x... Emu u.
flumwém.

$3., ,
‘7 ‘ v

rs... C it: V V . , .V , .
. . 7. 4 .
v.1. a...“ 5 , V V V
55...}! . , . . , V
V . , . .. . .., .
. V . , savage»...
V :22: .
. an...
, n'

3%.

 

 

2.3"...31
. .2.

 

. i
22...... .
3:13. ll.
.1..

 

”11.5..
(iv...

wag?” .-

 

2““

fl.,--‘” 1
. . _

mun

 

I‘
l

”x.:._'.x&1.nt1aI1 a. a: 3.2.»! 1...th

mags... . a I...1.-._V is V ,
. §§§$ V €V_.VT§§ 4%

. ..
,i :37 . 2.)
.. A...

   

      

11!. ‘7-

JIM. 3:35;}. 3:3

  

 

m.
o,-

      

  

‘1'!

a.
r I
n

 

 

 

 

 

‘-

This is to certify that the

dissertation entitled

Synthesis, Characterization and Properties
of Clay —Reinforced Thermoset Nanocomposites

presented by

Peter C. LeBaron

has been accepted towards fulfillment
of the requirements for

ph'n degree in 2001

 

Mafi’r professor “fl. m

Date 1 2 L3 \IO‘

MS U is an Affirmative Action/Equal Opportunity Institution 0- 12771

 

LIBRARY #‘
Michigan State
Universuty

 

 

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

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDue.p65-p.15

 

 

SYNTHESIS, CHARACTERIZATION AND PROPERTIES OF
CLAY — REINFORCED THERMOSET NANOCOMPOSITES
By

Peter C. LeBaron

A DISSERTATION

Submitted to Michigan State University in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department Of Chemistry

2001

ABSTRACT

SYNTHESIS, CHARACTERIZATION AND PROPERTIES OF CLAY -
REINFORCED THERMOSET NANOCOMPOSITES
By

Peter C. LeBaron

Nanocomposites are formed upon the uniform mixture of two or more
distinct phases over a nanometer length scale. Typically the components consist
of a stiff reinforcement phase dispersed in a polymeric matrix. Due to the
improved surface contact and interlacial properties Of nanocomposites, these
materials commonly exhibit superior mechanical behavior relative to
conventionally filled systems. The first example Of layered silicate materials
employed as a reinforcement phase was Shown by Toyota researchers in
exfoliating organOCIay in a nylon-6 matrix. The hybrid material exhibited a
dramatic improvement in thermal and mechanical properties, and ultimately
found practical use in under-the-hood applications in the automobile industry.

The present work focuses on nanocomposite syntheses of two distinctly
different polymeric systems, elastomeric polydimethylsiloxanes and rigid epoxies.
Each polymeric medium requires unique layered silicate organO-modification,
reaction conditions and sample treatment. These concepts are addressed along
with the mechanism of Clay exfoliation and an investigation on the final

properties.

Silicone rubbers gained remarkable toughness, as the stiffness,
elongation and tensile strength were notably improved upon the exfoliation of
Clay particles. Permeability Of the Silicone samples was not substantially
influenced by the presence Of Silicate layers, however, and this lead to further
investigations on the post — cure reactivity Of the silicone nanocomposites.

The nanocomposites formed in a glassy epoxy network also showed
substantial mechanical improvement compared to the unfilled epoxy. In the case
with epoxies, the amount and type Of organic modification used on the Clay was
scrutinized. Nanocomposites were synthesized from reduced organic — content

Clays and were noted to retain epoxy intercalation and exfoliation properties.

ACKNOWLEDGEMENTS

First and foremost, I am especially thankful to have had the Opportunity to
complete my Ph.D. under the best advisor in the business, Professor Thomas J.
Pinnavaia. He has served as a Close friend, providing helpful advice on some of
the more important aspects in life (relationships, health and fishing), along with
his excellent tutelage and guidance in research. I wish every graduate student
would have the Chance to work for such an intelligent, creative, kind preceptor.

I am also extremely appreciative for the support and patience my terrific
wife has shown me throughout my graduate career. Higher education is an
investment many do not understand, however She never complained (to me at
least) about the partial delay graduate school placed on our lives. For this
reason I would like to dedicate my thesis tO her, even though she really doesn’t
care for Chemistry and certainly won’t read any of this book.

Many former and current Pinnavaia group members have had an impact
on my life, but I would especially like to thank my original mentor, Zhen Wang,
and my current lab mate Costas Triantafillidis for their help and collaborations.
My Close friends Liu and KimbO have been wonderful sources of knowledge and
incite through the years, and I know someday they will be highly successful
industrial scientists. Other group members barely worth mentioning are In Park,
Mark Sislo, Abhi Karkamkar and Randy Hicks, whom I thank for eating pizza with
me. I would also like to acknowledge Jainisha Shah and Emily MCKimmy as they

were indispensable in helping me format my thesis, a job I am not cut out for.

iv

TABLE OF CONTENTS

LIST OF TABLES

LIST OF FIGURES

CHAPTER 1 INTRODUCTION

1.1

1.2

1.3

Nanocomposite Materials

1 .1 .1 Nanocomposite Concept

1 .1 .2 Nanocomposite Significance

1.1.3 Sol-Gel Synthesized Nanocomposites

1.1.4 Layered Silicate Nanocomposites

1.1.5 Smectite Clay Structure

1.1.6 Organoclay Structures and Properties

1.1.7 Organoclay - Polymer Interactions

1.1.8 Properties of Themioset — Clay Nanocomposites
1.1.9 Exfoliated Nylon-6 — Clay Nanocomposites
Recent Developments in Polymer — Clay Nanocomposites
Research Objectives

1.3.1 Current Issues and New Directions

1.3.2 Research Goals

1.3.3 References

11

14

17

20

22

22

23

25

CHAPTER2 CLAY NANOLAYER REINFORCEMENT OF A SILICONE

ELASTOMER: EXFOLIATION OF FLUOROHECTORITE IN

POLYDIMETHYLSILOXANE

' 2.1 Introduction 30

2.2 Experimental 33

2.3 Physical Measurements 38

2.4 Results 41

2.5 Discussion 49

2.6 Conclusions 75

2.7 References 76
CHAPTER 3 POST-CURE ION EXCHANGE OF EXFOLIATED
FLUOROHECTORITE CLAY IN

POLYDIMETHYLSILOXANE NANOCOMPOSITES

3.1 Introduction 79
3.2 Experimental 95
3.3 Results and Discussion 99
3.4 Conclusions 108
3.5 References 1 09
CHAPTER4 EXFOLIATION OF DIAMINE — MODIFIED CLAYS IN A
GLASSY EPOXY MATRIX

4.1 Introduction 11 1

vi

4.2 Experimental 1 15

4.3 Results and Discussion 119

4.4 Conclusions 151

4.5 References 1 52
CHAPTERS REDUCING THE ORGANIC MODIFIER CONTENT IN
EPOXY NANOCOMPOSITES: MIXED-ION

INORGANIC/ORGANIC MONTMORILLONITE AND
FLUOROHECTORITE HOMOSTRUCTURES

5.1 Introduction 155
5.2 Experimental 1 59
5.3 Results and Discussion 164
5.4 Conclusions 188
5.5 References 1 89
LIST OF TABLES
Table 1.1 Mechanical and Thermal Properties Of Nylon-6 20
Table 2.1 Silanol Terminated PDMS Monomers 34

Table 2.2 Mechanical Properties Of Silicone Elastomer - C15 FH 47

vii

Table 2.3

Table 3.1
Table 3.2

Table 4.1

Table 4.2

Table 5.1

Table 5.2

Organoclay Nanocomposites

Permeability and Solvent Uptake Properties of Silicone

Elastomera - C15 FH Organoclay Nanocomposites

Silanol Terminated PDMS Monomers

Copper Content Present After Post — Cure Ion Exchange
Composition and Basal Spacings Of the Inorganic and
Organic Clays used for the Preparation of the
Nanocomposites

Dynamic Mechanical Analysis and Thermal Analysis data
of the pristine epoxy polymer and the Clay-polymer

nanocomposite samples

Composition and Basal Spacings Of the Homoionic and

Mixed-ion Inorganic/Organic Clays

Dynamic Mechanical Analysis and Themal Analysis

Data of the Pristine Epoxy Polymer and the Clay-

Polymer Nanocomposite Samples

viii

48

96
1 03

125

141

167

183

Figure 1.1

Figure 1.2

Figure 1.3

Figure 1.4

Figure 1.5

LIST OF FIGURES

Depiction of the sol-gel approach to synthesizing
nanocomposites indicating the metal oxide filler phase
and the polymeric Silicone precursor.

Illustration Of the layered oxide framework Of smectite
clays. Clay layers consist of two tetrahedral sheets
sandwiching a central octahedral layer. Mn+ - szO
represents the interlayer exchangeable cation with its
coordination sphere of water molecules.

Illustration Of four possible organoclay structures based
on surfactant packing density. Given the same
surfactant, increasing the Charge density of the Clay will
move from a monolayer arrangement towards the paraffin
structure.

Schematic Of the different composite morphologies
possible when attempting to synthesize polymer - Clay
nanocomposites. Representative XRD patterns are
displayed for each type.

Schematic representation Of the tortuous path invoked on

perrneants by having dispersed Clay layers in a polymer

ix

13

16

matrix.

Figure 1.6 General synthetic route for synthesizing nylon - 6 19
nanocomposite materials. The caprolactam precursors
were first melted, then added to the organoclay to permit
intercalation (causing a 15 A basal spacing increase), and

finally polymerized at 250°C.

Figure 2.1 Schematic illustrating the organic ion exchange Of 35
fluorohectorite Clay using cetyltrimethylammonium

bromide.

Figure 2.2 General Silicone nanocomposite synthesis scheme. After 37
polymerization silicone samples were carefully removed

from the steel molds for final property assessment.

Figure 2.3 X-ray diffraction patterns of synthetic 42
hexadecyltimethylammonium fluorohectorite, abbreviated
C16FH, intercalated by PDMS polymers with molecular
weights Of 400, 1500-2000, and 4200. The pattern for the
initial organo Clay is included for comparison.

Figure 2.4 X-ray diffraction patterns for mixtures of 44
hexadecyltrimethylammonium fluorohectorite and PDMS

- 4200 at different stages Of cure. The curing agent was

Figure 2.5

Figure 2.6

Figure 2.7

Figure 2.8

tertraethylorthosilicate and the catalyst was tin(ll) octoate.
The inset indicates the presence Of an in-plane 060 Clay
peak in the exfoliated composite

Comparison of the stress - strain curves for pristine
PDMS and PDMS - C15FH nanocomposites containing

5.0 and 10 wt "/0 C15FH organoclay.

Schematic depicting potential intragallery surfactant

arrangements typically found in organomontmorillonites.

Inclined paraffin surfactant orientation found in higher
Charge density Clays, including fluorohectorite.
Cetyltrimethylammonium fluorohectorite has a paraffin
arrangement with an approximate angle Of inclination with

the Clay surface of 50°.

Schematic representation of the swelling of an
alkylammonium ion exchanged fluorohectorite by a long
Chain PDMS polymer (MW = 4200) containing terminal
OH groups. Shorter PDMS Chains with molecular weight
fractions Of 400 and 1750 may have only one end of the

Chain associated with the gallery surfaces.

xi

46

50

52

55

Figure 2.9

Figure 2.10

Figure 2.11

Figure 2.12

Photograph Of PDMS polymer - C16FH suspensions
containing 5.0 wt °/o organoclay. The tubes are filled in
left to right order with methyl terminated PDMS - 400 and
Silanol terminated PDMS - 400, - 1750, and - 4200,

respectively.

Tin octoate — catalyzed PDMS crosslinking. Although
there is some debate as to the exact mechanism, the
active species is widely thought to be the hydrolysis
product of the tin bis(2-ethylhexanoate) catalyst.

Photographs Of a pristine silicone elastomer formed from
PDMS - 4200 (upper specimen) and the corresponding
nanocomposite containing 5.0 wt "/0 C16FH organoclay
(lower specimen) after equilibration in a cyclohexane bath
and subsequent removal from the bath to allow for
solvent evaporation: (a) immediately after removal from
the bath, (b) after 15 seconds, and (C) after 20 minutes.
NO structural damage was noted for the nanocomposite
after complete evaporation Of the solvent.

Depiction of permeant flow through nanocomposite films.
A more tortuous path is found perpendicular to the film
when Clay particles are oriented parallel to the film
surface. Low surface tension fluids and thick films limit

the percentage of parallel -- to — surface oriented

xii

58

60

64

66

Figure 2.13

Figure 2.14

Figure 2.15

Figure 3.1

nanolayers.

TEM image Of Silicone nanocomposite Showing parallel
orientation Of nanolayers stemming from the same
tactoids. Prior to intercalation and exfoliation in silicone,
the organofluorohectorite galleries were separated by 2.7
A, and after completion Of crosslinking layers are 70-150

nm apart.

TEM image Of silicone nanocomposite Showing
neighboring tactoids with near orthogonal orientations.
Such an image is rare in Clay nanocomposites where
adjacent Clay tactoids are usually oriented parallel to one

another.

Schematic Of tin catalyst existing in different forms during
crosslinking of the silicone rubber. The moisture sensitive
tin carboxylate will hydrolyze more readily when in Close
contact with the fluorohectorite Clay due to the small
amount (26% by wt.) of water present at the Silicate

surface.

Schematics representing known reactive membrane and
selective gas separation

xiii

68

7O

74

82

Figure 3.2

Figure 3.3

Figure 3.4

Figure 3.5

Figure 3.6

Figure 3.7

procedures using functionalized aluminosilicates.

Schematic showing the reduced swelling observed in
nanocomposites due to the constrained polymer at the
interface Of the reinforcer.

XRD patterns of silicone nanocomposites cured under
different conditions. Accelerating the cure with premature
heating (top pattern) leads to a lower degree Of exfoliated

clay layers.

Schematic Of the procedure involved in synthesizing and
post-cure ion exchanging silicone nanocomposites.
Representation of the crosslinked silicone nanocomposite

before and after immersion in THF.

Copper content following post-cure ion exchanges for
silicone nanocomposites examined after successive THF
washes. Unleachable copper accounted for roughly 70%
Of the fluorohectorite ion exchange capacity in the filled
systems, while pristine silicone retained almost no copper

after washing.

Schematic representation Of the post-cure ion exchanges

xiv

85

88

91

94

100

102

Figure 3.8

Figure 4.1

Figure 4.2

Figure 4.3

using a CuCI2/T HF exchange bath and pure THF as the
wash. Uptake Of copper into both nanocomposite and
pristine samples was Observed, however upon purging

with THF only the nanocomposites retained copper.

Schematic illustrating copper ion exchanges carried out in
THF for different parent clay systems. Excess copper
successfully substitutes for both the inorganic cations in
lithium fluorohectorite as well as the majority Of surfactant
ions on organofluorohectorite exfoliated in a silicone
matrix. However, despite increasing the copper
concentration to 20 times the CEC of the Clay, organic
cations could not be appreciably displaced by Cu+2 in

THF.

Acid catalyzed intragallery ring - opening of an a epoxy
resin.

X-ray diffraction patterns Of the Li*-Fluorohectorite (Li-FH)
and organically modified FH-Clays (with diprotonated

diamines 0230, D400, D2000).

X-ray diffraction patterns of mixtures of the organic Clays
DZOOO-FH, the epoxy monomer EPON 826 and the curing
agent 0230: (A) after intercalation of EPON 826 in

XV

107

114

122

128

Figure 4.4

Figure 4.5

Figure 4.6

02000-FH at 50°C for 6 hrs, (B) after intercalation Of
EPON 826 for 24 hrs, (C) after addition of D230 in the
mixture Of step C, outgassing and heating at 75°C for 0.5
hrs, (D) after curing at 75°C, 3 hrs and 125°C, 3hrs (the
Clay loading was 6 wt. % Silicate basis). The inset shows
the presence of an in-plane 060 Clay reflection in the

cured nanocomposite.

TEM images of thin sections of the glassy epoxy
nanocomposites prepared from 02000-FH. The clay
loading Of the nanocomposite samples was 6 wt. %

silicate.

TEM images of thin sections of the glassy epoxy
nanocomposites prepared from 02000-PGW. The Clay
loading Of the nanocomposite samples was 6 wt. °/o
Silicate.

Indeed, the nanolayers of the lower aspect ratio 02000-

PGW Clay were more randomly dispersed in the polymer

matrix and existed in a more exfoliated arrangement.

X-ray diffraction patterns of EPON 826 - montmorillonite
PGW Clay composites and nanocomposites using: (A) H-
PGW, (B) 0230-PGW, (C) 0400-PGW and (D) 02000-
PGW Clays. NO intercalation occurred for the inorganic,
nor the 0230 and 0400-PGW Clays. The inset Shows the
presence of an in-plane 060 Clay reflection in the cured
02000-PGW nanocomposite.

xvi

131

133

135

Figure 4.7

Figure 4.8

Figure 4.9

Figure 4.10

Schematic portraying the three point bending device
used in Obtaining DMA data and the plot of DMA testing
parameters. The amount Of energy the material retums is
related to its storage modulus, the energy dissipated in
the internal motions of the polymer chains is related to its
loss modulus and the actual measured material lag in
experiencing the oscillatory force is the phase angle.

Dynamic Mechanical Analysis measurements indicating
storage modulus vs. temperature curves for (A) Pristine-A
epoxy polymer and for the nanocomposite samples
prepared from (B) H-PGW, (C) 02000-PGW, (0) 02000-
FH, (E) 0230-PGW, (F) 0230-FH. The Clay loading Of the

nanocomposite samples was 6 wt. % Silicate basis

Dynamic Mechanical Analysis measurements indicating
tan6 vs. temperature, for (A) Pristine-A epoxy polymer
and for the nanocomposite samples prepared from (B) H-
PGW, (C) 02000-PGW, (0) 02000-FH, (E) 0230-PGW,
(F) D230-FH. The Clay loading of the nanocomposite
samples was 6 wt. % Silicate basis

TGA curves Of the organic Clays (A) 02000-PGW and (B)
02000-FH, Of the nanocomposites prepared from (C)
02000-PGW and (0) 02000-FH Clays, and Of the Pristine-

xvii

139

143

145

149

Figure 4.11

Figure 5.1

Figure 5.2

Figure 5.3

A epoxy polymer (E). The Clay loading of the
nanocomposite samples was 6 wt. % silicate basis.

TGA curves of the organic Clays (A) 0230-PGW and (B)
0400-PGW, of the nanocomposites prepared from (C)
0230-PGW and (0) 0400-PGW Clays, and of the Pristine-
A epoxy polymer (E). The Clay loading Of the
nanocomposite samples was 6 wt. % silicate basis.

X-ray diffraction patterns of the homoionic and mixed-ion
inorganic/organic montmorillonite (PGW) and
fluorohectorite (FH) Clays prepared by ion exchange Of
the inorganic cations Na“, Li+ or H+ with onium ions of

diprotonated 02000.

X-ray diffraction patterns of mixtures Of the mixed-ion Clay
Li/020000,55-FH, the epoxy monomer EPON 826 and the
curing agent 0230: (A) after intercalation of EPON 826 in
Li/020000_65-FH at 50°C for 12 hrs, (B) after addition Of
0230 to the mixture of step A and outgassing at 25°C,
and (C) after curing at 75°C, 3 hrs and 125°C, 3hrs (the
Clay loading was 6 wt. % Silicate basis). The inset shows
the presence of an in-plane 060 Clay reflection in the
cured nanocomposite

TEM images of thin sections of the glassy epoxy

xviii

150

169

173

177

Figure 5.4

Figure 5.5

Figure 5.6

nanocomposites prepared from the mixed-ion Clays (a,b)
Li/020000,65-FH. The Clay loading Of the nanocomposite
samples was 6 wt. % silicate basis.

TEM images of thin sections Of the glassy epoxy
nanocomposites prepared from the mixed-ion Clays (a,b)
H/020000_5-PGW. The Clay loading of the nanocomposite

samples was 6 wt. % silicate basis.

Dynamic Mechanical Analysis measurements: (a) Storage
Modulus vs. temperature and (b) tan6vs. temperature, for
(A) Pristine-A epoxy polymer and for the nanocomposite
samples prepared from (B) H-PGW, (C) H/020000_25-
PGW, (0) Li/020000.65-FH, (E) 02000-PGW. The Clay
loading Of the nanocomposite samples was 6 wt. %

silicate basis .

TGA curves of the mixed-ion clays (A) H/020000,35-PGW
and (B) Li/020000,35-PGW, the nanocomposites prepared
from (C) H/020000_25-PGW and (0) 02000-PGW Clays,
and of the Pristine-A epoxy polymer (E). The Clay loading

of the nanocomposite samples was 6 wt. °/o silicate basis.

xix

178

181

187

Chapter 1
INTRODUCTION

1.1 Nanocomposite Materials
1.1.1 Nanocomposite Concept

Multi-matrix materials consisting Of a stiff reinforcement phase embedded
in a polymeric phase fall into the general Class Of composite materials. Due
primarily to the improvement in mechanical properties not obtainable by either
parent end member independently, such materials have found widespread
applications in nearly every facet of the modern world. For this reason,
accompanied with the drive to produce more environmentally sensitive materials
in place of energy consuming metal -based components, composites have been
intently studied recently and a variety Of new Classes have been explored. One
such relatively young category Of composite materials exhibits a Change in
composition and structure over a nanometer length scale."7 These nanO-scaled
hybrid materials, termed nanocomposites, have been Shown in the last decade to
afford remarkable property enhancements relative to conventionally scaled
composites.
1.1 .2 Nanocomposite Significance

The macroscopic effect Of downsizing the filler size Of composite materials

is remarkably manifested on the performance properties Of the final material.

Through decreasing the Size Of the reinforcement phase by three orders of
magnitude, which is typical in going from conventionally scaled composites to
nanocomposites, the surface area to volume ratio, and thus the interfacial
contact between the dissimilar components iS monumentally greater for the
nanO-scaled material. This results in a more efficient transfer Of potentially
damaging forces to the stronger filler phase, rendering the strained polymer
matrix more unaffected in nanocomposites compared to conventional
composites. Also important to note is the ability of a smaller reinforcement phase
to influence more of the polymer at much reduced loadings, giving rise to
significantly lighter materials. This is a primary driving force for the expansion Of
nanocomposite use, as the notion Of lower weight structural parts is of principal
concern in modem development of the automobile and aeronautical industries.
The need for large volume fractions Of conventional fillers to achieve the desired
Characteristics in the final material also adds to the overall cost, as typical
conventional reinforcers, such as fiber glass and graphite, are relatively
expensive materials, especially if they need to be modified prior tO addition to the
polymer. The layered silicates, on the other hand, are found ubiquitously in
nature and can be made polymer — ready through relatively inexpensive
purification means.°"°
1.1.3 Sol-Gel Synthesized Nanocomposites

The use Of naturally occurring layered silicates is not the only nanO-scaled
reinforcement phase used in strengthening polymer media. Another approach to

inorganic — organic hybrid materials involves the hydrolysis and condensation Of

metal oxide precursors from metal alkoxides in the presence of similarly
functionalized polymer precursors.°' ‘1 The most common polymer reinforced in
this way is polydimethylsiloxane, and a general depiction Of this synthetic

scheme is illustrated in figure 1.1.

Metal Oxide Polydimethylsiloxane

Filler Polymer
M(OR)4 + H2 HO-('Si(CH3)3)n-0H
' +
(RO)3 M-OH + ROH 51(01304

l

\ / /
: MD“: + ROH/HZO + ":‘Si-O-(SMCH3)3)n-O-Si:

M 2 Si, Al, Ti, Zr

 

 

 

Metal oxide — PDMS Nanocomposite

 

Figure 1.1 Depiction Of the sol-gel approach to synthesizing nanocomposites
indicating the metal oxide filler phase and the polymeric Silicone precursor.

1.1.4 Layered Silicate Nanocomposites

The morphology of layered silicates, which will be described more in depth
shortly, makes them particularly suitable for use as a reinforcement phase in
engineering nanocomposites. Their platy, high aspect ratio structure lends such
silicates towards better interfacial interactions at the polymeric boundary relative
to larger scaled, more spherical reinforcement phases. The Chemical makeup of
these Clay materials consists Of a dimensionally stable silicon - oxygen network
with four parallel oxide layers making up each individual silicate Sheet.12 The
robust, stiff nature of the Clay layers is not easily exploited in polymeric matrices,
however, due to the strong tendency of the layers to form face - to — face
stackings, or tactoids, that are difficult to separate into individual layers. This
dilemma can be thwarted by the use of a particular Class Of layered silicates
employed as the reinforcer, the smectite Clays. As will be discussed in section
1.1.5, there are a variety Of layered silicates that occur naturally, and these can
be distinguished from one another and separated into Classes based on the
Charge carried in each layer. Smectite Clays are in the middle of the Charge
density continuum, giving maximum swellability of the layers and thus providing a
means of disseminating the sheets. This occurs due to the fact that layered
silicates carrying too low a Charge density, ie talcs, reduces or nullifies (in the
case Of a neutral layer) the driving force to uptake solvent into the ionic gallery
regions, while too high a Charge density, ie micas, causes an electrostatic
interaction between adjacent layers that cannot be overcome and the layers are

essentially locked together. The ability to disseminate smectite tactoids through

solvent swelling is what has extended the practical applications of these minerals
to absorbants, lubricants, rheological fluids and, more recently, polymeric
reinforcers.
1.1.5 Smectite Clay Structure

Clay minerals are largely two - dimensional layered silicates with individual
particle sizes on the order Of ~ 2pm, depending on the specific type of Clay. The
aforementioned oxide network (composed of four discrete 001 planes) gives rise
to what is known as a 2:1 mica — type morphology with a layer thickness of 9.6 A.
In the notation, “2" indicates the presence of two tetrahedral silica - based sheets
that lie on either side and form a sandwich with “1” octahedral Sheet. The central
octahedral layer is composed Of edge — shared octahedra with the coordination
sites generally occupied by aluminum or magnesium, while the tetrahedral layers
are made up of comer — shared SiO4 units. Variations in the Clay charge density
stems from isomorphous substitution Of lower valent metal ions in either Of the
layers (e.g. AI+3 for Si+4 in the tetrahedral sheet, or Li+ for Mg”; Mg+2 or Fe+2 for
Al”3 in the octahedral sheet), or site vacancies giving rise to a net negatively
Charged layer. The gallery cations, ordinarily Na*, K“ or Ca”, are located
between the Clay layers and electrostatically balance the Charge. Figure 1.2

illustrates a typical smectic clay layer.12

9.6 A

 

 

 

0 Oxygen 0 Si/Al OOH .AllMg

Figure 1.2 Illustration of the layered oxide framework Of smectite Clays. Clay
layers consist of two tetrahedral sheets sandwiching a central octahedral layer.
MM - yH20 represents the interlayer exchangeable cation with its coordination

sphere Of water molecules.

Clay does not exist naturally as discrete layers, but instead stacks face — to —
face in turbostratic (only aligned in one direction) tactoids. While the presence Of
inorganic gallery cations makes the Clay hydrophilic and, in the case of the
smectites, easily swollen with water, the cations are readily exchangeable and
upon the ion exchange with carbon - based surfactants the galleries become
organophilic. The morphology and intercalation properties of these organic Clays
(organoclays) have been studied extensively and will be briefly covered here.
1.1.6 Organoclay Structures and Properties

Unlike inorganic clays, the greater volume Of the Charge balancing
surfactant molecules in organoclays yields multiple potential intragallery
morphologies. Depending on the Charge density Of the Clay and the alkyl chain
length Of the onium ion, a variety Of well-documented surfactant packing
orientations are possible. In general, the longer the surfactant chain length and
the higher the Charge density Of the clay, the further apart the Clay layers will be
forced. This is expected since both these parameters contribute to increasing
the volume occupied by the intragallery molecules. For low volume content
organics, the onium ions may lie parallel to the Clay surface as a monolayer, and
as the amount of organic in the galleries is increased the adopted conformations
exist as a lateral bilayer, a pseudo — trilayer or an inclined paraffin structure. At
very high surfactant concentrations, a lipid bilayer may be necessary to account
for such a large organic content between the Clay Sheets. Figure 1.3 indicates

the different organoclay arrangements possible for smectite Clays.13

 

Monolayer organoclay Bilayer organoclay
arrangement arrangement

P

 

Tri-Iayer organoclay Paraffin organoclay
arrangement arrangement

Figure 1.3 Illustration of four possible organoclay structures based on
surfactant packing density. Given the same surfactant, increasing the charge
density of the clay will move from a monolayer arrangement towards the paraffin
structure.

The orientations Of onium ion chains in organoclay were initially deduced
based on infrared and X-ray diffraction measurements. More recent modeling
experiments have provided further insights into the packing arrangements of the
alkyl Chains in the confined environments of layered silicates.14 Molecular
dynamics simulations were used to study molecular properties such as density
profiles, normal forces, Chain configurations and trans - gauche conformer ratios.
For the monO-, bi- and psuedo — trilayers, with respective basal spacings Of 13.2,
18.0 and 22.7 A, a disordered liquid - like arrangement of chains was preferred
in the gallery. In this disordered arrangement the Chains do not remain flat, but
instead overlap and CO-mingle with onium ions in opposing layers within the
galleries. However, for the trilayer orientation, the methylene groups are found
primarily with a span Of two atomic layers and only rarely do they continue into
the layer Opposite to the positive head group. AS anticipated, the onium head
group is also noted to reside nearer the Silicate surface relative to the aliphatic
portion Of the surfactant. The highest preference conformer is trans over guache
for the maximum surfactant Chain length just before the system progresses to the
next highest layering pattern. This is expected since the alkyl Chains must be
optimally packed under such dense surfactant concentrations.

The replacement of inorganic exchange cations by organic onium ions on
the gallery surfaces of smectite Clays not only expands the interclay distance, but
also serves to match the Clay surface polarity with the polarity of the polymer.
Given a sufficient polarity match between the polymer and clay galleries, the

organoclay galleries will swell further upon intercalation by the polymer.

10

Exemplary pre-polymers that have shown such behavior include caprolactam”,

16-18 19, 20

epoxides , silicones and polyolsz‘. For long Chain onium — exchanged
organoclays, the galleries swollen by these precursors typically show a d-spacing
indicative of a paraffin monolayer arrangement.
1.1.7 Organoclay - Polymer Interactions

Despite having an ideally matched system, where unreacted polymeric
precursors intercalate readily into organoclay galleries, there is still a large
possibility that these will not form a true nanocomposite. Only when the Clay
layers are forced apart and no longer interact through the onium Chains is an
ideal nanocomposite formed”. The complete dispersal, or exfoliation, of the clay
nanolayers yields composites with the highest degree Of property enhancement.
Composite materials retaining a regular gallery height equal to less than two
times the surfactant Chain length, then the amount Of polymer interacting with the
clay surface is limited, and intercalated composites are formed. Such materials
lack the desired degree Of layer expansion and as a result will have distinct
regions with either a very high or low to non-existent concentration Of
reinforcement fillers. This non-uniform dispersal of nanolayers limits stress
transfer throughout the composite, giving comparatively less than optimal
reinforcement. Still poorer, conventionally scaled composites are possible if the
polymer and smectic Clay exhibit only partial miscibility. Here the Clay persists as
tactoids of face — to - face stacked agglomerates throughout the polymer matrix.

This incomplete dispersal of the reinforcing phase inhibits ideal surface contact

between the polymer and clay”, creating large voids of pure polymer in the

11

composite. Figure 1.4 represents the different possible composite morphologies

and representative XRD patterns of each.

12

 

 

 

 

 

 

  

 

 

 

 

    

Disordered Exfoliated

r'
——- .- g

.—
w—vvw
if ‘7' I r

 

 

Relative Intensity

 
 
 

Ordered Exfoliated

 

 

 

 

Intercalated

13.7 A )1 conventional

1111111111

2 4 6 8 1
26) (Degrees)

    

 

 

 

 

  

 

 

 

Figure 1.4 Schematic of the different composite morphologies possible when
attempting to synthesize polymer - Clay nanocomposites. Representative XRD

patterns are displayed for each type.

13

Above and beyond intercalation of the polymer precursors, the ideal
exfoliated case requires an increased reaction rate in the confined gallery spaces
relative to the bulk crosslinking. This creates larger and faster forming
crosslinked polymer domains between Clay layers, thus providing a driving force
for layer expansion. In the majority Of successful exfoliated systems acidic sites
present in the Clay galleries provide the bias for intragallery polymerization.” 24
These can catalyze the crosslinking process, as is the case in the ring Opening
polymerizations of epoxy and caprolactam.

1.1.8 Properties of Thermoset - Clay Nanocomposites

The presence Of exfoliated clay layers in a polymer matrix has a dramatic
effect on the physical properties Of the filled system. Typically the most sought
after final nanocomposite Characteristics involve improved mechanical behavior.
Increasing the tensile strength, the elongation at break and the modulus
(stiffness) of polymer materials allows for a lighter weight alternatives to
structural metals and conventionally filled plastics. The nanolayers, however, do
not only provide a reinforcing affect to the polymer matrix. Other studies, for
example, focusing strictly on the ablative performance of polymer — clay
nanocomposites have also been explored.25 Here the nanocomposites are
subjected to extreme temperatures in the absence Of oxygen, and their ability to
form a protective ceramic Char on the exterior are determined. Applications in
the aerospace industry as protective shields serve as the driving force for such

studies. The impermeable nature Of the silicate layers have been shown to

dramatically lower the diffusion of small molecule perrneants in certain

14

nanocomposite materials.”29

In this case the Clay layers act as a barrier and
mandate a more tortuous and longer effective diffusion pathway through the
polymer matrix. A schematic representing this hypothesis for the reduction in

permeability in shown in figure 1.5.

15

 

 

 

 

 

\ \\

\
_ J— /

 

/ __

\ -:-—\

 

 

 

 

 

permeant flow

Figure 1.5 Schematic representation Of the tortuous path invoked on
penneants by having dispersed Clay layers in a polymer matrix.

Another polymeric property that can be improved by the introduction Of

3°34 This “inflammable” or self —

disseminated Clay particles is thermal stability.
extinguishing characteristic noted in certain nanocomposites likely stems from
the tortuous diffusion path noted above. Both the influx Of oxygen necessary for
combustion to occur, and the outflux Of small molecule degradation upon burning
are slowed due to the presence Of the Clay layers. Other properties that can be
influenced by the exfoliation Of silicate layers include rheological behavior,
minimizing solvent uptake and Optical properties.29' 35' 3°
1.1 .9 Exfoliated Nylon-6 - Clay Nanocomposites

The first example of successfully exfoliated layered silicates in a polymer
matrix was accomplished by researches at Toyota‘s' 24' 37' 3°. By ion exchanging
Na“ cations present on naturally occurring montmorillionite Clay with organic
amine surfactants, the intercalation of nylon - 6 precursors was possible. They
used an amino acid — modified organo-montmorillionite to intercalate cyclical
caprolactam, the precursor to thermoplastic nylon — 6. The seven membered
ring caprolactam monomers can be liquefied at 70°C, at which point they were
added to the organoclay, and a 15 A basal increase upon intercalation was
noted. Upon further heating, at a temperature of 250°C, the caprolactam begins
to ring Open and form nylon - 6. The organic modifier, a protonated 012 alpha,
omega amino acid, was thought to aid in the intragallery ring opening

polymerization by acid — catalysis. The Clay layers were almost completely

delaminated after completion of the polymerization, as noted by low angle X-ray

l7

diffraction. The general synthetic procedure for forming nylon — 6

nanocomposites is illustrated in figure 1.6.

18

Synthetic Procedure for Nylon — 6 Nanocomposites

 

+NH3(CHQ 11(JOOH—rnontmorillonite

 

 

 

NH

 

v
caprolactammonttmrillonite 250°C Nylon-6
intercalate (+15A) nanocomposite

 

 

 

 

Figure 1.6 General synthetic route for synthesizing nylon — 6
nanocomposite materials. The caprolactam precursors were first
melted, then added to the organoclay to permit intercalation (causing a
15 A basal spacing increase), and finally polymerized at 250°C.

19

The nylon - 6 nanocomposites exhibited substantial increases in mechanical
properties due to the reinforcing Clay particles. These improvements ultimately
lead to their use in under - the — hood applications in certain Toyota automobiles
as a timing belt cover. The mechanical properties of the hybrid material
compared to pristine nylon and unexfoliated Clay in nylon is described in table

1.1.

Table 1.1 Mechanical and Thermal Properties Of Nylon-6

 

 

 

 

Material type Wt. °/o Tensile Tensile Impact HDT (°C)
Clay strength modulus (kJ/mz) at 18.5
(Mpa) (GPa) kg/cm2
Exfoliated
Nanocomposite 4.2 107 2.1 2.8 145
(Clay layers)
Conventional
Composite (Clay 5.0 61 1.0 2.2 89
tactoids)
Pristine nylon 0 69 1.1 2.3 65

 

 

 

 

 

 

 

 

1.2 Recent Developments in Polymer — Clay Nanocomposites

Many Of the advances in nanocomposite technology over the past few
years have involved the extension of the concept into different polymer systems.
Each polymeric medium, whether it be polyimide, polyether, epoxy, or

polysiloxane, demands its own reinforcement modification and synthetic

20

procedures to successfully exfoliate Clay in the matrix. In general, the first
requirement that must be met is to ensure that the interclay regions are
compatible with the polymer precursors. This provides a driving force for
intercalation and swelling of the clay galleries. While many systems have the
potential for monomer intercalation, true nanocomposites are not formed due to
the lack Of clay expansion upon crosslinking. As aforementioned, the ability to
Chemically alter the environment inside the Clay galleries so as to significantly
bias the reaction rate over that Of the bulk is paramount in achieving ideal
nanocomposites.

Although statistically impressive to report, the rubbery polymer matrices
that have been reinforced through the years are somewhat limited in their
practical applications. A recent push to create nanocomposites from more rigid
parent matrices has improved the possibilities Of finding structural replacements
for steel and other alloys39' 4°. Unlike the balance of epoxy nanocomposite work,
the results expressed in this thesis are based on glassy systems.

One very practical and widely studied polymer nanocomposite system is
polypropylene. Polypropylene is a thermoplastic that is widely used in low —
stress Situations, however, by stiffening the matrix through nanolayer
reinforcement, researchers hope to extend its use to more demanding, load -
bearing applications. The dispersal Of clay nanolayers into polyolefin systems
proved to be a Challenge due to the extreme hydrophobicity of the polymer.
Initial attempts to create polypropylene - Clay hybrids were based on the

introduction Of a modified polypropylene with polar groups in attempt to mediate

21

the polarity between the Clay surface and bulk polypropylenef“ 42 However, an
organic solvent has to be used in order to facilitate the formation Of a modified
polypropylene intercalate. Only a limited degree Of Clay nanolayer dispersion
was Observed by this method. An alternative and more environmentally friendly

approach was developed later by the Toyota research group4°4°.

A mixture Of
stearylammonium - exchanged montmorillonite, maleic anhydride modified
polypropylene Oligomer and homopolypropylene was melt-processed to Obtain a
successful polypropylene - clay hybrid wherein a larger fraction Of the Clay
nanolayers were found to be exfoliated. The hydrolyzed maleic anhydride
polypropylene intercalated into the organoclay, expanding the galleries and
facilitating the incorporation Of unmodified polypropylene. Interestingly, the
density of maleic anhydride groups has a significant effect on the final
morphology and properties Of the composite. A mixture of roughly 3:1 maleic
anhydride polypropylene to organoclay was found to be the most effective in
forming hybrid composites. The hybrids exhibit improved storage moduli
compared to pristine polypropylene in the temperature range from T9 tO 90°C.
The significance Of nanolayer reinforcement in polypropylene is not as great as in
nylon 6, probably due to the lower degree Of exfoliation and the introduction Of a
large amount Of low molecular weight Oligomer.

1.3 Research Objectives

1.3.1 Current Issues and New Directions

Although the nanocomposite concept has only been studied for roughly a

decade, many significant developments pertaining to the importance Of the clay -

22

polymer interactions, the use Of different nanoscaled reinforcers, and the
quantification of how much reinforcement is possible, or how much organic
modification is necessary have been Observed. The Pinnavaia group has been
involved in several Of these issues, including the recent drive to elucidate the
type and extent of organic modification needed specifically in epoxy
nanocomposites.

Other recent avenues in nanocomposite technology have revolved more
on permeability reduction and packaging applications versus the mechanical
improvement Of engineering plastics. Although the physical fortification
properties will always be the primary goal Of nanocomposite materials, secondary
improvements, such as inflammability and solvent resistance, will continue
extend the applications of Clay - modified polymers.

Thermoplastic melt intercalation of liquid polymer into Clay galleries is
another field that has become attractive for study. Here the polymer is heated in
excess Of its melting point and added to organo - functionalized clay. The
mixture is then put through high shear in attempt to physically disperse the clay
from aggregated tactoids to near individual layers. The high shear is provided by
means of a twin — screw extruder, which forces the polymer melt and Clay
through a high pressure gap.

1.3.2 Research Goals

Two thermoset systems, silicones and epoxies, will be examined for their

capability to form exfoliated nanocomposites. Instead Of simply synthesizing the

nanocomposites and comparing final properties with unfilled polymers, the

23

exfoliation mechanism is to be considered along with thorough investigations on
the unreacted pre-polymer - Clay interactions. Although in the past typical
nanocomposite work has involved fully organo — exchanged layered silicates, the
reliance on organic modification for successful nanocomposites will be examined,
and composite materials lacking the customary surfactant molecules also will be
explored.

Specifically, the pristine polydimethylsiloxane rubber is in dire need of
mechanical improvement, and this is the area of primary concern. Silicones also
are extremely permeable, and nanocomposites with improved barrier properties
will also be addressed. Although silicone — Clay hybrid materials have been
previously studied, the reaction mechanism was unmentioned, and thus this work
sets out to elucidate the nature Of the clay layer expansion and the driving forces
for exfoliation. Related somewhat to the high permeability Of silicones Is the
susceptible nature to extensive swelling, and ultimately sample failure caused by
large uptake of organic solvent into the matrix. Of primary concern in this work is
to reduce this tendency in order to yield robust rubber materials that in the future
may find applications as sealants and gaskets where current silicones would fail.
While the high permeability of Silicone rubber is often thought to be a detriment,
this work will make use Of this fact and investigate post — cure nanocomposite
Chemical modification previously unexplored.

As aforementioned, the main issue with epoxy nanocomposites is that
previous work has focused primarily on subambient Tg networks that have limited

practical implications. One goal in this work is to study glassy systems and

24

improve their already impressive mechanical properties. Instead Of solely using
purely organic modified Clays to achieve this goal, various mixed ion inorganic —
organic Clay fillers will be explored.

Characterization Of the pre-polymerized and post - cured nanocomposites
will be performed primarily by X-ray diffraction, accompanied with Transmission
Electron Microscopy in order to fully understand the nature of clay layers in the

polymer matrix.

1.3.3 References
1. Novak, B.M.,. Advanced Materials, 1993. 5(6): p. 422-433.
2. Schmidt, H., J. Non-Cryst. Solids, 1985. 73: p. 681-691.
3. Mark, J.E., Polym. Eng. Sci., 1996. 36(6): p. 2905-2920.
4. Okada, A. and A. Usuki, Materials Science & Engineering C-
Biomimetic Materials Sensors and Systems, 1995. 3(2): p. 109-115.
5. Giannelis, ER and PB. Messersmith, PCT Int. Appl. 1996,
(Cornell Research Foundation, InC., USA). WO. p. 32 pp.
6. Ogawa, M. and K. Kuroda, Bulletin Of the Chemical Society of
Japan, 1997. 70(11): p. 2593-2618.
7. Pinnavaia, T.J., Beall, G.W., eds., 2000: Jonh Wiley & Sons Ltd,

New York.

25

8. Fukushima, H. and LT. 0rzal, PrOC. Annu. Meet. Adhes. Soc.,
2000. 23rd: p. 139-191.

9. Kojima, Y., et al., Journal of Polymer Science Part a-Polymer
Chemistry, 1993. 31(4): p. 983-986.

' 10. Gilman, J.W., et al., Proc. Int. Wire Cable Symp., 1997. 46th: p.
761-774.

11. Giannelis, E.P. and J.L. Keddie, 1993, (Cornell Research
Foundation, Inc., USA). Us. p. 9 pp.

12. Pinnavaia, T.J., Science, 1983. 220(4595): p. 365-371.

13. Lagaly, G., Solid State Ionics, 1986. 22(1): p. 43-51.

14. Hackett, E., E. Manias, and E.P. Giannelis, Journal Of Chemical
Physics, 1998. 108(17): p. 7410-7415.

15. Usuki, A., et al., Abstracts of Papers of the American Chemical
Society, 1 990. 200: p. 218-POLY.

16. Wang, Z., T. Lan, and T.J. Pinnavaia, Abstracts Of Papers of the
American Chemical Society, 1996. 211: p. 455-INOR.

17. Wang, Z. and T.J. Pinnavaia, Chemistry of Materials, 1998. 10(7):
p. 1820-1826.

18. Massam, J., et al., Book of Abstracts, 21Sth ACS National Meeting,
Dallas, March 29-April 2, 1998: p. PMSE-206.

19. LeBaron, PC. and T.J. Pinnavaia, Chem. Mater., 2001: p. A08
ASAP.

26

20. Burnside, 8.0. and E.P. Giannelis, Chemistry Of Materials, 1995.
7(9): p. 1597-1600.

21. Wang, Z. and T.J. Pinnavaia, Chemistry Of Materials, 1998. 10(12):
p. 3769-+.

22. LeBaron, P.C., Z. Wang, and T.J. Pinnavaia, Applied Clay Science,
1999. 15(1-2): p. 11-29.

23. Shi, H.Z., T. Lan, and T.J. Pinnavaia, Chemistry of Materials, 1996.
8(8): p. 1584-&.

24. Okada, A., et al., Mater. Res. Soc. Symp. Proc., 1990. 171(P0lym.
Based MOI. Compos.): p. 45-50.

25. Vaia, RA, et al., Applied Clay Science, 1999. 15(1-2): p. 67-92.
26. Kojima, Y., et al., Journal Of Applied Polymer Science, 1993. 49(7):
p. 1259-1264.

27. Kojima, Y., et al., Journal Of Materials Science Letters, 1993.
12(12): p. 889-890.

28. Yano, K., A. Usuki, and A. Okada, Journal of Polymer Science Part
a-Polymer Chemistry, 1997. 35(11): p. 2289-2294.

29. LeBaron, PC. and T.J. Pinnavaia. Abstracts Of Papers, 222nd ACS
National Meeting, Chicago, IL, United States, August 26-30, 2001, 2001:
p. lEC-041.

30. Gilman, J.W., T. Kashiwagi, and JD. Lichtenhan, Sampe Journal,

1997. 33(4): p. 40-46.

27

31. Gilman, J.W. and A. Morgan, Recent Adv. Flame Retard. Polym.
Mater., 1999. 10: p. 56-67.

32. Gilman, J.W., et al., Chem. Technol. Polym. Addit., 1999: p. 249-
265.

33. Gilman, J.W., et al., Polym. Mater. Sci. Eng., 2000. 83: p. 59-60.
34. Giannelis, E.P., Annu. Tech. Conf. - SOC. Plast. Eng., 1996.
54th(VOl. 3): p. 2998-3003.

35. Burnside, 8.0. and E.P. Giannelis, Journal Of Polymer Science Part
B-Polymer Physics, 2000. 38(12): p. 1595-1604.

36. Wang, 2., PC. LeBaron, and T.J. Pinnavaia, Addit. ’99, Int. Conf.,
8th, 1999: p. Paper10/1-Paper10/9.

37. Usuki, A., et al., Journal Of Materials Research, 1993. 8(5): p. 1174-
1178.

38. Usuki, A., et al., Polym. Prepr. (Am. Chem. SOC., Div. Polym.
Chem), 1990. 31(2): p. 651-2.

39. Massam, J. and T.J. Pinnavaia, Material Research Society
Symposium Proceedings, 1998. 520: p. 223-232.

40. Triantafillidis, 0.8., at al., Abstracts Of Papers, 222nd ACS National
Meeting, Chicago, IL, United States, August 26-30, 2001, 2001: p. lEC-
109.

41. Kurokawa, Y., H. Yasuda, and A. Oya, Journal of Materials Science

Letters, 1996. 15(17): p. 1481-1483.

28

42. Kurokawa, Y., et al., Journal of Materials Science Letters, 1997.
16(20): p. 1670-1672.

43. Kawasumi, M., et al., Macromolecules, 1997. 30(20): p. 6333-6338.
44. Kato, M., A. Usuki, and A. Okada, Journal Of Applied Polymer
Science, 1997. 66(9): p. 1781-1785.

45. Hasegawa, N., et al., Journal Of Applied Polymer Science, 1998.

67(1): p. 87-92.

29

Chapter 2
CLAY NANOLAYER REINFORCEMENT OF A SILICONE ELASTOMER:
EXFOLIATION OF FLUOROHECTORITE IN POLYDIMETHYLSILOXANE

2.1 Introduction

As has been demonstrated quite convincingly in a recent comprehensive
review,1 the insertion Of organic molecules into layered or porous inorganic hosts
often results in nanocomposite materials with properties dramatically different
from the parent end members. Nanocomposites differ from conventional
composites in that the mixing Of phases occurs over a much smaller length scale
in comparison to the micrometer length scale Of conventional composites.“
Employing a filler phase that has such a dramatic size reduction results in
heightened interfacial interactions between the polymeric matrix and the fortifying
inorganic support, thus facilitating a more efficient transfer Of an applied load.
When applied to the reinforcement Of engineering polymers the nanocomposite
concept can lead to compositions with properties far superior to conventional
composites at a much reduced filler loading. The weight savings is particularly
important for the automobile and aeronautical industries where 25-40°/o weight
addition of reinforcement is necessary to achieve the desired properties.

Layered-silicate Clays are particularly good candidates for the nanoparticle
strengthening of polymers due, in part, to their high surface area, platy

morphology and exceptionally stable oxide network.5 In certain respects the

30

layered silicate Clays share some of the same Characteristics as the aerogel
silicas and precipitated silicas used for the reinforcement of silicones and other
polymer systems. However, in comparison to small particle silicas°, Clays are
less costly and free Of health risks, adding to their attractiveness as a
replacement filler for silicones. Furthermore, the rich intercalation Chemistry
associated with smectic Clays can be used to facilitate the dispersion of the
nanolayers in a polymer matrix, a feat unobtainable with conventional silica
nanoparticles that tend to undergo phase-segregation in a Silicone matrix.7 The
generally spherical particle morphology inherent in aerogel silica is also a
detriment to reinforcement when compared to the high aspect ratio nanoplates of
individual clay particles.

Toyota researchers were the first to demonstrate that small quantities (<10
wt %) Of organoclays exfoliated in a Nylon-6 polymer matrix greatly improves the
thermal, mechanical, barrier and flame retardant properties Of the polymer.°'11 In

recent years, several other polymer systems have been investigated including

16-18

epoxies‘2'15, polyimides , polyurethanes”, polysiloxanes7'2° as well as various
thermoplastics2"3°. In earlier studies of polysiloxane - organoclay
nanocomposites, - dimethylditallow and - hexadecyltrimethyl ammonium

exchanged montmorillionites were dispersed in polydimethylsiloxane (PDMS)
with a molecular weights of 18,000 and 68,000, respectively, and crosslinked to a
rubber by reaction with tetraethylorthosilicate (TEOS). The former system
afforded an exfoliated Clay nanocomposite, whereas in the latter the Clay was

only partially exfoliated, as indicated by low angle X-ray reflection in the final

31

composite. Both clays, however, provided elastomer reinforcement, as well as
improved thermal and solvent barrier properties7'2°. In order to influence the
greatest portion Of the polymer matrix at the same Clay loading, the more
exfoliated nanocomposite materials are generally the ultimate goal.

In an effort to elucidate the swelling of organoclay galleries by PDMS, we
have examined the intercalation properties of a synthetic fluorohectorite in PDMS
polymers with molecular weights in range 400 to 4200. These smaller PDMS
polymers make it possible to follow the gallery expansion process quantitatively
by wide angle X-ray diffraction methods. Also, the synthetic fluorohectorite used
in this work has a substantially larger nanolayer aspect ratio (~2000) in
comparison to naturally occurring montmorillonite (~ 200). The higher nanolayer
aspect ratio should provide even better reinforcement and barrier properties, the
two performance properties in need of greatest improvement for silicone rubbers.
Having a larger individual particle size, the fluorohectorite also aids in monitoring
the true nature of the hybrid materials during and after polymerization by X-ray
diffraction, in agreement with the Scherrer Equation. Accordingly, we also have
prepared a crosslinked silicone rubber - exfoliated fluorohectorite nanocomposite
and investigated its tensile strength, dioxygen permeability and solvent uptake
properties in comparison to the pristine elastomer. The results Clearly
demonstrate the substantial reinforcement that is possible at Clay nanolayers
loading of 5 to 10 wt%, though the permeability Of the nanocomposites remains

high despite the larger nanolayer aspect ratio.

32

2.2 Experimental

Materials. PDMS polymers with average molecular weight values Of 400, 1750
and 4200 and tin(l|) 2-ethylhexaonate were obtained from Gelest, Inc.
Hexadecyltrimethylammonium bromide and all other Chemicals were purchased

from Aldrich Chemical CO. and used without further purification.

Polydimethylsiloxane (PDMS)
CH3 CH3
HO— (Si—OM— Sli—OH
(IDHa (I3H3

Tetraethylorthosilicate (TEOS)
(IDEt
EtO — Si— OEt

OEt

33

Table 2.1 Silanol Terminated PDMS Monomers

 

 

 

 

 

 

 

 

 

PDMS MW N value (Si(CH3)2-O repeat) Viscosity (cSt)
400 6 20
1 500-2000 23 45-85
4200 57 90-1 20
1 8000 243 700-800
Organoclay Synthesis

Lithium fluorohectorite (Corning, Inc.) with an anhydrous unit cell formula
Of Li1_12[Mg4,33Li1.12](Sig,oo)F4 was ion-exchanged with a two-fold stoichiometric
excess of hexadecyltrimethylammonium bromide. A mixture Of the surfactant
cations and a 1.0 wt % aqueous suspension Of the inorganic Clay was stirred at
room temperature for 12 hours. The flocculated organoclay was removed by
vacuum filtration and the excess surfactant was washed from the clay by re-
suspending the solid in water, filtering again, and then subjecting the solid to a
final Soxhlet extraction in ethanol for 24 hours. The Clay was then dried, crushed
and sieved to a particle size <106 um prior to use in the intercalation and
nanocomposite studies. The water and organic content Of the air-dried
organoclay, as determined by the weight loss at 150 °C and 500 °C, was 6 wt °/o
and 24 wt %, respectively. A general depiction Of the ion exchange is shown in

figure 2.1.

34

 

CTAB ion exchange

    
  

    

inorganic
fluorohectorite clay

Soxhlet wash Of
exchanged clay

     

organofluorohectorite clay

Figure 2.1 Schematic illustrating the organic ion exchange of fluorohectorite
clay using cetyltrimethylammonium bromide.

35

The need for organoclay washing is shown in figure 2.1 as the intermediate
organoclay product contains intersalated (intercalated salt) ion pairs trapped in
the galleries. This is evidenced by powder diffraction as a secondary 00l

shoulder at a slightly higher basal spacing than the pure (washed) organoclay.

Nanocomposite Synthesis

A predetermined amount of organoclay was added to the desired PDMS
pre-polymer and mixed for 12 hours. Agitation 0f the Clay-PDMS mixture was
accomplished by first manual stirring with a spatula followed by vortexing the
suspension for several minutes. A stoichiometric amount Of
tetraethylorthosilicate (TEOS) as a crosslinking agent and tin(ll) 2-
ethylhexanoate as a catalyst (TEOS : Sn = 4 : 1 by volume) were added to the
PDMS - organoclay mixture. The resulting thixotropic intercalate was then
transferred to an stainless steel mold and out gassed under vacuum.
Crosslinking Of the nanocomposite was carried out under vacuum at ambient
temperature for 12 hours, followed by an additional 12 hours at 100 °C remaining

under vacuum,a s indicated in figure 2.2.

36

PDMS monomer + organoclay stir

TEOSISn catalyst

 

Mixture transferred to
mold

 

 

 

   
      
 

1. room temp curing under
vacuum 12 hrs

2. 100 C curing under
vacuum 12 hrs

 

Final
nanocomposite

 

 

 

Figure 2.2 General silicone nanocomposite synthesis scheme. After
polymerization silicone samples were carefully removed from the steel molds for

final property assessment.

37

2.3 Physical measurements

X-ray powder diffraction (XRD) patterns were recorded on a Rigaku
rotaflex 2003 diffractometer equipped with Cu K,1| X-ray radiation and a curved
crystal graphite monochromator. X-ray analysis of liquid PDMS polymer —
organoclay mixtures were accomplished by dropping the suspension onto filter
paper mounted on a glass slide with two-sided tape, and blotting away the
excess liquid. This technique was imperative in allowing for the low-angle XRD
examination in a vertically oriented sample holder Of the liquid Clay-PDMS
mixture. After the addition of TEOS and the Sn catalyst to the Clay — PDMS
mixture, the mixture was likewise dispersed onto a filter paper slide and the X-ray
diffraction pattern was recorded during the polymerization process. While the
polymerization is generally carried out under vacuum, this is only to facilitate the
outflux of the condensation product ethanol, and the use of heat in non-vacuum
systems likewise results in polymerization. Crosslinked nanocomposites were
examined by mounting a rectangular flat specimen into an aluminum X-ray
sample holder. After data collection, the sample was turned over to collect
diffraction data from the Opposite side Of the casting in order to verify
homogeneous distribution of the Clay particles. This test of homogeneous
particle distribution has proven to be important in studying nanocomposites, as
poorly dispersed mixtures typically leads to settling of the Clay particles during
polymerization. In this latter case only the bottom surface Of the casting will
indicate the true nature of the Clay component. We strongly feel that mistakes in

this area by groups account for reported “exfoliated” systems which are more

38

likely a result of this settling nature of the more dense Clay and Clay-polymer
intercalates.

Tensile measurements were made under ambient conditions according to
ASTM procedure 03039 using an SFM-20 United Testing System. The dog
bone - shaped samples used in the tensile measurements were 28 mm long in
the narrow region, 3 mm thick and 3 mm wide along the center of the casting.
The dog bone mold was a two piece stainless steel Clamping — style mold which
was separated by aluminum foil to aid in the removal Of the rubbery bones. Prior
to pre-polymer and crosslinker addition into the casting, the mold was treated
with a non — stick release agent (Monocoat E-179) provided through Chem-
Trend, Inc. and used without further modification.

Oxygen permeability data were acquired on a Mocon Ox-tran 2/60 oxygen
permeability instrument with a test gas containing 5 vol % oxygen in nitrogen.
Test gas for the Ox-tran 2/60 ordinarily contains 100% oxygen, or 21% oxygen,
however the high permeability Of silicone polymers mandates lower oxygen
concentrations in order to achieve readable values (below the upper limit of
oxygen detection for the permeability instrument). Film samples with a thickness
between 0.7 - 1.0 mm, which were cast from an aluminum disk mold, were held
in the testing cells with adhesive foil masks. Using the foil masks instead Of
encompassing the sample holder entirely with a polymer film reduces the
exposed film area to the test gas, thus limiting the quantity Of oxygen permeated
through the film and ultimately reaching the detector. The Ox-tran 2/60 uses a

Coulox fuel cell - type oxygen sensor where incoming oxygen molecules

39

undergo a redox reaction and ultimately give an electric current proportional to
the amount of oxygen reacted. The half — cell cathodic and anodic reactions,

respectively, are shown below:

1/2 02 + H20 + 2e- -> 20H-

Cd+2OH~ —> 2e- + Cd(OH)2

Therefore, each oxygen molecule entering the Coulox sensor produces four free
electrons, and so in accordance with Faraday’s Law, one mole of oxygen at
standard conditions would result in four Faradays of current. In more practical
terms, one Faraday equals 96,500 Ampere-seconds, thus converting the oxygen
amount into more reasonable terms a cubic centimeter of oxygen over a 24
hours period gives rise to 1.99E‘4 amps of current.

Pristine and nanocomposite PDMS samples were soaked in cyclohexane
until they reached a maximum uptake of solvent. The samples were then
removed from the solvent and photographed after different stages of solvent
evaporation to record the mechanical damage to the polymer. Any organic
solvent can be used to demonstrate this concept, however cyclohexane has a
vapor pressure that allows for the damage to occur on a relatively fast time scale,
but not so rapid as to inhibit photographing the process.

TEM images were obtained on a JEOL JEM - 1OOCX Il microscope using
a CeBe filament and an accelerating voltage of 120 kV. Thin sectioned samples

were prepared by embedding the PDMS nanocomposites in a glassy epoxy

40

matrix and sectioning on an ultramicrotome. The thin sections (~80 nm) were

supported on 300 mesh nickel grids.

2.4 Results

The organoclay used in the synthesis of nanocomposites was a
hexadecyltrimethylammonium ion exchanged fluorohectorite, abbreviated C16FH.
This organo Clay, which was prepared by ion exchange reaction of Li“ - FH,
exhibited a dom basal spacing of 27.4 A. Solvating the organo clay in PDMS
polymers with average molecular weights in the range 400 to 4200 caused the
basal spacing of the clay to increase. As shown by the diffraction patterns in

Figure 2.3 the basal spacing increased with increasing Chain length of the PDMS

polymer.

41

 

 

 

 

 

 

 

8
76A 8
co
8
43A 4200
1750
38A --
2 A
8 400
__.-A.~______
C16FH
....1....1....1....T....1....1”..

 

 

1 2 3 4 5 6 7 8
zeioegrees)

Figure 2.3 X-ray diffraction patterns of synthetic
hexadecyltimethylammonium fluorohectorite, abbreviated C16FH, intercalated by
PDMS polymers with molecular weights of 400, 1500-2000, and 4200. The
pattern for the initial organo Clay is included for comparison.

42

The PDMS with MW = 4200 gave the largest spacing (76 A), corresponding to a
gallery height of ~ 66 A. Nanocomposites were synthesized using all of the
available PDMS monomers, however since PDMS-4200 showed the greatest
tendency to swell the galleries of the organoclay, this polymer was used to form
silicone rubber - organoclay nanocomposites that were studied for mechanical
properties as these nanocomposites will have more dispersed layers.

Silicone rubber nanocomposites containing up to 10 wt % organoclay
were prepared by crosslinking mixtures of PDMS-4200 and C16FH using
tetraethylorthosilicate (TEOS) as the crosslinking agent and tin(ll) 2-
ethylhexaonate as the catalyst. XRD patterns of a representative reaction

mixture at different stages of crosslinking are shown in Figure 2.4.

43

 

 

060 reflection of the

cured

  
   
 

 

 

l J l
60 61 62

 

cured nanocomposite

10 min at 80 °C

4200-FH Intercalate

 

 

JLLLilllllllllLJllllllLllAll] 11

1 2 3 4 5 6 7 8

 

29 (Degrees)

Figure 2.4 X-ray diffraction patterns for mixtures of
hexadecyltrimethylammonium fluorohectorite and PDMS — 4200 at different
stages of cure. The curing agent was tertraethylorthosilicate and the catalyst
was tin(ll) octoate. The inset indicates the presence of an in-plane 060 Clay peak
in the exfoliated composite.

44

Significantly, the 002 reflection increased from 38 A in the initial intercalate
to ~ 50 A after only 10 minutes of partial curing at 80 °C. At this incomplete
stage in the crosslinking process the nanolayers have been pushed apart by the
by ~ 90 A. The crosslinked nanocomposite (12 h, 100 °C) showed no evidence
for a 00l Clay refection, indicating that the nanolayers were optimally exfoliated.
Also, as shown by the inset in Figure 2.4, the Clay in-plane 060 reflection near 60
degrees 20 was observed in the cured composite, indicating the retention of
nanolayer crystallinity, even though they are highly dispersed.

Tensile data were Obtained for dog bone - shaped castings of the silicone
elastomers containing 0.0, 5.0 and 10 wt % organoclay loadings. These
organoclay loadings correspond to silicate loading of 0.0, 3.5 and 7.1 wt %,
respectively, after accounting for the organic content in fluorohectorite. Figure

2.5 provides the stress-strain curves for each composition.

45

 

10%FH

Load (lbs)

 

 

 

Extension (%

Figure 2.5 Comparison of the stress - strain curves for pristine PDMS and

PDMS - C16FH nanocomposites containing 5.0 and 10 wt % C16FH organoclay.

46

While figure 2.5 represents stress - strain data for three different silicone
samples, averaged tensile strength, modulus, and strain at break data for

multiple dog bones is summarized Table 2.1.

Table 2.2 Mechanical Properties of Silicone Elastomer‘I - C15 FH

Organoclay Nanocomposites

Organoclay Loading Strength Modulus Elongation

 

(wt%)° (Kpa) (Kpa) PM
0.0 152 1.8x 10'3 11
5.0 361 2.3 x 10° 31
10.0 880 3.5 x 10'3 52

Table 2.2 addresses a summary of non-mechanical nanocomposite properties
including oxygen permeability results and cyclohexane uptake values for the
pristine and Clay - reinforced elastomers. The perrneabillties represent the
transmission rates of oxygen at a concentration of 5.0% in nitrogen as a carrier
gas, and the cyclohexane uptake values are the loadings of liquid solvent

contained in the elastomer at equilibrium at 25 °C.

47

Table 2.3 Permeability and Solvent Uptake Properties of Silicone

Elastomer“ - C15 FH Organoclay Nanocomposites

 

Organoclay Loading (wt%) Oxygen PermeabilityE Cyclohexane Uptake

 

 

 

 

 

 

 

(ti/9)SI
0.0 6.12 x 104 2.55
5.0 1.07
1.5 5.75 x 104
8.0 4.56 x 10‘

 

 

a/ The elastomer was formed by crosslinking PDMS (MW = 4200) a with a
stoichiometric amount of TEOS.

b/ The organic and water content of the Clay was 6% and 24%, respectively.
0/ The units of permeability are cm‘3 mil m'2 day".

Grams cyclohexane/gram polymer

48

2.5 Discussion

The replacement of Li+ by hexadecyltrimethylammonium ions in the
galleries of fluorohectorite resulted in a basal spacing of 27.4 A. The difference
between the observed basal spacing and the estimated Clay layer thickness (9.6
A) indicated of gallery height of 17.8 A. This value is consistent with a paraffin -
like orientation of the surfactant Chains in the gallery region. If the carbon Chains
adopted an all anti conformation, the observed gallery height would correspond
to an angle of inclination of ~50° between the chain axis and the Clay surface.
However, the Chains are most likely kinked31 and oriented more vertically on the
gallery surfaces. Ordinarily, long chain ammonium ion surfactants adopt lateral
bilayer to pseudo trimolecular orientation in the gallery space of montmorillonite,
resulting in gallery heights below 10 A, as depicted in figure 2.6 along with a

monolayer arrangment.“33

49

 

Monolayer organoclay
arrangement

 

Bilayer organoclay
arrangement

H

Pseudotrilayer organoclay
arrangement

Figure 2.6 Schematic depicting potential intragallery surfactant arrangements

typically found in organomontmorillonites.

50

However, synthetic fluorohectorite has a higher layer charge (1.12 e' per 020 per
unit cell) in comparison to montmorillonite which usually has a layer charge
density of ~ 0.75 e' per 020 Per unit cell. Thus, the higher population of
surfactant Chains in fluorohectorite results in a paraffin - like packing of chains

and a comparatively large gallery expansion, shown in figure 2.7.

51

 

Paraffin organoclay
arrangement

Figure 2.7 Inclined paraffin surfactant orientation found in higher Charge
density Clays. including fluorohectorite. Cetyltrimethylammonium fluorohectorite
has a paraffin arrangement with an approximate angle of inclination with the clay

surface of 50°.

52

Somewhat surprisingly, the extent of Clay gallery expansion upon PDMS
intercalation is dependent on the molecular weight of the guest polymer (C.f.,
Figure 2.3). This is not the usual outcome Observed when other linear molecules
are intercalated into an organoclay. Ordinarily, the gallery height is determined
by the orientation of organo cation in the gallery and the critical diameter of the
intercalated polymer. The length of the guest species usually has no or little
effect on the gallery height. That is, the organoclay normally permits a fixed
volume of the pre-polymer to intercalate as dictated primarily by the polarity
match between the linear polymer, gallery surface, and gallery cation. The
gallery heights observed in the PDMS — fluorohectorite systems, however, are
determined by the molecular weight or, equivalently, the ratio of -OSi(CH3)2-
repeat units to terminal -Si(CH3)20H groups. As this ratio increases, the gallery
height increases.

The hydrophobicity of the organo cation surfactant undoubtedly plays an
important role in allowing the initial penetration of PDMS into the galleries.
However, the Silanol end groups of the polymer interact at least In part with the
gallery surfaces. These interactions most likely play a role in determining the
space - filling configuration of the PDMS. The importance of specific interactions
between the Silanol end groups and the gallery surfaces is verified by the
observation that PDMS polymers of the same molecular weight but terminated
exclusively by methyl groups do not intercalate into C16FH. As illustrated
schematically in Figure 2.8, restricting the Silanol end groups to specific contacts

with the gallery surface will influence the packing of long chain PDMS molecule

53

(e.g., PDMS-4200) in the gallery. Shorter PDMS chains that are comparable in
length to the organocation (e.g., PDMS-400, which contains ~ 6 -SI(CH3)20-
units) cannot bridge opposite gallery surfaces and at the same time account for a
10-A or more increase in gallery height in comparison to C16FH (e.g., compare

the diffraction patterns in Figure 2.3).

54

 

 

Figure 2.8 Schematic representation of the swelling of an alkylammonium ion
exchanged fluorohectorite by a long chain PDMS polymer (MW = 4200)
containing terminal OH groups. Shorter PDMS chains with molecular weight
fractions of 400 and 1750 may have only one end of the chain associated with

the gallery surfaces.

55

Thus, in the case of short PDMS Chains only one terminal Silanol group
may be linked to the gallery surface. The Charged head groups of the surfactant,
the siloxane oxygens of the Clay layer, and CO-adsorbed water molecules may all
contribute to the interactions of the terminal Silanol groups with the gallery
surfaces. Burnside and Giannelis have noted that a small amount of gallery
water was needed to intercalate a PDMS-18,000 polymer into an organo
montmorillonite. We have verified the need for gallery water to achieve PDMS
intercalation. The air-dried organoclay used in the present work contained ~ 6 wt
% water, as judged by the weight loss upon drying at 150 °C, but if the water is
removed by drying at elevated temperatures, then no PDMS intercalation was
observed. In view of these results, it is now clear that water is needed to
facilitate interaction Of the terminal silanols and the gallery surfaces, presumably
through hydrogen bond formation. Once the H-bonding sites are saturated,
PDMS intercalation is complete and there is no further swelling of the Clay
galleries.

Although the swelling of C16FH is best monitored quantitatively by XRD,
the dependence PDMS intercalation on molecular weight can be qualitatively
judged by the physical behavior of the Clay in PDMS suspensions. Figure 2.9 is
a photograph of four different C16FH - PDMS suspensions, each containing 5 wt
% organoclay. The tubes are filled in left to right order with methyl terminated
PDMS - 400 and silanol - terminated PDMS - 400, - 1750, and - 4200. The
methyl - terminated PDMS does not intercalate into the galleries, and the

unexpanded clay settles to the bottom of the tube. For the Silanol - terminated

56

PDMS derivatives, however, the nanolayers are separated by PDMS molecules,
allowing for the formation of a stable, scaffold - like suspension Of swollen Clay

tactoids with an equilibrium height that increases with the PDMS polymer

molecular weight.

57

 

Figure 2.9 Photograph of PDMS polymer - C16FH suspensions containing 5.0
wt % organoclay. The tubes are filled in left to right order with methyl terminated
PDMS - 400 and Silanol terminated PDMS - 400, - 1750, and - 4200,
respectively.

58

The Sn(ll)-catalyzed crosslinking of PDMS - 4200 by TEOS in the
presence C16FH organoclay leads to further expansion of the galleries and the
formation of silicone rubber - Clay nanocomposite (of, Figure 2.2). While
somewhat debated, the mechanism of condensative silicone polymerization in
the presence of a tin octoate catalyst is generally thought to involve an initial
hydrolysis of the tin carboxylate. The most widely accepted full catalytic

mechanism is depicted in figure 2.10.

59

1-3tQ\Si /OEt
E1! OEt

Sn OCOC H
( 7 15)2—->11ROS -OH/
+ H20 +

\\“‘W

OS\
{SCH H3 H:CH3

Figure 2.10 Tin octoate - catalyzed PDMS crosslinking. Although there is
some debate as to the exact mechanism, the active species is widely thought to
be the hydrolysis product of the tin bis(2-ethylhexanoate) catalyst.

60

The absence of an observable 001 diffraction peak in the cured elastomer
implies that the Clay nanolayers have been optimally separated (exfoliated) in the
elastomer matrix. This means that the intragallery crosslinking rate was
competitive with the extra gallery rate. For this to occur the catalyst and TEOS
should be favorably partitioned between the PDMS solution and the organoclay
galleries. Co-adsorbed water may also facilitate intragallery polymerization
through TEOS hydrolysis, especially in the confined interclay regions where
water associated with the quaternary amine head groups is more acidic.
Verification that the individual Clay nanolayers retained their structural integrity in
the exfoliation process, and that the layers were still perceptible to the X-ray
radiation was provided by the presence of a 060 in-plane reflection in the high 28
region near 60° (see inset in Figure 2.3). The importance of probing for the 060
reflection can not be stressed enough, as an amorphous XRD pattern can result
from a variety of Circumstances. Low Clay concentrations, or isolated Clay
tactoids in the nanocomposite both could give rise to minimal X-ray scattering.
Also, a settling of the more dense Clay and Clay-polymer intercalate phases prior
to polymer matrix gelation in certain systems, especially those which have limited
polymer intercalation, could lead to an anisotropic Clay dispersion and thus an
amorphous top side (polymer only) even in intercalated nanocomposite. For this
reason XRD patterns were taken of both top and bottom regions of the
polymerized silicone nanocomposites. The high degree of pre-polymer
intercalation and Significant intragallery reaction bias support the notion observed

in TEM imaging that the nanocomposites were indeed extensively exfoliated.

61

The uniform dispersion of Clay nanolayers in the nanocomposite affords a
marked improvement in the tensile strength, modulus and strain-at-break (see
Figure 2.3 and Table 2.1). Similar improvements in tensile properties have been
reported for silicone rubber - Clay nanocomposites derived from PDMS polymers
with molecular weights of 18,000 and 68,0007”. Through the absorption and
dispersal of an applied load, the nanolayers layers delocalize stress, as well as
toughen the matrix as indicated by the large increase in the strain-at-break. In
contrast, conventional composites ordinarily compromise toughness for tensile
strength as in general stiffer and stronger, but more brittle (less extendable)
composites are formed from macro-scaled reinforcement. It has been previously
noted‘4'2° that the Clay nanolayer reinforcement of elastomers dramatically
reduces the ability of the matrix to absorb solvents. An analogous reduction in
cyclohexane uptake was observed for the Clay nanolayer - reinforced silicone
elastomers prepared in the present work (of, Table 2.2). However, the
reinforcement effect goes far beyond simply limiting the uptake of solvent. The
Clay nanolayers can also dramatically reduce the structural damage that normally
occurs when the solvent is allowed to evaporate from a solvent - saturated
elastomer. As shown in Figure 2.11, the rapid evaporation of cyclohexane from
the pristine silicone rubber causes internal strain that fractures the matrix,
rendering the recovered elastomer useless. In contrast, the nanolayer -
reinforced composite undergoes less solvent swelling and loses solvent more
slowly, thus reducing internal strain and avoiding matrix fracture. The avoidance

of structural damage caused upon exposure to solvents may represent one of the

62

most important benefits of nanolayer reinforcement in polymer technology.
Polymers, and elastomers in particular, find uses in automotive and similar
applications as sealants and gaskets where exposure to petroleum - based
solvents is commonplace. Having the capability to influence large amounts of
the polymer matrix at relatively low loadings, exfoliated Clay - silicone

nanocomposites could find applications where typical composites fail.

63

 

Figure 2.11 Photographs of a pristine silicone elastomer formed from
PDMS - 4200 (upper specimen) and the corresponding nanocomposite
containing 5.0 wt % C16FH organoclay (lower specimen) after equilibration in a
cyclohexane bath and subsequent removal from the bath to allow for solvent
evaporation: (a) immediately after removal from the bath, (b) after 15 seconds,
and (c) after 20 minutes. No structural damage was noted for the
nanocomposite after complete evaporation of the solvent.

64

Although the tensile and solvent resistance properties of a Silicone rubber
can be greatly enhanced through Clay nanolayer reinforcement, the gas
permeability of the nanocomposites was only slightly affected. Even at an
organoclay loading of 8.0 wt % organoclay the oxygen permeability was only
reduced only by ~ 25 % (see Table 2.2). In contrast, previous work has shown
that dramatic reductions in permeability can be achieved for polyimide films and
other thin film forms of polymers.°‘"38 For instance, the exfoliation of 2% wt
synthetic mica in a polyimide film reduced the water vapor permeability 10 fold”,
and a 4.8 wt % dispersion Of Clay in poly(e-caprolactone) reduced the water
vapor permeability 5 fold.35

In general, a reduction in the permeability of a nanolayer reinforced
polymer film can be anticipated only if the nanolayers are highly aligned parallel
to the film surfaces. This greatly increases the tortuosity of the diffusion path,
causing the permeability to decrease with increasing aspect ratio of the
nanolayers. A schematic representation of the importance of parallel Clay
alignment perpendicular to the direction of permeant flow is depicted in figure

2.12.

65

 

 

_ _£——-_ / C/ I

—.—-_—\ ..——_/ \
: \/\

 

 

 

 

 

 

 

 

 

 

 

 

\ /\
_ _ —-' _ \ \
// ___....—-
parallel — to - surface Clay orientation random Clay orientation

Figure 2.12 Depiction of permeant flow through nanocomposite films. A more
tortuous path is found perpendicular to the film when Clay particles are oriented
parallel to the film surface. Low surface tension fluids and thick films limit the
percentage of parallel - to - surface oriented nanolayers.

66

Although synthetic fluorohectorite has one of highest nanolayer aspect
ratios among smectite clays, it has not been effective in providing substantially
reduced gas permeability in the case of silicone elastomers. Silicone rubbers
are so permeable that it was necessary to use relatively thick 0.7 - 1.0 mm
samples in obtaining permeability coefficients that were within the upper
detection limit of the permeability instrument. Under the conditions needed to
form these thick samples the surface tension forces are too weak to facilitate
nanolayer ordering parallel to the sample surface. This affect is amplified with
silicones as the weak intermolecular forces create an already low surface tension
of the silicone fluid that limits the favored parallel-to-surface ordering of the clay
layers. The relatively random orientation of the nanolayers was verified by
transmission electron microscopy. As shown by the TEM image in Figure 2.13,
the nanolayers generated by the exfoliation of a Clay tactoid tend to orient parallel

to one another.

67

 

Figure 2.13 TEM image of silicone nanocomposite showing parallel orientation
of nanolayers stemming from the same tactoids. Prior to intercalation and
exfoliation in silicone, the organofluorohectorite galleries were separated by 2.7
nm, and after completion of crosslinking layers are 70-150 nm apart.

However, as shown by the image in Figure 2.14, wherein nanolayers
emanating from different tactoids are nearly orthogonal to each other, there is
little or no tendency for the nanolayers to orient uniformly through out the matrix.
This is highly unusual in layered silicate nanocomposites in that expansive
regions of Clay tactoids ordinarily retain high degrees of face - to - face ordering,
even at distances of several hundred nanometers (as seen in the epoxy
nanocomposites addressed later). Thus, better methods for achieving nanolayer
orientation must be devised if substantially reduced permeability is to be

achieved for silicone - Clay nanocomposites.

69

 

Figure 2.14 TEM image of silicone nanocomposite showing neighboring
tactoids with near orthogonal orientations. Such an image is rare in clay
nanocomposites where adjacent Clay tactoids are usually oriented parallel to one

another.

70

The significance of TEM imaging in nanocomposites should not be understated.
Liberal use Of the phrase “exfoliated nanocomposites” results when powder
diffraction is solely employed as the elucidation technique in determining the final
morphology of Clay nanocomposites. In this particular Silicone system, an
amorphous XRD pattern does not begin to represent the extreme degree of
dissemination of the Clay layers. Other reported “exfoliated” nanocomposite
systems have average Clay - to — Clay separations of tens of nanometers, such
values are not much different than the original organoclay gallery heights. In
these cases it is merely the loss of a regularly repeatable Clay — to — Clay
distance giving rise to the absence of X-ray reflections. Slightly different
polymerization rates in adjacent galleries cause this scrambling of intragallery
distances witnessed in the majority Of thermoset nanocomposites. A mere
scrambling and slight expansion of Clay galleries is not the case in the
synthesized fluorohectorite - silicone nanocomposites. Clay layers initially
separated by ~2.7 nm in the organoclay ultimately become forced apart through
crosslinking to 70-150 nm in the final hybrid. The unprecedented degree of
exfoliation in fluorohectorite in the silicone matrix in this system is manifested in
the dramatic increase of tensile properties. Even at relatively low Clay loadings
(3-7% wt. silicate), tremendous affects are noted in the Young’s modulus, strain
at break, tensile strength and overall toughness of the hybrids versus the pristine
silicone. Such improvements are to be expected when dispersion of a

nanoscaled reinforcement phase occurs to such a large extent. Efficient transfer

71

of the applied stress at the highly exposed polymer - clay interfacial regions
readily permits the reinforcement phase to play its role as a load barer, and
allows the matrix to remain intact under forces that would cause failure of the
pristine silicone rubber.

Qualitatively, the successful exfoliation of clay layers could be noted early
stages of crosslinking. While monitoring the partially cured silicone materials by
powder diffraction, the loss of higher order reflections immediately occurred as
the first samples were analyzed. This is somewhat surprising when compared to
other thermoset systems where the registry of intercalated clay layers persist
tens of minutes into the polymerization. The rapid dispersion of fluorohectorite
particles, in particular, is especially noteworthy as the synthetic Clay particles are
much larger laterally than their naturally occurring counterparts, and thus
expected to be more cumbersome in migrating through the silicone matrix.
Reasons for the high degree of exfoliation in silicones likely revolve around the
presence of water molecules at the Clay surface previously mentioned. The need
for small amounts of surface adsorbed water was previously noted for the initial
intercalation of PDMS monomers, and its role as a catalyst for the condensation
of TEOS was also alluded to, however yet another important function of the
moisture found at the Clay surface can be conceived. As aforementioned, the
selective partitioning of tin catalyst into the galleries could account for an
increase in the intragallery polymerization rate relative to the bulk rate, however,
the high solubility of tin octoate in polydimethylsiloxane may restrict the selective

uptake of catalyst into the gallery regions. Instead, the catalyst that does

72

partition into the interclay space, even if it is at the same concentration as in the
bulk medium, is hydrolyzed to the active tin hydroxyl species more efficiently than
in the water — deprived bulk silicone. An illustration depicting this mechanism is

shown in figure 2.15.

73

Sn(OR)2

Sn(OR)22.

 

 

Sn(OR)2 :

Snmg  

Figure 2.15 Schematic of tin catalyst existing in different forms during
crosslinking of the silicone rubber. The moisture sensitive tin carboxylate will
hydrolyze more readily when in Close contact with the fluorohectorite clay due to
the small amount (2-6% by wt.) of water present at the silicate surface.

74

2.6 Conclusions

Silicone rubber has been reinforced by exfoliating a
cetyltrimethylammonium - exchanged fluorohectorite into the polymer matrix.
Unique molecular weight dependant intercalation of PDMS monomers into the
organoclay was observed by X-ray diffraction (38 A, 43 A and 76 A for PDMS-
400, 1750 and 4200, respectively) in the pre-polymerization analysis. The novel
end - to — end separation intercalation dependence of Silanol terminated PDMS
in organofluorohectorite arises from a bridging interaction between both ends of
the polydimethylsiloxane Chain and the surface absorbed moisture at adjacent
clay surfaces. Addition of the tin catalyst and TEOS crosslinking agent to the
intercalated clay mixture resulted in rapid dissemination of the layers, and
ultimately a 70-150 nm spacing between layers was Observed by TEM imaging.
The exfoliated layers improved tensile properties dramatically, as expected
increasing the modulus and also the tensile strength at break dramatically with
increasing Clay content. Somewhat surprisingly, the nanocomposites also
exhibited improved elongation at break with increased clay, a phenomenon not
ordinarily attained with the introduction of a rigid filler. The exfoliated Clay also
reinforced the silicone polymer when exposed to an organic solvent as the matrix
did not uptake as much solvent as pristine samples, and survived removal from
the organic bath unlike the pristine samples. Although the presence of exfoliated
nanolayers drastically improved the mechanical properties of silicones, the

oxygen permeability was not significantly influenced.

75

2.7 References

(1) Schollhom, R. Chem. Mater. 1996, 8, 1747.

(2) Messersmith, P.B.; Stupp, S. I. J..Mater. Res. 1992, 7, 2599.

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

(4) Giannelis, E.P. Adv. Mater. 1996, 8, 29.

(5) Pinnavaia, T.J. Science 1983, 220, 365.

(6) Wang, Y.M. Organosil. Mater. App]. Chin. 1992, 5, 11.

(7) Wang, S.; Li, 0.; Qi, 2. Key Eng. Mater. 1998, 137, 87.

(8) Usuki, A.; Kawasumi, M.; Kojima, Y.; Okada, A.; Kurauchi, T.;Kamigito, O J.
Mater. Res. 1993, 8, 1174.

(9) Usuki, A.; Kojima, Y.; Kawasumi, M.; Okada, A.; Fukushima, Y.; Kurauchi,
T.;Kamigito, O J. Mater. Res. 1993, 8, 1179.

(10) Kojima, Y.; Usuki, A.; Kawasumi, M.; Okada, A.; Fukushima, Y.; Kurauchi,
T.;Kamigito, O J. Mater. Res. 1993, 8, 1185.

(11) Kojima, Y.; Usuki, A.; Kawasumi, M.; Okada, A.; Kurauchi, T.;Kamigito, O J.
Appl. Polym. Sci. 1993, 49, 1259.

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

(13) Messersmith, P.B.; Giannelis, E.P. Chem. Mater. 1994, 6, 1719.

(14) Massam, J.; Pinnavaia, T.J. Mater. Res. Soc. Symp. Proc. 1998, 520, 223.

(15) Wang, Z; Pinnavaia, T.J. Chem. Mater. 1998, 10, 1820.

(16) Yano, K.; Usuki, A.; Okada, A.; Kurauchi, T.;Kamigito, O J. Polym. Sci, Part

A: Polym. Chem. 1993, 31, 2493.

76

(17) Yano, K.; Usuki, A.; Okada, A. J. Polym. Sci, Part A: Polym. Chem. 1997,
35, 2289.

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

(19) Wang, Z.; Pinnavaia, T.J. Chem. Mater. 1998, 10, 3769.

(20) Burnside, 8.0.; Giannelis, E.P. Chem. Mater. 1995, 7, 1597.

(21) Kawasumi, M.; Hasegawa, N.; Kato, M.; Usuki, A.; Okada, A.
Macromolecules 1 997, 30, 6333.

(22) Usuki, A.; Kato, M.; Okada, A.; Kurauchi, T. J. Appl. Polym. Sci. 1997, 63,
137.

(23) Kurokawa, Y.; Yasuda, H.; Oya, A. J. Mater. Sci. Lett. 1996, 15, 1481.

(24) Kato, M.; Usuki, A.; Okada, A. J. Appl. Polym. Sci. 1997, 63, 1781.

(25) Hasegawa, N.; Kawasumi, M.; Kato, M.; Usuki, A.; Okada, A. J. Appl.
Polym. Sci. 1998, 67, 87.

(26) Porter, T,L.; Hagerrnan, M.E.; Reynolds, B.P.; Eastman, M.P.; Parnell, R.A.
J. Polym. Sci, Part B: Polym. Phys. 1998, 36, 673.

(27) Akelah, A.; Most, A. J. Mater. Sci. 1996, 31, 3589.

(28) Vaia, R.A.; Jandt, K.D.; Kramer, E.J.; Giannelis, E.P. Macromolecules
1995, 28, 8080.

(29) Vaia, R.A.; Jandt, K.0.; Kramer, E.J.; Giannelis, E.P. Chem. Mater. 1996, 8,
2628.

(30) Vaia, R.A.; Giannelis, E.P. MacrOmo/ecules1997, 30, 8000.

(31) Legaly, G. Solid State Ionics 1986, 22, 43.

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

77

(33) Teppen, B. J.; Rasmussen, K.; Bertsch, P.M.; Miller, D.M.; Schafer, L. J.
Phys. Chem. 81997, 101, 1579.

(34) Messersmith, P.B.; Gianellis, E.P. Chem. Mater. 1993, 5, 1064.

(35) Messersmith, P.B.; Gianellis, J. Polym. Sci. Part A: Polym. Chem. 1995,
33, 1047.

(36) Yano, K.; Usuki, A.; Okada, A.; Kurauchi, T.; Kamigaito, O. J. Polym. Sci.
Part A: Polym. Chem. 1993, 31, 2493.

(37) Yano, K.; Usuki, A.; Okada, A. J. Polym. Sci. Part A: Polym. Chem. 1997,
35, 2289.

(38) Kojima, Y.; Usuki, A.; Kawasumi, M.; Okada, A.; Kurauchi, T.; Kamigaito,

O. J. Appl. Polym. Sci. 1993, 49, 125.

78

Chapter 3
POST-CURE ION EXCHANGE OF EXFOLIATED FLUOROHECTORITE CLAY
IN POLYDIMETHYLSILOXANE NANOCOMPOSITES

3.1 Introduction

Formation of an ideal nanocomposite requires an intimate mixing of the
inorganic reinforcement phase within a polymeric matrix such that the resultant
composite undergoes a compositional Change on a nanometer length scale”.
The use of organically modified smectite Clays as the support for various
thermoset polymers has proven an effective means for achieving this nanoscaled
reinforcement. In particular, the uniform dispersal of various inorganic fillers in
elastomeric silicones is one such system that has received notable attention
recently“. Specifically, we have Shown that the use of quaternary alkylamine-
modified fluorohectorite is an effective layered silicate for achieving an exfoliated
reinforcement phase in polydimethylsiloxane (POMS)°. In this case insertion of
the silanol-terrninated PDMS monomers into organoclay galleries invokes a
molecular weight - dependant swelling of the interclay regions, and, after
tetraethylorthosilicate (T E08) and Sn catalyst addition, the Clay layers are forced
apart further as gelation of the PDMS matrix occurs. In conjunction with the
unique intercalation behavior PDMS monomers (001 spacings of 38 A, 43 A and
76 A for monomers of average molecular weight 400, 1750 and 4200,

respectively), the tensile, thermal, solvent uptake and permeability properties

79

were also explored. While the presence of highly exfoliated nanolayers (Clay to
clay separations averaging 70-150 nm noted by TEM imaging) in silicones
yielded properties that were generally far superior relative to unfilled PDMS, the
oxygen permeability was only slightly reduced with the introduction of exfoliated
nanolayers.

The extreme permeability of silicones, and specifically the fluorohectorite -
PDMS nanocomposites, coupled with the individually exposed Clay layers in the
synthesized hybrid materials offer a unique opportunity to explore subsequent
Chemistry on the cured system. Upon immersion in an organic solvent the
crosslinked silicone polymer undergoes extensive swelling (comprehensive filled
elastomer swelling has been examined by Giannelis7), thus providing a potential
medium for accessing exchange sites of the exfoliated nanolayers. Once the
matrix is highly distended, introduction of alternative cations become feasible and
ultimately the formation of Clay nanocomposites lacking the customary organic
modifiers are possible. A secondary exchange of the quaternary alkylammonium
ions present on the Clay surfaces could prove to be Significant in improving the
mechanical properties of certain nanocomposites. It has been shown that the
presence of these low molecular weight organic modifiers can plasticize the
matrix through solvation of the polymer chains, and thus reduce the efficacy of
the organoclay reinforcement phase in the nanocompositeg. Furthermore,
location of the modifiers at the silicate surface interferes with the favorable and
mechanically important interfacial contact between the polymer Chains and the

10.11

reinforcement phase Containing 24% by weight organic content, the

80

modified clays in this particular system are likely compromising a portion of their
fortification due to this plasticizing nature and a disruption of the desired
interfacial interactions. Other clay - polymer systems require an even higher
loading of organic modifier in order to successfully disperse nanolayers in the
polymer matrix, and therefore could experience even more pronounced
mechanical improvements by eliminating these alkyl surfactants. Another
potential motivation for the post-cure ion exchange in nanocomposites is to
create selectively reactive membranes for use in liquid and gas separations.
Through discriminatory exchange for Chemically active cations, the adsorption,
coordination or selective reactivity of certain perrneants and reactants could be
biased by the reactive (catalytic or otherwise) membranes, and be employed in
separations or flow-through reactor systems. Two representative examples of
such chemistry involve the selective quadrapolar interaction of lithium and
calcium cations with N2 in 02/N2 mixtures (producing Oz-rich gas mixtures on the
post-diffusion side of the membrane)”, and the use of catalytically active cation -
containing zeolites in silicone membranes for the oxidative Cleavage of certain
alkanes (partitioning reactant and product concentrations on separate sides of
the film.)13 Schematics summarizing the above described processes are shown

below in figure 3.1.

81

.mo~mo___moc_E:_m “53:20:82 ago: 358005

53933 mam 0280.3 new ocmfiEoE 02.062 5505. 0.523059 moszocow _..m 0.59".

 

ocanEoE mEnE
9:53:00 I 0:63 0:253

 

 

$03605
8236
2028236

 

 

 

 

 

 

OEOON Counc— I «EOE.—
..EB ton—59:0 EE wsaa

   

 

9ng
NO I «Z

 

mam co: I NO

 

 

 

 

 

82

This study Involves introducing a copper salt delivered in tetrahydrofuran
(THF) to the crosslinked silicone systems. Solvent choice is particularly
important in achieving a post-cure ion exchange as ordinarily mutually exclusive
solvation properties are needed to successfully complete the replacement Of Clay
gallery cations within the silicone matrix. First off, the silicone polymer is highly
non-polar and swells when exposed to typical organic solvents such as toluene,
cyclohexane, Chlorinated hydrocarbons and straight Chain alkanes. Large
amounts of solvent Invade the crosslinked rubber and distend the samples to
several times their original volume as alluded to earlier. The extent of sample
expansion has been noted to vary directly with the amount and effectiveness of
filler content. This swelling phenomenon in composite materials has been
described by Kraus, who plotted the swelling ratio shown below versus the

amount of filler.

V volume of unswollen polymer matrix
r 7 volume of swollen polymer matrix

After accounting for the volume content of the unswellable filler, correlations
between the curves of different filler types revealed that the better reinforcement
phases also had the greatest effect of reducing solvent uptake. High surface
area fillers having strong interfacial interactions with the polymeric phase matrix
create regions in the matrix known as “bound polymer". Here physisorbed or

Chemisorbed polymer in Close proximity with the reinforcement phase has a more

83

limited free volume and extendability, thus restricting the swelling capability.
Non-reinforcing fillers lack the presence of significant amounts of bound polymer
at the interface, and interestingly can actually undergo a greater solvent uptake
than in the unfilled system. In this case, large vacuoles arise at the interface of
the non-reinforcing filler and these rapidly fill with solvent and expand further.
Conventionally filled elastomers, such as kaolin - filled styrene butadiene rubber,
commonly swell more than the unfilled rubber due to this detaching polymer at
the filler interface when exposed to solvent. A schematic representing the affect
of bound polymer and the detrimental filling affect of a poor reinforcement phase

is shown in figure 3.2.

84

00.0.50. 05 .0 000.05 05 H0 .0E200

0050.060 05 0. 0:0 005003.880: 5 0020000 05:02.0 000300. 05 053020 0305028 «.0 0.59".

 

 

 

 

08:00.05 005.05 .000
53> 05000500 _0co_E0>c00

 

 

 

 

 

 

 

E0200 95:02.0

 

 

 

 

05:00.05 _0_00t05 0.00.00.
:52, 0200050080: 00n0__2xw

 

 

 

 

E0200 lm5=0>>0

 

 

 

 

85

The silicone nanocomposites synthesized in this work obey the indirect
relationship between solvent - influenced volume increase and the
nanocomposite Clay content in accordance with favorable reinforcement phase —
polymeric phase interfacial interactions. This supports the mechanical
improvement evidence identified in Chapter 2 that the nanolayers are not only
well dispersed in the silicone matrix (influencing a larger portion of the polymer),
but they are also having favorable interactions with the polymer at the interfacial
region (leading to the bound polymer - driven reduced swelling). Indeed, even at
clay loadings of 1-2% by weight, the nanocomposites exhibit a reduced swelling
compared to the pristine silicone on the order of 10-20%. Rather uniquely,
organofluorohectorite — synthesized nanocomposites generated with identical
clay loadings, but cured under different conditions gave rise to a disparity In the
amount of solvent equilibrated into their matrices. While all samples cured at
room temperature under vacuum absorbed the same volume of solvent at
particular clay loadings, those nanocomposites that were prematurely heated
(~100°C) were capable of uptaking more organics. This can be understood by
examining the X-ray diffraction patterns in the two dissimilarly cured samples.
Amorphous patterns arise from all room temperature cured systems, indicating a
more exfoliated nanocomposite, while rapid heating of the ungelled samples
causes a premature gelation that immobilizes Clay layers before they can be fully
dispersed. Figure 3.3 shows two XRD patterns, one of the typically cured PDMS
nanocomposite, the other of a prematurely heated nanocomposite that retains

some registry between Clay layers. Detrimental implications of non-exfoliated

86

Clay tactoids arise from inefficient surface area exposure of the Clay particles,

and thus a limited reinforcing effect.

87

prematurely heated nanocomposite
3500

 

111111111

Intensity
086188“
5333533

111L1111111Li111111111111

 

 

 

IfTIIIIIUITIUTIrIII[IUIIIUIIIIIIII

1 2 3 4 5 6 7 3
20

room temperature cured nanocomposite

3500

 

111

141111

888
8888

Intensity

3 Fri

8
111L111111111111111J11111

8
o

 

 

CD

 

 

TTUU'IIIIIIUUUIITIIIIUIIITIIIIII

1 2 3 4 5 6 7 8
20

Figure 3.3 XRD patterns of silicone nanocomposites cured under different
conditions. Accelerating the cure with premature heating (top pattern) leads to a
lower degree of exfoliated Clay layers.

88

Aside from the importance of swelling polydimethylsiloxane, which exposes the
exfoliated Clay layers to the liquid medium, the solvent also must be capable of
dissolving the salt in order to allow for the solvated cations to penetrate into the
silicone matrix and ultimately approach the Clay exchange sites. The use of
tetrahydrofuran as a post-cure ion exchange bath is an ideal Choice as it both
swells the organophilic silicone matrix and also has sufficient polarity and
coordinative ability to allow for copper salt dissolution. Other solvents explored in
achieving both the swelling of the silicone matrix and dissolution of various salts
included acetone, ethanol, isopropanol, acetonitrile, chloroform, and mixtures
thereof. While the copper salt has appreciable solubility in the more polar
solvents, these solutions lack the capability of infiltrating the silicone rubber.
Employing THF solves this problem, and therefore all post-cure ion exchanges
were carried out in a THF — CuClz bath. As touched upon later, copper was
selected as the replacement cation for several reasons, however a variety of
copper (ll) salts were experimented with for delivering the cations. Although
ordinarily not a discriminating factor, the balancing anions did affect the end ion
exchanges in certain cases. Copper acetate is one such example where the
persistence of significant amounts of acetate after several washings was
detected by Infrared Spectroscopy. Though evidence for copper ion exchange
was noted in these systems, it appeared that the cetyltrimethylammonium
acetate salts formed after copper replacement were more soluble in and
therefore less mobile through the silicone. Just as the THF was selected as the
ideal solvent, the readily available and most successful copper source was

determined to be halide salts. An Illustration of the general post - cure ion

89

exchange procedure utilizing THF as the swelling agent, and CuClz as the

exchange salt is represented in figure 3.4.

90

0050000600000 000050

05000090 00_ 0.30.000: 000 05500000 5 0020>5 0.300090 05 .0 0:00.008 «.0 0.32....

 

  

‘ 103685886:
05500000000000
8.0.088

_ 05000500000:
. A 000.0000
_ 00000.5 “=10. .0008 00090

  
   

 

 

 

 

0030\th

 

020.00 5.
E05.

05.200028: ( V 020090000000

00.0.0005
- 00005.0 @550 E::00> 000050 005.000

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0.000000003000090 IT 00.0000.
000050

 

 

 

 

91

This study represents the first demonstration involved in probing the Clay
exchange site accessibility of a polymer — Clay nanocomposite. The potential
contributions for examining the fluorohectorite - Silicone hybrid system for this
previously unexplored phenomenon were touched on above, but the main focus
of this preliminary work is to first prove that the smectite surfaces can be
accessed even when housed in an entangled polymer matrix. The Challenges of
such a procedure are obvious, however, the unique properties of silicone rubbers
make the unconventional ion exchange achievable. The ability of silicone
rubbers to absorb and transmit large molecules through the matrix hinted at the
notion that a post-cure ion exchange was possible provided an adequate
solvation medium could be isolated. It is the weak intermolecular interactions
between neighboring polydimethylsiloxane Chains that facilitate the migration of
macromolecules into and through the matrix. Weak Chain — to - Chain forces
create large free volumes not ordinarily associated with organic elastomers, nor
any carbon based polymer systems. The lack of crystallinity does not discount
silicones as dimensionally stable, physically robust elastomers, however. In fact,
silicones have a more Chemically inert and thermally stable Si-O backbone
relative to carbon based polymers, and the bond strength is almost twice that of a
carbon — carbon bond (160 kcaI/mol versus 86 kcanol). The backbone also
has a much lower barrier to rotation (1.0 kcal/mol versus 3.0 kcanol), and this,
along with the reduced intermolecular interactions, creates a highly random
crosslinked polymer that can uptake large amounts of solvent, thereby extending

the siloxane chains and rendering the swollen matrix liquid-like. In this state the

92

desired migration of exchangeable cations to the exfoliated clay layer sites is
quite feasible. Figure 3.5 schematically shows the influence on
polydimethylsiloxane chains by equilibrating the crosslinked polymer network

with an organic solvent.

93

.015 5 00.0.00.
. :5 .000 000 0.000 050000500000 000050 0050:0090 05 00 0000000500.

500 LIL.

 

0.0 0.8.".

94

The occurrence of two distinct polymer regions, one at the interface and another
in the bulk, gives rise to anisotropic swelling of the matrix. Although less solvent
is partitioned into the bound polymer at the interfacial regions, the overriding
permeability of polydimethylsiloxane still permits infiltration of the solvated
‘ copper, as described in the results section. When equilibrated with Cu/T HF, the
silicone samples take on the color of the THF - copper solution as described
later. The nanocomposite color Change when copper is present provided
qualitative feedback and this, along with the added accuracy of monitoring
copper metal content in particular via elemental analysis, made it a practical
cation choice for the demonstration-of-concept experiments in silicone

nanocomposite post-cure ion exchanges.

3.2 Experimental

Materials. PDMS polymers with average molecular weight values of 400, 1750
and 4200 and tin(ll) 2-ethylhexaonate were obtained from Gelest, Inc.
Hexadecyltrimethylammonium bromide and all other Chemicals were purchased

from Aldrich Chemical Co. and used without further purification.

Polydimethylsiloxane (PDMS)

95

CH3 CH3
HO— (Sli-O)N- Sil-OH
CH3 CH3

Tetraethylorthosilicate (TEOS)

CIDEt
EtO - Si- OEt

OEt

Table 3.1 Silanol Terminated PDMS Monomers

 

 

 

 

 

 

PDMS MW N value (Si(CH3)2-O repeat) Viscosity (cSt)
400 6 20
1 500-2000 23 45-85
4200 57 90-1 20
1 8000 243 700-800

 

 

 

96

 

Organoclay Synthesis

Lithium fluorohectorite (Coming, Inc.) with an anhydrous unit cell formula
of Li1,12[M94.aaLi1_12](Sig,oo)F4 was ion-exchanged with a two-fold stoichiometric
excess of hexadecyltrimethylammonium bromide. A mixture of the surfactant
cations and a 1.0 wt % aqueous suspension of the inorganic Clay was stirred at
room temperature for 12 hours. The flocculated organoclay was removed by
vacuum filtration and the excess surfactant was washed from the Clay by re-
suspending the solid in water, filtering again, and then subjecting the solid to a
final Soxhlet extraction in ethanol for 24 hours. The clay was then dried, crushed
and sieved to a particle size <106 um prior to use in the intercalation and
nanocomposite studies. The water and organic content of the air-dried
organoclay, as determined by the weight loss at 150°C and 500°C, was 6 wt %

and 24 wt %, respectively.

Nanocomposite Synthesis
The nanocomposites were synthesized through initial swelling of

organofluorohectorite by the PDMS monomers, followed by addition of TEOS
and tin octoate catalyst to commence crosslinking. A predetermined amount of
organoclay was added to the desired PDMS pre-polymer and mixed for 12 hours.
Agitation of the Clay-PDMS mixture was accomplished by first manual stirring
with a spatula followed by vortexing the suspension for several minutes. A
stoichiometric amount of tetraethylorthosilicate (TEOS) as a crosslinking agent
and tin(ll) 2-ethylhexanoate as a catalyst (T EOS : Sn = 4 : 1 by volume) were

added to the PDMS - organoclay mixture. The resulting thixotropic intercalate

97

was then transferred to an stainless steel mold and out gassed under vacuum.
Crosslinking of the nanocomposite was carried out under vacuum at ambient
temperature for 12 hours, followed by an additional 12 hours at 100 °C remaining

under vacuum.

Post-cure Ion Exchange

Nanocomposite and pristine crosslinked silicone samples were exposed to a THF
bath containing copper cations for the ion exchange of cetyltrimethylammonium
present at the exfoliated clay surface. Nanocomposite immersion in copper salt-
containing THF was carried out at a concentration of 6.0*10'° mole per gram of
silicate in the nanocomposite samples (10 fold excess calculated by Clay CEC).
This dark green solution was allowed to penetrate the samples and ultimately
swell them to several times their initial volume once replete with solvent. The
equilibrated samples remained in copper - THF solution for 24 hours, after which
they were transferred to a pure THF wash. At this point the samples had taken
on the dark green color of the copper - THF solution, and an immediate leaching
of the excess adsorbed copper into the first THF wash was observed.
Successive THF washes, each 10 ml per gram of silicone sample, were
employed to remove the unbound copper and potentially exchanged quaternary
alkylamine salts. Elemental analysis studies indicated that the removal of
unbound copper was completed after roughly 2 washes, however a total of 10

THF washes were employed to ensure an equilibrated state.

98

 

3.3

imn

sar

rel

the

IQ!

8V

I'IO

Ull

UI'I

rel

89

na

3.3 Results and Discussion

Cured pristine silicone and silicone nanocomposite samples were fully
immersed in a copper Chloride - THF solution. Once equilibrated, the swollen
samples had taken on the dark green color of the solution. This color was
retained after allowing the solvent to evaporate (a process which destroyed the
pristine Silicone), indicating that strong coordination of THF to copper persisted in
the samples even in the unswollen state. Washing the samples with pure THF
resulted in rapid removal of the unbound copper salt, as noted above. Indeed,
evaporation Of THF after the third wash, unlike the first two, did not result in any
notable residual copper/alkylammonium salt. Therefore, the high solubility of
unbound copper Chloride in THF is thought to promote the removal of
unexchanged copper as well as potentially exchanged alkylammonium salts
relatively efficiently. Figure 3.6 depicts the copper content present after
sequential THF washings for pristine (0%), 5% and 10%wt. silicate

nanocomposite samples.

99

 

 

15% FH nano
.10% FH nano
A 5% FH nano

.pristine

 

 

 

 

 

 

 

+

£
L A
l #1 T‘ 14

 

 

 

Copper content (*103 ppm)

 

 

 

 

 

 

 

Figure 3.6 Copper content following post-cure ion exchanges for silicone
nanocomposites examined after successive THF washes. Unleachable copper
accounted for roughly 70% of the fluorohectorite ion exchange capacity in the

filled systems, while pristine silicone retained almost no copper after washing.

100

The final nanocomposites retained a pale sky blue color after THF washings,
Identical to Cu*2-exchanged fluorohectorite (synthesized for comparison), while
the pristine silicones returned to their original transparent state. A brief depiction

of this significant result is shown schematically in figure 3.7.

101

000000 00500. 0050005000000 00. 200 ".I... .52. 059:0 000:

.0>0>>00 0020000 002. 00.00.00 05.000 000 050005000000 £00 0.5 .00000 .0 00.0.0: .000? 0.... 00 0...... 0.30
000 500 0000090 0105030 0 050: 00000090 00. 0.003000 00. .0 0000000000. 0000.000w hm 0.390

 

 

 

 

 

00.200 00.200 .0550
0.200 00.0.00 00.05.0050 0.200 05.0.0
0 A ._. __ 1 0
000.5 “.1... - .V 01.3030
0:000E000000 02000050000:
00.0.0.5:0050 00.05.0050 02000050000:
“.1... 01.00.05
.I 1 _ . 1 E

 

.1 :0 000.0000 . 0000090
00090 .0 0002. 00. 0.005000

102

The predicted values, assuming complete ion exchange of Cu“2 based on a
cationic exchange capacity of 121 moo/1009 (experimentally determined) for
fluorohectorite, are listed in Table 3.2 along with the observed final copper

Content for several nanocomposites.

Table 3.2 Copper Content Present After Post - Cure Ion Exchange

 

 

 

 

 

Theoretical Cu Actual Cu % ion
(ppm? (PW!)b exchange
Pristine 0 53 —
5% PH 1330 891 67
10% FH 2660 1747 66
15% PH 3990 2884 72

 

 

 

 

 

 

a - theoretical Cu+2 content as calculated using an experimentally
determined fluorohectorite CEO of 121meq/100g

b - residual copper content determined by ICP elemental analysis

The direct correlation between clay content and remnant copper concentration
after THF soakings reveals the presence of strong copper to clay binding. The
high solubility of cupric halides in THF dispels the possibility of insufficient
washing giving rise to the residual, unleachable copper analyzed in the final clay
containing samples. Unlike the reverse scenario, quantitative copper exchange

for cationic alkylamines occupying the organoclay exchange sites is not

103

anticipated in any solvent. More generally, the attempted ion exchange of any
inorganic cations for already present organic surfactant cations on high charge
density surfaces is highly unfavorable. Here the van der Waals interactions
between organic surfactant chains creates a more stable intercalated organoclay
phase relative to the inorganic clay. This leads to an unfavorable displacement
by the inorganic cations despite introducing a large excess of copper salt. The
higher the charge density of the exchange medium, the denser the surfactant
stacking, and thus the thermodynamically favorable organic entanglement
becomes more difficult to infiltrate by alternative cations. In fact, attempting a
cation exchange starting with pure cetyltrimethylammonium fluorohectorite
suspended in THF, and introducing a 20 fold excess of copper salt (based on
CEO of the clay) for 48 hours yields an identical XRD pattern to the original
organoclay, and elemental analysis studies indicate only trace amounts of copper
substitution into the final exchanged clay. However, THF is not to be precluded
as a viable solvent for general smectite clay ion exchange as Cu+2 -
fluorohectorite clays were successfully synthesized by introduction of 10 fold
excess CuClz to a parent inorganic fluorohectorite (Li+-exchanged) suspension in
THF. As hypothesized, the persistence of bound copper in the post-cure
nanocomposite ion exchanges occurs at a much higher fraction than pure
organoclay exchanges, on the order of 70%, using a 10 fold equivalence of
copper cations. The highly dispersed clay layers in exfoliated silicone
nanocomposites appear to facilitate exposure of the incumbent quatemary

surfactant cations to the copper - containing solvent relative to the exchange

104

performed with pure organoclay. Additionally, a more complete ion exchange of
copper for alkylammonium ions in the post-cure nanocomposite could result from
solvation of the C16 surfactant chains by the silicone polymer. The favorable
silicone - onium interactions (as supported by the large driving force for PDMS
monomer intercalation into the organofluorohectorite) disrupt onium - onium
contact at the silicate surface that, in the pure organoclay - copper exchange,
likely limits the abstraction of surfactant ions. This onium — onium disruption is
seen in certain solvated and intercalated organoclay species where the inclined
paraffin arrangement (001 spacing of 27.4 A for CTA-FH) is expanded to a
completely vertical orientation (~32 A solvated), and an interdigitated vertical
orientation (38 A intercalated with 400 MW PDMS). In the latter case,
polydimethylsiloxane monomers solvate the onium chains and permit their
enthalpic favorable contact with one another to be disrupted in favor of silicone -
onium interactions. Silicone — onium contact is not restricted to the clay surface,
therefore displacement of the organic cations by copper followed by subsequent
migration of the surfactant salts ultimately out of the silicone matrix into the THF
bath is a potential route not viable in the copper ion exchange of pure
organoclay. After washing, the comprehensive copper extraction in the clay-free
silicone, coupled with the direct correlation of clay content to the amount of
bound copper, indicates that the copper found in the nanocomposites does not
arise from residual incorporation or trapping in the polydimethylsiloxane matrix.
A schematic representation of the biased ion exchange in the nanocomposite

versus the pure organoclay, along with a depiction of the successful

105

fluorohectorite ion exchange of Li+ for Cu“2 as carried out in THF is represented

in figure 3.8.

106

Li+ FH C16 FH C16FH nanocomposite

9.? 33
E? ii

Li Li

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

THF/CuClz THF/CuClz THF/CuClz

 

 

 

V V V

 

 

 

 

 

 

a" Cu
E 67:

 

 

 

 

 

 

 

 

 

 

Figure 3.8 Schematic illustrating copper ion exchanges carried out in THF for
different parent clay systems. Excess copper successfully substitutes for both
the inorganic cations in lithium fluorohectorite as well as the majority of surfactant
ions on organofluorohectorite exfoliated in a silicone matrix. However, despite
increasing the copper concentration to 20 times the CEO of the clay, organic
cations could not be appreciably displaced by Cu+2 in THF.

107

3.4 Conclusions

In summary, crosslinking of a PDMS - swollen cetyltrimethylammonium
fluorohectorite clay—silicone suspension yields exfoliated nanocomposites as
outlined in chapter 2. Utilizing the high free volume and large diffusion coefficient
inherent in silicones, post-cure ion exchanges of the dispersed nanolayers were
performed. Post-cure ion exchange was accomplished by immersion of the
nanocomposites in a CuCl2/T HF solution, followed by pure THF purging of the
distended samples in order to remove unbound copper and potentially
exchanged alkylamine salts. The amount of unleachable copper varied directly
with clay content of the nanocomposites, while pristine silicone samples retained
essentially no copper after washing. In the clay containing samples, remnant
copper concentration accounted for roughly 70% of the predicted ion exchange
capacity for the exfoliated clay layers. Performing ion exchanges of the identical
clay (CTA-FH) outside the silicone matrix in THF resulted in only trace
exchanges of copper. While inorganic - inorganic ion exchanges were
successfully completed in THF, organic cations could only be exchanged for
inorganic cations (copper) when the organoclay was exfoliated in a PDMS matrix.
The accessible cationic clay sites in a crosslinked nanocomposite afford an
opportunity to explore selective membrane and flow-through gas or liquid
separation chemistry formerly unavailable with typical organo-funtionalized

silicate nanocomposites. The success of a post-cure ion exchange with silicone

108

nanocomposites likely stems from both the high permeability of the matrix,
allowing a relatively unhindered influx of exchangeable solvated cations, along
with the highly exposed exfoliated nanolayers, limiting confined galleries that
would inhibit accessibility of the cationic clay sites. Extension of post-cure
nanocomposite ion exchanges into other polymer systems will be dependant
upon how permeable and swellable the polymer matrix is, and also how well
exfoliated (and thus exposed to the exchange medium) the clay layers are in the

final nanocomposite.

3.5 References
(1) Pinnavaia, T. J. Science 1983, 220, 365-371.
(2) Messersmith, P. B.; Stupp, S. J. Mater. Res. 1992, 7, 2599.
(3) Okada, A.; Usuki, A. Mater. Sci. Eng. C-Biomimetic Mater. Sens.
Syst. 1995, 3, 109-115.
(4) Giannelis, E. P. Adv. Mater. 1996, 8, 29-&.
(5) Wang, 8.; Long, C.; Wang, X.; Li, 0.; Qi, Z. J. Appl. Polym. Sci.
1998, 69, 1557-1561.
(6) Burnside, S. D.; Giannelis, E. P. Abstr. Pap. Am. Chem. Soc. 1996,
212, 19.
(7) Burnside, S. D.; Giannelis, E. P. J. Polym. Sci. Pt. B—Polym. Phys.
2000,38, 1595-1604.

(8) LeBaron, P. C.; Pinnavaia, T. J. Chem. Mater. 2001, ACS ASAP.

109

(9) Shi, H.; Lan, T.; Pinnavaia, T. J. Chem. Mater. 1996, 8, 1584-1587.
(10) Balazs, A. 0.; Singh, G.; Zhulina, E. Macromolecules 1998, 31,
8370-8381.

(11) Lyatskaya, Y.; Balazs, A. C. Macromolecules 1998, 31 , 6676-6680.
(12) Coe, C. G. Access Nanoporous Mater., [Proc. Symp.] 1995, 213-
229.

(13) Langhendries, G.; Baron, G.; Vankelecom, |.; Parton, R.; Jacobs, P.

Catal. Today 2000, 56, 131-135.

110

Chapter 4
EXFOLIATION OF DIAMINE - MODIFIED CLAYS IN A
GLASSY EPOXY MATRIX

4.1 Introduction

Polymer layered silicate nanocomposites (PLSNs) represent one of the
most interesting classes of organic - inorganic hybrid materials, with an
increasing potential both in fundamental research and in commercial
applications.“15 The PLSNs exhibit improved physical and performance
properties compared to the unmodified polymers and the conventional
composites, mainly due to their unique phase morphology and interfacial
properties induced by the highly dispersed silicate nanolayers in the polymer
matrix. These effects are minimized in the conventional composites where the
inorganic additives have dimensions in the macroscopic (um) length scale. As a
result greater improvement of the polymers properties can be achieved at a
considerably lower concentration of the inorganic reinforcement phase in PLSNs
compared to conventional composites. Loadings of only up to ~5 wt. "/0
exfoliated silicate nanoparticles in PLSNs generally result in significant
enhancement of a number of physical properties, including mechanical properties
(modulus, strength, thermal expansion coefficient), barrier properties, thermal
stability, resistance to solvent swelling, flammability resistance and ablation

performance.

111

A wide variety of layered materials are potentially well suited for the formation
of organic — inorganic nanocomposites because their Iamellar elements have
high in-plane strength and stiffness accompanied with a high aspect ratio.
However, the smectite clays16 (e.g. montmorillonite, hectorite and synthetic
smectites) along with other layered silicates are the materials of choice for
polymer nanocomposite design, mainly because of their rich intercalation
chemistry which allows them to be chemically modified and to become
compatible with the polymer precursors. Many robust reinforcement fillers,
including high surface area silicas and various aluminosilicates. certainly have
the facility of strengthening a polymeric matrix, however, their lack of ion
exchange capability renders such materials highly limited as universal polymer
fillers. Additionally, even though they can be readily synthesized, layered
smectites occur ubiquitously in nature and can be purified at relatively low cost.

Toyota researchers demonstrated the first example in which an organic
montmorillonite clay was exfoliated in a thermoplastic nylon-6 polymer, resulting
in a commercialized product with enhanced thermo-mechanical properties. The
organic clay was synthesized by ion-exchanging the cations of the parent clay
with more hydrophobic organic onium ions.”19 In the next 10 years research
focused on issues relating to the most suitable organic modifier for the clay and
the best conditions for the formation of the nanocomposites. A major portion of
this research was explored in studying nanocomposites formed by thermoset
polymers, more specifically epoxy resins, and modified smectite clays,""°"2"“"2‘*29

along with other layered silicates, such as silicic acids (e.g. magadiite).3° It has

112

been previously shown that acidic primary onium ions, when they are ion-
exchanged for the inorganic cations of the parent layered silicates, catalyze the
intragallery epoxide polymerization process in the presence of diamine curing
agent.25'3° Parameters like the chain-length of the organic onium ion, the relative
acid strength (primary, secondary, tertiary) and the clay layer charge density,
which dictates the concentration of the organic modifier in the galleries, play an
important role in achieving optimum nanolayer intercalation by, and ultimately
exfoliation in the polymer. The catalytic activity of the acidic onium ions is more
pronounced when the alkyl chain is of particular size and functionality to
intercalate the largest amount of monomer. Matching the polarity of the
intragallery region with the polymer in this way provides a compatabilzed
environment that promotes a maximum polymerization rate between the clay
layers by permitting a sufficient concentration of reactive monomers to exist
within the acid catalyzed region between clay layers. In a recently reported
study,22 it was shown that the use of a hydroxyl - substituted quaternary
ammonium modifier in a montmorillonite clay resulted in a high degree of clay
player exfoliation. In this case the strong hydrogen bonding between the
surfactant hydroxyls and the oxirane ring in the resin provided a similar
intragallery reaction rate bias that facilitated the dissemination of the
organofunctionalized clay particles. Figure 4.1 schematically represents the

influential role of acidic surfactant molecules found in the intragallery regions.

113

 

wT/zw :/
+
0A» m: :\20
V II m...
007000 +20.
z:N
\/\ mwz
:02
W A“

 

 

0.05.0 0.. >00 00.0230 3:000:03. 1:0 I 0003.00 0. 0: 0 000x< «00.3.

In an effort to optimize the overall formation process of the of thermoset
epoxy — clay nanocomposites, the work presented in this chapter focuses on the
synthesis of a new type of potentially advantageous organic smectite clays. The
direct incorporation of Jeffamine (multifuntionalized primary amines provided by
Huntsman Chemical) curing agents inside the galleries of montmorillonite and
fluorohectorite clays in the form of diprotonated diamines through an ion-
exchange process, is presented as an alternative, commercially attractive way of
preparing epoxy — clay nanocomposites. Different degrees of clay nanolayer
intercalation/exfoliation could be achieved, depending mainly on the chain-length
and the concentration of the diamine surfactant molecules in the clay galleries.
While a substantial improvement in the mechanical properties of all
nanocomposites synthesized via this route was found, the ultimate goal of
achieving the highest degree of clay exfoliation is stressed, as these materials
should be most functional commercially in replacing heavier glass — reinforced

engineering plastics.

4.2 Experimental

Materials. The epoxide resin used for synthesis of the pristine glassy
epoxy polymer and the epoxy - clay nanocomposites was a diglycidyl ether of
bisphenol A (DGEBA), more specifically, Shell EPON 826 with MW ~369 and

epoxy equivalent weight ~182:

115

o r c on ,00 ... 0
1 V“ J“ 4(“ ' * -cu —-cH- CH
cn,-cn—CH,—o—~—< SAC; warren—cu, o-—< Wm. . r-o , .

n=000%x100%)

The curing agent was a polyoxypropylene diamine, Jeffamine D230

(Huntsman Chemical) with MW ~230:

HZN'CHCH 2f OCH sz}XNH 2
CH3 CH3

Jeffamines D2000 (MW ~2000, n=33.1), D400 (MW ~400, n=5.6) and D230
(MW ~230, n=2.6) were used for the synthesis of the organic clays, as is outlined
below. All other chemicals used in this work were purchased from Aldrich
Chemical Co. and used without further purification.

Organic Clay Synthesis

An industrially purified Na*-montmorillonite clay (PGW, Nanocor Inc.) with
CEC = 120 meq/100 g, was used to produce the H“ and Li“ forms of the
montmorillonite clay by typical ion-exchange procedures with diluted HCI (for H”-
clay) or LiCI (for Li+-clay) aqueous solutions. A synthetic Li+-fluorohectorite clay
(FH, Coming Inc.) with CEC similar to that of the PGW montmorillonite was used
as the parent fluorohectorite clay in this study.

The organic clays were prepared by ion-exchanging the inorganic clay

with diprotonated diamines, Jeffamines D2000, D400 and 0230. Jeffamines in

116

slight excess of the stoichiometric ion-exchange of the clay were diluted in
aqueous HCI solution; a small excess of H“ was also used to ensure the
formation of diprotonated diamines. The acidic Jeffamine solution was then
blended for 1-2 min with an aqueous inorganic clay suspension. The blended
mixture was further stirred for 48 hrs at ambient temperature. In the case of
0230, a second sample was prepared by ion-exchanging the parent clay using
double the stoichiometric amount of diamine, in order to achieve higher degree of
ion exchange. The ion-exchanged clays were separated by centrifugation,
washed 3-4 times with deionized water (until free of Cl') and once with EtOH,
previous to air-drying at room temperature. The dried products were ground and
sieved to a particle size < 270 mesh (53 pm). All the inorganic and organic
montomorillonite (PGW) clays prepared are listed in Table 1, together with

compositional and structural data.

Preparation of Epoxy - Clay Nanocomposites

The pristine glassy epoxy polymer was formed by mixing the epoxy monomer
(EPON 826) with the curing agent (Jeffamine D230) at 50°C for ~30 min,
outgassing of the liquid mixture at room temperature and curing in RTV silicone
rubber molds at 75°C for 3 hrs and, subsequently, at 125°C for an additional 3
hrs, under nitrogen (to avoid oxidation and browning of the polymer at the higher
cure temperatures).

In the preparation of the epoxy - clay nanocomposites, the modified clay was

added to EPON 826 - curing agent 0230 mixture at 50°C by stirring for 30 min,

117

before outgassing and curing at 75°C and 125°C (procedure P-1), similar to the
pristine polymer formation. Alternatively, the epoxy EPON 826 was initially
mixed with the clay at 50°C for a certain period of time and then the curing agent
D230 was added (procedure P-2). In the case of the organic clays synthesized
with the three different Jeffamines (D2000, D400, D230), the moles of onium ions
present in the organic clay were counted as contributing crosslinking agents to
the curing process, supplementary to the amine groups of Jeffamine D230 (the
main curing agent for the formation of the glassy epoxy polymers). The amount
of diamine crosslinker arising from the organic modifier of the clays was always
less than 4 % of the total amine groups necessary for complete cross-linking of
the epoxy monomers. The clay loadings of the nanocomposites prepared were
1, 3 and 6 wt. °/o silicate (meaning the content of organic in the functionalized
clays was normalized and actual reinforcing silicate weights were reported).
Physical Measurements

X-ray diffraction (XRD) patterns were obtained on a Rigaku rotaflex 2008
diffractometer equipped with Cu Ka X-ray radiation and a curved crystal graphite
monochromator, operating at 45 KV and 100 mA. The diffraction patterns were
collected between 1 and 12° at a scanning rate of 1°/min. Samples of the liquid
(gel-like) epoxy - clay mixtures or of the uncured epoxy/curing agent - clay
mixtures were prepared by applying a thin film on filter paper mounted on glass
slide. Cured composite samples were prepared by mounting a rectangular flat

specimen into an aluminum holder.

118

Transmission Electron Microscopy (T EM) images were obtained on a JEOL
JEM-1OOCX ll microscope with a CeBe filament and an accelerating voltage of
120 W, a beam diameter of ~5 pm and an objective aperture of 20 pm. The
TEM samples were prepared by supporting thin sections (80-100 nm) of the
nanocomposite samples onto 300 mesh nickel grids.

Dynamic Mechanical Analysis tests were performed on a DMA 2980
Dynamic Mechanic Analyzer (TA instruments) in the Three Point Bending mode
at a frequency of 1 Hz and amplitude of 20 pm, from 25°C to 140°C at a heating
rate of 4°C/min. The pristine epoxy polymer and the nanocomposite specimens
were rectangular bars with dimensions 60 mm x 13 mm x 3 mm.

Thermogravimetric analyses (TGA) were performed using a Cahn TG System
121 Analyzer. The powdered clays, the pristine epoxy polymer and the
nanocomposite samples tested, were heated to 800°C at a heating rate of
5°C/min under N2 flow.

The composition of the inorganic and the organic clays was determined by a
combination of Energy Dispersive Spectroscopy (EDS), CHN and TGA analysis.
4.3 Results and Discussion
Synthesis of Diamine Smectite Clay lntercalates.

Insertion of organic onium ions in the layered silicates by ion - exchange
procedures results in an expansion of basal separation between the silicate
nanolayers. Various potential configurations of the organic modifier give rise to
an array of observed basal distances as outlined in chapter 1, and noted by .

Legaly and Giannelis?"32 When alkylammonium ions with a 012-018 carbon

119

chain are ion - exchanged for the inorganic cations in typical montmorillonite
clays (CEC ~90-100 meg/100 9), they adopt a lateral bilayer (~18 A) intragallery
structure. The same surfactant molecules present in the higher charge density
synthetic fluorohectorite clay (CEC ~120-130 meq/100 9) gives rise to an inclined
paraffin - like structure (~20-30 A, depending on the alkyl length). The psuedo -
trilayer structure shown in chapter 1 is also a possible intermediate surfactant
orientation, however, due to the regular basal increase witnessed when moving
to larger exchanged alkyl chains in fluorohectorite, it is more likely an inclination
of consecutively steeper angles accounting for the incremental clay layer
expansion noted. A lipid bilayer configuration at higher onium ion concentrations,
perhaps induced upon the introduction of excess neutral amines, is also
possible.“'33 However, synthesizing nanocomposites using clays containing these
relatively high concentrations of alkylammonium ions yields undesirable dangling
alkyl chains in the final matrix. Here the gallery .‘ contained onium ions
contribute to the crosslinking stiochiometry of the epoxy groups, and have a
detrimental effect on the final properties by rubberizing and solvating the resin
with the low molecular weight, mono-functionalized amines ordinarily employed
in epoxy nancomposites.3°

The utilization of the diprotonated primary diamines (Jeffamines) as
organic modifiers of the clay instead of typical octyl and dodecyl amines
eliminates this undesirable “polymeric dead end” effect. The XRD patterns of the
fluorohectorite clay exchanged by diprotonated Jeffamines D230, D400 and

02000 are shown in figure 4.2 (organoclays denoted 0230-FH, D400-FH,

120

DZOOO-FH). Similar XRD patterns of the respective PGW - diamine exchanged

clays were observed, though not shown here.

121

Intensity

Figure 4.2

 

 

 

 

46.0 A

23.2}.

 

 

 

 

20 (degrees)

X-ray diffraction patterns of the Li"-Fluorohectorite (Li-FH)

and organically modified FH-clays (with diprotonated diamines D230, D400,

02000).

122

From the spacings of the dam basal reflection it can be suggested that the short
chain, low molecular weight Jeffamine 0230 (n=2.6, the average number of
polypropylene oxide units in the chain) adopts a lateral monolayer structure (13.8
A), as expected from the short end — to - end distance in the diamine. When the
slightly longer polypropylene oxide chain - containing Jeffamine D400 (n=5.6)
was used, the basal spacing of the exchanged fluorohectorite was ~17.5 A. This
spacing could be either explained by a lateral bilayer orientation, or with a more
complex argument that will be addressed in chapter 5. Ion - exchange of the
parent clay with the high molecular weight diprotonated Jeffamine 02000
(n=33.1) resulted in a d-spacing of ~46 A, indicative of a structure where the long
alkyl oxide chains of D2000 adopt an inclined paraffin - like configurationf Unlike
their purely carbon backbone — containing cousins, the polypropyleneneoxide
(PPO) diamines do not have the same solubility properties, nor conformational
restrictions most clay chemists are accustomed to. For example, neutral alkyl
chain diamines are not soluble in water, while the clear solutions obtained from
aqueous dissolution of 0230 and D400 give off noticeable heats of mixing. The
high solubility of these PPO diamines results in a lowered micellular driving force,
meaning complete organic insertion into all of the clay exchange sites is less
likely than if synthesizing typical organoclay with purely carbon — containing
surfactants. Typically the ion exchange of cationic alkyl surfactant in smectite
clays involves the simultaneous influx of large quantities of onium ions, as
observed in the instantaneous flocculation of formed hydrophobic clay tactoids

upon surfactant addition. In the PPO case, there are not sufficient chain — to —

123

chain interactions in solution to permit the rapid and complete ion exchanges
ordinarily anticipated in synthesizing organoclays. For this reason, somewhat
incomplete ion exchanges are the norm for PPO — based cations, as indicated in
the compositional and basal spacing data of the diamine - exchanged clays given

in table 4.1.

124

Table 4.1 Composition and Basal Spacings of the Inorganic and Organic
Clays used for the Preparation of the Nanocomposites

 

 

. . parent organic modifier a basal spacings
Inorganic — inorganic
organic clays clay

onium diprotonated dom dooz

ion (°/o) diamine (A) (A)
H-PGW Na-PGW - - 13.4 -
Li-PGW Na-PGW - - 12.4 -
Li-FH Li-FH - - 12.3 -
D2000-PGW Na-PGW 80 D2000 45.5 21 .8
D400-PGW Na-PGW 90 D400 17.2 -
0230-PGW Na-PGW 85 D230 13.8 -
D230/high-PGW" Na-PGW 95 0230 13.8 -
D2000-PH Li-FH 80 D2000 46.0 23.2
D400-FH Li-FH 85 D400 17.2 -
0230-PH Li-FH 80 D230 13.8 -
0230/high-FH b Li-FH 95 0230 14.0 -

 

a The inorganic clays were ion-exchanged with diprotonated diamines D2000,
D400 and D230 (Jeffamines, Huntsman Chemical); the Na+ content of all the ion-
exchanged clays was below 5 %

° The amount of D230 used for the ion-exchange was twice the amount

necessary for stoichiometric exchange of Na“ or Li+ ions with onium ions

Epoxy-Clay Nanocomposite Formation from Diamine Smectite Clay

lntercalates.The complexity of a three - component system in making a

125

thermoset epoxy — clay nanocomposite has been well described recently by Vaia
and his co-workers.22 The preferred sequence of component mixing, the use of
solvents to reduce the viscosity of the system, the temperature and time of
curing, and finally the fabrication techniques could have a significant effect on the
quality of the final nanocomposite material. With regard to the preparation of the
mixture before curing, the compatibility of the organic modifier with both the
epoxy monomers and the curing agent should be considered. Strong cases can
be made for both addition of pure epoxy resin to the clay (followed finally by
Jeffamine), and also for premixing the curing agent and resin prior to organoclay
addition. Epoxy resin by itself will only be available to react with the diamines
present between the clay layers, therefore a longer mixing time with smaller
polymer domains will be possible. Contrarily, premixing both polymeric
components before adding the organoclay lowers the viscosity initially (allowing
for easier mixing), and ensures sufficient curing agent content in the interclay
regions. In the case of the diprotonated diamines used in this study, successful
intercalation by the pre-polymer, and exfoliation of the clay in the final polymer
were accomplished via both routes. Therefore, it appears that the conditions are
optimum for the epoxy monomers not only to migrate inside the galleries, as
evidenced by the intercalated X-ray patterns, but also the protonated alpha -
omega diamines effectively catalyze the intragallery polymerization. However, in
order to avoid gelling in the premixed epoxy — curing agent scenario, care should
be taken that the mixing of the polymeric components is not overly lengthy or

heated. An alternative way to achieve premixing could be the use of low - boiling

126

point solvents for viscosity reduction and facilitation of the epoxy monomers
intercalation in the clay galleries.

The above described reaction pathways between the epoxy and the onium
ions of the diprotonated diamines require the polymer precursors to diffuse inside
the clay galleries. This was clearly the case for both the D2000-PGW and the
D2000-PH organic clays, as indicated in the alluded to intercalation studies.
Figure 4.3 represents XRD patterns of DZOOO-FH clay in which the epoxy
monomer was progressively intercalated resulting in d-spacing values of ~59 A
after 6 hrs of mixing and ~64 A after 24 hours mixing at 50°C. After addition of
the short chain diamine D230 curing agent necessary for the formation of the
rigid epoxy matrix, gelation of the epoxy matrix occurs and polymer domains
become larger, thus forcing apart the clay layers. A broad second order peak
with d-spacing of ~30 A dominated in the XRD pattern after heating the mixture
at 75°C for 30 minutes. As crosslinking continued, the basal reflection continued
to expand, noted in the second order peak as it became broader and shifted
towards lower angle 28. The XRD pattern of the final, cured nanocomposite
becomes completely amorphous, revealing the loss of registry between clay
layers. As a check to ensure that the clay is still perceptible to the X-rays, the in
— plane 060 reflection given off by individual clay layers is also included in figure

4.3 as an inset peak near 60° 28.

127

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m (D) 59 60 61 62 63 64“
C
g. _ __ f >fi
(I)
c
.93
E
(B)
u -(A) _ m
l l l l
1 3 5 7 9 11
20 (degrees)
Figure 4.3 X-ray diffraction patterns of mixtures of the organic clays

DZOOO-FH, the epoxy monomer EPON 826 and the curing agent 0230: (A) after
intercalation of EPON 826 in DZOOO-FH at 50°C for 6 hrs, (B) after intercalation of
EPON 826 for 24 hrs, (C) after addition of 0230 in the mixture of step C,
outgassing and heating at 75°C for 0.5 hrs, (D) after curing at 75°C, 3 hrs and
125°C, 3hrs (the clay loading was 6 wt. % silicate basis). The inset shows the
presence of an in-plane 060 clay reflection in the cured nanocomposite.

128

It is interesting to note that after pre-intercalation of the epoxy monomer the
Iamellar ordering of the D2000-PH organic fluorohectorite-clay was not damaged,
although the increase in d-spacing (~20 A) resulted in a relatively high value (~65
A) compared to other organic layered silicates in similar epoxy resins.”30 This
could be attributed to electrostatic interactions between the onium ions of the
alpha-omega diamine (in a paraffin-like arrangement) and the negatively charged
clay layers, capable of retaining the integrity of the nanolayer tactoids. In the
case of the D2000-PGW organic montmorillonite clay, the degree of Iamellar
ordering was lower upon mixing with the epoxy monomers, since both the (001)
and (002) reflections in the XRD pattern (not shown) were relatively broad
compared to those in the XRD pattern of DZOOO-FH. The lower aspect ratio of
PGW (~200) compared to FH (~2000) also likely accounts for the higher
tendency of the montmorillonite nanolayers to lose higher orders of reflection
more easily, even during pre-intercalation of the epoxy monomers. Also
important to note is the fact that synthetic clays have a very uniform distribution
of charge compared to naturally occurring clays, and this can lead to a more
regular arrangement of ion exchanged surfactants and intercalated pre-polymers
between the layers. Regardless, the initial organoclay basal spacing and ease of
intercalating epoxy resin for both types of clays examined are promising results
which should not discount either as potential nanoscaled fillers for reinforcement
of an epoxy polymer.

The above suggested mechanism of nanolayer exfoliation is in accord with

the findings from the TEM images of the nanocomposites prepared from D2000-

129

FH (figures 4.4a and b) and D2000-PGW (figures 4.4c and d), at clay loading of 6

wt. °/o silicate.

130

 

Figure 4.4 TEM images of thin sections of the glassy epoxy nanocomposites
prepared from D2000-FH. The clay loading of the nanocomposite samples was 6
wt. % silicate.

131

The DZOOO-FH nanocomposite consisted mainly of tactoids with
disordered, highly intercalated layers of high aspect ratio (2 2000), with basal
spacing between 5—15 nm. Individual, shorter layers could also be found on the
edges of the primary dispersed particles and most frequently in domains with
relatively low number of layers. It appeared that the smaller fluorohectorite
layers were more easily disseminated into the matrix, and accordingly it would
seem that that much smaller naturally occurring montmorillonite should show a

better dispersion of the clay layers.

132

 

Figure 4.5 TEM images of thin sections of the glassy epoxy nanocomposites
prepared from D2000-PGW. The clay loading of the nanocomposite samples was
6 wt. % silicate.

133

Indeed, the nanolayers of the lower aspect ratio D2000-PGW clay were more
randomly dispersed in the polymer matrix and existed in a more exfoliated
arrangement.

The XRD patterns of nanocomposites formed from D230 and D400
exchanged clays indicated that no significant intercalation occurred before or
during the polymerization process, as it can be seen for the D230-PGW and

D400-PGW organic clays in figure 4.6.

134

 

 

    

 

I I I I I 1

(D) 59606162 6364

 

Intensity (arb. units)

 

 

 

 

28 (degrees)

Figure 4.6 X-ray diffraction patterns of EPON 826 - montmorillonite PGW clay
composites and nanocomposites using: (A) H-PGW, (B) 0230-PGW, (C) D400-
PGW and (D) D2000-PGW clays. No intercalation occurred for the inorganic, nor
the D230 and D400-PGW clays. The inset shows the presence of an in-plane
060 clay reflection in the cured D2000-PGW nanocomposite.

135

This behavior is expected for 0230-clays due to the relatively short PPO
chain of the 230 molecular weight diamine, as there is little driving force for
solvation by epoxy resin into the more polar galleries. However, in the case of
D400, the d-spacing of both organic clays was ~17.5 A, similar to that of a
primary 018 alkylammonium exchanged montmorillonite, hinting that there. may
be sufficient driving force to permit intercalation of the epoxy resin. Octadecyl
primary amine - montmorillionites have been shown in other work, and also
positively tested here for comparison, to readily intercalate 826 resin. The fact
that D400-clays do not allow for epoxy intercalation could be attributed to the
situation alluded to above where the electrostatic interactions between either end
of the two protonated amines in each diamine and the clay layers are strong
enough to inhibit any significant swelling by essentially pinning the galleries shut.
Somewhat surprisingly, the composite samples prepared by 0230-PGW and
D400-PGW clays exhibited very good modulus properties and the 3 mm thick
specimens which were prepared for DMA tests had a light brown color and were
optically transparent, despite retention of the organoclay basal spacing. This
indicates that while epoxy resin was unsuccessful at penetrating between the
layers, the separation and suspension of small PGW tactoids could be
accomplished. The larger particle size fluorohectorite clays were not as easily
separated into individual tactoids, and the composites showed a distinct
separation of the phases, causing them to be opaque.

In contrast to the non-intercalated composite samples formed from D230-

and D400-exchanged PGW, the use of the inorganic H*- and Li+-clays resulted in

136

samples where nearly 20 wt. % of the added clay (6 wt. % silicate) settled at the
bottom of the composite specimen. Inorganic clay, it seems, even in the small
particle sized PGW, does not become separated into smaller tactoids as in the
case of D-230 and D-400 PGW.

As aforementioned, the nanolayers of a D2000-clay intercalate could be
exfoliated in a glassy epoxy matrix first through intercalation by the resin,
followed by the crosslinking reaction. Moreover, pre-intercalating the 2000
molecular weight diamine in an inorganic H"-clay before the addition of the epoxy
resin was a viable means to synthesize the effective D2000-organoclay avoiding
the costly and time - consuming step of synthesizing the organic clay in solution
and washing. Here the acid - base reaction provides the driving force for
insertion of the large Jeffamine D2000, creating an in situ organic clay synthesis

that can be accomplished by using an otherwise inactive H*-clay.

Properties of the Nanocomposites. The affects of clay nanolayer
reinforcement on the mechanical properties of the D230 - cured 826 glassy
epoxy polymer were investigated by Dynamic Mechanic Analysis (DMA). DMA is
a powerful analysis tool for probing the stiffness, glass transition temperature and
heat dependant behavior of polymeric materials. While DMA can be used to
analyze complex dynamic rigid and viscoelastic polymer chain behavior, it can be
described simply as monitoring a material’s response to an applied oscillatory
force.34 Using a three point bending mode, the sample is supported at either end

and a constant force is applied at the center of the material. In conjunction with

137

the steady force, a dynamic pulsed load is also applied at a known frequency and
amplitude. Through monitoring the deflection of the material, the delay of
deflection and the resistance to deflection one can plot temperature — dependant
curves that track two different stiffness, or moduli, of the material. The DMA
technique provides two different moduli, an elastic modulus (E) which is called
storage modulus and is related to the ability of the material to return or store
energy and an imaginary (loss) modulus (E’) which relates to the ability of the
material to lose or disperse energy. In simple terms, one can envision dropping
a ball, and measuring the height that the ball returns on the rebound, its storage
modulus, and the difference of the dropped height to the returned height, the loss
modulus. The ratio E”/E’ is called tan delta (tan6) and typically reaches a
maximum at the glass transition (T9) of the polymer. The following schematic,
figure 4.7, depicts the sample orientation, the three point bending sample holder

and the curves describing the energy applied, lost and returned in the DMA test.

138

 

 

 

 

Force

1' j
T'p' '
a. .

 

 

 

 

 

 

 

 

 

 

 

 

Flextural 3 POiIlt Bending static fOI'OC
Time
7 E lost to
internal motion
A
elastic response

 

 

 

Figure 4.7 Schematic portraying the three point bending device used in
obtaining DMA data and the plot of DMA testing parameters. The amount of
energy the material returns is related to its storage modulus, the energy
dissipated in the internal motions of the polymer chains is related to its loss
modulus and the actual measured material lag in experiencing the oscillatory

force is the phase angle.

139

Care should be taken when independently acquired DMA data are compared
since different DMA modes can be applied, such as the flexural (ie. three point
bending) or torsional mode. The curves in figures 4.8 and 5b show the
dependency of the storage modulus and the tan6 (E’/E”) on temperature
obtained from the three point bending DMA tests performed on the
nanocomposite samples of this study. As would be expected, increasing the
heat makes a glassy matrix more plastic, and, as the chains become more
mobile, the dampening effect on the applied force is increased, giving a rise in
the loss modulus. Coincident with the increase is loss modulus is a reduction in
storage modulus with increased heating as the material loses stiffness
(resistance to a load) and becomes more deformable. The data in Table 2 refers
to the storage modulus values of the pristine polymer and the nanocomposite
samples at 40°C (glassy region below T9) and at 120°C (rubbery region above

T9), and to the Tg values of the samples.

140

 

 

Table 4.2 Dynamic Mechanical Analysis and thermal analysis data of
the pristine epoxy polymer and the clay-polymer
nanocomposite samples "

samples storage modulus (E) Tg Thermal
stability d
at 40°C at 110°C (tan6) (TGA data)
(GPa); (MP3); (°C), :1: (°C)
10 /o 10 /o 2
Pristine epoxy-A 2.9 18.4 82.6 373
Pristine epoxy-B b 3.0 20.1 75.5 351
H-PGW 2.9 22.1 82.8 354
Li-PGW 2.7 19.9 82.3 356'
Li-FH 3.1 25.2 85.2 351
3:3)00-PGW (1 Wt % 3.5 21.5 81.2 345
3:3)00'PGW (3 w“ % 3.5 22.1 76.8 341
DZOOO-PGW 3.8 21.5 69.8 336
D2000-PGW ° 3.6 24.0 75.1 339
D400-PGW 3.6 23.2 84.2 346
D230-PGW 3.5 28.3 84.5 358
D2000-PH 3.9 30.5 73.7 323
D400-PH 3.9 32.1 79.1 340
D230-PH 4.0 45.4 84.7 345

 

a The nanocomposite samples were named from the organic clay used; the clay
loading of the samples was 6 wt. % silicate basis, unless otherwise stated

b Substitution of 4 °/o mole D2000 (organic modifier of clays) for 0230 (curing
agent) in the pristine epoxy polymer

° Nanocomposite sample prepared by pre-mixing of epoxy monomers with the
organic clay (P-2); all the rest of the nanocomposite samples in Table 2 were
prepared by direct addition of the clays in the epoxy - curing agent mixture (P-1)

141

d Temperature at which the weight loss in TGA was 1.5 %

The storage modulus in both the glassy and rubbery regions is higher for
the nanocomposite samples prepared from the diamine exchanged organic clays
(6 wt. % silicate loading), compared to the pristine epoxy polymer. The storage
modulus data for a selected portion of the nanocomposites studied is shown in

figure 4.8.

142

 

 

 

 

 

. .\('D) (1'3)\

4000- (E) (P)
a , .
E (A)'
% 3(XX)- 90 110
“D
o .

2
902000-
8
9 -
(I)

3

 

 

 

 

O
..
_
2
q
.

2D 40 60 80 1(1) 120 140

Figure 4.8 Dynamic Mechanical Analysis measurements indicating storage
modulus vs. temperature curves for (A) Pristine-A epoxy polymer and for the
nanocomposite samples prepared from (B) H-PGW, (C) D2000-PGW, (D)
D2000-FH, (E) 0230-PGW, (F) D230-PH. The clay loading of the nanocomposite

samples was 6 wt. % silicate basis.

143

The most commonly accepted means for obtaining the glass transition
temperature of a plastic material is by taking the maximum value of the loss to

storage modulus ratio. This ratio, denoted tan 6, is plotted in figure 4.9.

144

 

1.2

0.8 -

Tand

0.4 ~

 

 

 

 

25 45 65 85 105 125
Tenperature ((J C)

Figure 4.9 Dynamic Mechanical Analysis measurements indicating tan6 vs.
temperature, for (A) Pristine-A epoxy polymer and for the nanocomposite
samples prepared from (B) H-PGW, (C) DZOOO-PGW, (D) D2000-FH, (E) D230-
PGW, (F) D230-PH. The clay loading of the nanocomposite samples was 6 wt. %

silicate basis.

145

As for the D2000-PGW nanocomposite, a 25 °/o increase in storage modulus
was noted in both glassy and elastomeric regions. The composite samples
formed from the D230-PGW and D400-PGW clays showed similar increase of
the storage modulus, although the clays in those samples were not intercalated
by epoxy resin. A 35 °/o (glassy region) and 65 °/o (rubbery region) increase in
storage modulus was measured for the 02000-PH nanocomposite sample, while
the increase was even higher for the non-intercalated D230- and D400-PH
composite samples, especially in the rubbery region at 75-150 % in the increase.
It is important to note that while the 0230 and D400 clays did not form ideal
exfoliated nanocomposites, drastic improvements in storage modulus,
particularly in the higher temperature regime, are to be expected relative to the
02000 clays. This is due to the fact that even though intercalated, and ultimately
exfoliated, the 02000 clays contain approximately 4% mol 2000 molecular weight
Jeffamine molecules along with the 0230 curing agent added to crosslink the
matrix. While the large 02000 diamines compatabilize the intragallery region
with the epoxy resin and permit intercalation, they also crosslink with the epoxy
and create rubbery regions in the glassy matrix. While this likely toughens the
epoxy polymer, especially due to the locale of the 02000 molecules near the
reinforcing platelets, the storage modulus and general stiffness of the materials
will be reduced at higher temperatures. Also, the Tg of the polymer will be
sacrificed with increasing 02000 content for similar reasons.

The increase in storage modulus is in the same range found in previous

related worksf’m‘m9 but it should be mentioned that the absolute modulus values

146

of the pristine epoxy polymer of this work (e.g. ~3 GPa below T9 and ~18 MPa
above T9) are higher than those previously reported.

The glass transition temperature (T9) of the nanocomposites remained nearly
the same compared to the pristine epoxy polymer. The small amount of 02000
introduced to the matrix from the onium ions of the clay galleries acted as a
rubberizer and caused a small decrease in T9, as alluded to earlier, however,
when the nanocomposites formed from the 02000-exchanged clays were
compared to a pristine polymer in which the same amount of 02000 was used as
curing agent (~ 4 °/o mole of the total curing agent) the Tg remained relatively
unchanged.

One interesting and unexpected find that may reveal information about the
crosslinking and mobility of reactants in a gelling epoxy network pertains to the
drop in T9 (5°C) noted between the same nanocomposites synthesized by
different procedures (P1 versus P2). By first allowing the pure epoxy resin to mix
with the 02000 organoclay, followed by addition of the 0230 curing agent, and
even distribution of curing agent throughout the entire system will occur. This is
different than if the epoxy and 0230 are premixed prior to organoclay addition. In
the latter case, an equal distribution of diamine will be found prior to clay
addition, however once clay is added, the intercalation of stiochiometric amounts
of 826 and 0230 results in a Jeffamine - heavy mixture in the regions between
the clay layers. Here uncrosslinked chains can result when an anisotropic
partitioning of curing agent occurs in the samples. An alternative explanation for

the significant Tg drop in the premixed D230 — 826 is that the polymer has begun

147

to react prior to clay addition, and although XRD patterns indicate intercalation
still occurs, the driving force for exfoliation of the layers could be reduced as the
bulk polymerization already has been initiated.

Thermogravimetric analyses of the samples provided additional information
on the structural characteristics of the nanocomposites. From the TGA curves in
Figures 4.10 and 4.11 and the related data in Table 2, it is evident that the
thermal stability of the nanocomposites was not sacrificed by the presence of the
long chain 02000-organic modifier. This verifies that the surface onium ions of
the diamine are crosslinking with the epoxy and being incorporated into the
polymer network, unlike free protonated primary amines that ordinarily undergo
thermal decomposition several hundred degrees before the crosslinked epoxy
matrix. Interestingly, the stability of the composites with the non-intercalated
0230- and D400-exchanged clays was also high, indicating that possibly small
amounts of epoxy molecules were able to crosslink within the clay galleries,

although swelling was not detected in these clays by XRD.

148

 

00
O
A

l

8

 

 

Weight loss (%)
S

 

 

 

 

 

 

Temperature (° C)

Figure 4.10 TGA curves of the organic clays (A) 02000-PGW and (B) 02000-
FH, of the nanocomposites prepared from (C) D2000-PGW and (D) 02000-FH
clays, and of the Pristine-A epoxy polymer (E). The clay loading of the
nanocomposite samples was 6 wt. % silicate basis.

149

 

 

 

 

 

 

 

 

 

100 ' a. (A)
. /
80 - \ {
,3 ‘ (B)
it: 60 - \ (C)
§
.5
g 40 -
‘ (D)
20 -
- (E) / _
O I I I I l I l l
O 200 400 600 800
Temperature (0 C)

Figure 4.11 TGA curves of the organic clays (A) D230-PGW and (B) 0400-
PGW, of the nanocomposites prepared from (C) 0230-PGW and (D) D400-PGW
clays, and of the Pristine-A epoxy polymer (E). The clay loading of the
nanocomposite samples was 6 wt. "/6 silicate basis.

150

4.4 Conclusions

In summary, the use of Jeffamine - modified clays have been successful
in synthesizing glassy epoxy nanocomposites. In particular, the 02000-clays
were the first example of a difunctionalized organic - modified layered silicate
capable of intercalating epoxy resin. Typical diprotonated diamine clays are
limited in their capability due to the affect they can have in electrostatically
pinning neighboring clay layers by bridging the gallery. Partial reaction XRD
analysis revealed the basal expansion that occurred upon epoxy crosslinking.
Final nanocomposites were determined by XRD to be partially exfoliated, and as
a result significant improvements in the mechanical properties due to the
dispersed nanolayers was determined by DMA. Nanocomposites containing 6%
wt. silicate showed marked improvement in storage modulus (35% at 40°C).
Also important to note is that although diamine - modified organoclays thermally
degrade at 250°C, when crosslinked into the epoxy network, the premature
degradation of primary amines is no longer detected and the nanocomposites do
not lose any thermal stability. While typical syntheses of the organoclays used in
this work are problem-free, commercially significant shortcuts can be taken in
producing the epoxy - swellable clay powder. XRD studies indicate that addition
of the neat liquid 02000 to proton — exchanged inorganic clay powders results in
an uptake of the basic diamines into the acidic clay galleries, thus negating the
need for solvent, filtering and washing. The use of primary diamines both as

organic modifiers of clays and as curing agents provide an alternative,

151

commercially attractive way of fabricating epoxy — clay nanocomposites with

improved interfacial and therrno/mechanical properties.

4.5 References

(1) Vaia, R. A.; Giannelis, E. P. MFIS Bull. 2001, 26, 394-401.

(2) Pinnavaia, T. J., Beall, G.W., eds. Polymer - Clay Nanocomposites;
Jonh Wiley & Sons Ltd, New York, 2000.

(3) Vaia, R. A., Krishnamoorti, R., eds. Polymer Nanocomposites;
American Chemical Society: Washington, DC, 2001.

(4) LeBaron, P. C.; Wang, Z.; Pinnavaia, T. J. Appl. Clay Sci. 1999, 15,
1 1-29.

(5) Giannelis, E. P. Adv. Mater. 1996, 8, 29-&.

(6) Okada, A.; Usuki, A. Mater. Sci. Eng. C-Biomimetic Mater. Sens.
Syst. 1995, 3, 109-115.

(7) Lagaly, G. Appl. Clay Sci. 1999, 15, 1-9.

(8) Chin, l. J.; Thum-Albrecht, T.; Kim, H. 0.; Russell, T. P.; Wang, J.
Polymer 2001, 42, 5947-5952.

(9) Hsiue, G. H.; Liu, Y. L.; Liao, H. H. J. Polym. Sci. Pol. Chem. 2001,
39, 986-996.

(10) Kommann, X.; Lindberg, H.; Berglund, L. A. Polymer 2001, 42,

4493-4499.

152

(15)

(15)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

Ahuja, S. K.; McGriff, R. B. Abstr. Pap. Am. Chem. Soc. 2000, 220,
61-COLL.

Lu, J. K.; Ke, Y. C.; Qi, Z. N.; Yi, X. S. J. Polym. Sci. Pt. B-Polym.
Phys. 2001, 39, 115-120.

Curliss, D. B. Abstr. Pap. Am. Chem. Soc. 2000, 219, 43-POLY.
(14) Gam, K. T.; Sue, H. J. Abstr. Pap. Am. Chem. Soc. 2000,
219, 223-PMSE.

Kang, J. H.; Lyu, S. G.; Sur, G. S. Polym.-Korea 2000, 24, 571-577.
Pinnavaia, T. J. Science 1983, 220, 365-371.

Usuki, A.; Kawasumi, M.; Kojima, Y.; Okada, A.; Kurauchi, T.;
Kamigaito, O. J. Mater. Res. 1993, 8, 1174-1178.

Usuki, A.; Kojima, Y.; Kawasumi, M.; Okada, A.; Fukushima, Y.;
Kurauchi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1179-1184.
Kojima, Y.; Usuki, A.; Kawasumi, M.; Okada, A.; Fukushima, Y.;
Kurauchi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1185-1189.
Zerda, A. S.; Lesser, A. J. J. Polym. Sci. Pt. B-Polym. Phys. 2001,
39, 1137-1146.

Ke, Y. C.; Wang, L.; Qi, Z. N. Acta Polym. Sin. 2000, 768-773.
Brown, J. M.; Curliss, 0.; Vaia, R. A. Chem. Mat. 2000, 12, 3376-
3384.

Kommann, X.; Lindberg, H.; Berglund, L. A. Polymer 2001, 42,

1303-1310.

153

(24)

(25)

(25)

(27)

(28)

(30)
(31 )

(32)

(33)

(34)

Ke, Y. 0.; Lu, J. K.; Yi, X. S.; Zhao, J.; Qi, Z. N. J. Appl. Polym. Sci.
2000, 78, 808-815.

Lan, T.; Kaviratna, P. 0.; Pinnavaia, T. J. Chem. Mat. 1995, 7,
2144-2150.

Lan, T.; Pinnavaia, T. J. Chem. Mat. 1994, 6, 2216-2219.

Wang, M. S.; Pinnavaia, T. J. Chem. Mat. 1994, 6, 468-474.
Massam, J.; Pinnavaia, T. J. Material Research Society
Symposium Proceedings 1 998, 520, 223-232.

(29) Messersmith, P. B.; Giannelis, E. P. Chem. Mat. 1994, 6,
1719-1725.

Wang, Z.; Pinnavaia, T. J. Chem. Mat. 1998, 10, 1820-1826.
Lagaly, G. Solid State Ion. 1986, 22, 43-51.

Hackett, E.; Manias, E.; Giannelis, E. P. J. Chem. Phys. 1998, 108,
7410-7415.

Wang, 2., Massam, J., Pinnavaia, T.J. In Polymer - Clay
Nanocomposites; Pinnavaia, T. J., Beall, G.W., eds., Ed.; Jonh
Wiley & Sons Ltd, New York, 2000, pp 127-149.

Menard, K. P. Dynamic Mechanical Analysis: A Practical

Introduction; CRC Press LLC, 1999.

154

Chapter 5
REDUCING THE ORGANIC MODIFIER CONTENT IN EPOXY
NANOCOMPOSITES: MIXED-ION INORGANIC/ORGANIC
MONTMORILLONITE AND FLUOROHECTORITE HOMOSTRUCTURES

5.1 Introduction

The remarkable enhancement in performance properties of engineering
plastic nanocomposites over conventional and non-filled systems provided the
polymer industry with a new perspective of reinforcement in thermoplastic and
thermoset polymers for a variety of practical applications. The rich intercalation
and ion-exchange chemistry of the smectite clay layered silicates enabled them
to be successfully used as polymer - strengthening additives through organo-
functionalization of the gallery regions, diversifying their industrial applications
beyond water-based thickeners, cation-exchangers, catalysts, adsorbents,
rheological modifiers, lubricants, etc. Polymer Layered Silicate Nanocomposites
(PLSNs) exhibit improved mechanical properties (modulus, strength, heat
distortion temperature, thermal expansion coefficient), barrier properties, thermal
stability, resistance to solvent swelling, flammability resistance and ablation
performance compared to unmodified polymers and conventional composites,
mainly due to their unique interfacial properties induced by the highly dispersed,
comparatively diminutive sized silicate nanolayers throughout the polymer

matrix."15

155

The beneficial high in-plane strength and high aspect ratio of the Iamellar
layered materials on the nanocomposite properties can be maximized by the use
of smectite clays‘° (e.g. montmorillonite, hectorite) in part due to their ease of
organic modification. This invokes a highly compatible, hydrophobic intragallery
environment that allows for insertion of polymer precursors, leading to an ideal
situation for optimum nanolayer exfoliation upon polymerization. Additionally,
even though they can be readily synthesized, smectite clays are ubiquitous in
nature and can be purified at relatively low cost.

The pioneering work of Toyota researchers in the early 90’s demonstrated the
first practical application of a nylon - montmorillonite clay nanocomposite. In this
case the modified materials had highly improved mechanical properties, including
a near 80°C increase in heat deformation temperature, and this permitted their
use as a timing belt cover for the automotive industry.”19 In the original Toyota
approach, the hydrophilic inorganic cations on the clay nanolayer surface were
ion-exchanged with a protonated amino acid in order to both match the polarity of
the caprolactam polymer precursor and provide an acidic gallery region to
catalyze the ring opening polymerization. In the next 10 years, the in situ
polymerization of pre-polymers in organoclay galleries has been successfully
applied for the preparation of exfoliated clay nanocomposites of polyimide,
epoxy, acrylonitrile rubber, polypropylene, polystyrene, polyether and
polysiloxane matrices.”30 Epoxy — clay nanocomposites have attracted the

majority of research interest amongst the thermoset polymers while modified

156

845203527314" and layered silicic acids, such as magadiite,35 were

smectite clays
found to be the most appropriate inorganic additives for this system.

Organic alkylammonium ions have been used as effective modifiers of the
inorganic clays acting both as compatibilizing agents for the intercalation of the
epoxy monomers and as acidic catalysis sites for facilitating the epoxy ring-
opening with amine curing agents.”35 In general, the chain-length of the organic
onium ion, the relative acid strength (primary, secondary, or tertiary amines) and
the clay layer charge density, which dictates the concentration of the organic
modifier in the galleries, all play an important role in achieving optimum
nanolayer exfoliation. In a recently reported work,32 it was shown that a hydroxyl-
substituted quaternary ammonium modifier in a montmorillonite clay also induces
a high degree of clay layer exfoliation, behaving in this system much like the
acidic amines in forming hydrogen bonds with the epoxy rings.

While the use of fully exchanged organoclays has been successful in
improving the desired epoxy properties, the goal of this work was to find new
avenues for achieving even better mechanical behavior with more advanced and
cost effective organo-funtionalized clays. The previous chapter focused on the
first diamine - modified clays as successful swelling and ultimately exfoliating
agents in synthesizing a thermoset nanocomposite. The next step, which will be
addressed in this chapter, was eliminating the need for large amounts of organo-
modifiers in the smectite clays. One way of reducing the amount of the costly

organic cations (alkyl amines) is to have the charge balancing cations consist of

mixtures organic and inorganic cations within the same clay galleries. Such

157

mixed exchange cation forms have been termed “homostructured” by the
Pinnavaia group, as each gallery is compositionally equivalent (exhibiting uniform
XRD patterns and intercalation properties). Mixed-ion homostructure synthesis
using alkylammonium ions and Na+ or H+ cations was first described by
Pinnavaia and his co-workers in a series of related patents.”38 The clays used
in the examples of the referenced patents were mainly of montmorillonite and
hectorite-type, while the organic modifiers consisted of various amines, mainly
primary octadecylamine (ODA). It was shown that an onium ion to proton ratio of
65:35 in montmorillonite clay was sufficient to induce similar epoxy resin swelling
properties as with the homonionic organic clay.

As alluded to previously, chapter 4 reports the use of diamine smectite
clay as advantageous fillers in epoxy - clay nanocomposites. Here the primary
diamine played a dual role as organic modifier of the clay and cross-linking agent
in the nanocomposite, thus eliminating the negative effects ordinarily introduced
by monofunctionalized amine modifiers”. The use of diprotonated diamines as

an organic modifier nullifies the “dangling I chains”35

caused by the
monofunctionalized amines via inducing polymeric dead ends as the surfactants
ring-open the epoxy monomer. However, nanolayer exfoliation was
accomplished only when a sufficient long-chain diamine (molecular weight
~2000) was employed as the clay modifier. As the 02000 diamine (Jeffamine,
provided by Huntsman Chemical) was the only difunctionalized organic modifier

to permit epoxy resin intercalation, this was selected as the primary surfactant of

choice in synthesizing clay homostructures. The mixed-ion clays were

158

characterized and tested as fillers in a thermoset glassy epoxy polymer matrix
(Shell epoxy resin 826 cured with Jeffamine 0230). The observed improvement
of mechanical properties (monitored by Dynamic Mechanical Analysis) and the
high thermal stability of the nanocomposites formed from a mixed-ion clay with
an onium ion to proton ratio of ~1:3, along with the clearly shown nanolayer
exfoliation of a 1:1 mixed-ion clay supported the notion that minimization of the
organic modifier content in clays used for nanocomposite synthesis is readily
obtainable. Mixed-ion clays allow for the synthesis of more cost-effective
nanocomposites that have reduced long chain diamine crosslinker content, thus

resulting in higher glass transition temperature materials.

5.2 Experimental

Materials. The epoxide resin used for the formation of the pristine glassy
epoxy polymer and the epoxy — clay nanocomposites was a diglycidyl ether of
bisphenol A (DGEBA), more specifically, Shell EPON 826 with MW ~369 and

epoxy equivalent weight ~182:

159

q

’ 0" i CH,_.. 0

O -' --—\
CH;'CII"CH¢"0"‘* flflgowfia’CH‘CHz'L‘od r\i‘:\,>---('.:""‘l ‘t:->—0 CH3” ”CH CH;

\_
L W C", 4" CH3

n=0@0%)1U0%)

The curing agent was a polyoxypropylene diamine, Jeffamine 0230

(Huntsman Chemicals) with MW ~230:

HZNICHCH 2f OCH 20H} NH 2
X
CH3 CH3
x = 2.6
Jeffamines 02000 (MW ~2000, x=33.1), 0400 (MW ~400, x=5.6) and
0230 (MW ~230, x=2.6) were used in the synthesis of the homoionic and mixed-
ion inorganic/organic clays, as is described below. All other chemicals used in

this work were purchased from Aldrich Chemical Co. and used without further

punficafion.

Synthesis of Homoionic and Mixed-ion Inorganic/Organic Clays. An
industrially purified Na*-montmorillonite clay (PGW, Nanocor Inc.) with CEC =
120 meq/100 g, was used as a precursor to the H” and Li+ forms of the
montmorillonite clay by typical ion-exchange procedures with diluted HCI (for H*-

clays) or LiCI (for Li+-clays) aqueous solutions. A synthetic Li+-fluorohectorite

160

clay (FH, Coming Inc.), with similar CEC to that of the PGW montmorillonite was
used as the parent fluorohectorite clay in this study.

The homoionic organic clays and the mixed-ion inorganic/organic clays
were prepared by ion-exchanging the inorganic clays with diprotonated diamine
(Jeffamine) D2000. Typically, an aqueous inorganic clay suspension was
blended for 1-2 min with an acidic HCI aqueous solution of the Jeffamine, using a
small excess of H+ to ensure the formation of diprotonated diamines. The
blended mixture was further stirred for 48 hrs under ambient conditions with a
magnetic stirrer. For the homoionic organic clays, the amount of 02000 used
was in small excess of the stoichiometric ion-exchange of the inorganic clay,
while for the mixed-ion clays the mole ratio of onium ions to inorganic cations in
the mixture was gradually reduced in order to prepare clays with different
degrees of organic ion-exchange. For comparison purposes, a mixed-ion PGW
clay was synthesized by ion-exchanging the H*-PGW with ODA, as this was the
primary clay of choice in prior epoxy work. The ion-exchanged clays were
separated by centrifugation, washed 3-4 times with deionized water (until free of
Cl', as noted by lack of AgCl precipitate when AgNOa was added to the
supernatant) and washed once with EtOH prior to air-drying at room temperature.
The dried clay products were ground and sieved to a particle size <270 mesh (53
pm). The percent ion-exchange of the inorganic cations by onium ions and the
basal spacing of all the homoionic and mixed-ion clays produced are shown in

Table 5.1.

161

Preparation of Epoxy - Clay Nanocomposites. The pristine glassy epoxy
polymer was formed by mixing the epoxy monomer (EPON 826) with the curing
agent (Jeffamine 0230) at 50°C for ~30 min, outgassing of the liquid mixture at
room temperature and curing in RTV silicone rubber molds at 75°C for 3 hrs and,
subsequently, at 125°C for an additional 3 hrs under nitrogen.

For the preparation of the epoxy - clay nanocomposites the clay was added
to the epoxy EPON 826 - curing agent 0230 mixture at 50°C by stirring for 30
minutes before outgassing and curing at 75°C and 125°C (procedure P- 1), similar
to the pristine polymer formation. Alternatively, the epoxy EPON 826 was initially
mixed with the clay at 50°C for a certain period of time and then the curing agent
0230 was added (procedure P-2). Pre-mixing of the clay with the epoxy
monomer was also done in the presence of low—boiling point organic solvents,
eg. acetone, ethanol (EtOH) and isopropyl alcohol (IPA). The epoxy-clay-solvent
mixture was stirred at 25°C for 30 min, after which the solvent was then
evaporated for 1 hr at 25°C and 30 min at 50°C under stirring and for 1 hr at 75°C
under vacuum. The amount of solvent used was 25 wt. % of the epoxy weight
and it was kept constant throughout the first 30 min of the epoxy - clay reaction.

In the case of the organic clays and mixed-ion inorganic organic clays, the
moles of onium ions present in the clay were counted as contributing to curing
process, supplementary to the amine-groups of Jeffamine 0230, which was the
main curing agent for the formation of the glassy epoxy polymers. The amount of
onium ions due to the organic modifier of the clays was always less than 4 % (by

moles) of the total amine-groups necessary for complete cross-linking of the

162

epoxy monomers. The clay loading of the nanocomposites prepared was 6 wt.

% (based on silicate content, not including the weight contribution of the organic).

Characterization and Testing of Clays and Nanocomposites. X-ray
diffraction (XRD) patterns were obtained on a Rigaku rotaflex 2008
diffractometer equipped with Cu Ka X-ray radiation and a curved crystal graphite
monochromator, operating at 45 KV and 100 mA. The diffraction patterns were
collected between 1 and 12° at a scanning rate of 1°/min. Samples of the liquid
(gel-like) epoxy-clay mixtures or of the uncured epoxy-curing agent-clay mixtures
were prepared by applying a thin film on filter paper mounted on glass slide.
Cured composite samples were prepared by mounting a rectangular flat
specimen into an aluminum holder.

Transmission Electron Microscopy (TEM) images were obtained on a JEOL
JEM-1OOCX II microscope with a CeBs filament and an accelerating voltage of
120 W, a beam diameter of ~5 um and an objective aperture of 20 pm. The
TEM samples were prepared by supporting thin sections (80-100 nm) of the
nanocomposite samples onto 300 mesh nickel grids.

Dynamic Mechanical Analysis tests were performed on a DMA 2980
Dynamic Mechanic Analyzer (TA instruments) in the three-point bending mode at
a frequency of 1 Hz and amplitude of 20 pm, from 25°C to 140°C at a heating
rate of 4°C/min. The pristine epoxy polymer and the nanocomposite specimens

were rectangular bars with dimensions 60 mm x 13 mm x 3 mm.

163

Thermogravimetric analyses (TGA) were performed using a Cahn TG System
121 Analyzer. The powdered clays, the pristine epoxy polymer and the
nanocomposite samples tested were heated to 800°C at a heating rate of
5°C/min under N2 flow.

The composition of the inorganic and the organic clays was determined by

Energy Dispersive Spectroscopy (EDS), CHN and TGA analysis.

5.3 Results and Discussion

Synthesis of Mixed-Ion Inorganic/Organic Diamine Smectite Clay
lntercalates. The basal spacing of homoionic organic clay depends on the chain
length of the organic alkylammonium ions and the charge density of the layered
silicate. Both of these factors contribute to the different configurations of the
organic modifier inside the clay galleries (as outlined in chapter 1).""“"41 When
almost all inorganic cations are ion-exchanged by onium ions (degrees of ion-
exchange 2 80 %) the homogeneity of the organic modifier content in the
galleries depends on the uniform charge distribution on the surface of the
nanolayers, which is higher in the synthetic clays compared to the natural clays.
Having a very uniform charge distribution allows for homogenous dispersion of
the cations, giving rise to a very consistent interclay spacing and invariant
swelling behavior from tactoid to tactoid. However, even with synthetic clays the
synthesis of a mixed-ion clay could result in inhomogeneous distribution of the

organic cations if the ion exchange is not carried out with some care. Pinnavaia

164

at 3136’” developed a method of synthesizing high quality mixed-ion
homostructured clays by utilizing water/alcohol mixtures for the ion-exchange
bath along with prolonged ion-exchange times. In this way phase-segregated
structures that consisted of separate inorganic-rich and organic-rich regions
(readily notable by XRD) were minimized and a more homogeneous ion
exchange within all the galleries could be achieved. In this case, however, the
tendency for large organic surfactants to aggregate into micelles and undergo ion
exchange in clusters is lessoned by the higher solubility of polypropylene oxide —
based surfactants in water relative to a pure alkyl amine.

The target of the present study was to synthesize mixed-ion inorganic/organic
smectite clays with a very low onium ion content (onium ion to proton (O*/H*)
ratio ~ 1:3), which would still show a homogeneous distribution of the organic
modifier between the nanolayers and permit the intercalation of epoxy resin. The
use of a relatively long-chain, primary alpha-omega diamine (Jeffamine 02000)
allowed the synthesis of true homostructures with a single 001 reflection. When
alkylammonium ions with a 012-018 carbon chain were stiochiometrically ion
exchanged for the inorganic cations in montmorillonite clays with a CEO ~100-
120 meq/100 9, they adopted a lateral bilayer (~18 A) or an inclined paraffin-like
structure (~20-23 A). The mixed-ion homostructured clays, prepared by partial
ion exchange with C18A onium ions, had basal spacings of 18.1, 17.4 and 14.2 A
for O*/H+ ratios of 65:35, 50:50 and 35:65, respectively.38 However, when the

diprotonated diamine 02000 was used for the ion exchange of the inorganic

165

clays, the basal spacing of the mixed-ion clay with O“/H+ ratio 1:3 was 16.9 A, as

it can be seen from the data in Table 5.1.

166

Table 5.1 Composition and Basal Spacings of the Homoionic and Mixed-

ion Inorganic/Organic clays

 

 

. . . a .
Inorganic/organic _ parent. organic modifier basal spacrngs
clays inorganic
clay

Onium diprotonated dam dooz

ion (%) diamine (A) (A)
H-PGW Na-PGW - - 13.4 -
Li-PGW Na-PGW - - 12.4 -
Li-FH Li-FH - - 12.3 -
H/0200025-PGW H-PGW 25 02000 1 6.9 -
H/0200035-PGW H-PGW 35 02000 17.8 -
Li/0200035-PGW Li-PGW 35 02000 1 7.6 -
H/0200050-PGW H-PGW 50 02000 43.7 18.8 b
02000-PGW Na-PGW 80 02000 45.5 21 .8
Li/0200065-FH Li-FH 65 02000 45.3 21 .3
02000-FH Li-FH 80 02000 46.0 23.2
H/C18Ao,65-PGW H-PGW 65 C18A 17.5 -

 

a The inorganic clays were ion-exchanged with diprotomated diamine 02000
(Jeffamine, Huntsman Chemical); the Na” content of all the ion-exchanged clays
was below 5 %

° The d-spacing of that broad peak (~19 A) is related to both the second order
reflection and to a second phase of the mixed-ion clays as is discussed in the

text

167

Furthermore, from the X-ray patterns of the D2000-exchanged clays in Figure
5.1, it can be seen that there is no indication of an inorganic-rich phase (d-

spacing ~12-14 A) in all the mixed-ion clays.

168

 

D2000-FH

 

43.7 15.1 A

 

Intensity (arb. units)

 

 

 

 

 

 

 

26 (degrees)

Figure 5.1 X-ray diffraction patterns of the homoionic and mixed-ion
inorganic/organic montmorillonite (PGW) and fluorohectorite (FH) clays prepared

by ion exchange of the inorganic cations Na”, Li+ or H+ with onium ions of
diprotonated 02000.

169

The basal spacing for both the homoionic 02000-PGW and 02000-FH
organic clays was ~46 A, and multiple orders of reflection could be seen in these
XRD patterns, although the higher aspect ratio of the fluorohectorite gives more
intense peaks and more distinct higher orders of reflection. The configuration of
the long-chain 02000 diamine inside the clay galleries is an inclined paraffin-like
structure, as expected for such a large onium modifier. Interestingly, the mixed-
ion Li/020000,55-FH clay with O“/Li+ ratio of 65:35 had a very similar d-spacing to
the homoionic organic clays, indicating that the end - to - end separation of the
diamine modifier is near a maximum value in both instances. The mixed-ion clay
H/020000,5-PGW with 07H” ratio of 1:1 showed a basal spacing of ~44 A,
indicating a more parallel surfactant orientation to the clay surface. The second
order reflection was relatively broad and had a d-spacing in the ~19 A range.
This region could represent both the 002 reflection and a monolayer diamine
arrangement that occurs at roughly the same spacing. More important was the
lack of higher angle 12-14 A 001 distances, which represent unexchanged
inorganic portions of the clay sample. At lower degrees of ion exchange, e.g.,
O“/H+ ratios of ~3:7, the d-spacing of the mixed-ion clays was ~17-18 A. Even at
this low organic content a uniform lateral monolayer structure was observed
without the inorganic clay peak. The configuration of 02000-diamine inside the
galleries of the mixed-ion clays could be described by mixtures of lateral bilayer
and folded-monolayer or inclined-folded Iayer-to-layer structures, depending on
the concentration of the diamine in the galleries. The “folded” or “coiled”

structures are unique to polypropylene — based surfactants as such a backbone

170

permits more random orientation due to the flexible, easy to rotate nature of the
carbon - oxygen bond. Lower concentrations (e.g., for onium ions/inorganic
cations s 35:65) could be represented better with a lateral bilayer structure while
at higher loadings of organic modifier (e.g., for onium ions/inorganic cations 2
65:35) the inclined-paraffin configuration of the diamine molecules would be
preferred. For comparison purposes, a mixed-ion H/C18Ao,65-PGW was
synthesized with O“/H+ ratio of 65:35; this clay showed a basal spacing of ~17.5
A. As expected, the smaller onium exchanged in this case adopts a monolayer
arrangement even at relatively high loadings. Table 5.1 contains basal spacing
and compositional data for all the homoionic and mixed-ion montmorillonite

(PGW) and fluorohectorite (FH) clays synthesized in this work.

Epoxy-Clay Nanocomposite Formation from Mixed-Ion Inorganic/Organic
Clays. Exfoliation of the clay nanolayers in a polymer matrix first requires the
polarity of the interclay region to match that of the polymer precursors, and
subsequently that the intragallery polymerization rate is equal to, or better yet
greater than, that of the bulk polymer. The chemical modification of the clays
with organic modifiers, mainly acidic primary onium ions, provides with the
appropriate hydrophobic intragallery environment for initial intercalation of the
epoxy resin, and their acidic nature aids in catalytically ring opening the epoxy
molecules intercalated into the galleries. The above mechanism can lead to
more improved nanocomposites through using diamines, in particular, as the

organic modifiers as they eliminate the undesirable effect of the free, dangling

171

chains invoked on the epoxy by mono-amines that can not be crosslinked into

the polymer matrix.2°'3""39

However, large loadings of the relatively high
molecular weight diamine could induce rubberizing effects to the rigid, glassy
epoxy polymer matrix and have similar effect to that of the dangling chains of the
mono-amines. It would thus be desirable both from the mechanical point of view
and for economical considerations to minimize the organic modifiers of the clay
additives, yet still achieve high degrees of nanolayer intercalation/exfoliation and
improved properties of the epoxy - clay nanocomposites.

The clay nanolayer intercalation and ultimate exfoliation upon mixing the
epoxy monomer (EPON 826) with the Li/020000,65-FH mixed-ion clay, and after

final curing of the nanocomposite is shown in the stacked X-ray patterns of

Figure 5.2.

172

 

 

 

 

(C) 59 60 61 62 63 64

“‘ AA-‘AAAL- ‘-
—v —'v w—‘vvwv

75.5 A

 

Intensity (arb. units)

 

 

 

 

26 (degrees)

Figure 5.2 X-ray diffraction patterns of mixtures of the mixed-ion clay
Li/D2000665-FH, the epoxy monomer EPON 826 and the curing agent 0230: (A)
after intercalation of EPON 826 in Li/020000_55-FH at 50°C for 12 hrs, (B) after
addition of 0230 to the mixture of step A and outgassing at 25°C, and (C) after
curing at 75°C, 3 hrs and 125°C, 3hrs (the clay loading was 6 wt. % silicate
basis). The inset shows the presence of an in-plane 060 clay reflection in the
cured nanocomposite

173

Pre-mixing of the epoxy monomers with Li/02000665-FH mixed-ion clay
(Oi/Li+ ~65:35) at 50°C resulted in a disordered intercalation of the nanolayers
with a basal spacing reaching ~75 A (pattern A). Subsequent addition of the
diamine curing agent at the same temperature induced rapid intragallery
polymerization and expansion of the nanolayers after a short period of time
(pattern B). Upon completion of the curing process at higher temperatures, the
nanolayers remained highly separated resulting in the absence of Bragg peaks in
the small-angle region (pattern C), similar to case of the homoionic 02000-PH
organic clay.39 The mixed-ion clay H/C18Ao_55-PGW, which was synthesized by
ion-exchanging the protons with C18A-onium ions instead of the diprotonated
diamine 02000, resulted in a highly intercalated/exfoliated nanocomposite
sample, similar to the Li/02000055-FH mixed-ion clay.

The formation of a high quality thermoset epoxy - clay nanocomposite
material with homogenous dispersion of the individual nanolayers throughout the
polymer matrix is not only dependent on the catalytic acidity of the organoclay,
but on the fabrication conditions as well.32 The sequence of mixing the
components of the thermoset epoxy — clay nanocomposite system (clay/diamine
curing agent/epoxy monomers), the use of solvents to reduce the viscosity of the
system, the temperature and time of curing, and finally the fabrication and
molding techniques could have a significant effect on the degree of nanolayer
dispersion. Unlike the rubbery epoxy systems, epoxy monomer pre-intercalation
inside the clay galleries before addition of the curing agent typically results in

higher degrees of nanolayer exfoliation. This is likely more pronounced in the

174

glassy epoxy polymer synthesis due to the smaller, more nucleophilic 0230
curing agent that can prematurely react with the epoxy resin prior to intercalation,
as covered in chapter 4. However, in order to avoid gelling of the interclay
regions, care should be taken that pre-mixing of the diamine-modified clay with
the epoxy monomers does not take place at relatively high temperatures (higher
than 50°C) and for a very prolonged period (more than 24 hrs). Heating the
epoxy - clay mixture to 50°C is necessary for keeping the viscosity relatively low
and boosting the diffusion of the epoxy monomers inside the galleries. An
alternative way to reduce the viscosity, and thus increase the diffusion of the
epoxy, would be the implication of low-boiling point solvents, such as acetone,
ethanol and isopropyl alcohol to the epoxy — clay mixture.

In the case of the nanocomposite samples prepared by mixed-ion clays
with lower onium ions to protons ratios (eg. 35:65 and 25:75) a broad, low
intensity peak with d-spacing ~17-18 A was present in the XRD patterns,
indicating different degrees of limited nanolayer intercalation, in contrast to the
relatively narrow peak with d-spacing ~13.7 A of the nanocomposite sample
formed by the inorganic proton clay (PGW) indicating no intercalation. The
qualitative information obtained from observing the rheological and optical
properties of the clay - epoxy mixture can be a powerful tool in predicting the
nature of the clay in the polymer medium. For instance, the inorganic clay, no
matter the sieved particle size, will not invoke as large a viscosity change when
added to the resin, and given a short amount of time will settle to the bottom of

the resin. An immediate large viscosity increase and unsettling nature of the

175

~17-18 A - spaced low organo-content clays preliminarily indicates that these
materials will show superior mechanical properties relative to the pristine and
inorganic clay-filled epoxy systems.

TEM images of the nanocomposite samples prepared from the mixed-
ion Li/D20000,65-FH fluorohectorite and H/D200065-PGW montmorillonite clays
are shown in Figures 5.3a,b and 5.4c,d, respectively. The clay loading in both

samples was 6 wt. % (silicate basis).

176

 

Figure 5.3 TEM images of thin sections of the glassy epoxy nanocomposites
prepared from the mixed-ion clays (a,b) Li/D2000055-FH. The clay loading of the
nanocomposite samples was 6 wt. % silicate basis.

177

 

Figure 5.4 TEM images of thin sections of the glassy epoxy nanocomposites
prepared from the mixed-ion clays (a,b) H/D200065-PGW. The clay loading of
the nanocomposite samples was 6 wt. % silicate basis.

178

The Li/D20000,65-FH nanocomposite consisted mainly of clay tactoids with
disordered, highly intercalated layers separated by 5-20 nm. Individual, shorter
layers could also be found on the edges of the primary dispersed particles and
most frequently in domains with relatively low number of layers. The high aspect
ratio synthetic clay particles proved difficult to exfoliate fully into the polymer
matrix, although the observed expansion of clay layers adequately exposes the
clay surface to the polymer matrix, and no regions of tightly bound, non-
intercalated tactoids were evident. Similar images were noted for the homoionic
02000-PH organic clay, as addressed in chapter 4. The importance of equally
exfoliated clay layers in both homoionic organoclay and partially exchanged
mixed-ion clay cannot be understated, as this is the first example of a reduced-
organic system that competes with a pure organic layered silicate.

A more disordered intercalated or partially exfoliated state was found in the
H/020000,5-PGW nanocomposite, where most of the nanolayers were separated
by more than 10 nm space, still generally keeping parallel orientations within
larger domains. The more expanded layer - to - layer separation noted in the
natural clay is as expected given the much reduced aspect ratio, resulting in

easier migration within the crosslinking polymer.

Properties of the Nanocomposites. Dynamic Mechanical Analysis
(DMA) tests were applied in order to investigate the effect of the homoionic and
the mixed-ion clay nanolayer reinforcement on the mechanical properties of the

glassy epoxy polymer. The DMA technique provides two different moduli, an

179

elastic modulus (E) which is called storage modulus and is related to the ability
of the material to retum or store energy when an oscillatory force is applied to the
polymer specimen and an imaginary (loss) modulus (E’) which relates to the
ability of the material to lose energy. The ratio E”/E’ is called tan delta (tan6) and
typically reaches a maximum at the glass transition (T9) of the polymer. A more
in-depth explanation of the DMA theory and mechanics is provided in chapter 4.
The DMA tests of this study were conducted in a three-point bending (flexural)
mode while other modes, like the torsional mode, are also commonly used.32
The curves in figure 5.5a and 5.5b show the dependency of the storage modulus
and the tan6 (maximum value at the glass transition temperature) for
representative nanocomposite samples prepared by homoionic and mixed-ion

clays, at a clay loading of 6 wt.% (silicate basis).

180

 

 

 

 

 

Storage Moduhrs (MPa)

A

 

 

 

 

 

120

 

Tand

 

 

 

 

20 40 60 80 100 1 20
Temperature (°C)

Figure 5.5 Dynamic Mechanical Analysis measurements: (a) Storage Modulus

vs. temperature and (b) tan6 vs. temperature, for (A) Pristine-A epoxy polymer

and for the nanocomposite samples prepared from (B) H-PGW, (C) H/02000025-

181

PGW, (D) Li/D20000_65-FH, (E) 02000-PGW. The clay loading of the
nanocomposite samples was 6 wt. % silicate basis .

The data in Table 2 refers to the storage modulus values of the pristine polymer
and the nanocomposite samples at 40°C (glassy region, below T9) and at 120°C

(rubbery region, above T9), and to the Tg values of the samples.

182

Table 5.2 Dynamic Mechanical Analysis and Thermal Analysis data of
the pristine epoxy polymer and the clay-polymer

nanocomposite samples "

 

Samples storage modulus (E’) Tg thermal
(tan6) stability (TGA
data) d

at 40°C At 110°C
(GPa),:r: (MPa),:r: (°C), :1:

 

10% 10% 2 (°C)
Pristine epoxy-A 2.9 13-4 82.6 373
Pristine epoxy-B b 3.0 20-1 75.5 361
H-PGW 2.9 22.1 82.8 354
Li-PGW 2.7 19.9 82.3 356
Li-FH 3.1 25.2 85.2 351
l-l/0200025-PGW 3.7 26.0 80.2 342
H/0200035-PGW 8.8 26.8 79.0 339
H/D200035- P2 c 3.6 38.3 85.1 342
Li/0200035-PGW 3.5 24.1 81.7 342
H/D200050-PGW 3.3 22.7 78.0 339
H/0200050-PGW - P2 3.2 28.0 82.6 341
H/D200050-PGW - 35.2
P2/acetone 3'6 83'1 339
H/0200050-PGW - 32.4
P2/EtOH 3-4 83.1 330
H/D200050-PGW — 31.6
P2/IPA 3-6 80.4 347
02000-PGW 3.8 21 -5 69.8 336
02000-PGW — P2 3.6 24.0 75.1 339
Li/0200065-FH 4.0 27.6 78.0 328
02000-PH 3.9 30.5 73.7 323
H/C18Ao,65-PGW-P2 3.5 34.2 84.4 340

 

183

a The nanocomposite samples were named from the inorganic/organic clay
used; the clay loading of the samples was 6 wt.% silicate basis

° Substitution of 4 % mole 02000 (organic modifier of clays) for 0230 (curing
agent) in the pristine epoxy polymer

° P2 refers to the nanocomposite samples prepared be pre-mixing of epoxy
monomers with the clays; all the rest of the samples in Table 2 were prepared by
direct addition of the clays in the epoxy - curing agent mixture (P1)

d Temperature at which the weight loss in TGA was 1.5 %

The storage modulus and the T9 of the nanocomposite sample formed by the
inorganic H-PGW montmorillonite clay were similar to those of the pristine epoxy
polymer. The inability of the inorganic clay to swell with epoxy resin created two
distinct regions in the epoxy material, with the majority of the clay settling to the
bottom of the specimen by the time crosslinking was completed. Despite poor
mixing with the epoxy, an interesting finding was that there was no damage by
oxidation or premature thermal decomposition as would be expected due to the
presence of protons on the clay. While those protons trapped in the interior,
unwetted clay galleries should not contribute to the acid-catalyzed thermal
decomposition, the exterior protons ordinarily would be expected to affect the
thermal decomposition as is noted in the TGA of proton — containing
organoclays. The important difference in the inorganic clay -epoxy case is that a

large amount of neutral amine (Jeffamine curing agent) is added to the mixture,

184

and these basic molecules can sequester the free surface protons, and eliminate
their detrimental affect on the premature thermal degradation of the epoxy
network. This is a significant find in that if methods are developed to pre-
exfoliate inorganic clay layers through disruption of the preferred face - to — face
stacking, then inexpensive proton clays could be employed as the reinforcement
phase.

The nanocomposite sample formed by the homoionic 02000-PGW organic
clay showed a ~25 % increase in storage modulus in both the glassy and rubbery
region, while a 35% and 65 % increase, respectively, was found for the 02000-
FH organic clay, indicating the improved reinforcing properties of the high-aspect
ratio fluorohectorite nanolayers compared to the montmorillonite nanolayers.
Direct comparisons ofthe 0230 - Epon 826 pristine epoxy (no normalizing
addition of 02000 present) to the nanocomposites indicates that there is a
sacrifice in the glass transition temperature in the final hybrid systems. This is
not an accurate comparison, however, as the small amount of 02000 present on
the organoclay in the nanocomposites contributes to the crosslinking of the
epoxy, and a more direct comparison of the normalized pristine epoxy (including
the same ~4 % mole D2000 content) reveals nearly no change in T9.

Not surprisingly, a similar or even higher increase in storage modulus without
any significant changes in T9 could be achieved with the mixed-ion clays, even
for the nanocomposite sample prepared from a mixed-ion clay with a OVH" ratio
of 1:3 (fig. 5.2 and table 5.2). This is due to the reduction of high molecular

weight D2000 content in the final nanocomposite, while still permitting

185

intercalation of epoxy and partial exfoliation of the reduced-organic clays.
Specifically, the nanocomposite sample formed from the H/D20000_25-PGW clay
showed a 30% and 40% increase of storage modulus in the glassy and rubbery
region respectively, the sample prepared from H/D20000,5o-PGW exhibited a 25%
and 70% increase, the sample formed from H/ODA0,65-PGW showed 20% and
85% increase, and the sample formed from Li/02000055-FH showed 40% and
50% increase. The T9 for all these samples ranged from ~78 - 84°C compared
to 83°C of the pristine glassy epoxy polymer. The increase in storage modulus is
in the same range found in previous studies with homoionic organic clays,2°'32'34
but it should be mentioned that the absolute modulus values of the pristine epoxy
polymer of this work (e.g. ~3 GPa below T9 and ~18 MPa above T9) are higher
than those previously reported.

As alluded to earlier, therrnogravimetric analyses of the mixed-ion clays
showed that when the protons were present in the sample, the decomposition of
the organic modifier, diamine 02000, occurred at lower temperatures compared
to an homoionic organic clay or to a mixed-ion clay in which the inorganic cation
was Li+ instead of H” (curves A and B in Fig. 5.6). However, the above proton —
catalyzed decomposition was not transferred to the glassy epoxy — clay
nanocomposite samples, as the thermal stability of the samples formed by the

H*/02000 mixed-ion clays was uninfluenced despite the presence of protons as

noted in the TGA curves shown in figure 5.6 and the data in table 5.2.

186

 

 

 

 

 

 

 

Weight loss (%)

 

 

 

 

0 200 400 600 800 1000
T enperature (° C)

Figure 5.6 TGA curves of the mixed-ion clays (A) H/020000_35-PGW and (B)
Li/02000035-PGW, the nanocomposites prepared from (C) H/020000,25-PGW and
(D) 02000-PGW clays, and of the Pristine-A epoxy polymer (E). The clay

loading of the nanocomposite samples was 6 wt. % silicate basis.

187

The high thermal stability of the nanocomposites prepared from both the
homoionic and the mixed-ion diamine-modified clays suggested that the surface
onium ions of the diamine were incorporated into the polymer network, resulting
in nanocomposites with improved interfacial and physical properties. This is
apparent as the onset of organic degradation was higher in the nanocomposites
containing protonated amine — modified clay relative to the protonated amine clay

use".

5.4 Conclusions

The thermoset glassy epoxy — clay nanocomposites prepared from
reduced organic content, mixed-ion smectite clays in this work have been shown
to be successful in synthesizing exfoliated systems. Most significantly, the ability
of the inorganic — containing homostructured clay to swell with epoxy resin, just
as is the case with homoionic organoclay, represents the first example of a more
cost effective, non-fully organic clay suitable for nanocomposite synthesis.
Additionally, the presence of protonated diamines, while lessoned in content,
have sufficient acidic catalytic nature to bias the intragallery polymerization and
ultimately force apart the clay layers. Impressive mechanical improvements of
the nanocomposites were noted, as with the homoionic diamine — modified clays
covered in chapter 4, however, by reducing the content of large molecular weight
modifiers, the reduction in glass transition temperature and the overall
rubberizing affect of the 02000 curing agents was greatly diminished. The

mixed-ion smectite clays, in which the organic modifier was reduced by nearly

188

 

70%, could be used in various thermoset or thermoplastic polymer matrices for a

more economical and effective means of fabricating polymer - clay

nanocomposites.

5.5

References

(1) Vaia, R. A.; Giannelis, E. P. MRS Bull. 2001, 26, 394-401.

(2) Pinnavaia, T. J., Beall, G.W., eds. Polymer - Clay Nanocomposites;
Jonh Wiley & Sons Ltd, New York, 2000.

(3) LeBaron, P. C.; Wang, Z.; Pinnavaia, T. J. Appl. Clay Sci. 1999, 15,
1 1-29.

(4) Lagaly, G. Appl. Clay Sci. 1999, 15, 1-9.

(5) Krishnamoorti, R.; Vaia, R. A.; Giannelis, E. P. Chem. Mat. 1996, 8,
1728-1734.

(6) Giannelis, E. P. Adv. Mater. 1996, 8, 29-&.

(7) Okada, A.; Usuki, A. Mater. Sci. Eng. C-Biomimetic Mater. Sens.
Syst. 1995, 3, 109-115.

(8) Chin, l. J.; Thum-Albrecht, T.; Kim, H. 0.; Russell, T. P.; Wang, J.
Polymer 2001, 42, 5947-5952.

(9) Hsiue, G. H.; Liu, Y. L.;‘Liao, H. H. J. Polym. Sci. Pol. Chem. 2001,

39, 986-996.

189

 

(10) Kommann, X.; Lindberg, H.; Berglund, L. A. Polymer 2001, 42,
4493-4499.

(11) Ahuja, S. K.; McGriff, R. B. Abstr. Pap. Am. Chem. Soc. 2000, 220,
61-COLL.

(12) Lu, J. K.; Ke, Y. C.; Oi, Z. N.; Yi, X. S. J. Polym. Sci. Pt. B-Polym.
Phys. 2001, 39, 115-120.

(13) Curliss, D. B. Abstr. Pap. Am. Chem. Soc. 2000, 219, 43-POLY.
(14) Gam, K. T.; Sue, H. J. Abstr. Pap. Am. Chem. Soc. 2000, 219, 223-
PMSE.

(15) Kang, J. H.; Lyu, S. G.; Sur, G. S. Polym.-Korea 2000,24, 571-577.
(16) Pinnavaia, T. J. Science 1983, 220, 365-371.

(17) Usuki, A.; Kawasumi, M.; Kojima, Y.; Okada, A.; Kurauchi, T.;
Kamigaito, O. J. Mater. F195. 1993, 8, 1174-1178.

(18) Usuki, A.; Kojima, Y.; Kawasumi, M.; Okada, A.; Fukushima, Y.;
Kurauchi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1179-1184.

(19) Kojima, Y.; Usuki, A.; Kawasumi, M.; Okada, A.; Fukushima, Y.;
Kurauchi, T.; Kamigaito, O. J. Mater. Res. 1993, 8, 1185-1189.

(20) Lan, T.; Kaviratna, P. 0.; Pinnavaia, T. J. Chem. Mat. 1995, 7,
2144-2150.

(21) Usuki, A.; Kato, M.; Okada, A.; Kurauchi, T. J. Appl. Polym. Sci.
1997, 63, 137-139.

(22) Vaia, R. A.; Jandt, K. 0.; Kramer, E. J.; Giannelis, E. P. Chem. Mat.

1996, 8, 2628-2635.

190

(23) Burnside, S. D.; Giannelis, E. P. Chem. Mat. 1995, 7, 1597-1600.
(24) Lan, T.; Kaviratna, P. D.; Pinnavaia, T. J. Chem. Mat. 1994, 6, 573-
575.

(25) Messersmith, P. B.; Giannelis, E. P. Chem. Mat. 1994, 6, 1719-
1725.

(26) Lan, T.; Pinnavaia, T. J. Chem. Mat. 1994, 6, 2216-2219.

(27) Wang, M. S.; Pinnavaia, T. J. Chem. Mat. 1994, 6, 468-474.

(28) Kojima, Y.; Fukumori, K.; Usuki, A.; Okada, A.; Kurauchi, T. J.
Mater. Sci. Lett. 1993, 12, 889-890.

(29) Yano, K.; Usuki, A.; Okada, A.; Kurauchi, T.; Kamigaito, O. J.
Polym. Sci. Pol. Chem. 1993, 31, 2493-2498.

(30) Fu, X.; Qutubuddin, S. Polymer2001, 42, 807-813.

(31) Zerda, A. S.; Lesser, A. J. J. Polym. Sci. Pt. B-Polym. Phys. 2001,
39, 1137-1146.

(32) Brown, J. M.; Curliss, 0.; Vaia, R. A. Chem. Mat. 2000, 12, 3376-
3384.

(33) Ke, Y. 0.; Lu, J. K.; Yi, X. S.; Zhao, J.; Qi, Z. N. J. Appl. Polym. Sci.
2000, 78, 808-815.

(34) Massam, J.; Pinnavaia, T. J. -Materiial Research Society
Symposium Proceedings 1998, 520, 223-232.

(35) Wang, Z.; Pinnavaia, T. J. Chem. Mat. 1998, 10, 1820-1826.

(36) Pinnavaia, T. J.; Shi, H.-z.; Lan, T. In US; (Board of Trustees

Operating Michigan State University, USA). Us, 1999, p 24 pp.

191

(37) Pinnavaia, T. J.; Shi, H.-Z.; Lan, T. In PCT Int. Appl.; (Michigan
State University, USA). Wo, 2000, p 102 pp.

(38) Pinnavaia, T. J.; Wang, Z. In US; (Board of Trustees Operating
Michigan State University, USA). Us, 2001, p 25 pp.

(39) Triantafillidis, C. S.; Sislo, M. R.; LeBaron, P. C.; Pinnavaia, T. J.
Abstracts of Papers, 222nd ACS National Meeting, Chicago, IL, United
States, August 26-30, 20012001, lEC-109.

(40) Lagaly, G. Solid State Ion. 1986, 22, 43-51.

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

7410-7415.

192