisfied.

The presence of some tetrahedral charge on fluorohectorite-like clay layers
inhibits the capability of the clay to form C16H33NBu3+ and Na+ mixed ion
heterostructures. Although requirement 1 is satisfied upon the intercalation of half an
exchange equivalent of C1¢5H33NBu3+ ions, a heterostructure is still not formed.
Tetrahedral layer charge is preferably compensated by metal ions. Thus, these metal ions
can always be found on both sides of a clay layer. This initiates mixing of cation species
within interlayers. Moreover, an onium ion that is hindered in the approach to the gallery
surface may prefer to compensate octahedral layer charge opposite of a tetrahedral charge
location, which already has a charge compensating metal ion. This also explains why
monolayer onium ion intercalates are formed when these clays are treated with half an

exchange equivalent of C15H33NBu3+ ions.

6.5 Conclusions.

A model is developed that describes the parameters of mixed ion heterostructure
formation. Two general conditions were formulated that concern the onium ion geometry
and the distribution of cationic species within interlayers. The model does explain the
discussed intercalation results that were described in the previous chapters.

The model also permits speculation why natural clay minerals with a layer charge
comparable to fluorohectorite do not form mixed ion heterostructured clays with half an
C16H33NBu3+ onium ion fraction, f0 = 0.50. The presence of tetrahedral layer charge

inhibits the clay heterostructure formation capability. Most likely, onium ions take a

160

preferential position on exchange sites opposite to tetrahedral layer charge locations.
Next, all remaining onium ions are accumulated in segregated domains on both sides of
the clays layers by means of the van der Waals interactions between the onium ion alkyl
tails. Thus, onium ion domain assemblies are distributed throughout all interlayers. A

mixed ion homostructure is obtained instead of a mixed ion heterostructure.

161

Hrc

"llllllllllllllflinwgiflfir