.-.! fiuuisqfldflmfisfi a... _ ...- .

... ...S .. ... - .5... ...... ... Eu... . I - -.

o. ...... ...... ..unnpflmflb...‘ .. .. . Bibi...” .-. . ... 17.1.... ..u....~......._o...n...§mk§........ u. £2.45.fi.x.u.fi

.@$ “.2, ,- {a in - . n. . . . kfivdrmarflndfmwaa . u ".Jfi.‘ 3w.

... .. .. . .33.... 24M. .

I :b p n 1... r 4 .

fiwsé .... 3..-... .

....«m. 4... n... .

. . . L. . i W.
.... ...“... .05.». Wyn-“H
... .... «......Ar ... ..us
.. ......w.,.... .

..
L .H- a, ...4....

.0
“(.4 v..
.

...... ......nNH..u~.. . . .
. 2h...
. .

. ... . ...... ....u...... . I. .
. ... . .ufihmn. ruvrMMvfim.
. 1‘ “.01... u .‘OCDV.’>\'I

. cu! NM... lfldrfflflfafl.
IA! 0 n o M A l .

“.....Vfiuwfiww...
.r ..-. ....... .
tofllnnflnl ” I.
... ,. .4 \. “Willihfilvllll ‘ I (\lvv ¢H'.‘Irl.. ,tvfilav-‘l
-l.. .IWI nu - a v ‘rbr.lvlev.t.|0vN\Ol9V0ll .IidF. It: 1.“?th Ix...
. . . 18.. . , l..u.-.... .. ..Iipn. ...... «L5H...
. ..- -.l . . .... . . up... In. . . . ...» .(
... ... . : ..U...«.I..|.U......o . . . . . . .. . . . v I! 14$
.. . Irk—'l4.vli V . . . . n .. .. nAcAl‘I.’ .lpolostllfl1 I. I
.. I .I. .u .A‘-.. v0 .... u v A . . .s . cit III In ||. r l Gar
.. .. . v. . . .. . . ...-.1 .u .. . ...: .... . ..0‘{.a}a4fifio..vol or...» k."
. . .. ...... .. ... - . . . . I ... .. ...n. ... .-urr 3%.- 4.0!... . . HRH-Md.»
. . .. . . .. . . ... .. ... .... . ...... vplu..... .
I IV .I ..v I .lv. .V 7.. V . I .4 v 1 >1! . IOOQII Ll.
. I..- . 03"“. A” . u, u. 3 up; ... . ... . ... I u ... u v; u: . .
.. ...I..- Jr... Inulk. .....HXIf ho .. . . Z .- . . . . . . .
. . w -.....1. ".4 financiz. \w.....%..1::... .. . . . . . . . . . . .
. . .. I .....qu ..........1.h.....v.o. 4.... ..."...5: . -
, . \I .0007. -..... 1 .... ..v n ,
......Ilhu. .... ..LI y vb ...
. t I ......»3... 30.0.3...
.. . “mitt... ... « ...,uunflflvuh...hlit -
9| HIW‘
. . all... n v

(0
III

-.. :0. v

n --.- .

......“ .-.
. . bu.......v..l...u.. fiwfll.
- n..t.h[!.ihv.l.o
. 4 .. . 3 ...: 1-..»..1..... v...
. . ..- . . .
I. u n v. .
. . 3 . .
u I A. . A II
y.‘ .0 . A l V I vl V I. I: 0
....I 10.51.. “145,13 .nduovoufinl. lava-v.1. “.5 y
I I‘fl‘. I. .7. ~""II~1C t5!“§~'7¢l§'.l."‘.'|qh f.
. . ....v...d..........l.v.h.......t....d.-€..4.. .. 5.-....
IV. ''''''''''' It‘ll“ J“..ll|v.fll".""‘.“w
3|.-. ”.....flynflt.‘l\.' N... PO“:
. ... I..-
\Iylulwowhnflnhh :l... u 3
V. {IIIQH I I. I put-\vlv--.
(61.5....{W

1......va
v .4 . ...‘0

quv
l
x
#1: g.
WW "

'Y

. v .lti 4.

to.|..o§‘k‘bllv .
- 1.1.7- p} 1.1. '5

’1" ul..|.|vbloup.l
. . ...V

‘ .l'"\l

ll'lulll
t, 4' t‘t' I
. 4.9."? . 4......
....II.... ......12... .. Illa... l...
.....-.. ...... (trial... rlfitiill:
{1.1%.}. dung!
t. ....Iunnuuuflhl...“
.1 ti...

 

.. ,. .... £31--..041,|'rhb§0.
. . ..- .........l.. ...

7—}

 

‘ ...... llllllljlljll .111... “

3 1293 O1

 

 

This is to certify that the

dissertation entitled

The Rheology and Morphology of Reactively
Compatibilized Polymer Blends

presented by

Himanshu Asthana

has been accepted towards fulfillment
of the requirements for

 

Ph . D . degree in Chemical Engineering

 

   

 

 

”91”»C <iytlfirprofessor '
Date ’aI/I X [/6 7 "'l/

MSU is an Affirmuliw Anion/Equal Opportunity Institution

 

0-12771

‘_

‘~

~ V—w V‘- ‘, —‘ —v-_—7-,

‘p.— —‘ ‘_.—~—v—_w_

vW.r‘—-'—‘ van—Wm

 

LIBRARY

Michigan State
Unlvorslty

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINE return on or before date due.

DATE DUE um. DUE. DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1]” W14

THE RHEOLOGY AND MORPHOLOGY OF REACTIVELY
COMPATIBILIZED POLYMER BLENDS

BW

Himanshu Asthana

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Chemical Engineering

(1997)

ABSTRACT
THE RHEOLOGY AND MORPHOLOGY OF REACTIVELY
COMPATIBILIZED POLYMER BLENDS
By

Himanshu Asthana

The study was conducted to investigate the effect of progressive extents Of interfacial
reaction on the rheology and morphology of reactively compatibilized polymer blends.
Although the interfacial reaction is known to produce a finer morphology, its effect on
interfacial tension has not been quantified in earlier studies. The values of equilibrium
interfacial tension in blends with progressive extent of reaction were determined in the
present study from rheological measurements on blends of maleated polypropylene and
nylon 6. Nylon 6 was melt blended in a twin screw extruder with one of several grades of
maleated polypropylene which contained progressively increasing extent of maleation to
study reactive compatibilization via direct reaction. The results show that the equilibrium
interfacial tension falls due to interfacial reaction. However, the corresponding
morphological Observations indicate that a minimal extent of interfacial reaction (and
hence a small reduction in equilibrium interfacial tension) is sufficient to bring about a
change in the particle size. The dynamic shear measurements show that interfacial
reaction imparts additional rigidity to the blend.

Polypropylene and polystyrene were similarly melt blended with a pre-made
elastomeric -[S-EP]- di block copolymer. The data show that the equilibrium interfacial

tension falls due to the addition of the compatibilizing agent from 5 mN/ m in non

compatibilized blends to 1 mN/m in non compatibilized blends. However, unlike the
directly reacted blends, increasing the amount of the compatibilizing agent did not reduce
the interfacial tension. Also, in this case the cross over Of the blend storage modulus with
the components is at a much higher frequency compared to the directly reacted blends.
Due to the solubilization of the matrix in the external block copolymer, the storage
modulus curves are non-monotonic. In the directly reacted blends, the storage modulus

increased progressively with greater extent of maleation and reaction.

In Loving Memory

of

Shashi Asthana

ACKNOWLEDGEMENTS

First and foremost I wish to express my sincere appreciation for the help and guidance
that Dr. K. Jayaraman gave me during the course of PhD. Especially in the early stages,
his help in defining the problem and abstracting a fundamental aspect was significant.
Next, I wish to thank my parents, my brother Rupam and his wife Monica and my wife
Roli for their patience with the Ph.D. experience. This thesis would not have ever been
completed without the moral and intellectual support of Ashim Dutta, Doug Backes and
Sanjay Mishra. They have been the best of colleagues that one can expect. I wish them
luck in their professional and personal lives and hope for their friendship in the coming

years.

TABLE OF CONTENTS

LIST OF TABLES ....................................................................... ix
LIST OF FIGURES ..................................................................... xi
LIST OF SYMBOLS .................................................................... xvii
CHAPTER 1
INTRODUCTION ........................................................................ l
1.1 DEVELOPMENT OF MORPHOLOGY .............................. 2
1.1.1 Drop Deformation ................................................ 2
1.1.2 Break-up of threads by growth of Rayleigh instabilities 4
1.1.3 Coalescence of drops ............................................. 5
CHAPTER 2
PAST WORK ... ............................................................................. 8
2.1 SYSTEMS COMPATIBILIZED BY IN-SITU REACTION BETWEEN
CONIPLEMENTARY GROUPS ............................................ 8
2.2 SYSTEMS COMPATIBILIZED BY THE ADDITION OF AN
EXTERNAL AGENT .......................................................... 13
CHAPTER 3

THE EFFECT OF INTERFACIAL REACTION ON THE INTERFACIAL
TENSION IN REACTIVELY COMPATIBILIZED NYLON 6- MALEATED

POLYPROPYLENE BLENDS ............................................................ 16
3.1 INTRODUCTION .............................................................. 16
3.1.1 Emulsion Models ...................................................... 17

3.2 EXPERIMENTAL .............................................................. 22
3.2.1 Materials ............................................................... 22

3.2.2 Rheological ............................................................. 23

3.2.3 Morphological .......................................................... 23

3.3 RESULTS AND DISCUSSION ............................................... 24
3.3.1 Blend Morphology ..................................................... 24

3.3.2 Rheology ................................................................ 24
3.3.3 Interfacial tension values ............................................. 27

3.3.4 Physical phenomenon ................................................. 32

vi

3 .3 .5 Estimates of surface shear modulus ................................. 35
3.4 CONCLUSIONS .................................................................. 40

CHAPTER 4

THE EFFECT OF PROGRESSIVE EXTENT OF INTERFACIAL REACTION
ON THE RHEOLOGY AND MORPHOLOGY OF NYLON 6 - MALEATED

POLYPROPYLENE BLENDS .............................................................. 74
4.1 INTRODUCTION ............................................................... 74
4.2 BACKGROUND ................................................................. 77
4.2.1 Emulsion Models ....................................................... 77
4.2.2 Determination of the Interfacial Tension using
Emulsion Models ........................................................ 80
4.3 EXPERIMENTAL ................................................................ 82
4.3.1 Materials ................................................................. 82
4.3.2 Blending .................................................................. 82
4.3.3 Rheological ............................................................... 83
4.3.4 Morphological ........................................................... 83
4.4 RESULTS AND DISCUSSION ................................................. 84
4.4.1 Blend Morphology ....................................................... 84
4.4.2 Rheology .................................................................. 84
4.4.3 Determination of the interfacial tension ............................. 87
4.4.4 The issue of extent of coverage ........................................ 91
4.5 CONCLUSION .................................................................. 93
CHAPTER 5

THE EFFECT OF COMPATIBILIZATION OF POLYPROPYLENE AND
POLYSTYRENE WITH A DI-BLOCK COPOLYMER ON THE RHEOLOGY

AND THE MORPHOLOGY OF THEIR BLENDS ..................................... 130
5.1 INTRODUCTION ................................................................ 130
5.2 EXPERIMENTAL ................................................................ 133

5.2.1 Materials ............................................................... 133
5.2.2 Rheological .............................................................. 133
5.2.3 Morphological ........................................................... 134
5.3 RESULTS AND DISCUSSION ................................................ 135
5.3.1 Blend Morphology ....................................................... 135
5.3.2 Rheological Observations .............................................. 135
5.3.3 Physical Events at the interface and the
value of interfacial tension ............................................. 138
5.3.4 The issue of extent of coverage ........................................ 142
5.4 CONCLUSION .................................................................... 143

vii

CHAPTER 6

CONCLUSIONS AND RECOMMENDATIONS .................................... 173
6.1 CONCLUSIONS ............................................................... 173
6.2 RECOMMENDATIONS AND FURTHER WORK .................... 175

BIBLIOGRAPHY ............................................................................ 177

viii

.iln

 

LIST OF TABLES

Table 3.1: The comparison between the models of Oldroyd and of Choi and Schowalter

for an emulsion of Newtonian fluids [ Graebling and Muller, 1990 ] ..................... 41
Table 3.2: The volume average radii of different blends .................................... 42
Table 3.3: The zero-shear viscosities and the corresponding relaxation times of the
components of the blend. ........................................................................ 43
Table 3.4: The data used for determination of interfacial tension from

thermodynamics. . ................................................................................. 43
Table 3.5: The polar and dispersive component of the component phases [Wu, 1987;
Paul, 1978]. ....................................................................................... 43
Table 3.6: A comparison of the parameters of the models of Oldroyd and of Choi and
Schowalter for emulsion of Newtonian fluids. ............................................... 44
Table 3.7: The interfacial tension values for different blends. ............................ 44

Table 3.8: The values of TD, M, and GP based on the Newtonian limit. The value of
interfacial tension used is 4 mN/m for non-reactive and l mN/ m for

reactive blend. ..................................................................................... 45
Table 4.1: Emulsion Rheology models available in the literature. ......................... 95
Table 4.2: The volume average radii of the dispersed phase in the various blends. ..... 96
Table 4.3: The material properties of the materials used in this study. .................. 96

Table 4.4: The values for reactive sites in the materials used in the study. .............. 97

Table 4.5: The values of the extent of coverage of the interface in
different blends. .................................................................................. 97

ix

Table 5.1: The properties of the components used in the study. ......................... 145

Table 5.2: The volume average radii of the disperse phase in the
various blends. .................................................................................. 145

Table 5.3: The polar and dispersive component of polystyrene and polypropylene [Paul,
1978]. ............................................................................................. 145

Table 5.4: The values of the parameters used in to estimate the interfacial tension using
the thermodynamic theory [Rudin 1993; Immergut et. al., 1989]. ....................... 146

Table 5.5: The variation of the extent of coverage and the equilibrium interfacial tension
with progressive addition of the compatibilizing agent. .................................... 146

LIST OF FIGURES

Figure 3.1: The storage and loss moduli for a blend of Newtonian and viscoelastic fluids.
Notice the appearance of a secondary plateau in viscoelastic blend which is a result of

interfacial effects [ Graebling et. al., 1993 ]. .................................................. 46
Figure 3.2a: Micrograph showing the morphology of B3S/ PP (90/ 10). ................. 47
Figure 3.2b: Micrograph showing the morphology of B3 8/ PP (80/ 20). ................ 48
Figure 3.2c: Micrograph showing the morphology of B3S/ PP (70/ 30). ................. 49
Figure 3.2d: Micrograph showing the morphology of B3S/ 3150 (90/ 10). ............ 50
Figure 3.2c: Micrograph showing the morphology of B3S/ 3150 (80/ 20). .............. 51
Figure 3.2f: Micrograph showing the morphology of B3 8/ 3150 (70/ 30). ............... 52
Figure 3.3a: Comparison of the storage moduli of the components of the blend. ........ 53
Figure 3.3b: Comparison of the loss moduli of the components of the blend. ......... 54

Figure 3.4a: Comparison of the storage moduli of the non-reactive blends for varying
weight fractions. .................................................................................. 55

Figure 3.4b: Comparison of the loss moduli of the non-reactive blends for varying
weight fractions. ................................................................................... 56

Figure 3.5a: Comparison of the storage moduli of the reactive blends for varying weight
fractions........... ................................................................................. 57

Figure 3.5b: Comparison of the loss moduli of the reactive blends for varying weight
fractions. ............................................................................................ 58

Figure 3.6: Comparison of the storage moduli of the PA6/ PP (90/ 10) blend with the
components. ........................................................................................ 59

xi

Figure 3.7: Comparison of the storage moduli of the PA6/ PP (80/ 20) blend with the
components. ........................................................................................ 60

Figure 3.8: Comparison of the storage moduli of the PA6/ PP (70/ 30) blend with the
components. .................................................................................... 61

Figure 3.9: Comparison of the storage moduli of the PA6/ PP-MA (90/ 10) blend with
the components .................................................................................. 62

Figure 3.10: Comparison of the storage moduli of the PA6/ PP-MA (80/ 20) blend with
the components. ................................................................................. 63

Figure 3.11: Comparison of the storage moduli of the PA6/ PP-MA (70/ 30) blend with
the components. ................................................................................. 64

Figure 3.12: Comparison of the storage and loss moduli values obtained from the Choi
and Schowalter mOdel with that obtained experimentally for PA6/ PP (80/ 20) (D) = 4
mN/ m, R = 8 pm). ............................................................................... 65

Figure 3.13: Comparison of the storage and loss moduli values obtained from the
Oldroyd model with that obtained experimentally for PA6/ PP (80/ 20) (1'0 = 4 mN/ m, R
= 8 pm). .......................................................................................... 66

Figure 3.14: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP (70/ 30) (I'° = 4 mN/ m, R
= 12 um). .......................................................................................... 67

Figure 3.15: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP (80/ 20) (I‘0 = 4 mN/ m, R = 8 um). 68

Figure 3.16: Comparison of the storage and loss moduli values obtained from the model
with that obtained, experimentally for PA6/ PP (90/ 10)
(I*’=4mN/m,R=4um). ...................................................................... 69

Figure 3.17: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP-MA (90/ 10)
(1"0 = 1 mN/ m, R = 0.5 um). .................................................................. 70

Figure 3.18: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP-MA (80/ 20)
(1'°=1mN/m,R=0.5um). .................................................................. 71

Figure 3.19: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP-MA (90/ 10) (1'0 = l mN/ m, R = 0.5 pm,
[30 = 0.5 mN/ m, cob = 0.2 rad/ s). ............................................................... 72

xii

Figure 3.20:

Comparison of the storage and loss moduli values obtained from the model

with that obtained experimentally for PA6/ PP-MA (80/ 20) (W = 1 mN/ m, R = 0.5 pm,

80 = 0.5 mN/ m, no,3 = 0.2 rad/ s). .............................................................. 73
Figure 4.1a: Micrograph showing the morphology of B3/ PP (90/ 10). ................. 98
Figure 4.1b: Micrograph showing the morphology of 33/ PP (80/ 20). ................ 99
Figure 4.1c: Micrograph showing the morphology of B3/ PP (70/ 30). ................ 100
Figure 4.1d: Micrograph showing the morphology of B3/ 3001 (90/ 10). .............. 101
Figure 4.1e: Micrograph showing the morphology of B3/ 3001 (80/ 20). .............. 102
Figure 4.1f: Micrograph showing the morphology of B3/ 3001 (70/ 30). .............. 103
Figure 4.1g: Micrograph showing the morphology of B3/ 3002 (90/ 10). .............. 104
Figure 4.1h: Micrograph showing the morphology of B3/ 3002 (80/ 20). .............. 105
Figure 4.1i: Micrograph showing the morphology of B3/ 3002 (70/ 30). .............. 106
Figure 4.1 j: Micrograph showing the morphology of B3/ 3150 (90/ 10). .............. 107
Figure 4.1k: Micrograph showing the morphology of BB/ 3150 (80/ 20). .............. 108
Figure 4.11: Micrbgraph showing the morphology of B3/ 3150 (70/ 30). .............. 109
Figure 4.2a: Comparison of the storage moduli of the components of the blend. ..... 110
Figure 4.2b: Comparison of the loss moduli of the components of the blend. .......... 111
Figure 4.3a: Comparison of the storage moduli of the non-reactive blends for varying
weight fractions of the dispersed phase. ...................................................... 112
Figure 4.3b: Comparison of the loss moduli of the non-reactive blends for varying

weight fiactions of the dispersed phase. ...................................................... 113

Figure 4.4: Comparison of the storage moduli of the B3/ 3001 blend for varying weight
fractions of the dispersed phase. ............................................................... 114

Figure 4.5: Comparison of the storage moduli of the 80/ 20 blends for different dispersed

phases. ......

....................................................................................... 115

xiii

Figure 4.6: Comparison of the storage moduli of the B3/ PP (80/ 20) blend with the
components. ..................................................................................... 116

Figure 4.7: Comparison of the storage moduli of the B3/ 3001 (80/ 20) blend with the
components. ..................................................................................... 117

Figure 4.8: Comparison of the storage moduli of the B3/ 3002 (80/ 20) blend with the
components. ..................................................................................... 118

Figure 4.9: Comparison of the storage moduli of the B3/ 3150 (80/ 20) blend with the
components. ..................................................................................... 119

Figure 4.10: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for 83/ PP (80/ 20) blend (I'° = 8 mN/ m, R = 10.7

um). ................................................................................................ 120

Figure 4.11: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for B3/ 3001 (90/ 10) blend (1‘0 = 7 mN/ m, R = 0.8
pm). ................................................................................................ 121

Figure 4.12: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for B3/ 3001 (80/ 20) blend (1") = 7 mN/ m, R = 0.8
pm). ................................................................................................ 122

Figure 4.13: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for 83/ 3001 (70/ 30) blend (1'0 = 7 mN/ m, R = 0.6
pm). ................................................................................................ 123

Figure 4.14: Comparison of the G’im values obtained from the model with that obtained
experimentally for 33/ 3001 (80/ 20) (I‘0 = 7 mN/ m, R = 0.8 pm). ...................... 124

Figure 4.15: Comparison of the G’int values obtained from the model with that obtained
experimentally for B3/ 3002 (80/ 20) (I'° = 7 mN/ m, R = 0.8 um). ...................... 125

Figure 4.16: Comparison of the G’m values obtained from the model with that obtained
experimentally for 83/ 3150 (80/ 20) (1'0 = 4 mN/ m, R = 1.5 pm). ...................... 126

Figure 4.17: Comparison of the G’ values obtained from the model with that obtained
experimentally for 83/ 3150 (80/ 20) (I0 = 4 mN/ m, R = 1.5 pm, [30 = 0 mN/ m). 127

Figure 4.18: Comparison of the G’ values obtained from the model with that obtained
experimentally for 83/ 3150 (80/ 20) (D) = 4 mN/ m, R = 1.5 pm, [30 = 0.1 mN/ m).. . . 128

xiv

Figure 4.19:

The plot showing the reduction of equilibrium interfacial tension with

progressively increasing extent of maleation. ................................................ 129
Figure 5.1a: Micrograph showing the morphology of PP/ PS (90/ 10). ............... 147

Figure 5.1b: Micrograph showing the morphology of PP/ PS (80/ 20). ............... 148

Figure 5.1c: Micrograph showing the morphology of PP/ PS/ 1702 (90/ 10/ 2). ...... 149
Figure 5.1d: Micrograph showing the morphology of PP/ PS/ 1702 (90/ 10/ 4). ...... 150
Figure 5.1e: Micrograph showing the morphology of PP/ PS/ 1702 (90/ 10/ 6). ....... 151
Figure 5.1f: Micrograph showing the morphology of PP/ PS/ 1702 (80/ 20/ 2). ....... 152
Figure 5.1g: Micrograph showing the morphology of PP/ PS/ 1702 (80/ 20/ 4). ....... 153
Figure 5.1h: Micrograph showing the morphology of PP/ PS/ 1702 (80/ 20/ 6). ....... 154
Figure 5.2a: Comparison of the storage moduli of the components of the blend. ...... 155
Figure 5.2b: Comparison of the loss moduli of the components of the blend. .......... 156

Figure 5. 3: Comparison of the storage moduli of the 90/ 10 blends with different
amounts of the compatibilizing agent. ...................................................... 157

Figure 5.4: Comparison of the storage moduli of the 80/ 20 blends with different
amounts of the compatibilizing agent. ...................................................... 158

Figure 5.5: Comparison of the storage moduli of the PP/ PS (90/ 10) blend with its
components. ..................................................................................... 159

Figure 5.6: Comparison of the storage moduli of the PP/ PS/ 1702 (90/ 10/ 2) blend with
its components. .................................................................................. 160

Figure 5.7: Comparison of the storage moduli of the PP/ PS/ 1702 (90/ 10/ 4) blend with
its components. .................................................................................. 161

Figure 5.8: Comparison of the storage moduli of the PP/ PS/ 1702 (90/ 10/ 6) blend with
its components. .................................................................................. 162

Figure 5.9: Comparison of the storage moduli of the PP/ PS (80/ 20) blend with its
components. ...................................................................................... 163

XV

Figure 5.10: Comparison of the storage moduli of the PP/ PS/ 1702 (80/ 20/ 2) blend
with its components. ........................................................................... 164

Figure 5.11: Comparison of the storage moduli of the PP/ PS/ 1702 (80/ 20/ 4) blend
with its components. ............................................................................ 165

Figure 5.12: Comparison of the storage moduli of the PP/ PS/ 1702 (80/ 20/ 6) blend
with its components. ............................................................................ 166

Figure 5.13: Comparison of the storage moduli obtained from the model with that
obtained experimentally for PP/ PS (90/ 10) (D) = 5 mN/ m, R = 1.12 um). ............ 167

Figure 5.14: Comparison of the storage moduli obtained from the model with that
obtained experimentally for PP/ PS (80/ 20) (I‘° = 5 mN/ m, R = 1.86pm). ............. 168

Figure 5.15: Comparison of the storage moduli obtained from the model with that
obtained experimentally for PP/ PS/ 1702 (90/ 10/ 2) (1") = l mN/ m, R = 0.9 pm). 169

Figure 5.16: Comparison of the storage moduli obtained from the model with that
obtained experimentally for PP/ PS/ 1702 (90/ 10/ 4) for different values of equilibrium
interfacial tension (R= 0.8 pm). ................................................................ 170

Figure 5.17: Comparison of the (32,, obtained from the model with that obtained
experimentally for PP/ PS/ 1702 (90/ 10/ 6) (1"0 = l mN/ m, R = 0.6 pm). ............. 171

Figure 5.18: The variation of the equilibrium interfacial tension with varying amount of
the compatibilizing agent. ..................................................................... 172

xvi

LIST OF SYMBOLS

Ca : Capillary Nmnber.
d : Diameter of the drops, m.
d“ : Diameter of the drops in the unreacted system, m.
d, : Diameter of the drops in the reacted system, m.
D : Deformation of the drop.
G’ : Storage modulus of the material, N/ m2.
G” : Loss modulus of the material, N/ m2.
G; : Complex modulus of the blend, N/ m2.
G; : Complex modulus of the dispersed phase, N/ m2.
Gm' : Complex modulus of the matrix, N/ m2.
p : Viscosity ratio, nd/nm.
pr : Viscosity ratio in the reactive system.
pu : Viscosity ratio in the unreactive system.
q : Rate of growth of the instability.
R2 : Radius of the filament at t = 0.
R, : Volume average radius of dispersed phase, m.
t : Time, s.
t, : Time for break-up of the filament, s.
tcoll : Time for collision of the drops, 3.
tint : Time for interaction of the drops, 5.
tam : Time for drainage of the drops, s.

tpm : Time for process of collision of the drops, 5.

tm : Time for retraction of the drops, 5.

Greek Symbols:

a : Amplitude of disturbance, m.

(10 : Amplitude of the disturbance at time t = 0.

[3° : Sum of surface shear modulus and surface dilatation modulus, N/ m2.
[3,. : Surface shear modulus, N/ m2.

[3; : Surface dilatation modulus, N/ m2.

1],, : Viscosity of the dispersed phase, Pa-s.

y : Shear rate, Us.

I" : Equilibrium interfacial tension, N/ m.

I‘, : Interfacial tension of the reactive system, N/ m.
I“, : Interfacial tension of the unreactive system, N/ m.
it : Wavelength of the disturbance, m.

A, : Relaxation time, s.

2., : Retardation time, 8.

Ad : Relaxation time of the dispersed phase, 5.

km : Relaxation time of the matrix, 5.

AD : Beginning of secondary plateau of the blend, 5.

xvii

3”?

696333333
a 9-3 0.0- c

d

an.

'-l'-l

: End of secondary plateau of the blend, 5.

: Beginning of primary plateau of the blend, 5.
: Tabulated function.

: Sum of viscosity of component phases, Pa-s.
: Zero-shear viscosity, Pa-s.

: Viscosity of the blend, Pa—s.

: Viscosity of the dispersed phase, Pa-s.

: Viscosity of the matrix, Pa-s.

: Volume fiaction of the dispersed phase.

: Volume fraction of the matrix.

: Time scale of retraction of the filament, s.

: Shear stress, N/ m2.

: Torque in mixer in reactive system, N-m.

: Torque in mixer in unreactive system, N-m.

xviii

INTRODUCTION

 

Chapter 1

 

It is a common industrial practice to mix two or more polymers to make a new
material whose properties are an optimum combination of the individual components.
When two polymers are mixed, one of the phase (the minor phase) forms the dispersed
phase and the other component (the major phase) forms the continuous or the matrix
phase. The shape and size of the dispersed phase plays a critical role in the final
properties and the. processing behavior of the blend [Wu, 1983]. The main issue in
blending of polymers is to control the particle shape and size (distribution). Most
polymers phase separate on mixing due to the thermodynamic considerations [Sperling,
1992]. Thus, there is a need to compatibilize the components so that they do not phase
separate on mixing and at the same time offer control over the particle size of the
dispersed phase. Most often, this requires promoting dispersive mixing by the reduction
of interfacial tension between the component phases. This may also lead to enhanced
adhesion between-the component phases which is a result of reduction of interfacial
tension between the two phases. There are two common techniques employed to achieve
this [Xanthos, 1994]. They are:

1. Addition of an external compatibilizing agent to the mixture.
2. Carry out an in-situ reaction between the complementary groups on the individual

components.

2

These lead to smaller particle sizes, a narrower particle size distribution and improved
adhesion. The following section is a brief discussion of the steps involved in the
development of morphology in emulsions. A direct analogy can be made to the

morphology of polymeric blends.

1.1 DEVELOPMENT OF MORPHOLOGY

The simplest problem to study the development of morphology is that of a break-
up of a drop suspended in an infinite continuous phase. There are three main steps
involved during this process:
1. Drop deformation.
2. Drop dispersion.

3. Coalescence of drops.

1.1.1 Drop Deformation.
After the seminal work of Taylor [Taylor 1934; Taylor 193 8], several researchers
have studied the deformation of a drop suspended in another fluid [Cox, 1969; Grace,

1979; Bentley, 1986; Stone, 1994; Tjahjadi et. al., 1996]. The relation governing the

deformation of the drop is:

3

_ [(19p+16)/(l6p+16)] 1 1
D ‘C" [(19pCa/40)2 +11“2 ( ° )

6=7r/4+05tan'1(19pCa/20) (1.2)

where the capillary number Ca is defined as

I‘d ' md
r'.=11%.— (1.3)

 

Ca:

The main results of these studies are :

1. The Cam, for drop break-up is lowest at a viscosity ratio of unity.

2. Beyond a viscosity ratio of 3.0-3.5, shear is incapable of dispersing the drop. For
extensional flow this limit is very high.

3. There is a critical capillary number Cacrit beyond which the drop bursts. In effect,
beyond Cam, interfacial tension forces can not balance the viscous forces.

4. Extensional flow is more conducive to drop dispersion as compared to shear flow.

It is seen that the equilibrium interfacial tension plays a direct role through the
capillary number. For a given flow field, lower interfacial tension facilitates drop

deformation (or dispersion) by the action of viscous forces. In the initial stages of

4

blending when the particle sizes are large (Ca >> Cam) viscous forces dominate
interfacial forces and thus long stretched threads are formed. The mechanism of break-up

of these threads is by Rayleigh instabilities as discussed in the following section.

1.1.2 Break-up of threads by growth of Rayleigh instabilities

Rayleigh presented the first study of the problem of capillary instability of a
cylindrical column [Rayleigh, 1878]. The salient feature of the hypothesis is that the
disturbance, assumed sinusoidal grows if the wavelength of the disturbance it > 21cRo.
This situation leads to an increase in the interfacial energy/ area which is
thermodynamically unfavourable. To reduce the interfacial energy the disturbance grows
at an exponential rate such that the thread breaks up into several smaller drops. Larger
number of drops causes the energy to distribute over a larger interfacial area, causing a
reduced interfacial energy/ area and hence thermodynamic stability. The growth of the

disturbance is exponential and given as:

a = a. exp<qt> (1.4)

In Equation 1.4, or is amplitude of the disturbance, or, amplitude at t=0, t being the time

and q the rate of growth.

= floor, p)

1.5
277.1% ( )

The parameter Q in Equation 1.5 is a tabulated function [Tomotika 1935; Tomotika

1936]. Experimental investigations have confirmed the validity of these relations

5
[Rumscheidt and Mason, 1962; Elmendorp, 1986]. From the relations presented above,

the time for break-up of the thread is given as [ Rumscheidt and Mason, 1962]:

1 0.81R
tb =: '— |'{ 0:| (1.6)
q “0

 

The preceding discussion brings out two important parameters for the dispersion
of a cylindrical thread - tb and CGcrit. Ca > cacrit is a pre-requisite for break-up,
which is accomplished only if t > tb. It can be seen that the interfacial tension plays a
critical role in this step also. The Equation 1.5 shows that the rate of growth of the
disturbance is directly proportional to the equilibrium interfacial tension. A lower
interfacial tension means that the disturbance grows at a slower rate. In effect, a lower
interfacial tension translates to improved adhesion between the two phases. In addition,
the above discussion also shows that the residence time (in case of batch mixers) and / or
residence time distribution (in case of continuous reactors) is important during
processing. This governs the time of break-up allowed. Besides, the residence time

(distribution) also affects the coalescence phenomenon directly.

1.1.3 Coalescence of drops

With increasing concentration of the dispersed phase, the probability of
interaction of the drops with each other increases. This leads to collision. But, does each
collision necessarily lead to a coalescence of drops? Janssen and co-workers have studied

this problem and found that the process is governed by four time scales [J anssen and

Meijer, 1995].

6

1 , time of collision tcoll , average time after which the drop collides.
2 , time of interaction tin, , duration of collision.
3 , time of drainage of the film tdmin.

4 - time of process, tprocess .

The equilibrium interfacial tension is directly involved in the drainage time.
Depending on the mobility of the interface it has differing impact. But, in general a

reduced interfacial tension leads to an increased drainage time. This suppresses

coalescence which prevents smaller drops from forming larger drops and ultimately leads

to a reduced particle size.

The study by Janssen et. al. also showed that there is a range of processing
Parameters whichfavor coalescence [Janssen and Meijer, 1995]. Coalescence
preferentially occurs in regions of small deformation rates, which provide large
interaction times. In this process the nature of the interface (of the drops) plays a critical
role. Higher interface mobility promotes film drainage increasing probability of
coalescence. Externally added compatibilizing agents or in-situ reaction reduce the

interface mobility [Janssen and Meijer, I 995]. This promotes higher drainage times

hindering coalescence and stabilizing the morphology.

The preceding discussion illustrates the importance of interfacial tension in the
various steps involved in the break-up of a drop. It is a critical step involved in the
development of morphology of polymeric blends. The present work is an attempt to

understand the specific role that interfacial tension plays in reactive polymer blends. An

7

important parameter that characterizes the interfaces is the interfacial tension. The
following chapter illustrates the past work that has been done to understand the role of
equilibrium interfacial tension in polymeric blends and its role in development of

morphology.

PAST WORK

 

Chapter 2

 

The preceding chapter showed the importance of interfacial tension in the
development of morphology. Also, the different equations and experimental observations
show that a reduced interfacial tension promotes drop dispersion. This fact has been used
to promote compatibilization in polymeric blends. It is achieved either by addition of an
external agent or by direct reaction between the complementary groups on the blend
components [Xanthos, 1994]. In both cases, an emulsifying action occurs at the interface
which leads to a reduced interfacial tension. This chapter discusses the past work that has
been done to understand the role of interface and interfacial tension in the development of

morphology of polymeric blends.

2.1 SYSTEMS COMPATIBILIZED BY IN-SITU REACTION BETWEEN
COMPLEMENTARY GROUPS

It is widely accepted that interfacial reaction promotes dispersive mixing by the
reduction of interfacial tension. Experimental evidence is well documented in the
literature which shows that the interfacial tension is reduced by the addition of an external
agent [Cho et. al., 1996; Elemans et. al., 1991]. However, there is no systematic study on
the effect of an in-situ reaction between the polymers on the interfacial tension between
them. The study by Wu on a system of nylon 6 blended with non-reactive and reactive

(functionalized with less than 1% carboxylic acid ) EP-rubbers in a twin screw extruder is

considered a significant step in this direction [Wu 1987]. Wu presented interfacial
tension values based on statistical mechanical theory of polymers. The interfacial tension
was related to the Flory-Huggins x parameter which in turn was related to the interfacial
thickness. The basis of this was the theoretical development by Helfand and Sapse

[1975].

F0 0:21”2 at L’I (2.1)

Based on the regression analysis of the experimentally measured thickness of the

interface in model systems Wu arrived at the following relation.

r0 = 7.6L‘0'“ (2.2)

Wu then made measurements of the interfacial thickness in nylon 6 - reactive rubber
blends and by using the Equation 2.2 proposed that the interfacial tension drop in a
reactive system could be as much as 30 fold (from 8.8 mN/ m in non-reactive to 0.25 mN/
m in reactive systems). Wu attributed this reduction in interfacial tension to the
emulsifying action of the graft copolymer formed in-situ due to the reaction [Wu 1983].
Although a useful result, it is still not a direct measurement of the interfacial tension in
reactive system. It is an important purpose of this study to provide experimentally
measured values of interfacial tension. Also, in the same study Wu proposed that the

droplet breakup behavior of polymeric viscoelastic drops is fundamentally different from

10

those of Newtonian fluids. On the basis of some empiricism he derived the following

relation.

9%,ng = g = 4 pio'84 (2.3)

The ‘+’ sign applies to systems having the viscosity ratio greater than unity while ‘-‘ sign
applies when it is less than unity. The viscosity values are prior to the reaction. Although
it is a good starting point for estimating particle sizes based on processing conditions, it
is not afimdamental result. Two important parameters were estimated in Wu's analysis.
1. interfacial tension in reactive blends.

2. shear rate.

This study focuses on attempting to remove the discrepancy with regard to the interfacial
tension. Serpe and co-workers [1990] used Wu's relation in their work with
polyethylene-polyamide blends and found that for differing compositions, their curves
followed a V-shape trend parallel to Wu's curves but all the points did not fall onto a
single curve. They modified Wu's relation as follows.

i rid

Ca' = F0 (l —4(¢d¢m)°'8) = 4 p103" (2.4)

 

1n Equation 2.4, 6', and 4),, are the volume fractions of the dispersed and the continuous
phases respectively. Note that the matrix viscosity has been replaced by the blend
viscosity. Such disagreements necessitate further investigations into the role of interfacial
tension on the blend behavior. This work aims to do this.

Scott and Macosko [1995, Intern. Polym. Proc] investigated the morphology of
reactive blends of nylon with fimctionalized EP rubbers. They found that the size of the
dispersed phase decreased continuously as the extent of reaction increased. They

proposed that difierences between the actual size and that predicted by W u's relation are

11

due to changes brought about by interfacial tension and coalescence. The shear stress
was estimated from T, the torque on the mixer because the shear rate is difficult to
determine for the complex flow field. Based on Wu’s relation the differences between the
reactive and non-reactive blends obey the following relation.

i _L p(T.) 0'“
d ’T. p(T.)

I‘

(2.5)

In the empirical relation of Equation 2.5 the differences due to interfacial tension alone
have not been delineated. Now, one can also modify the above relation as follows

(directly from Wu's relation)

6}, _L(i 77...), [gr (2.6)

Afier making estimates of the shear stresses and (if) interfacial tension values are
available, one can comment on the effects due to reaction. This would be addressed in

this work.

In a related study Scott and Macosko [1995, Poly. Eng. Sc.] blended non-reactive
and reactive EP rubbers (functionalized with Maleic anhydride) with non-functionalized
and functionalized Polystyrene (functionalized with vinyl oxazoline). The non-reactive
blends showed poor interfacial adhesion and large particle sizes of the dispersed phase.
On the other hand, the reactive blends exhibit good adhesion and smaller dispersed
domain sizes. Reactive blends were shows to have a higher morphological stability. In
addition, increasing the oxazoline content lead to decreased dispersed phase particle size.

This may not be the case always. Borggreve and Gaymans [1989] carried out a study on

12

blends of nylon 6 with EPD rubber. They observed that increasing the maleic anhydride
content of frmctionalized rubber from 0.13 to 0.89 did not have a significant impact on
the dispersed phase sizes. A similar observation was made by Scott and Macosko [1995,
Polym. Eng. Sc.]- This indicates that in reactively compatibilized systems the
development of morphology may depend on factors other than reduction of interfacial
tension.

In a similar study Hosoda et. al. [1991] carried out a study of morphological
changes in nylon 6 - polypropylene (maleated and unmaleated) blends. Their
observations show that the graft copolymer formed after the reaction resides at the
interface. Similar observations were reported by Fayt et. al. [1986]. Hosoda and co-
workers observed that the thickness of the interface lies between 50-100 A. Moreover, as
the grafted maleic anhydride content in the maleated polypropylene increased, the
average particle size of the dispersed phase decreased while the interface thickness
remained constant. They concluded that the interface stability per graft copolymer
molecule is constant and independent of the degree of the reaction between nylon 6 and
maleated polypropylene. The study of Nishio et. al. [1991] on nylon 6 and maleated
polypropylene shows that the interface area per unit volume increased linearly with
increase in the grafied copolymer content. They hypothesized that a unit area of the
interface per unit volume is occupied by a certain amount of the copolymer independent
of the sample. The copolymer located at the interface behaves as an emulsifier with a

constant concentration per unit interface area.

13

The brief discussion shows that the interface plays a critical role in the
determination of morphology. If one has to understand it holistically, then the interplay
of interfacial (tension) forces and viscous forces has to be understood. This in turn is
dependent on the processing conditions (shear rate, temperature for example). Moreover,
in polymers the situation is further complicated by the elasticity of the components. This
work will address the interfacial tension segment of the problem. How does morphology

vary/ depend on the interfacial reaction between the polymeric species?

2.2 SYSTEMS COMPATIBILIZED BY THE ADDITION OF AN EXTERNAL
AGENT

The components of the blend can be compatibilized by the addition of an external
agent. This compatibilization may be of a physical or chemical nature. In physical
compatibilization, the compatibilizing agent does not react with the component phases.
Instead, there are groups on the compatibilizer which are physically similar to the
chemical nature of the components. On the other hand, in chemical compatibilization, a
chemical reaction‘occurs between the external agent and the component phase(s). For
example, maleated polypropylene would chemically compatibilize neat polypropylene
and nylon 6 [lde and Hasegawa, 1971]. The maleic anhydride group on the maleated
polypropylene can react with amine of the nylon 6. A brief discussion of some relevant

past work in this area follows.

l4

Okarnoto and Inoue [1993] carried out a study on poly-(e-caprolactone) blended
with two different kind of functionalized rubbers. They were coupled (chemically
compatibilized) with an external agent. The aim of that study was to understand the
development of morphology and relate it to the extent of interfacial reaction. They
concluded that as the residence time in the mixer is increased, the average particle size
decreases until a given value and then becomes a constant. The exact behavior depends
on the amount of coupling agent added. Expectedly, the average size is less for larger
amounts. At the same time, the specific interfacial area increases and saturates at a given
value. The interesting observation was regarding the interfacial thickness. As observed
by Hosoda et. al. [1991], the interfacial thickness assumed a constant value irrespective of
the residence time. Based on their observation with two different kinds of rubber, they
hypothesized that coupling reaction is faster at thinner interface, which leads to a faster
rate of size reduction in such systems.

Lim and White [1994] studied externally compatibilized blends of polyethylene
and nylon 6 in a modular twin screw extruder. The main purpose of their study was to
relate the development of morphology along the extruder length. It was found that the
rate of decrease of phase morphology scale increases rapidly along the screw length by
the addition of the compatibilizing agent while at the same time leading to a finer
ultimate morphology. They showed that besides the functionality on the compatibilizing
agent, the processing conditions and the properties of the components make a significant
difference on the resultant morphology. This observation is supported by the study of Lee

and Yang [1995]. 1 They prepared blends of polypropylene with nylon 6 by three different

15

mixing processes; single step blending, two-step blending with reactive premixing and
two-step blending with non-reactive premixing. They found that the single step mixing
proved to be the most effective for scaling down the morphology.

Now, the studies cited so far have not studied the effect of interfacial tension on
the morphology development by making actual measurements of the interfacial tension.
An important focus of the present study is to study this aspect of the problem and attempt
to provide this information. Also, as preceding discussion shows, the rheology of the
blends is effected by the interfacial reaction. This work will systematically study the
effects of reaction on morphology development and rheology of the reactively

compatibilized polymeric blends.

THE EFFECT OF INTERFACIAL REACTION ON THE

INTERFACIAL TENSION IN REACTIVELY COMPATIBILIZED

NYLON 6 - MALEATED POLYPROPYLENE BLENDS

 

Chapter 3

 

3.1 INTRODUCTION

Polymers are thermodynamically incompatible. As a result they phase separate on
blending [ Sperling, 1992 ]. Compatibilization of polymers is carried out to circumvent
this problem. It is achieved primarily in two ways. Firstly, by adding an external
compatibilizer which is compatible with the blend components. Secondly, by an in-situ
reaction of the complementary groups on the components. Both techniques
compatibilize by directly influencing the interface between the polymers. Reduction of
equilibrium interfacial tension plays a significant role in the process of compatibilization.
It has been shown that addition of an external compatibilizer leads to a reduction in
equilibrium interfacial tension [ Elemans et. al. 1990 ]. An important aim of this study is
to investigate the effect of interfacial reaction on the interfacial tension in blends
compatibilized via reaction of complementary reactive groups. The values available in
the literature are indirect and based on empirical relations [ Wu, 1987 ]. The relation
requires the use of morphological parameters. In recent studies it has been shown that
besides equilibrium interfacial tension, suppression of coalescence also plays an
important role in the development of morphology [ Sundraraj and Macosko, 1993;

O’Shaughnessy and Sawhney, 1996 ]. That is, the morphological parameters alone are

16

17

insufficient to characterize the equilibrium interfacial tension. Thus, there is a need for
directly measured values of equilibrium interfacial tension.

The reaction product located at the interface leads to a fundamental change in the
rheological behavior of the blend as has been shown in this study and Chapter 4. Due to
the ‘occupied interface’ and the physical links’ that are established between the
component phases, the elasticity of the system is enhanced. In light of this observation it
can be concluded that the composition of the interface is critical in determining the
rheological behavior of the blend.

The materials chosen in this study contain additives which are incorporated in the
polymers as processing aids. They are usually low molecular weight materials. Being
low molecular weight materials they have a thermodynamic drive to rise to the interface.
Another question being investigated in this study relates to the effects of the presence of

these low molecular weight materials at the interface? Can they behave as surfactants?

3.1.1 Emulsion Models

The rheological behavior of the polymeric blends has been explained on the basis
of emulsion models. Three emulsion models have usually been applied to polymeric
systems. They are due to Oldroyd [ Oldroyd, 1953 ], Choi and Schowalter [Choi and
Schowalter, 1975 ] and Palieme [Palieme, 1990 ]. The models of Oldroyd, and of Choi
and Schowalter were formulated for a mixture of Newtonian fluids with monodisperse
spherical inclusions. Interestingly, both models predict a non-zero storage modulus of

the blend especially in the low frequency region. This is a direct result of the interfacial

18

tension between the blend components. It leads to long time relaxation processes of the
dispersed phase which are of the order of mechanical relaxation of the drop shape

[ Scholz et. al., 1989 ] as shown in Equation 3.1.

R 77..

2,, ~f(k) 1.. (3.1)

These models show that the rheological behavior, especially the storage modulus is
sensitive to the interfacial tension in the low frequency region.

The general form of the models of Oldroyd and of Choi and Schowalter is

 

._ 02(41 ‘42)
G _ a 1+w2112 (3'2)
G — a 1+w2212 (3‘3)

The definition of the parameters 7.0, 1,, k, and n, for the models have been shown in
Table 3.1. These models have been used by several workers to determine the interfacial
tension in the polymeric systems [ Scholz et. al., 1989; Graebling and Muller, 1991;
Gramespacher and Meissner, 1992 ].

To account for the viscoelastic properties of the component phases, Palieme
proposed an emulsion model for a mixture of two viscoelastic fluids [ Palieme 1990 ]. It
accounts for the distribution in particle size and interactions and is applicable for a wide
range of volume fractions of the dispersed phase. The model explicitly takes into account
the rheological behavior of the component phases and the interfacial tension. The

relation is shown in Equation 3.4.

l9

 

 

 

 

 

 

 

1 32—”
. , 2,. D,
G, =0, ¢ E (3.4)
1_ZI I
L i D, _
where,
a t e O o
. . . . 4sfldr° 3255(r°+rd) sr° . . 2rd . . 45 . .
Ei=2(Gd—Gm)(l9Gd+léGm)+ R2 + R2 +7606,+2o,,,)+—R—(23Gd-160,,,)+ Rs(130d+80,,,)
l
|__(3.5)
and
. . . . 485;!” 324055;) 40r° . . 2r; . . 4p; . .
Di=(ZGd—3Gm)(l9Gd+l6Gm)+ R2 + R2 + R Gd+Gm)+T(23Gd+3ZGm)+T(l3Gd+1201")
(3.6)

 

 

An important feature of this model is its treatment of the interfacial tension as a
sum of two parts. A static part which is the equilibrium interfacial tension I0 and a
frequency dependent complex part 8(0)). In turn, [3'((o) consists of two complex moduli -
the surface dilatation modulus and the surface shear modulus. Both these properties are
characteristics of the interface. The surface dilatation modulus is a result of the non-
uniformity of the interfacial tension over the interface while surface shear modulus is the
resistance of the interface to the deformation. The preceding models lacked any

parameter(s) besides the equilibrium interfacial tension which characterized the pr0perties

20

associated with the interface. These parameters attain significance as it has been shown
that in externally compatibilized systems, the interface is no more unirnolecular layer
thick. It is a region of finite thickness with its own associated properties [ Germain et. al.,
1991 ]. In reactively compatibilized systems also a reaction product is being formed at
the interface. The ‘bare’ interface is being ‘occupied’. This leads to an interface of finite
thickness which has the potential of altering the behavior of the blend significantly.

Thus, the need to incorporate the properties of the interface in the model. Palierne’s
model is a step in this direction. It should be pointed out that Oldroyd’s model is retained

from Palierne’s model. Under such circumstances,

1+ 3Z¢,H,(m)
G’<w) = G.‘.(w) 1_ 2: M (w) (3.7)

 

where,

_ (41“0 / mag; 50;) +(G; - @160; +19G;)

H. — o 0 ‘ . . .
((0) (4OF0 /Ri)(Gm '1' Gd) + (204 ‘1' 3G1» )(1600' +1904)

i

 

(3.8)

Based on the above model the terminal relaxation time of the emulsion whose
components are represented by single relaxation time Maxwellian model is given by

Equation 3.9 [Graebling and Muller, 1990 ].

21

 

Id”) 2 (x—1)
= '" 2 3.9
T ”+124an (19k+16)2’”+ "' ( )

The parameter IN is the terminal relaxation time of an emulsion of two Newtonian fluids
whose viscosities are same as the viscosities of the viscoelastic phases. It is given by

Equation 3.10.

 

2
TN = gen") (19k+l6) (3.10)

8 r° ¢d50(k+1)2+5¢d(5k+2)(k+1)—3¢j(5k+2)2

The basis of the equilibrium interfacial tension measurements is the secondary plateau in
storage moduli which is a result of the interfacial forces. Thus, to determine interfacial

tension from the rheological measurements, it is essential to capture as much secondary

plateau as possible.

22

3.2 EXPERIMENTAL

3. 2. 1 Materials

Nylon 6 (PA6) ( BASF Ultramid, B3 SQ661 ), neat polypropylene (PP) ( Montell
Polyolefins, Profax 6501 ) and maleated polypropylene (PP-MA) (Uniroyal Chemicals,
PB3150 ) were used in this study. The Nylon 6 was reported to contain lubricants for
processing ease. The Maleic anhydride content of the PB3150 was reported to be 0.8 wt.
% as per the manufacturer. PA6 was blended with PP and PP-MA respectively. PA6 was
the matrix and FF (or PP-MA) were the dispersed phases. Blending was carried out in a
ZSK-30 twin screw extruder at a temperature of 230 °C for all the zones. To minimize
the effects of moisture, the materials were dried under a nitrogen blanket for 6-8 hours
prior to blending. The extrudate strands were pelletized and dried. Three blend
compositions of differing weight fractions of PP and PP-MA (10, 20, 30 weight percent
respectively) were prepared. A density of 1.14 g/cc for PA6 and 0.93 g/cc for
polypropylene was used. In the reactive system the amine group of the PA6 reacts with
the maleic anhydride of the polypropylene [ Ide and Hasegawa, 1974 ]. The blends will
be referred to in the text with the components and their weight fractions in the

parenthesis.

23

3. 2.2 Rheological

Rheological characterization was carried out on an RMS-800 rheometer (from
Rheometrics Scientific, Inc.) using a 50 mm parallel plates arrangement. The disks were
prepared by compression molding the pellets in a Carver press under a force of 6 tons and
230 °C. The pellets were pre-dried in a vacuum oven for 10-12 hours. The instrument
oven was purged with dry nitrogen during measurements to avoid degradation. A
frequency range of 0.05 - 50 rad/ s and strains of 10-15% were chosen. It was ensured

that the testing was carried out in linear viscoelastic limits.

3.2.3 Morphological

The samples for morphological examination were prepared by pre-notching the
disks used in rheological measurements, beating them to 230 °C in the rheometer oven
and placing them directly in the liquid nitrogen at that temperature. Those were then
fractured under liquid nitrogen. This ensured that the morphological information used in
the analysis was as close to the one under investigation. These samples were then
examined in Phillips Electroscan 2020 environmental scanning microscope. Water vapor
at a pressure of 2-3 Torr was used as the imaging gas. The size of the disperse phase was
measured directly from the micrographs. The volume average radii were calculated for

the blends and used in the analysis.

24

3.3 RESULTS AND DISCUSSION
3. 3.1 Blend Morphology

Figures 3.2a through 3.2c show the morphology of the non-reactive blends while
Figures 3.2d through 3.2f show the morphology of the reactive blends. For the non-
reactive blends the particles are spherical with a broad particle size distribution. Their
size increases progressively with increasing volume fraction of the dispersed phase. Such
a behavior can be attributed to coalescence. On the other hand, in case of reactive blends,
the particle size is reduced, the particle size distribution is narrowed and the effect of
volume fraction is minimal. Coalescence is reduced in reactive blends due to the
increased stability and the reduced mobility of the interface [ Sundraraj and Macosko,
1995; O’Shaughnessy and Sawhney, 1996; Janssen and Meijer, 1995 ]. Table 3.2

summarizes the volume average particle sizes.

3. 3.2 Rheology
Table 3.3 shows the zero-shear viscosity no and the corresponding relaxation time

of the components. The zero-shear viscosity was determined from the flat portion of the

11’ vs a) curve. The values were confirmed by the following relation.

'7. =(limw—>0)9La()w—) (3.11)

25

Relaxation time was determined by

G'(60)
(0277

0

 

2 = (limo) —) 0) (3.12)

Figures 3.3a and 3.3b show the storage and loss moduli of the blend components
respectively. The storage modulus of PA6 is lower than those of the dispersed phases PP
and PP-MA over the frequency range under investigation. The lower relaxation time of
the PA6 is significant to the study since it minimizes the effects of the relaxation of the
matrix. It should be noted that the relaxation time of the dispersed phases are almost
same in the non-reactive and reactive systems.

Figures 3.4a and 3.4b compare the storage and loss moduli for the non-reactive
blends respectively. The storage moduli curves are characterized by a distinct plateau.
This plateau becomes more pronounced as volume fraction of the dispersed phase
increases. This is in accordance with the predictions of the model [ Graebling et. al.,
1990 ]. Figures 3.5a and 3.5b show a similar comparison for reactive systems. These
blends are also characterized by a plateau in storage moduli curves which occurs at a
lower frequency compared to the non-reactive blends. The full width of this plateau has
not been captured due to limits of the instrument. According to the model, the plateau
moves to lower frequency if
0 the interfacial tension is reduced

0 viscosity ratio is increased

26

The effect of the viscosity ratio ‘k’ ( from 1.7 in the non-reactive system to 0.7 in
reactive system) should be to shift the plateau to higher frequencies. However, the shift
is toward the lower frequency. Thus, the change in position can be attributed to a
reduction in equilibrium interfacial tension ( due to interfacial reaction ). Before
proceeding to quantify the extent of change in equilibrium interfacial tension, it is in
order to observe the rheological behavior of the blend with respect to its components.

Figures 3.6 is a comparison of the storage moduli curves of the PA6/ PP (90/ 10)
blend with its components. The storage modulus of the blend follows the trend of the
dispersed phase till a certain frequency after which it falls between them. This effect is
due to the interfacial tension forces. At lower frequencies the relaxation of the dispersed
droplets dominate the rheological behavior. The shape retaining interfacial tension forces
dominate the viscous forces. At higher frequencies the effect of equilibrium interfacial
tension is reduced and the visco-elastic properties of the components, especially the
matrix dominate the behavior. Similar behavior was observed for PA6/ PP (80/ 20) and
PA6/ PP (70/ 30) blends as shown in Figures 3.7 and 3.8 respectively.

In the case of PA6/ PP-MA (90/ 10) blend the storage modulus follows the
dispersed phase behavior till a certain frequency after which it is higher than the
components. Refer to Figures 3.9. But, as the weight fraction of the dispersed phase was
raised (to 80/ 20 and 70/ 30 respectively), the behavior changed. The storage modulus of
the blend was consistently higher than the components over the complete frequency range
under investigation. This is shown in Figures 3.10 and 3.11 respectively. This
observation shows that there is a fundamental change brought about in the rheological

behavior of the blend as a result of the interfacial reaction. As the reaction proceeds, the

27

‘bare interface’ is being ‘occupied’ by the reacted moiety. The interfacial reaction leads
to the formation of a product located at the interface which has a significant effect on the
rheological behavior of the blend In PA6/PP-MA (90/ 10) system the observations show
that the process (of imparting elasticity) has just started. To sum up,

0 In non-reactive systems G’b > G’d at low a), and G’b < G’d at high to.

o In reactive systems, G, > G’d over the complete frequency range under investigation.

At this stage two important questions need to be answered.
1. What is the extent of the reduction in interfacial tension due to reaction, if any?
2. How does one explain the increased storage modulus in reactive blends over the
complete frequency range?
To determine the solution to these, one needs to determine the value of equilibrium

interfacial tension.

3.3.3 Interfacial Tension Values

The thermodynamic models which explain the behavior of the interface can be
used for providing the initial estimates of the values. The most notable work in this
direction is due to Helfand and Tagami [ 1971 ]. According to these researchers the
following relation can be used for determining the equilibrium interfacial tension between

two asymmetric polymer melts.

28

I... = mmnyzv +4. +106. —r.> ]

2 6 tin/3. (3'13)

The parameter x is estimated from the Hilderbrand solubility parameters. The relation is

 

(6,, - (5,)2 (3.14)

The temperature dependence of x in Equation 3.14 leads to the temperature dependence
of I”. A major limitation of this equation is the lack of data on x in the literature. The
values for the materials used in this work have been shown in Table 3.4 [ Brandrup and
Immergut, 1989 ]. These values yield an equilibrium interfacial tension value of 28 mN/
m.

Another independent estimate for the value of interfacial tension can be made
from the polar and dispersive components of the individual phase. For nylon 6 and

polypropylene, the,values are listed in Table 3.5 [ Wu, 1987 ]. The equation used to

estimate the equilibrium interfacial tension is

4n" 1“," 4I“1”I‘,”
1“," + r,” r,” + 1‘,”

 

r,, = r,° +r,° — (3.15)

This yields a value of 10 mN/m. The two values have a difference of ~1 50%. Which of

them is correct? In Chapter 4 it has been shown that the equilibrium interfacial tension

29
between nylon 6 and polypropylene is 8 mN/ m. This value agrees well with the value of

10 mN/ m estimated from the polar and dispersive components of the individual phases.

In this work rheological technique was employed to determine the value of
equilibrium interfacial tension in the system under investigation. Equation 3.7 and 3.8
were used. Storage moduli curves were used for this purpose. As discussed in the
Introduction section, these are most sensitive to the changes in the rheology brought
about by the effect of interfacial tension and morphological parameters, especially in the
low frequency region. The model curves were generated from the experimentally
obtained storage and loss moduli curves for the blend components. The morphological
observations were made directly from the micrographs. The curves generated from the
model were matched with the experimentally obtained curves. The ratio (1'0/ R, was
used as the variable to fit the model curve to the experimental curve. The secondary
plateau and the frequency region below it were the main focus of attention while varying
the ratio.

The Figures 3.12 and 3.13 show a comparison between the model and the
experimentally obtained G’ and G” curves based on the models of Oldroyd and of Choi
and Schowalter for PA6/ PP(90/10) blend. Since the models were developed for
emulsion of Newtonian fluids, the model curves are characterized by a single transition
only. Thus, the limited use of these models. Table 3.6 shows the values of the relaxation
time it, and the retardation time A? from the relations of Table 3.1 for the two models.

Since these models are incomplete, there is a need to use Palierne’s model which

accounts for the viscoelastic properties of the components.

30
Figure 3.14 shows the result of such a fit for PA6/ PP (70/ 30) by using Palierne’s

model. A value of 4 mN/m for equilibrium interfacial tension yields a good fit between
the models and the experimentally obtained curves. This value gave good results for
PA6/ PP (80/ 20) and PA6/ PP (90/ 10) too, as shown in Figures 3.15 and 3.16.
However, in the 90/ 10 system there was a discrepancy in the lower frequency region.
The model and experimental curves do not match well in this region. A similar
observation was made by Graebling and co-workers [ 1993 ]. This can be attributed to
the polydispersity in the particle size of the dispersed phase. The relaxation time of the
monodisperse emulsion increases with particle radius. Due to the polydispersity in
particle sizes, there is a resultant dispersity in the relaxation times, which leads to this
discrepancy. Such effects would be most pronounced toward low frequency region where
the long time relaxation processes are dominant.

The results thus far indicate that the interfacial tension between nylon 6 and non-
reactive polypropylene under investigation is 4 mN/ m. This value is 50% lower than a
similar reported value of 8 mN/ m in nylon 6 - polypropylene system determined by
similar technique as shown in Chapter 4. What is the cause of this difference? The main
difference between the two studies is in the nature of the nylon 6. Although both have the
same molecular weight (M, of 18,000), but the nylon 6 in this study contained low
molecular weight lubricating agents. Due to thermodynamic considerations these have a
tendency to migrate to the interface region. As a result they have the potential of acting
as surfactants and hence reduce the equilibrium interfacial tension. This is supported by

the theory presented by Broseta and co-workers [1990 ]. They showed on theoretical

31

grounds that the equilibrium interfacial tension is lowered by the presence of small chains

at the interface according to Equation 3.16.
72,2
FOzF£[l-—+....:l (3.16)

In this equation wn signifies the degree of incompatibility.

Now, what happens when the system is reactive? Figure 3.17 shows the fit
between the model curve and the experimentally obtained storage moduli curve for PA6/
PP-MA (90/ 10) system. A value of 1 mN/m for I“ was used. There is an acceptable fit
in the low frequency region, while in the high frequency region, the model curve falls
below the experimentally obtained curve. The situation deteriorates further in the case of
higher volume fractions. Figure 3.18 shows the case for PA6/ PP-MA (80/ 20). There is
no match even for values as low as 0.1 mN/ m. Similar observation was made in PA6/
PP-MA (70/ 30) blend. This means that at this stage, an additional phenomenon besides
the role of equilibrium interfacial tension reduction seems to be coming into play. There
is an enhanced elasticity in the system due to reaction. This additional elasticity is
volume fraction dependent as well as frequency dependent. The model (using the
equilibrium interfacial tension alone) provides lower values of the moduli as compared to
the experimentally obtained curves. The enhanced elasticity seems to be playing an
increasingly important role in the reactive systems and needs to be accounted for in the
model. This needs a closer look and understanding. It is worthwhile to focus on the

physical events occurring in the blends to understand the phenomenon.

32

3.3.4 Physical Phenomenon

The position of the secondary plateau is an important indicator of the interfacial
events in the polymer blend. But, what is the physical process that leads to this plateau?
The physical events in this region are an interplay of the interfacial forces and the viscous
forces. In the low-frequency region of the dynamic behavior of the polymer melts the
long-time relaxation processes dominate. The mechanical relaxation of the droplets after
deformation is one such phenomenon. Interfacial tension plays an important role in this
behavior. It has been shown that the time required for the deformed droplets to return to
their original shape is of the same order as the mechanical relaxation times [ Scholz et.
al., 1989 ]. The parameter it], in Figure 3.1 is the shortest relaxation time of the emulsion
corresponding to the relaxation of the droplets back to the original spherical shape.
Below this frequency (i.e. higher relaxation times) the interfacial forces dominate the
viscous forces. The long time relaxation processes dominate in the terminal region.

In the region TD to M, the interfacial forces and viscous forces are of the same
order. The shape deforming viscous energy is being spent in overcoming the resistance
offered by the shape retaining equilibrium interfacial tension. This leads to time-scales of
relaxation that result in a secondary plateau in the storage modulus curve which lead to a
secondary plateau. A similar phenomenon has been observed in dispersed systems where
the energy is spent in overcoming Van der Waals kind of forces. These forces cause a
yield stress kind of phenomenon [ Matsurnoto et. al., 197 5 ]. The equilibrium interfacial

tension causes a similar resistance. It should be reminded that a well-defined plateau as

33
per the model occurs in the blend if the component phases were assumed to be ideal

Maxwellian elements with a single relaxation time. However, in real systems, there is a
distribution of relaxation times.

Beyond 2... there is enough energy in the system to overcome the interfacial
tension resistance and flow of the materials start. The role of long time relaxation
processes is reduced. The short time relaxation processes start playing an increasing role,
as in the transition zone. Effects due to equilibrium interfacial tension alone do not fall in
this category.

The discussion presented till now is valid for a ‘bare interface’. In such a
situation, the interfacial force competes with viscous forces. The interfacial tension
forces are well-defined as the interface between the matrix and the dispersed phase is too.
However, if the interface is ‘occupied’, say, due to the products of the interfacial reaction
then the effects due to this region will also participate in the inter-play of forces. The
interface is not demarcated as sharply. It has a finite thickness and is occupied by a new
product (of reaction). This should contribute to the rheological and morphological
behavior of the blend. The extent of impact should depend on the ‘extent of coverage’ of
the interface, i.e., how much product is at the interface. This is supported in the
rheological observations made for the reactive system. As the volume fraction of the
dispersed phase is increased, the deviation from the model predictions also increase.
O’Shaughnnesy and Sawhney [ 1996 ] have shown theoretically that after a critical extent
of the coverage of the interface, the reaction is ‘switched-ofl’ . They showed that as the

reaction products crowd the interface, the interface thickness becomes larger than the

34

unperturbed chain dimensions. The interface can no longer be considered ‘uni-
molecular’ layer thick.

It has been shown by F ayt and co-workers [ 1986 ] that in the externally
compatibilized systems the copolymer resides at the interface. This is a physical layer
with its own associated visco-elastic properties. It has a characteristic relaxation time
(and spectrum) of its own which enhances elasticity. Thus, it provides resistance to
deformation. We believe that in the system under investigation this phenomenon is
occurring. The layer around the dispersed phase in reactive blends acts as a reinforcing
agent which supports the stress transfer mechanism. This leads to a good stress transfer
from the matrix to the particles which increases the elasticity of the system and hence the
storage modulus. On the other hand, in non-reactive blends no such layer is present to
offer additional resistance which maintains the storage moduli values within those
estimated by the models developed for ‘bare’ interface.

In the micrographs for non-reactive blends (refer to Figure 3.2a through 3.2c), it is
clear that there is a lack of good adhesion between the spherical particles and the matrix.
This causes poor stress transfer from the matrix to the particles. In reactive blends
(Figures 3.2d through 3.2f), the adhesion is improved. There are physical and chemical
links between the two phases due to interfacial reaction. The properties of the graft
copolymer layer govern the behavior of the blend. The interface cannot be treated as
‘bare’ anymore. It is occupied by the reacted moiety.

But, how does one quantify the enhanced elasticity? The following section
addresses this issue by incorporating the surface shear modulus of the interface [ Palieme,

1990 ]. This leads to the resistance to the deformation of the interface. In other words,

35

this is a cause for the additional elasticity which is seen in the storage moduli of the
reactive blends. To use the surface shear modulus, estimates of this value had to be made

as direct measurements of this property are not possible yet.

3.3.5 Estimates of surface shear modulus

A parallel was drawn between the reactive blends and lightly cross-linked rubbers.

In this formalism [ Perry, 1961 ] (refer to Figure 3.20),

.5; = .3; +135; (3-17)
For or}.B <1
fl; =fl0
.. 3.18
r. = 13.022. ( ’
For (01191
a. = fl." = [3. Jen/1,. (3.19)

The Equations 3.17 thru 3.19 show that two parameters are crucial - Bo and 7‘3- In the

theroy of rubber elasticity, Bo (N/mz) is the equilibrium modulus in the range of

36
infinitesimal deformations. To draw a parallel in case of the modulus associated with the

interface (N/m), it can be thought of as a product of a bulk modulus and a characteristic
length. This bulk modulus could be different from either of the individual components
and also different in each blend depending on the volume fraction. On the other hand, the
choice for the characteristic length falls clearly on the interfacial thickness. Experimental
evidence suggest, the interfacial thickness attains a constant value in the early stages of
mixing [ Okarnoto et. al., 1993 ]. Work by Hosoda and co-workers [ 1991 ] shows that
the interfacial thickness is ~50 °A (50 x 10"0 m). Increasing extents of reaction increases
the amount of reactive copolymer in the interface region. That is, an effect of increased
reaction should be enhanced elasticity. In fact, this is what is observed. The deviation
from the base model increases as the volume fraction of the maleated polypropylene is
increased.

On the other hand, TB is a characteristic relaxation time. It corresponds to a
frequency until which the elastic recoil is accomplished after the removal of stress
[ Ferry, 1961 ]. In the theory, 2.” refers to the longest relaxation time possible in the
system. The parameter 1,, is a good candidate for this (refer to Figure 3.1). In Figure 3.8
it is seen that at a frequency of ~0.2 rad/s the secondary plateau ends and the effect of
equilibrium interfacial tension starts to diminish (the region after it? in Figure 3.1). This
yields a value of 21, to be 5 s.

The result of incorporating a surface shear modulus is a change in the form of

Hi(w) of Equation 3.8. The new equation assumes the form [ Palieme, 1990 ]:

37
0 t O t C O t O 0 2 O ‘ t
(41‘ lR-XZG +50 )+(G —G )(16G +190 )+(16,6 F /R )+(2,6 /R)(l3G +8G )
1 m d d m m d s s d m
o i O O t O
(401‘ /R))(Gm +Gd)+(26d +3Gm)(l6Gm

 

Hi(w) = t t 0 2 O O O
+19Gd)+(32/331‘ IR )+(4,134r IRXlBGd +126m)

(3.20)

Figures 3.19 and 3.20 show the result of incorporating a value of [30 for the 80/ 20 and 90/
10 reactive systems. For 70/ 30 reactive system this parameter did not improve the
situation significantly. It should be reminded that these values are approximations only.
There are no experimental values available in the literature at this stage. If the interfacial
thickness is assumed to be ~50 °A, these values of [30 yield a bulk modulus of ~10‘5 N/m.
Table 3.7 shows the results of the values of the equilibrium interfacial tension and the
estimated values of the surface shear modulus.

Now, Wu has proposed a relation based on empirical grounds which can
be used to make estimates of the particle sizes if the properties of the materials and the

processing conditions are known [ Wu, 1987 ]. It is shown in Equation 3.21.

”"1761 =4p1084

1., (3.21)

38

If this equation is used to determine the ratio of particle sizes in non-reactive and reactive
systems with the same matrix properties and processing conditions the relation obtained

is

 

 

d p034 F0
21—”: f0'84[ ) (3.22)

The subscript u and r signify non-reactive and reactive systems respectively. The ratio
du/d, from the above relation is 3.00. However, the actual ratios vary from 8 to 24
depending on the volume fraction. Thus, there are effects other than reduction in
equilibrium interfacial tension which play a role in development of morphology of
polymeric blends [ Sundraraj and Macosko, 1993; O’Shaughnessy and Sawhney, 1996 ].
Also, it is interesting to observe the behavior of the blend as per the theoretical
description in the Newtonian limit. As suggested by Graebling and co-workers [ 1993 ],

in the Newtonian limit, the following approximations are useful.

 

 

 

~ Rum) (19k +16)((2k + 3) -2¢(k — 1))

’1” ”(41"0 [ 10(k+1)—2¢(5k+2) 1 (3‘22)
° 1

G” T (2072—45) [(2]: + 3) - 2¢(k —1)]2 (3°23)

39

 

1%. =3§lg(k.x,¢) (3.24)
where,
k X _ 3(1- ¢)(1— X) ((2k + 3) + 3¢(k —1))((2k + 3X) — 2¢(k - X))]
g( ’ ’¢) ' (2k+3)—2¢(k—1) + ((2k+3)—2¢(k—1))2

| (3.25)

Based on these equations and the data in Table 3.2 and 3.3, the XD, 7.? and GP obtained are
presented in Table 3.8.

The discussion above shows that interfacial tension reaction leads to a reduction
in interfacial tension and is accompanied by an enhancement in elasticity. Based on the
estimates in the value of [30, the elasticity enhancement is increased as the volume of the
dispersed phase is increased. This is similar to the phenomenon compatibilized by
external agents. Okamoto and co-workers [ 1993 ] postulated that increasing the extent of
reaction leads to the accumulation of the products at the interface. The observations in
this work show that this seems to be the case and that it results in an enhancement in
elasticity. Also, the role of the equilibrium interfacial tension is decreased once this
‘finite thickness’ layer is formed around the dispersed phase. An interesting question that
arises then is that what happens to the interfacial properties as the reactivity of the system
is progressively altered. Does the interfacial tension go down in steps as the extent of
reaction is increased? Or does it depend on the reactivity of the system? The results of

such a study have been reported in Chapter 4.

40
CONCLUSIONS

It has been shown in this chapter that the interfacial reaction between nylon 6 and
polypropylene leads to a reduction in the equilibrium interfacial tension. The value of
interfacial tension drops from 4 mN/ m in non-reactive system to l mN/ m in reactive
system. This reduction in equilibrium interfacial tension is accompanied by an
enhancement in elasticity in the reactive blends. In addition, it is observed that the
presence of the low molecular weight lubricating agents reduces the interfacial tension
due to their presence at the interface as compared to the similar system without any
agents.

Observations show that the equilibrium interfacial tension alone is insufficient to
account for the rheological behavior over the complete frequency range. Toward the
lower frequencies and until a certain volume fraction, the results are in agreement with
the model while in the higher frequency range the agreement is not good. A possible
cause of this behavior is that with the progress of the reaction, the interface is being
‘occupied’ with the reaction product which imparts additional elasticity to the system.

This has been accounted for by considering surface shear modulus in addition to the

interfacial tension.

. 41
Table 3.1: The comparison between the models of Oldroyd and of Choi and Schowalter

for an emulsion of Newtonian fluids [ Graebling and Muller, 1990 ].

 

 

 

Oldroyd Choi and Schowalter
'1' (5k + 2) 2 (5k + 2)2 (5k + 2) 2 5(51 + 2)2
I + ¢——- + —— 1+ — + ——
21!: +1) 10(k +112 2(k +1) so. + 1)2
7" [ (19k +16) ] [ 5(19k +16) ]
20 1+ ¢——— 20 1 + ¢——-—
5(k + l)(2k + 3) 4(k + 1)(2k + 3)
4’2 3(l9k +16) 3(191 +16)
20 1— ¢ 210 1+ ¢--———
10(k + l)(2k + 3) 4(k + 1)(2k + 3)

 

M (nmR)[(19k +16)(2k + 3)]
r0 400: +1)

 

42

Table 3.2: The volume average radii of different blends.

 

 

Material RV
(um)
PA6/ PP (90/ 10) 4
PA6/ PP (80/ 20) 8
PA6/ PP (70/ 30) 12
PA6/ PP-MA (90/ 10) 0.5
PA6/ PP-MA (80/ 20) 0.5
PA6/ PP-MA (70/ 30) 0.5

 

Table 3.3: The zero-shear viscosities and the corresponding relaxation times of the

components of the blend.

 

 

Material no Relaxation time, A
( Pa-s) (s)
PA6 690 0.01
PP l l 50 0.3 5

PP-MA 490 0.28

 

43

Table 3.4: The data used for determination of interfacial tension from thermodynamics.

 

 

6, pi x 10 'fi 131 x 10'”
(cal/cc)"2 (monomer/cc)
PA6 13.6 6.06 6.25
PP 8.3 8.9 6.67

 

Table 3.5: The polar and dispersive component of the component phases [Wu, 1987;
Paul, 1978].

 

 

1“ 1"F 1'"
(mN/ m) (mN/ m) (mN/ m)
PA6 29.6 9.9 19.7

PP 21.0 0 21.0

 

44

Table 3.6: A comparison of the parameters of the models of Oldroyd and of Choi and

Schowalter for emulsion of Newtonian fluids.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Oldroyd Choi and Schowalter
A'1 A? ‘1. A'0 A] 2’2 T111
lo
PA6/PP (90/ 10) 1.9 2.0 1.8 865 1.9 2.8 2.0 944
PA6/PP (80/20) 3.9 4.5 3.1 1070 3.9 7.3 4.4 1380
PA6/PP (70/30) 5.9 7.0 4.0 1305 5.86 13.3 7.5 2010
PA6/PP-MA (90/ 10) 0.6 0.7 0.6 835 0.6 1.0 0.7 890
PA6/PP-MA (80/20) 0.6 0.8 0.5 1000 0.6 1.4 0.8 1220
PA6/PP-MA (70/30) 0.6 0.8 0.4 1187 0.6 1.8 0.9 1680

Table 3.7: The interfacial tension values for different blends.

 

 

Material I” [30 T

(mN/ m) (mN/m) (s)

PA6/ PP (90/ 10) 4 0 0.3
PA6/ PP (80/ 20) 4 O 1.3
PA6/ PP (70/ 30) 4 0 3.0
PA6/ PP-MA (90/ 10) 1 0.5 0.2
PA6/ PP-MA (80/ 20) 1 0.5 0.3
PA6/ PP-MA (70/ 30) 1 0.5 0.5

 

45

Table 3.8: The values of TD, 8,, and GP based on the Newtonian limit. The value of

interfacial tension used is 4 mN/m for non-reactive and 1 mN/ m for reactive blend.

 

 

A6 M G,
(S) (8) (Pa)

PA6/PP (90/10) 2 0.77 63
PA6/PP (80/20) 4.5 1.0 67
PA6/PP (70/ 30) 7.5 1.2 70
PA6/PP-MA (90/10) 0.72 0.33 238
PA6/PP-MA(80/20) 0.8 0.31 462
PA6/PP-MA(70/30 0.9 0.3 672

 

46

 
    

 

 

 

 

’3 . : """ 5 " '
e’ i E :
52 a a a
c: i E 5
51/16 5 10., 5 10,,
Frequency, (rad/s) 5

 

Figure 3.1: The storage and loss moduli for a blend of Newtonian and viscoelastic fluids.
Notice the appearance of a secondary plateau in viscoelastic blend which is a result of

interfacial effects [ Graebling et. al., 1993 ].

47

 

 

Figure 3.2a: Micrograph showing the morphology of B3 S/ PP (90/ 10).

48

 

Figure 3.2b: Micrograph showing the morphology of 338/ PP (80/ 20).

49

 

Figure 3.2c: Micrograph showing the morphology of B3S/ PP (70/ 30).

50

 

 

Figure 3.2d: Micrograph showing the morphology of B3S/ 3150 (90/ 10).

51

 

Figure 3.2c: Micrograph showing the morphology of B3S/ 3150 (80/ 20).

 

 

Figure 3.2f: Micrograph showing the morphology of B3S/ 3150 (70/ 30).

G', P:

53

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00E+05 .
100304 M__ # , G'(PA6). Pa
0 G‘(PF).Pa
_._G'(PP-MA).Pa
1.00803.~_ _ _ _.__ _, ‘ .
l
1
1.005402% __ , p”
l .
I . ' I)"
- 0
1.00901 .2 _ __ -_1_._._
Q I
o | r.
0 I
I’- ' 1
1.005100... _ 7.x 1 ‘ ,3 j 1
:' 1‘ ' i
‘. ‘ . 1
1.00501 ._ ._ A [An _ 1 _ L r.—
l
. 1
1.00502 .i i .
1.00502 1.005-01 1.00E-t00 1.00E+01
Freq,radls

1.005102

Figure 3.3a: Comparison of the storage moduli of the components of the blend.

 

 

 

 

 

 

 

 

 

 

 

 

 

1.005105 , 1
, G"(PA6). Pa 5 ‘
. G'<H=1.Pa ! ‘ . 3.
100154041 —'—G'(Fp'm)'Pa - --*_ r". -
' 7’7 E ‘_ l ." -
i l . :Iffl
.2 ‘
l . A
o . ‘
. l O ‘
°' 10015103 * . ° 1 ‘ r/
.- ' ‘*’”'—“*"’4' o [T 7
0 ‘ . O ‘ 1);
l . 0 ‘II’.
‘ . ‘ 1 ’ .
l O . . 5,".7- i ii
1 0 ,
1.005102 ..-..- -1 _ ..7L_:;.H'} HL___ 1
o . 3"] i
I
,1 1 l
i’ i ’
[ 1
1.001301 , 1
1.00502 1.00501 1.005100 1.001301 1.005102

Freq, radls

Figure 3.3b: Comparison of the loss moduli of the components of the blend.

55

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100504 . T 1
’ l
1 1 . r‘
, 6(8381FP(90/10)). Pa 1 1 . l’"
. G(B:33/PP(80/20)). Pa 1 j",
_._ G‘ (333/ P (70/ 30)). Pa ‘ 3"
1.00803 4»——___ ..__ _. .’;L
l ‘ . . .
1 . :z! |
1 '1'" l
o :I- ‘ 1
‘ . 9 .I A ‘
a l A
0; 1.00502 ._____ 2,2,. 5, 5,. _ ._, .1577:
.0 ,I ‘ l
1 ,r' .‘ 1
x: . 1
I. . ‘
r1 - , 1
I . ‘ 1
1.00501 .___,__. '1' l , ”L“ f
o ‘ ‘ ‘ a [
A 1
1 1 [
‘ 1
. 1 1
1.00500 L , ,
1.00502 1.00501 1.00500 1.00501 1.00502
Freq, rad/c

Figure 3.4a: Comparison of the storage moduli of the non-reactive blends for varying
weight fractions.

56

 

 

 

 

 

 

 

 

 

1.00505 1 1 1
, G" (3331143501 10)). Pa 1 l ,
. G" (1333/ PP(BO/ 20)), Pa 1 I , , 3"
1.00504 .._ -... G“(B3$/PP(7OI30)), Pa 1 . g:
1 ; l/t’
1 g]:
1 l j I l
1 L j}
r .3 1
.. 1.00503. _____ E .4 E E g,"
i9 1‘ . :1-
1 . r
1 . - r 1
1 . 0,11’ ‘ ‘
I
1 na’
1 .1" l 1
1.00502 5... 2 -_ __-2:’ 1 ,
’A
:4 l 1 1
A l l “
1 1 1
1.00501 1 l 1
1.00502 1.00501 1.00500 1.00501 1.00502
Freq, radls

Figure 3.4b: Comparison of the loss moduli of the non-reactive blends for varying
weight fractions.

 

 

 

1.00505 1
, G' (333/ 3150 (90/ 10)). Pa
. G‘ (838/ 3150 (80/ 20)). Pa

1 00504 ._
_._G' (838/ 3150 (70/ 30)). Pa

 

1
1.00503 ...;_2 __ 214% __

 

'x:

 

 

 

l l’ ’ ‘

1.005.024—g—#_—-+_ -’ e g A
III’ 0 Al
i .I-’ . . A l
..r" . 1
I’ o ‘
O . ‘

1.00801 .5____~ ‘

 

 

 

 

 

 

1.005101 1.00802

Freq, rad/s

Figure 3.5a: Comparison of the storage moduli of the reactive blends for varying weight
fractions.

58

 

 

 

 

 

 

 

 

 

1.00505 ; ,
1 l 1
l .:
, G"(B3813150(90/10)). Pa ‘ 3 t-
. 61(83813150(80/20)), Pa If!"
1.00504 ...5_2_ _._ G" (833/ 3150 (70130)). Pa
1 1 (1")
1 ' f-
1 l U’!’ y
1 1 II!" 1
.. ,.«.“- l
“1 1.00503 .-.-.. _. _ A E -L .1: 2‘1 4
6 1 .2“: 1 1
1 _,4’: ‘ 1
l l’o I ‘ l [
(1" ‘ 1 ‘
.r 1‘ , 1
1.00502 ...__.-E _ ,_./_._:n1__ A]
: ‘ ' l
1 1 1
1 1
1 1 1
l .
1.00501 1 l 1
1.00502 1.00501 1.00500 1 00501
Freq, rad/s

1.00E+02

Figure 3.5b: Comparison of the loss moduli of the reactive blends for varying weight

fractions.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00505 1 1
1 1
1 1
. 6(838). Pa ' ,
1.005004 .._.___ ___ e
o 6' (PP). Pa . . ’
_._G'(B3$IPP(90I10)). Pa , . ‘ .
. 0
1 1 o ' . ‘
1.005031.#__.5__.,_ -,1.,m__._._- .._1_ . ' if?
1 1 . o . ; 1 ‘
1 1 .. II,”- . 1
o ‘ ‘ 1
1.00502 .____.m-,__1n__ .91," .
I 1 O I’- A 1
o. 1 . :III" 1 ‘ ‘
ED 1 . . l’ 1 A 1
1 II A ’
1.00501 .a___ a _ .55...- -__.___. 1_- 1-
£5" 11 1
‘1', A ‘
i, 1 ‘ 1 1
1005001555 _ 51* _.A 1__._51_..w 1
1 ‘ 1 1 1
. 1 1 1
1.00501 .0._ - .- .551- a f - , .1_.____ -
A 1 1 1
1
1 1
1.00502 1 1 fi1
1.00502 1.00501 1.00500 1.00501 1.00502
Froq.radls

Figure 3.6: Comparison of the storage moduli of the PA6/ PP (90/ 10) blend with the
components.

60
1.00E+05

 

1

1
1
1.00504 ._ . G'(B$S).Pa L .9-
. 61141.5 1
_._ G‘ (838/ PP(80/ 20)), Pa '

1.005003 ..___. _ ..

 

 

      

 

 

 

1.00602 ._ -

 

 

6', Pa

1.00E+01 __-__ _,

 

 

1.008011__ _. ,

 

 

 

 

1.00502 1.00501 1.00600 1.00801 1.00E+02
Freq. radio

Figure 3.7: Comparison of the storage moduli of the PA6/ PP (80/ 20) blend with the
components.

1 .00E+05

61

 

1 .00E-0-03

A

_._ G‘ (838/ PP(70/ 30)). Pa

1
G‘ (33S), Pa
6' (m. Pa

 

 

 

 

 

 

1.00502 .___r -_ _ éfl—f
1 .
O
O

 

 

 

 

 

 

 

 

 

 

 

 

. 1
°' 1
E9 fr“. . ' 0
1.006001.___# _" 1 o ‘ ‘
o . ‘ 1
o 1 ‘ '
. A
o 1 .
1 DOE-+00 .___ - fl- 1.— ‘ 1
1 A 1
A 1
A ‘ 1
1.00501 .___i -_ fix 1
A 1 1
1.00802 1 1
1 .005-02 1 00501 1.00E+00 1.00501
’ Freq, radio

1 .00E+02

Figure 3.8: Comparison of the storage moduli of the PA6/ PP (70/ 30) blend with the

components.

62

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00505 1
1.00504 ...—5 . G(B3S). Pa
. 6(3150). Pa
_._ G‘ (338/ 3150(90/10)). Pa
1.00503 .__.___ __.__fl1 1
1 1
1 1
1.00502 ‘ 1
1 ,I"
a ." 1
.- 1 .I
0 1 .r 1
1.00801 4____7)_dfi_ 11 .1" 4
1 r" 1
I! ‘ ‘
I . ‘
1.00500 .____._-.' a ' 1 _‘ 1
O
1 A
1 .
1.00501 ..___5_ j 1 1
. 1
1
1.00502 1 , ,
1.00502 1.00501 1.00500 1.00501 1.00502
Froq.radls

Figure 3.9: Comparison of the storage moduli of the PA6/ PP-MA (90/ 10) blend with the
components.

63

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00505 1 1
1 1 1 ..
100E404...______ ‘ G(B3$).Pa 1 1
. 6(3150). Pa 1 , g 2
_._G'(BSSI3150(80/20)),Pa 1 1 . .1.
1.00903 .___.-.._ _ 5% _ 1 .J’.’ ' o . ‘ ‘
1 1 If . o . ‘ ‘
1 I . ‘
1 . . 1
100502 1 ”,1”. . 1 1
- W~———-— — —~ *1; — ,u’". .. e ‘—a
a ' ' 1 1 1
'. 1 I), ’ . ‘ 1
6 100501 _ 17.1”" °. 1. ‘ 1
. - — A- ':I . . I ‘ g 1
1 . A ‘ 1
. 9 1 1
1.00E+00 .-_ + g W .,_--1.___._ T‘ h__ ;__ 1
. 1 A 1 1
1 1 ‘ 1 1
1.00501 .____r _._.-_‘_1..;;_. E ‘. . 1 1
. 1 1
1 1 1
1.00502 1 1 1
1.005-02 1.00501 1 DOB-00 1.008101
Freq, radio

1.0051102

Figure 3.10: Comparison of the storage moduli of the PA6/ PP-MA (80/ 20) blend with

the components.

 

 

 

 

 

 

 

 

 

 

 

 

1.00E+O5 1 1
1 1
‘ G'(B3S),Pa
1.00304 ._# . G' (3150). Pa
_._G' (838/ 3150 (70/ 30)), Pa
1 1
1.005403 .____ a __ __s _. i 1
1
1 {if-
1.00E+02 l? _ - -1 _, __._.¢.___1,
1 .’.’ .1
a 1 r" . ' 1
6 .’.’ 1 . . l
1.00E+O1 ‘_______ _- .. -.. . _ .-. f .
1 . ‘ 1 _.
I Q ‘ 1 1
° . 1 1
.1 A 1 1
100300._____ __._.-';-a_ ‘ A
o ‘ 1
A
A 1
A 1
‘. 1 1
1.00501 .._.. ___ W ., _‘_‘.1 _7 _ fl __4._ ,_ 1_fi
A 1 1
1
1.00502 1 J .
1.00502 1.00501 1.00800 1.00E+01 1.008002
Fteq,radle

Figure 3.11: Comparison of the storage moduli of the PA6/ PP-MA (70/ 30) blend with
the components.

1 DOE-005

65

 

1.00E+04 1___

 

A

G’(model).Pa —

G" (model). Pa

—-— 6' (exp). Pa

 

 

 

 

 

 

 

 

 

1.00303 . 1
«- 1
“- 1
‘° 1
1.00E-I-02. -..—“Wu... . 1
r {.fi A A A A A 1 A
:I. ‘{;’-/‘AL AA AAAATAA AAAA
1 ‘ Ir 1
. 1,-’ 1
I 1 1
A ’1
1.00E+01 1h___ _ ‘ ';1 1 1
I, 1 1 1
1 1 1
1 1 1
1.001500 1 1 1
1.00502 1.00501 1 00300 1 006001
Freq,radls

1.00E+02

Figure 3.12: Comparison of the storage and loss moduli values obtained from the Choi
and Schowalter model with that obtained experimentally for PA6/ PP (80/ 20) (1'0 = 4

mN/ m, R= 8 pm).

66

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00E+05 1 _I
1
1
1 1 1
1 O
1 1 1 . .
1 o :2/
1 1 . :z'
1.00E+04.u___ ‘ G‘(model),Pa ._-.._nfl_. '2'?”
. G“ (model). Pa 1 «‘1’ .l'
../
_._ 6' (exp), Pa 1 (o -’
_.._ G" (exp). Pa (3’
| 1 /3’
1 [.../u
1.005103 1*_u__n__ . m r;
V ,4; ,I’
' 1 /° °
1‘- ; .4 ‘
0 1 .4 ' 1
." o O
0 1 ’8’. I 1
1’: I’-
1.00E+02 .0..- _ i _ W (.3. 1
.ls/V1 {“A‘AAAAAAAAAAAAAAA
1 I’2 ‘ .
./A 1
.IA 1 1
I" 1 1
l1 1 1
1.005401 ._ _ i “4; 1 _- 1
l' 1 1 . 1
I ‘ 1 1 1
A 1 1
1 1
1 1
1.005100 1 . .
1.00502 1 DOE-01 1.005100 1.00801 1.00E+02
Freq,radls

Figure 3.13: Comparison of the storage

and loss moduli values obtained from the

Oldroyd model with that obtained experimentally for PA6/ PP (80/ 20) (1'0 = 4 mN/ m, R

= 8 pm).

67

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.005105 1 1 1
1
1 1 1
1 1 ’0’:
‘ G rmde , Pa . 0’
10054041 * 1 I) #_ 1 (:3,
f . 6‘ (exp). Pa 1'3” .
o “’./. A .
_._ 6' (model). Pa 1 ”30,. z .
_._ G" (exp). Pa 1 ('5? 3 3
i 1 ’Ojo/. ‘ 3
' 1 /:/° . '
1.005103 t____ 7 , i. +_V_ ____ .;2/ w?
1 W . i
a ./:;.’ . O
:- 1 .,.$°/ o .
SD. 052;./ o 2 - 1
0 ’3’. o 2 ‘
1.005102 9..., * r :53’ e 2 t 1
#5:] . 0 ° 9. 1
‘o" 5 ° 1
_ 6
: . '
g
0 1
O A 1 1 1
1.00E+O1 4»—_.* -‘_.__‘ 1. _____ _
. 1 1
1.00300 1 , 1
1.00502 1.00501 1.0015100 1.00901 1.00502
Freq, radls

Figure 3.14: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP (70/ 30) (1"0 = 4 mN/ m, R = 12 um).

68

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00805 1
1
1 his”;
1 3’.
1.00804 .__ , G‘(model).Pa .__r__m 1 0’3’
0
. G" (exp). Pa 1 ’11" : .
...._ G" (model). Pa ’9". o :
I ’9. . ‘
_._G' (exp), Pa ’9, 2 r
’V
1 £73 2 2
1.005403 ___“ ,_*1 .#FC [$3 . i 1
C 9’8] . A
/ I 1
a 1 ./;I. o A 1
s‘ .5./ o A
o 1 /.§o . 0 A
." 0’. . A
e ’35. . ‘ ‘ 1
1005102 41,6" . 2 ‘ ;
. .._*, F7 _,,_/’fi 2:1 1
/°/. 0 2 1
’I' o
. 1 , 1
1 -’
o ‘ 1
Q A
1.00801.___-.____, #11 _____1 1
. . 1 1
O
‘ 1
‘ 1
1.00800 1 1 1
1.00502 1.00501 1.00800 1.00801 1.00802
Freq, radio

Figure 3.15: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP (80/ 20) (1") = 4 mN/ m, R = 8 pm).

69

 

 

 

 

 

 

 

 

 

 

1.00805 1 1 1
1 1 ’0‘:
6' model . Pa 1 . .43"
1.005104 ._ ‘ 1 1 _ - - ,3
. G'(exp). Pa 1 ’0': .
_g— G" (MEI), Pa ’3'” . 2 A
’0'
-... G" (exp). Pa 1 (0’ . g ‘
I 1 ’3’ o t ‘
100503 “_f __ n .. C -1... f. _ ___..___-1#’,x’ .1» .
g 1 ’91’. . ‘ 1
'1 1 ’9’. 1 . O ‘
I I. 1 . ‘ A
SD. 1 I”. 1 o . ‘
‘910015102. -0 _ I". 1.2‘ 1
- —-— -— — —1,;s~ ,1 1 1
Ii" ' 1
3 1 0 1
8’ 1 o
'. . 1
1 2 Q 1 1
1.00801 .C_, k t __1__ _:__ 1 1
: o
1 O A .
. A
o 1 1
1
" 1 1
1.00800 1 1 1
1.00502 1.005-01 1.00800 1.00801 1.00802
Freq, rad/s

Figure 3.16: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP (90/ 10) (11") = 4 mN/ m, R = 4 pm).

70

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00505 1 1
I 1
1 1
1 , 0’.
1 1 x’
1 .’./‘ .’./.
1.005104 4__ ‘ mm” Pa 1 Ali’.’ . '
. 6' (exp). Pa 1 I.) ”,0 . o
...... G" (model). Pa 1 I.) .I.’ o ‘ .
_._ G" (exp) Pa ./° .I” . ' ‘
I 1 ’0’ .I.’ '1 ‘ ‘
1.00903 1___ 1 ‘0’” ,o’“ . . . ‘
Or I. A
g 1 /°/././"’. 0 . A ‘
t' 1 o . o’.’ . a ‘
6 1 /./ ./ g A
o" O O .
o ././.’.’ ‘ ‘
./ ’ .
1.00E+02._ W _ (17. ‘ _
./ .x . 0
’0’1 A . 1
o 1 o 1
1 . 1 1
1 ° 1 1
1.00901 ‘____ # ,11 A, ___.3_-_ 1 1
1 2 ' 1
1 9 1
1 2
g
2
2
* 1 1
1.00900 1 1 1
1.00502 1 .00501 1 00900 1.00E+01 1.00E+02
Freq, radio

Figure 3.17: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP-MA (90/ 10)(T*1 = l mN/ m, R = 0.5 pm).

71

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00805 . 1
1 1 '
1 1 1
1 1 /°’.’.
1 1 o’.’. ’3’:
I
10015104 ._ ‘ 61m” P“ _ ._ __ _._1 Ii» (.34
. 6' (exp). Pa 1 l,/ ’3’.
_._ G" (model). Pa 1 ~’° ’0’: . ‘
' ./0/ .I. . ‘
_._G' (exp).Pa / ,I o .
0’. 0’ 0 I
1 /./ o", A ‘
. x" ,o’ 0 3
1.00803 .m__-.i# 1m I. (9 . A a 1
fl 1 ’0’ ’0’ O ‘ ‘
l ’0 .I. . ‘ A
. /° 0’ . A -
:0 1 /. ’0’ A
- _ . ./. .I. Q
0 / I ‘
1/’ .I. o 1
1.00802 “___” .4? ,I e +
/ 0’ .
. ./1 o ‘ 1 1
0’ 1 . A
2 ‘ 1 1
1 ° ‘
. a 1
g o 1 1
1.00801 ..___,__. _ ..2 m 1
, g 1 1
'1 1 1 1
1 1 1
1m 1 1 1
1.00502 - 1.00501 1.00800 1.00801 1.00802
Freq, radio

Figure 3.18: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP-MA (80/ 20) (1'0 = 1 mN/ m, R = 0.5 pm).

72

 

 

 

 

 

 

 

 

 

 

1.00805 . 1
’0
1 1 0”.’.
I 1 0”] l.’.
1 005.04 -1 ‘ 6‘ I). a -—---——~——v————_‘ 1———. —————~——gé.f.’. . .
O G (exp)! Pa . I.’ 0’ O
1 /. ’Il’ . ‘
_._ G" (MOI). pa 1 /./¢ .10 . o ‘ A
I /. O, A
—.—G' (exp), % I.’.’.’. '1 . ‘ A
1 o
1 ’./.’.I. . O A A
1.00803 ..__ i. - % 11min__ ___77 . ‘ 1
. ’0 ’0 . A 1
n' 1 ./° ./. 1
3" 1 ’0’ ’0’ . : ‘
ED" 1 0’. 0’. . ‘
(D 1 /°’ I" ‘ 1
o o 2 1
1.005102.._.-_. A- ___.1/ l.’ 1 g .
o 0
°/ I’ll o 91 1
. _ .
0’ 1 2 1
j . 1 1
1 1 ' 1 1
1.005101 ._~-_.__ ___- _. - 1
1 . 2 1
2 ‘ 1
.2 1
‘" 1 1
1.005100 1 1 fl
1.00502 1.005-01 1.00800 1.00801 1.00802
Freq, rad/s

Figure 3.19: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP-MA (90/ 10) (1"0 = 1 mN/ m, R = 0.5 um,
[30 = 0.5 mN/ m, (11)l3 = 0.2 rad/ s).

73

 

 

 

 

 

 

 

 

 

 

 

 

 

1.005405
1 1 1
1 .le
l 1 .l.’;: I: .
1 00804 . ‘ G (model). Pa 1 I.’;/." '
. 6‘ (exp), Pa 1 I,” 0,” A , ‘
o I A
_._ G" (MOI). pa 1 ’././.’./O 0 o ‘ .
—0— G" (9X9). Pa /°’.,0’. . o ‘ ‘
I . . ./. . ll
1 ’0’ l.’ . ‘ ‘
1.00E+03 ‘__._ .+ -.. 4._ ,- ,_‘__-_7¢_VO .
. I. ’11 ‘
I 1 ’0 /. O
a. /. I. C ‘
3" 1 ./. .x' o A
0- IO/ l.’ . ‘
ED /° .l' 0 ‘
/. .’ O ‘
100802 ._____,_7_ T.’ . A
. /.v 0" A
0 ./1 . t
0’ 1 2 A
l 2
1 . - 1
1 , ‘ ‘ 1
1.00E+01 ._ _ __21 1 1
. A 1 1 1
A 1 1 1
1 1
1005100 1 1 1
1.00502 1.00501 1.005000 1.00E+01 1.00E+02
Freq, radio

Figure 3.20: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for PA6/ PP-MA (80/ 20) (1" = 1 mN/ m, R = 0.5 pm,
Bo = 0.5 mN/ m, c0B = 0.2 rad/ s).

THE EFFECT OF PROGRESSIVE EXTENT OF INT ERFACIAL REACTION

ON THE RHEOLOGY AND MORPHOLOGY OF NYLON 6 - MALEATED

POLYPROPYLENE BLENDS

 

Chapter 4

 

4.1 INTRODUCTION:

Blending of polymers is a popular method of improving their end-use properties
[Paul et. al, 1988]. However, most polymers are thermodynamically incompatible which
leads to their phase separation on blending. Compatibilization of polymer blends is
carried out to reduce the degree of incompatibility and stabilize the system. This can be
achieved by addition of a pre-made block copolymer to the system or by carrying out an
in-situ reaction between the complementary groups of the blend components. Both
techniques promote dispersive mixing which leads to a reduced particle size.

Experimental evidence suggests that the addition of premade block copolymers
leads to a reduction of interfacial tension [Cho et. al., 1996; Elemans et. al., 1990]. In the
area of reactive blending experimental results showing the quantitative effects of
interfacial reaction on interfacial tension are limited. In a study on blending of nylon 6
with non-reactive and reactive rubbers Wu [1987] predicted that the interfacial tension
could drop from 8 mN/m in non-reactive system to 0.25 mN/m in reactive system. The
prediction was based on a statistical mechanical theory of polymer interfaces due to

Helfand and Tagami [1971]. The theory predicts the relation shown in Equation 4.1.

74

l
0 - 4.1
F at L ( )

where I’° is the equilibrium interfacial tension and L is the interfacial thickness. The
Helfand theory was developed for a ‘bare interface’. That is, an interface which has not
been occupied by a copolymer or a reacted moiety. On the basis of interfacial thickness

and interfacial tension in model experimental systems Wu arrived at the following

modified result.

11
L036

Foot

 

(4.2)

This empirical rule was then combined with measurements of interfacial thickness in
nylon 6 - reactive rubber blends to obtain a value of 0.25 mN/m for interfacial tension in
these systems. One of the aims of this study is to obtain values of interfacial tension

independent of thickness in reactively blended polymer systems.

An important effect of the interfacial reaction is the modification of the interface
from a ‘bare interface’ to an ‘occupied interface’. As the extent of interfacial reaction
increases, the interface is progressively occupied. Fayt et. al. [1989] have reported that
the addition of a block copolymer leads to a broader interface. The interface is not
unimolecular thick, but has a finite dimension. In a recent theoretical paper
O’Shaughnessy and Sawhney [1996] conclude that besides reducing interfacial tension,
an ‘occupied interface’ also suppresses coalescence. This is a direct result of steric
hindrance provided by the ‘occupied interface’ and leads to a stabilization of the

morphology. The experimental evidence presented by Sundraraj and Macosko [1995]

76

corroborates this conclusion. This study will investigate the effect of increased
occupation of the interface by successively increasing the extent of reaction. Thus, to

sum up, there are two important issues being investigated in this work.

0 What is the effect of interfacial reaction on the interfacial tension in polymeric

blends?

o What is the effect of ‘progressive crowding’ at the interface on the rheological

behavior of the blend?

In this study, nylon 6 has been blended with polypropylene which has been
maleated to different extents. Increasing extents of maleation lead to increasing extents

of interfacial reaction and hence ‘progressive crowding’ at the interface.

77

4.2 BACKGROUND
4. 2. I Emulsion Models

The classic theory of rheology of emulsions focuses on dilute emulsions of
spherical, Newtonian drops - see e.g. Frankel and Acrivos [1967] and Choi and
Schowalter [1975]. The parameters in this theory are the capillary number Ca and the

viscosity ratio k. They are defined in Equations 4.3 and 4.4 respectively.

77“, 7 d _ Viscous Forces

 

 

C = — 4.3

a To Interfacial Forces ( )

k _ 37¢ _ Viscosity of the dispersed phase (4 4)
77m — Viscosity of the matrix '

These two parameters determine the deformation of the drop D, the ratio of the major axis
to minor axis of the distorted spherical drop. Frankel and Acrivos point out that the bulk
stress in a flowing emulsion can be predicted well only if the shapes of the dispersed

phase agree well with the predicted shapes.

Theories and computational results on concentrated emulsions have been reported
by Palieme [1990] for dynamic shear with very small deformation from spherical; by
Loewenberg and Hinch [1996] for shear flows with appreciable departures from spherical
shape for the dispersed phase; and by Sangani and Mo [1994] for shear flows of emulsion
with nearly spherical dispersed phase. The first two cover only a moderate concentration

range - up to 0.3 while the last study accounts for lubrication effects that arise in very

78

concentrated suspension and emulsions. The Palieme theory has an added distinction of
being formulated for viscoelastic constituents. However, it is incorrect to use the Palieme
model for large deformation response of emulsions. Table 4.1 is a summary of the

available theory and computations of emulsion rheology in the literature.

Of these, three models have been applied widely to polymeric systems to explain
their rheological behavior. They are due to Oldroyd [1953], Choi and Schowalter [1975],
and Palieme [1990]. The equilibrium interfacial tension leads to long time relaxation
processes affecting the dynamic moduli in the low frequency region. These long time
relaxation processes are associated with the mechanical relaxation of the dispersed phase
[Scholz et. al., 1989]. The important physical parameters governing the relaxation
processes are the viscosity ratio k, the particle size R, the equilibrium interfacial tension
1", the matrix viscosity nm and AD, the relaxation time for the drop shape.

R77...
r0

 

'11) ~ f (k) (45)

The model due to Palieme accounts for the viscoelastic nature of the component
phases, non-dilute emulsion and the particle size distribution. These factors were not
accounted for in the models of Oldroyd and of Choi and Schowalter. The relation for the

complex modulus of the blend is

 

" (4.6)

 

 

 

79
where,

t t A A .
484 r" 3216(r°+16) sr° . . 219 . . 41;
A v t o d s d d s

i i l I I

1__ (4.7)

 

. t
(130d+80,,,)

and

t O t .
. . . . 4sfldr° 3219.1r°+13d) 40r° . . 216d . . 45,
R. R. R R, R.
l

i l 1

The sub-script ‘i’ refers to the i'h particle fraction. As Equations 4.6 thru 4.8 show, this

 

 

O 0
(13Gd+120,,,)

 

(4.8)

model explicitly accounts for the rheological properties of the component phases, the
volume fraction, the morphology and the interfacial tension. An interesting feature of
this model is its treatment of the interfacial tension as a sum of two parts - an equilibrium
interfacial tension 1"0 and a frequency dependent complex part B’(00). In turn, B°(c0)
consists of two complex moduli - surface dilatation modulus and surface shear modulus.
Both these properties are characteristics of the interface. The surface dilatation modulus
is associated with a non-uniform interface while the surface shear modulus is the
resistance of the interface to the shear deformation. The surface shear modulus is related
to the layer thickness of the interphase. If, B'(0)) equals 0, the model takes the reduced

form shown in Equation 4.9.

80

1+3Z¢,H,(m)
‘ = ‘ " 4.9
G(co) G"'(w)1-ZZ¢.-H.(w) ( )

 

where,

(4r° / R,)(2G;, + 50;) +(G; - 6;)(16G;+19G;)
(40r° / R,)(G;, + 6;) + (26; + 36;)(160; +19G;)

 

H.- (w) = (4.10)

4. 2.2 Determination of the Interfacial Tension fiom Emulsion Rheology

Several groups have used these models to study the rheology of polymeric blends
and determine the equilibrium interfacial tension [Scholz et. al., 1989; Graebling et. al.,
1991; Gramespacher et. al. 1992; Graebling et. a1, 1993]. The models of Oldroyd and of
Choi and Schowalter describe only the low frequency region of the dynamic storage
modulus curve. Graebling and co-workers [1993] showed that Palierne’s model maybe
used to describe the dynamic moduli over a broader frequency range. If each of the blend
components is described by a single relaxation time Maxwell model, and if a
monodisperse particle size distribution is assumed, Equation 4.9 leads to the dynamic
behavior depicted in Figure 1 of paper by Graebling et. al. [1993]. The appearance of the
low frequency plateau is striking. This is associated with the shape relaxation of the
disperse shape domains as shown by AD in Equation 4.5. The location and magnitude of

the secondary plateau determines the accuracy with which interfacial tension can be

81

estimated from such data. The location of the secondary plateau is sensitive to some

emulsion parameters as itemized below.
a As the viscosity ratio k is raised, the secondary plateau moves toward lower storage

moduli and lower frequency.

. As the interfacial tension is reduced, the secondary plateau shifts toward lower

frequencies and its width increases. Particle size has the opposite effect.

. As the volume fraction of the disperse phase is increased, the secondary plateau

becomes more pronounced in width and shifts toward higher frequencies.

. As the ratio of component relaxation times X is decreased, the secondary plateau

becomes ill-defined.

82

4.3 EXPERIMENTAL
4. 3.1 Materials

Nylon 6 ( Ultrarnid B3 from BASF Corporation), neat polypropylene ( Profax
6501 from Montell Corporation ) and three different grades of maleated polypropylene (
PB3001, PB3002, PB3150, from Uniroyal Chemicals ) were used in the study. The three
grades of maleated polypropylene contained different amounts of maleic anhydride
functionality on polypropylenes of different molecular weight distribution. The maleic
anhydride content of PB3001, PB3002, PB3150 was reported to be 0.15 wt. %, 0.3 wt. %

and 0.8 wt. % respectively.
4. 3.2 Blending

Blends were prepared with nylon 6 as the matrix and different polypropylenes as
the dispersed phase. To minimize the effects of moisture, the materials were dried under
a nitrogen blanket for 6-8 hours at a temperature of 120 °C. Each pair of blends
contained 10 wt%, 20 wt% and 30 wt% of the disperse phase respectively. Blends were
prepared in a ZSK-3O twin screw extruder at a temperature of 230 °C for all the zones.
The extrudate strands were pelletized and dried. The blends will be referred to in the text

with the components and their respective weight fractions in the parenthesis.

83

4. 3.3 Rheological

Rheological characterization was carried out on an RMS-8OO rheometer (from
Rheometrics Scientific, Inc.) using a 50 mm parallel plate arrangement. The disks were
prepared by compression molding the pellets in a Carver press under a force of 6 tons and
230 °C. The pellets were pre-dried in a vacuum oven for 10-12 hours. The instrument
oven was purged with dry nitrogen during measurements to avoid degradation. A
frequency range of 0.05-50 rad/s and strains of 10-15% were applied during the
measurements. A strain sweep was carried out to determine the viscoelastic limits on the

strain.

4. 3. 4 Morphological

The samples for morphological examination were prepared by placing the pre-
notched disks in the RMS 800 oven, heating them to 230 °C and placing them directly in
liquid nitrogen. Those were then fractured under liquid nitrogen. These samples were
then examined in a Phillips Electroscan 2020 environmental scanning microscope. Water
vapor at a pressure of 2-3 Torr was used as the imaging gas. The measurements were

made directly from the micrographs and the volume average radii were calculated.

84

4.4 RESULTS AND DISCUSSION
4. 4. I Blend Morphology

Figures 4.1a thru 4.11 show the micrographs of the different blends. Table 4.2
summarizes the volume average radii of the disperse phase in the blends. For the non-
reactive blends it is observed that the particles are spherical with a broad particle size
distribution. Also, their size increases progressively with increasing volume fi'action of
the dispersed phase. This behavior is attributed to the effects of coalescence. On the
other hand, in the case of reactive blends the particle size is reduced, the particle size
distribution is narrowed and the effect of volume fraction of the dispersed phase is
minimized. The data show that the particle size in reactive blends does not vary
significantly with progressive extent of reaction or with volume fraction of the disperse
phase. This suggests that the stability imparted to the interface is independent of the
extent of coverage. Also, in reactive blends the probability of formation of micelles is
low as compared to the blends compatibilized by an external agent. In externally
compatibilized systems the compatibilizer may phase separate above a critical
concentration. Chapter 5 discusses the rheology and morphology of externally

compatibilized blends.

4. 4.2 Rheology

Table 4.3 shows the zero-shear viscosity no and the corresponding relaxation time
of the components. It should be noted that the relaxation time of the matrix is

significantly lower than that of the disperse phase. This is important to identify the

85

relaxation changes brought about by the changes occurring at the interface. The values
were determined using the dynamic shear data. Figures 4.2a and 4.2b show the storage
and loss moduli of the blend components. It is worth pointing out that the molecular
weights of the different maleated polypropylenes are not the same. To observe the broad
rheological changes brought about by the interfacial reaction, two important issues need

to be addressed at this stage:

0 What is the effect of increased weight fraction of the disperse phase, for a given

disperse phase component ?

0 What is the effect of increased extent of maleation (causing progressive extents of

reaction) for the same weight fraction of the disperse phase ?

Figures 4.3a and 4.3b show a comparison of the storage and loss moduli of the
non-reactive blends with increasing weight fraction of the disperse phase. In Figure 4.3a,
the width of the secondary plateau and the associated plateau modulus become more
pronounced with increasing weight fraction of the dispersed phase. As the weight
fraction is increased, the G’ values around the low frequency also increase. These
features are in accordance with theoretical predictions [Graebling et. al., 1993]. It is also
observed that at higher frequencies, the curves converge. That is, the effects due to the
interface contribution are reduced. Also, as shown in Figure 4.3b, there are no significant
changes in the loss moduli behavior. This is also in accordance with the theory. The

remaining discussion shall focus on trends in the storage moduli.

86

The effect of the increased weight fraction of the disperse phase in the reactive
blend is illustrated in Figure 4.4. It is the same as Figure 4.3a but for B3/ 3001 blend
which is also the least reactive. For B3/ 3001 (80/ 20) blend, there is a lift at low
frequencies. Also, the curves indicate that an additional elasticity is imparted to the
blends with increasing weight fraction of the disperse phase. A comparison with the plots
for non-reactive blends shows that the frequency range over which this additional
elasticity is significantly enhanced is greater in reactive blends. It appears that the
interface in B3/ 3001 blends has a pronounced effect over a larger frequency range as
compared to the non-reactive blends. This pattern was observed in all the reactive blends.

In the following sections we shall examine the possible reasons for this.

Next, we examine how blend rheology is affected by the increased extent of
maleation (and hence the increased extent of reaction) if the weight fraction of the
dispersed phase is maintained constant. Figure 4.5 shows a comparison of the storage
moduli for the 80/20 blends with varying extents of maleation in polypropylene. An
increased extent of maleation (and interfacial reaction) leads to progressively higher
elasticity over a larger frequency range. When compared to the non-reactive blend it is
seen that the elasticity of the non-reactive system is higher as compared to the elasticity
of the reactive system toward the lower frequencies, but 10wer at the higher frequencies.
This can be explained on the basis of the Equation 4.5. It shows that the longest
relaxation times correspond to the mechanical relaxation of the dispersed phase [Scholz
et. al., 1989]. The size of the disperse phase in the non-reactive blend is much larger

compared to the size in reactive blends. The larger size of the droplets leads to longer

87

relaxation times and larger storage modulus. However, at higher frequencies, where the
long time relaxation processes play a less significant role, the size of the droplets does not
control the relaxation behavior. It is the product of the interfacial reaction which will

control the relaxation phenomenon.

The Figure 4.6 compares the storage modulus of the B3/ PP (80/ 20) blend with
that of its components. It is observed that the storage modulus of the blend lies above the
respective component values until a certain value, after which it falls between them.
Figure 4.7 shows a similar comparison for B3/ 3001 (80/ 20) blend. The storage modulus
of the blend falls between those of the components for the entire frequency range. In the
B3/ 3002 (80/ 20) blend (Figure 4.8), the blend storage modulus falls between those of
the components for the major part of the frequency range but above the component values
at high frequencies. This trend is further enhanced in B3/ 3150 (80/ 20) blend (Figure
4.9), where the blend modulus is consistently higher than those of the components over
the entire frequency range. That is, the interfacial reaction causes a fundamental change
in the rheological behavior of the blend. Progressive extents of reaction lead to a
progressive increase in elasticity. In the following section the effect of interfacial
reaction on the interfacial tension shall be discussed in greater detail which is an

important aim of this study.
4. 4.3 Determination of the interfacial tension

To determine the value of the interfacial tension, Equation 4.9 and 4.10 have been

used. Figure 4.10 shows a comparison of the storage and loss moduli obtained from the

88
model with the experimentally obtained values for the non-reactive B3/ PP (80/20) blend.

The agreement between the model and the experimental value is good for a value of 8

mN/m. This value gives a good fit for 90/ 10 and 70/ 3O non-reactive blends also.

Now, what is the effect of interfacial reaction on the interfacial tension in reactive
blends? Figure 4.11 shows a comparison of the model and the experimentally obtained
values of the storage and the loss moduli in B3/ 3001 (90/10) system. This is the least
reactive system as it is the least maleated. A value of 7 mN/m gives a good fit between
the model and the experimentally obtained values. In B3/300l (80/20) blend, a
secondary plateau toward the low frequency appears, as shown in Figure 4.4. Figure 4.12
shows a comparison of the model and the experimentally obtained values for 33/ 3001
(80/ 20) blend. They do not agree well, especially in the lower frequency range. The
interfacial reaction has created a fundamental change in the behavior of the reactive
system as compared to the non-reactive system. The change is imparting an additional
elasticity to the system which is not accounted for by the model in the form it has been
used until now. This discrepancy is enhanced as the weight fraction of the 3001 is
increased firrther. The effect of elasticity enhancement is so pronounced that there is
disagreement over the complete frequency range for any value of equilibrium interfacial
tension in all the reactive blends for 70/ 3O blends as shown in Figure 4.13 for B3/ 3001
(70/ 30) blend. In further discussion, the discussion shall be limited to 90/ 10 and 80/ 20

blends.

A more discriminating fit maybe obtained by extracting the contribution of the

interface to the dynamic storage modulus. To delineate this contribution, the following

89
quantity was defined and computed from the storage moduli curves for both the predicted

and experimental values.

6;... = GA... — 1116.; +11 — 1116;] (4.11)

It should be stated at this stage that the storage modulus in the low frequency limits is the
main focus of discussion as this is most susceptible to the changes in the interface. Figure
4.14 compares G’int obtained from model prediction and from experimental data for the
B3/3001 (80/20) curve in the low frequency range. It is seen that there is reasonable
agreement in the lower frequency range for an equilibrium interfacial tension value of 7
mN/ m which is the same as in Figure 4.11. Similar comparison was carried out for B3/
3002 (80/ 20) and B3/ 3150 (80/ 20) blends as shown in Figures 4.15 and 4.16 to obtain

equilibrium interfacial tension values of 7 mN/ m and 4 mN/ m respectively.

To explain the discrepancy observed in the data above, it is important to
understand the physical events occurring at the interface with progressive extent of
reaction. As the interfacial reaction occurs, the interface goes from being a ‘bare
interface’ to an ‘occupied interface’. Increased extent of reaction leads to a ‘crowding’ at
the interface. This increased occupancy of the interface and the links established due to
reaction cause the interface to take progressively long times to relax thus causing an

enhanced elasticity.

Thus, it is crucial to account for the contributions being imparted by the interface.
Until this stage, the only property characterizing the interface has been the equilibrium

interfacial tension. This does not account for the ‘crowding’ effects explicitly. In this

90

respect the modifications of Palierne’s model are useful [Palieme, 1990]. It accounts for
the surface dilatation modulus Bd'(c0) and the surface shear modulus B,’(0)). The origin
of these two quantities lies in the non-uniformity of the interface and the resistance to the
shear deformation respectively. In this study it will be assumed that the interface is
uniformly occupied by the reaction products and the surface dilatation modulus can be
ignored. However, the reaction product does resist shear deformation, and the surface
shear modulus B,'(0)) should be accounted for. The quantity B,‘(0)) itself consists of the
storage and loss moduli - B’,(c0) and B”,(00) respectively. However, there are no
experimental techniques available to measure these quantities. In the present work, the
estimates shall be made from the theory of lightly cross-linked rubber [Ferry, 1961]. The

reaction products are thought to behave as lightly cross-linked rubber. In the framework

ofthis theory,
For 01))»B < 1
’B“. = fl" (4.12)
fl. = 13.44,,
For 00711, > 1
13.; =4; arm (4.13)

The two important parameters in these equations are [30 and k1, (see to Figure 4.15). The
quantity Bo is the low frequency, long time limit of G’ and 2.1, corresponds to the

frequency until which elastic recoil of the system occurs. For 7‘13 the transition point (in

91
the experimental curve) till which the additional elasticity is exhibited is a good choice.

It is the inherent elasticity of the system. Incorporating the quantity B,°(0)) in the
Equation 4.9 leads to improvement in the agreement between the experimental and model
curve in reactive blends. This has been shown in Figure 4.18 and 4.19 for B3/ 3150 (80/
20) blend. Similar improvements were observed in other reactive systems. Table 4.2
shows the values of the equilibrium interfacial tension and the approximate value of the

B0 for the different blends.

The progressive extents of reaction leads to progressive ‘coverage’ of the
interface. This leads to a reduction in equilibrium interfacial tension and an enhancement
in elasticity. But, the important question at this stage is that how much coverage of the
interface is critical to cause a reduction in interfacial tension? This will be discussed in

the following section.
4. 4. 4 The issue of extent of coverage

Table 4.3 shows the material properties of the materials used in the study. From
these values, the values in Table 4.4 have been derived. It is seen that the amount of
amine functionality available for interfacial reaction is much more than the amount of
maleic anhydride. This is important as it ensures that the interfacial reaction goes to
completion and isnot limited by the amount of reactive sites available. That is, whatever
maleic anhydride is in the interface is completely reacted. Also, it is seen that in the
disperse phase the number of maleic anhydride groups per chain available for reaction

increase from 1 per chain in PB3001 to 3.4 in PB3150. This is a significant difference

92

and will help in delineating the effect of progressively increasing extent of reaction.
Assuming complete reaction and uniformity of -MAH chains in the bulk of the dispersed

particle and the interface, it can be estimated that

 

 

 

no. of g - moles of - MAH in the interface _ (Volume of interfaceXno. g - moles of - MAH 1

kg - mixture _ Volume of particle kg - mixture

Assuming the particle radius to be R and the interface thickness to be AR,

 

Volume of interface 472'R2AR (3M)

Volume of the particle ~ 47:123 R
3

 

Let,

 

0- no. ofg-mole of -MAH
kgPP-MA

If, w is the weight fraction of dispersed PP-MA in the blend

no. ofg-mole of -MAH _
kg - mixture —

 

CW

Using above,

no. of graft chains of amine - MAH in the interface PAR)
= cw A

kg - mixture R

47:18
Area occupied by the particle interface per kg " mixture = 4 7rR3 (101]
__ d

 

93

Therefore,

_ Area occupied by grafts in the interface/ kg - mixture _ 1
_ no. graft chains of amine - MAH in the interface/ kg - mixture - pdARcN A

 

The value of 2 was determined for the respective reactive blends and have been shown in
Table 4.5. As the‘extent of maleation increases, 2 reduces as it should, because the
number of grafts increases due to increased reaction as the number of maleic anhydride
group available per chain increases from 1 to 2 to 3.4 respectively. Combining this with
the estimates of the equilibrium interfacial tension it is seen that it drops continuously
with increasing reaction and increasing occupation of the interface. The reactive blends
with progressively increasingly extents of interfacial reaction show departure from the
model at higher frequencies also. This means, that the product of the interfacial reaction
plays a significant role in the rheological behavior of the blend even at higher

frequencies. Its effect is not limited to low frequencies alone.

4.5 CONCLUSION

This study has investigated the effects of interfacial reaction on the interfacial
tension in reactive polymeric blends. It has been shown that interfacial reaction leads to a
reduction in particle size of the dispersed phase and a reduction in interfacial tension.
Morphological observations show that a minimal amount of interfacial reaction is
required to reduce the particle size. Also, increased reaction does not necessarily cause
progressive reduction in particle size. Interfacial reaction progressively drops from 8

mN/ m in non-reactive blends to 7 mN/ m to 4 mN/ m in reactive blends with different

94
extents of reaction. The number of reactive sites for grafls per maleated polypropylene

chain goes from lto 2 to 3.4 respectively in these systems. Rheological observations
show that besides a reduction in interfacial tension, there is an enhancement in the
elasticity of the reactive blends as compared to the non-reactive blends. This is due to the
fundamental difference brought about by the product of the interfacial reaction. To
account for this, surface shear modulus has been used as proposed by the model of

Palieme [1990].

In order to understand the effects of interfacial reaction on the behavior of the
reactive blend, the effects of the visco-elastic nature of the polymeric components need to
be eliminated. In the present study, the disperse phases were of varying visco-elastic
properties. It is suggested that studies be carried out with disperse phases of same visco-

elastic properties, but varying amounts of reactive sites.

95

Table 4.1: Emulsion Rheology models available in the literature.

 

 

Type of Limit of Limit of Theory or Authors
Flow Concentration Deformation Approach
Shear-time <0.02 =0 Oldroyd
1 varying (Spherical
drops)

Shear-time <0.02 <0.05 Asymptotic Frankel and
varying expansion Acrivos
Steady Shear 0.02 <O.6 Boundary Kennedy and

Integral Pozrikidis
Steady Shear 0.1 < 0.05 Method of Choi and
Reflections Schowalter
Oscillatory <O.3 Very small Self Palieme
Shear consistent
cell
Steady Shear <0.3O <0.6 Boundary Lowenberg and
Integral Hinch
periodic
B.C.
Steady Shear up to max <0.05 Multipole Sangani and
1 packing expansion Mo
periodic

B.C.

 

96
Table 4.2: The volume average radii of the dispersed phase in the various blends.

 

 

Blend Radius I0 [30
(um) (mN/ m) (mN/ m)
B3/PP (90/10) 8.75 8 -
B3/PP (80/20) 10.7 8 -
B3/PP (70/30) 16.5 8 -
B3/3001(90/ 10) 0.8 7 -
B3/3001(80/20) 0.8 7 0.1
B3/3001(70/30) 0.6 7 0.1
B3/3002(90/ 10) 0.5 7 -
B3/3002(80/20) 1.1 7 -
B3/3002(70/30) 0.7 7 -
B3/3150(90/10) 0.6 4 0.1
B3/3150(80/20) 1.5 4 0.1
B3/3150(70/30) 1.2 4 -

 

Table 4.3: The material properties of the materials used in this study.

 

 

Material 1],, Relaxation time M, N
(Pa-8) (8) (kg/ kg-mole)
B3 860 0.002 18,000 160
PP l 150 0.35 54,000 1286
3001 3000 1.1 76,000 1714
3002 1970 1 .3 63,000 1500

3150 490 0.28 42,000 1000

97

Table 4.4: The values for reactive sites in the materials used in the study.

 

 

Material n
(no. -MAH/ chain)
B3 15.74 (g-mole amines/ kg-B3) -
3001 0.015 (g-mole -MAH/ kg-3001) 1
3002 0.030 (g-mole -MAH/ kg-3002) 2
3150 0.080 (g-mole -MAH/ kg-3150) 3.4

 

Table 4.5: The values of the extent of coverage of the interface in different blends.

 

 

2x10" 22/nx1018 I],

(m’/ grafts) (m’I chain -PP-MA) (mN/ m)
33/ 3001 24 24 7
B3/ 3002 12 6 7

B3/ 3150 4.5 1.3 4

 

98

 

Figure 4.1a: Micrograph showing the morphology of B3/ PP (90/ 10).

99

 

Figure 4.1b: Micrograph showing the morphology of B3/ PP (80/ 20).

 

 

Figure 4.1c: Micrograph showing the morphology of B3/ PP (70/ 30).

101

 

 

Figure 4.1d: Micrograph showing the morphology of B3/ 3001 (90/ 10).

102

 

 

Figure 4.1e: Micrograph showing the morphology of B3/ 3001 (80/ 20).

103

 

 

Figure 4.1f: Micrograph showing the morphology of B3/ 3001 (70/ 30).

 

Figure 4.1g: Micrograph showing the morphology of B3/ 3002 (90/ 10).

l05

 

Figure 4.1h: Micrograph showing the morphology of B3/ 3002 (80/ 20).

106

 

Figure 4.1i: Micrograph showing the morphology of B3/ 3002 (70/ 30).

107

 

Figure 4.1 j: Micrograph showing the morphology of B3/ 3150 (90/ 10).

108

 

 

 

Figure 4.1k: Micrograph showing the morphology of B3/ 3150 (80/ 20).

109

 

 

 

Figure 4.11: Micrograph showing the morphology of B3/ 3150 (70/ 30).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00905 T
l
1.00E+04 “— _ ‘ 6' (B3), Pa
. 6' (PP). Pa
+9 (3150). Pa
—o—G' (3001), Pa
1.00903 _ 0 GMT” P“
l
0 1
9. 1.00902 .94 _. ,. h. ___,
0 j ,«o
O
I’loo
Id? 0
(’0 o ’1'! .
1.005401 -1~- —— — O_+_O__ A
0 Of A
O A
' o. l- ,
I
. I
1.00900 _- -.-. '%_._. =
I A ‘
. l
" i
A .
‘ l
1.005-01 _ V r
1 DOE-02 1 DOE-01 1.005“ 1.00501 1 .OOE+02

Freq. lldl.

Figure 4.2a: Comparison of the storage moduli of the components of the blend.

lll

1.00E+05

 

 

. G" (83). Pa \ ‘ .
Q G" (PP), Pa l ‘ "Q
—-— G" (3150). Pa 1 “3.530
_... G” (3001). Pa . . .r 5
1.00E+04 “___ o G" (amzx Pa —— +— ’.a¢ $06——

 

 

 

 

 

 

 

 

 

 

a
:0 1.mE+03 ‘P' - '
0
0
1.00902 .9 5 4.4.3.." _; h
A I’ l l
. A "I' l ‘
' ‘ .l 1
a II I
.’ l ‘
1.00901 , l ,
1 DOE-02 1 DOE-01 1 .00E+OO 1 .OOE+O1 1 .00E+02

Freq, radio

Figure 4.2b: Comparison of the loss moduli of the components of the blend.

112

 

 

 

 

 

 

 

 

 

 

 

1.00904 . I
1 l l ”I"!
A
, G'(33/PP(90/10)).Pa I. ', .
A
. G‘(83/FP(80/20)). Pa § {41,
_ 33/131D 70/30 .Pa ! t ‘
1.00903 19 ___ "m ( ” . ,3 .
1 l ((- . ‘
I l . . A i
L. l’o . A ‘
, . . l
1 1 C
a ’ 1%.: ‘ ‘
91100902.--. 9... .. ’. ,0 l1 l
0 i .1. . . . l l
1 I, . ‘
I I, A i
i, O I g
I 0 ‘
,- 0 A l \
I’. 9 ‘ 1 I
1.00901 . ___ ___. __'_l_._:_. l l
e j ‘ i 1
o i ‘ ;
. ’ i 1
. ‘ l 1
1.00900 1 l 1
1.00502 1.00501 1.00900 1.00901 1.00902
Froq,radls

Figure 4.3a: Comparison of the storage moduli of the non-reactive blends for varying
weight fractions of the dispersed phase.

113

 

 

 

 

 

 

 

 

 

1.00905
1
. G"(B3/PP(90/10)).Pa
. G"(83/F‘P(80I 20)).Pa
10090" ~——— -... G' (33/ PP(70/ 30)), Pa
1
&
. 1.00903 .99. a , 9—
8
o 1 ’g ,.
l 1'
r !
1 A
l o ‘
1 g‘
0
1.0090239 * {2. W h _ _
fl} HT ‘
2 I 1
. - 1 .
l l 1
1.00901 l J l
1.005-02 1.00501 1.00900 100901
Freq,radls

1.00E+02

Figure 4.3b: Comparison of the loss moduli of the non-reactive blends for varying
weight fractions of the dispersed phase.

114

 

 

 

   

 

 

 

 

 

 

 

 

 

1.00905 1 ‘
1 1 1
1 1 1
1-°°E*°4-——- , G'(83/3001(90/10)).Pa — 1 l 1
. G‘(83l3001(80120)).Pa 1 |
1 o
_._ G' (83/ 3001170130», Pa 1 1
1.00903 .___ 1 g 1"
1 l 1
‘ 1
I .1 O ‘ A
0; 1.00902.__ A , V ,. J —..; ,.’ .47., l
b 1 l’ o . . ‘ ‘j i
. A
”IT/ll. . O ‘ A ‘ 1
1.00901 .9. _. , -r- ...-_,__ ._ 1 1
O ‘ 1
. i A L 1
O ‘ ‘
1.00900.A- * , .‘t- ,o-,,_h_-_u L -
1 1 1
1.00501 , . ,
1.00502 1.00501 1.00900 1.00901 1.00902
Froq,radls

Figure 4.4: Comparison of the storage moduli of the B3/ 3001 blend for varying weight
fractions of the dispersed phase.

115

 

 

 

 

 

 

 

 

 

 

 

 

1.00905 , 1 1
1 1 l
1 1 1
, G’(83I PP(80/ 20)). Pa 1 1 «...
1005.04, . G'(8313001(80l20)), Pa 1 '1 was"; .
_._G(Bal 3002(80/20)). Pa 1 , '4: . I . ‘
_._G'(83/ 3150(30/ 20)). Pa 1 fl ’ . ‘
l l of . T ‘ ‘
l O A
1.00903 .3 _ __ _ . gm _ o_o_. L A
o/ O A T
1 o ' ‘ ‘ .
a 1 , o ‘
fl. /. O A
as «’2‘ 1
((2
1.00902 .9 A n - v .1-4 1 1 .-
} A/g/ . x r
1 ‘It/OI' ‘
A’:/. '/- I
V’ . :’ : i
1.00901 ._. _ “(£144 1
.l‘ . C I
. 1f -
o f’T l
., i l
1.00E+00 % I
1.00502 1.00501 1.00900 1.00901
Freq, radls

1.005402

Figure 4.5: Comparison of the storage moduli of the 80/ 20 blends for the different

dispersed phases.

116

 

1.005405 1 1
. 61831.9 1 1
. empe- ' 1
”0504‘ __.._G‘(B3/PP(80l20)),Pa 1 1 ,

 

1.00E+03 ..

 

 

 

1.00E+02 .

 

6', Pa

1.00E+01 .

 

 

 

 

 

 

1.00900 ..__ _- __. ._.._‘_ ,1 n A
1 ‘ ‘ 1 1
1

A
1 1
1.00501 , ‘ , , r

1.00502 1.00501 1.00900 1.00901 1.00902
Freq.radlo

Figure 4.6: Comparison of the storage moduli of the B3/ PP (80/ 20) blend with the
components.

1 DOE-+05

117

 

1.00804 .9.._ .

1.00E+03 .

6', Pa

1.00E+O1 .

1.00E+00 . __

 

1 DOE-01

1.00E+02 _

 
 

1

1
‘ G'(B3),Pa

 

  
  

. G‘ (3001). Pa
_._ 6' (33/3001 (BO/20)). Pa

 

 

 

 

 

 

 

 

1 DOE-02

1.00501 1.006400 1.006001
Freq, rad].

1 .OOE+02

Figure 4.7: Comparison of the storage moduli of the B3/ 3001 (80/ 20) blend with the

components.

1 .00E+05

118

 

 

 

 

 

 

 

 

 

 

 

 

 

1
1
1
1.00E+O4 -9— ‘ 6' (B3). pa ~
. G' (3002). Pa ‘ .
—-_ 6' (8313002 (80120)). Pa .
1.00903 .99 .9 9 999919999 ___ 9
1
C 1 .
5': 1.00902 .9 __...__
o
O I
r ..-’
. 1
. ._ . 9 .93 1
100901 . 1;"
A
, A
l’ 1
1.0090099 9 9 _9 ___.9 9 99 ___
1 - ‘ 1
A
. ‘ 1 1
1.00901 _ 1
1.00902 1.00901

1.00E 1.00E+O1
Fm. .1919.

 

1 .00E+02

Figure 4.8: Comparison of the storage moduli of the B3/ 3002 (80/ 20) blend with the

components.

119

 

 

 

 

 

 

 

 

 

 

 

 

1.005005 .
1
1
. G'(B3).Pa
100904999 . G'(3150).Pa
_._ G' (813/3150 (BO/20)). Pa
1
1.00803 . ._ 99 9 9999-9 9 99 999
I
9: 1.0090299 9 9 9 9 .99_ /
(D 1 I," G
M . '
. o '
1. 1.19.9’99 __._9. . A
0090 ”999 ..
1 . 1
1 ° ‘
. A
o 1 ‘
1009001999. ....9._ .9.9 99._9_
O A
1A
. ‘ 1
1.00501 . _ 9: , '
1.00502 1.00501 1.00E+00 1.00E+01 1 .00E+02
Freq,radls

Figure 4.9: Comparison of the storage moduli of the B3/ 3150 (80/ 20) blend with the
components.

120

 

 

 

 

 

 

 

 

 

 

 

 

1.00905 1 .
1 1 1
1 1
’2
1 1 1
l - 2’.
1.00304 .9999 ‘ G'( I), a 1 l .;:/.’ ‘ 5
I
. G’ (exp). Pa 1 [2,350 3 t
9,_ G" (model). Pa ”5232’ . g 3
...._ G" (exp). Pa ’35:) 6 .
1.00903 .9999 -999 9 9 9 ’53" . . 1
a 1 ’89:” o 1
A Q
a. 1 ’95:,“ o g 1
q 1 .’el 0 2
b 1 a”? :1 2 ‘
1.00902 99.9.9 1’!” - o ’
I. '
o’t/T o . 1 1
o/ 1 o . 1 1
O
1 g ‘ 1 1
1.00901 .99 9 9 92:19 __9 i . 1
. . . 1 T
‘ 1
. f 1 1
1 1
1.05.00 1 , 1
1.00902 1.00901 1.00900 1.00901 1.00902

Freq, rad/s

Figure 4.10: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for B3/ PP (80/ 20) blend (1'0 = 8 mN/ m, R = 10.7

um).

121

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00E+05 1 1
1 1
1 1 1 ’0’.
1 .I.
1.009041%. . G (nodal). Pa __Mflw; 1”...” . o
. G (Exp), a 1 ’3’.’. . . .
_._ G" (MOI). Pa 1 .’,.’8’ . O
“I .
1.00903 .__ -°-G"(°"p)' P“ 1 z’z’. A ° 1
1 """.11 . . ' 1
1 3,3,. 1 3 . 1
fl 1 3’3’ 1 . ‘ 1
l “’3’ 1 5 . ‘
i9. 1.00E+02 ‘._-* A A _ _e'sfi...___ .1 é 1
o" “’0 1
0 o O 1 1
1 . ° 1
1 2 1 1 1
1.00801 .L____1__...Vi__._1, __ . g 1 1
I g ‘ 1 1
: 1
1 2 1 1
100900.? f v i ._._2_ 1‘-v,+,_
2 ’ 1
. 1
' 1
1.00501 1 1
1.00502 1.00501 1.00E+00 1.00E+01 1.00E+02
Freq. radls

Figure 4.11: Comparison of the storage and loss moduli values obtained fi'om the model
with that obtained experimentally for Bill 3001 (80/ 20) blend (F0 = 7 mN/ m, R = 0.8
pm).

122

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.005105 1 1 1
1 1
1 1 ”5“”:
1.00E+O4 ._ ‘ 6' (m1). pa __w ...._1- ”page" {6 6
. G (exp). Pa 0,59%” 1 ‘ ‘ a
...—G" (model). Pa 95:?" ‘ 3 6
l.’
1.00E+O3 .__ -... G" (exp). Pa ___ ,«e‘V’ .5 ’ 1
1 o’:$:’ ‘1 : .
.59," ' 1
are
a 1 5:93,“ . 3 ‘ 1
11’ 1.005102 .___.._ « ___. '1'" e '1
0. e" 1 ' 1
is I . 2 1
1 ’ ‘
. - 1 1
1 g A 1.
1.006001 1* _f _.. -*—-—,~-:"*—1—-- _._ _1
o ‘ 1
o 1 A
1 1
1.005100 ._._ fl -i_+._11_____t#1fi
‘ 1
1 1
1.00501 1 1
1.00502 1.00501 1.001300 1.001501 1.00302

Freq, radio

Figure 4.12: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for B3/ 3001 (80/ 20) blend (I‘0 = 7 mN/ m, R = 0.8
pm)-

123

 

 

 

 

 

 

 

 

 

 

 

 

1.00E+05 1
1 1 1
,0
1 1 053:8’:
G‘(model) Pa ”:35” ' ' ‘
1.00304 ._.h_ 1 . -# . fig?!» - g
. G‘(exp). Pa 1 ,«zfii’l - i ‘
1”.’ 1
_._ G'1(rmdel), Pa .’;::,. . . : ‘ A
—.— 6" (exp). Pa ’04:" ‘
o"/.’ o 1
1.001:—103._.C.w .11 1.1.1.111 -24?" 0 1 ‘ 1
Q I I’lz’.' C . ‘ 1
n. /. 0’ I» A
1 I. .’
3" 1 ,0 .l O A
o 1 .l.’.’ C A
.. l 0 o
b ,4,“ ' 1 1
0 I A
1.00E+02 1 _____ 6A1. . _o____‘ 1
1 . A 1 1
1 o A 1
1 C ‘ 1 1
O
o 1 ‘ 1 1
1.00901 ._._. -- .-..J. 17;“ __ 1
1 1 1
. 1 1
. 1
1.005100 ‘ T 1 1
1.00502 1.00501 1.00E+00 1.00801 1 .00E+02
Freq. radio

Figure 4.13: Comparison of the storage and loss moduli values obtained from the model
with that obtained experimentally for BB/ 3001 (70/ 30) blend (1'0 = 7 mN/ m, R = 0.8
pm).

124

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.005102 1 1 1 1
1 1 1 fl

8.0015101 .. ___ .1. _ _____1._ ..fi 1 1

i 1 '

6005.01 “_ _‘ _5 —.— G'iflt (exp). Pa _ 1 1 / 1
_._ G'int (model). Pa 1 ' 1

1 1 1 /'1/ 1

4.00301 “__nm __1__ -____ _n-,__1 1

, 1

a 2 005101 1 1 1 / 1
:s' - -__,,--h___1__ 17 1
o , . ’ .
0.0051001. _ 71* i If -1_ I J1: . 1
“'I:-.. w 1 /1 1

‘0‘ \I 1 1 .
4.005101 . 0 .:.<1__. ._ .17/_ 1 1
.\:§'<o/ 1

1 1 1

4.005401 .._ .__ h 1 k\1\ 11 1.

1 '\1~I——I1/ 1

45.005101 1 1 1

 

 

 

 

0.00E+00 5.00501 1.00E+00 1.50E+00 2.00E+00 2.50E+00 3.00E+00

Frequo ncy. rad! s

Figure 4.14: Comparison of the G’int values obtained from the model with that obtained
experimentally for 33/ 3001 (80/ 20) (l‘° = 7 mN/ m, R = 0.8 pm).

4.00E+02.._ _ V - Ca -.-, 7,, 7-, u

 

 

 

1 1 1 1 1
3.505102 .._ _. - aga , a, n1, -,___+fi 1 1 1
. 1 1. / 1.
3.005102 .. ———--1—————:~——- _._-__.i- n /1
1 1
2.505102 .. _-_____

 

 

2.00E+02 ... _

 

1
1

_._ G'int (exp). Pa ' 1‘ f
1 /

 

 

 

 

 

 

 

 

& f
.‘s’ 1
i9 1.505102 . #1
1 ,
1 1
1.00802 .. . 1
5.005101 . 1 1
1
0.005+00 .. 1 __11_ 1
1 1 1
45.005101 1 1 1

 

0.00E+00 5.00501 1.00E+00 1.50E+00 2.00E+00 2.505900 3.00E+00

Froque ncy. rad! s

Figure 4.15: Comparison of the G’int values obtained from the model with that obtained
experimentally for B3/ 3002 (80/ 20) (1"0 = 7 mN/ m, R = 0.8 pm).

126

7.00302

1 1 1 1
s 1
6.00E-t-021F _. ._ _ .. , _ - *1"—_*5g 1 .1

_.—iGint (model), Fla 1
5‘°°E’°2‘1" — _._Gint(exp).Pa * 1 / /
1 1 /
4.005r02.5 5 - , ,. , 1 .

3.00902“! ,

2.005102“; ,, . ,5. _17/4 ,_

 

 

 

 

 

 

 

-1
\
“w

 

 

 

 

1.005102% 5, - - -k.._ 1
l/. ‘ 1 1

--'./1 1 1

0.00500 1&55' 1 1 1

 

0.00E+00 5.00E-01 1.00E+00 1.50E+00 2.00E+00 2.50E+00 3.00E+00

Froquo ncy. rad! a

Figure 4.16: Comparison of the G’int values obtained from the model with that obtained
experimentally for B3/ 3150 (80/ 20) (1'0 = 4 mN/ m, R = 1.5 pm).

127

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00E+05 1 1 1
' 1
1 “o
1 .,;::<’:
6‘ mode ,Pa ’0’
1.005104 1__. ‘ 1 I) 1 .7 ,3 {O . '
. G'(exp). Pa 1 ,o’ ii: . . ‘
/ 0’
_._ G" (model). Pa ' .,o’. .1". 4 A ‘
—._ G" (exp). % ’9’.’ I.’.’.’ Q ‘ 1. A
. ’0’” .,O’. O ‘ A 1
1 1. a A
1.00E+03 "—5. -.._ -_ .1- __ .’° ., 01‘ -
I 1 .l .1 2 A
n: 1 ./°’./.’ 0
:9 1 .l.’.l.’ . .
a," . [1:0 ’0’ . L
/ 0
1.005102 .._ -._~ _ -7;’=’°1 2 1
0’ 1 2 1
1 Q 1 1
1 o A
1 . ° . 1 1
1 A
1.00301 ._._._.._ _ 5...;T1___._ L 1
O ‘ 1 1
1 .
. 1 1 1
A
1.00m , 1 1 1
1.00802 1.00501 1.00800 1.00801 1.00E+02
Freq, radio

Figure 4.17: Comparison of the 6’ values obtained from the model with that obtained
experimentally for B3/ 3150 (80/ 20) (1"0 = 4 mN/ m, R = 1.5 um, B0 = 0 mN/ m).

128

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00E+05 1
1 .a.
’0’.’.’.
1 ’0’ (:1. ,z’. .
1.00E+O4 “Hg ‘ 6' (model). Pa ’3’" 740;... A
V .l. ”I. . O A
. G‘(exp).Pa ’0’.’ .’./ . ‘ A
_ ' a I. .’ 0 ‘
kgm’bdenpa . ’0..’./ .011‘A
__0— (exp), 1 ’./‘1/.’.’. . C ‘ ‘
1.005103 1 ,o_ ‘0’ . . .
' 1 I. /.f O A -
CI 1 I. I. A
a 1 0’0 . C 2
a. 1 ’0’ ’0’ .
G 1 /. .1. 0
o" ./. .’ .
G /./1’.’ . O
0
1.005102 a-w-flwzz_’.z1fl_l__; . 1
o 1 2 - 1
1 . 2 1
1 . ‘ 1
1 - . 1
O A 1
1.00901 ._._._*.___ _. g 1 1
: ‘ 1 1
1 1
1 1
1.00300 1 1
1.00502 1.00501 1.00300 1.00E+01 1.00E+02
Freq, radio

Figure 4.18: Comparison of the G’ values obtained from the model with that obtained
experimentally for B3/ 3150 (80/ 20) (W = 4 mN/ m, R = 1.5 um, [50 = 0.1 mN/ m).

129

 

 

 

 

 

 

 

0.108

 

 

0.

c (g-mole -MAHI kg PP-MA)

Figure 4.19: The plot showing the reduction of equilibrium interfacial tension with
progressively increasing extent of maleation.

THE EFFECT OF COMPATIBILIZATION OF POLYPROPYLENE AND
POLYSTYRNE WITH A DI-BLOCK COPOLYMER ON THE RHEOLOGY AND

THE MORPHOLOGY OF THEIR BLENDS

 

Chapter 5

 

5.1 INTRODUCTION

Most polymers are thermodynamically immiscible. As a result, they phase
separate on mixing. In order to circumvent the problem of phase separation, the
component polymers are compatibilized. This is achieved by addition of an external
compatibilizer or by carrying out an in-situ reaction between the complementary groups
of the components. The resulting blend has a fine dispersion of the minor phase in the
major phase. The particle size (distribution) of the blend depends on the processing
conditions and the interfacial tension between the components. The governing parameter
is the capillary number given in Equation 5.1. It is a dimensionless ratio of the shape

deforming viscous forces and the shape retaining interfacial tension forces.

d
Ca = fl"— (5.1)

F 0
According to this relation, a reduced equilibrium interfacial tension leads to a
higher capillary number which promotes break-up of drops. In polymeric blends this is
achieved by the addition of an external compatibilizer or the in-situ reaction which leads
to a reduction in interfacial tension [Graebling et. al., 1993; Chapters 3 and 4]. The
Chapters 3 and 4 show that besides a reduction in interfacial tension, interfacial reaction

also leads to a fundamental change in the rheology of the blend. An additional elasticity

I30

13 l
is imparted to the system. They attributed this effect to the product of the interfacial

reaction. In this study, a similar effect has been observed with externally compatibilized
systems. The compatibilizer is located at the interface or it phase separates as micelles.
In either case, it brings about a change in rheology of the blend [Brahimi et. al., 1991;
Germain et. al., 1994] .

Graebling .et. al. [1993] carried out a study on rheology of polystyrene blended
with PMMA compatibilized by an external agent. In the same study, results for PDMS
blended with POE in presence of an external compatibilizer were presented. It was
shown that equilibrium interfacial tension drops due to the presence of the compatibilizer.
The study by Brahimi et. al. [1991] shows that a blend of polyethylene and polystyrene
conforms to the rheological model of Palieme, but the compatibilized blend does not.
However, the HIPS used in that study is itself a two phase system and can alter the
behavior of the blend. The present study uses neat polystyrene to eliminate the effects
which could have been caused by the dual phase nature of HIPS itself. Germain et. al.
[1994] also observed a disagreement between the model and the experimentally obtained
values. This discrepancy was attributed to the properties associated with the interface. In
a non-compatibilized blend, the interface is 2-5 nm thick. However, as the study by Fayt
et. al. [1986] shows, in a compatibilized blend, the compatibilizing agent occupies the
interface and thickens it. The thickness of this interface was estimated around 10 nm.
The properties associated with the material in this layer affect the rheology of the blend.
In light of this, Germain et. al. [1994] attempted to explain the blend rheology on the

basis of a core-shell model. They said that the properties of the dispersed phase play a

132

minimal role in the blend rheology. Since the compatibilizer is located at the interface,

its properties will dominate those of the disperse phase. The results with this assumption

improved, but there was still disagreement with the model behavior. The aim of this

work is to systematically study the effect of progressive addition of the compatibilizing

agent on the rheology of the blend. The main focus would be understand the changes

brought about in the rheology of the blend by the compatibilization and determine the

reason for those changes. In order to do this, the role of the interface and the equilibrium

interfacial tension will be investigated. The model of Palieme [1990] will be used to fit

the observed rheology of the blend with appropriate values of the equilibrium interfacial

tension. Equations 5.2 and 5.3 describe the model.

where,

1+ 3Z¢,H,(w)
G (on) = 6.1a» 1 _ 22111111101) (5.2)

 

_ (41“0 / R1.)(2G,;,+SG‘:1)+(G‘:1 —GI;,)(16G; +19G;)

H. — o O ‘ . . .
(0)) (401—0 /R,)(G,,. +Gd)+ (20d +3GM)(I6G,,, +19Gd)

I

 

(5.3)

l 3 3
5.2 EXPERIMENTAL

5. 2. 1 Materials

Polypropylene ( Profax 6523 from Montell Polyolefins) was used as the matrix
phase. It was blended with polystyrene (Huntsman Polycom) which was the disperse
phase. A di-block copolymer -S-EP- (Kraton G1702 from Shell Chemicals) was used as
the compatibilizing agent. The propylene units on the di-block are physically compatible
with the polypropylene and styrene units are compatible with the polystyrene. The
molecular weights of the respective components have been shown in Table 5.1.

Blends were prepared with 10 wt. % and 20 wt. % of the polystyrene as the
dispersed phase and 0 wt%, 2 wt.%, 4 wt% and 6 wt% compatibilizing agent. Blending
was carried out in a ZSK-3O twin screw extruder at a temperature of 180 °C for all the
zones. The extrudate strands were quenched in a water bath and pelletized. These were
then air dried. The density used for calculations was 0.93 g/ cc for polypropylene and the

compatibilizer and 1.11 g/ cc for polystyrene [Immergut et. al., 1989]

5. 2.2 Rheological

Rheological characterization was carried out on an RMS-8OO rheometer
(Rheometrics Scientific, Inc., New Jersey, USA) using a 50 mm parallel plate
arrangement. The disks were prepared by compression molding the pellets in a Carver
press under a force of 6 tons and 180 °C. The instrument oven was purged with dry

nitrogen during measurements to avoid degradation. A frequency range of 0.05-50 rad/s

134

and strains of 10-15% were applied during the measurements. A strain sweep was carried

out to determine the viscoelastic limits on the strain.

5. 2. 3 Morphological

The samples for morphological examination were prepared by placing the pre-notched
disks in the RMS 800 oven, heating them to 180 °C and placing them directly in liquid
nitrogen. Those were then fractured under liquid nitrogen. These samples were then
examined in a Phillips Electroscan 2020 environmental scanning microscope. Water
vapor at a pressure of 2-3 Torr was used as the imaging gas. The measurements were

made directly from the micrographs and the volume average radii were calculated.

135

5.3 RESULTS AND DISCUSSION
5. 3. 1 Blend Morphology

Figures 1a and 1b show the micrographs for the non-compatibilized blends while
Figures 10 thru 1i show the micrographs for the compatibilized blends. In all cases, the
particles are spherical. The volume average radii of the various dispersed phase particles
are shown in Table 5.2. The particle size increases in case of non-compatibilized blends
as the polystyrene content is raised from 10 wt% to 20 wt%. This effect is due to
coalescence. In non-compatibilized blends the interface is highly mobile and during
blending the probability of the droplets coming in contact with each other and coalescing
is high. On compatibilization, this effect is reduced as can be observed from the data in
Table 5.2. This behavior can be attributed to the reduced mobility of the interface in
presence of the compatibilizer (at the interface) [ Macosko et. al., 1996 ]. A reduced
mobility leads to a decrease in the probability of coalescence [Janssen et. al., 1993].
Besides, another effect of the presence of the block copolymer at the interface is to reduce

the interfacial tension. This in turn, imparts stability to the interface and reduces the

drive to coalesce.

5. 3. 2 Rheological Observations

Figures 5.2a and 5.2b show a comparison of the storage and loss moduli of the
components of the blend respectively. It can be seen that the storage modulus of the

polystyrene (the disperse phase) is higher than that of the polypropylene (the matrix

136
phase). As shown in Table 5.1, the viscosity of polystyrene is about 4.5 times the

viscosity of the polypropylene. This ensures that the dispersed phase particles are
spherical. The storage modulus of the compatibilizer is higher in comparison to the
matrix and the dispersed phase. Also, as can be observed from Figure 5.2a, the curve is
relatively flat over the frequency range under investigation while those for polypropylene
and polystyrene rise steadily with increasing fiequency. It appears that the compatibilizer
is in the rubbery plateau region and has a high degree of elasticity compared to

polystyrene and polypropylene, in the frequency range under investigation.

Using the experimental data, the first effort will be to gather the main changes
brought about by the addition of the compatibilizer to the components. Figure 5.3 is a
comparison of the storage moduli of the compatibilized and non-compatibilized 90/ 10
blends. It is observed that the storage modulus of the PP/ PS/ 1702 (90/ 10/ 2) falls
below PP/ PS (90/ 10) and then rises for the 90/ 10/ 4 and 90/ 10/ 6 blends. Also, the
difference between 90/ 10/ 4 and 90/ 10/ 6 is small. In case of PP/ PS (80/ 20) blend, 80/
20/ 2 rises, then drop occurs for 80/ 20/ 4 and then rises again for 80/ 20/ 6 curve. Refer
to Figure 5.4. Brahimi et. al. [1991] also observed a trend in their study on
compatibilization of HDPE with HIPS depending on the compatibilizer they used. For
the tapered block copolymer, the storage moduli fell as the extent of compatibilizer added
was increased while for a straight block copolymer, the storage moduli decreases and
then increases. Asthana and J ayaraman on the other hand have observed a distinct trend
in their studies on the compatibilization of nylon 6 with polypropylene by in-situ reaction

[Chapter 3 and Chapter 4]. The observations of Chapter 3 and Chapter 4 show that as the

137

extent of reaction and the volume fraction increased, the storage moduli also increased
progressively. What is the reason for the drop in the modulus value in externally
compatibilized blend, while in in-situ reactively compatibilized blend the elasticity is
enhanced progressively with the extent of reaction? This can be understood if the

physical events occurring at the interface are analyzed.

Figure 5.5 is a comparison of the storage moduli of the PP/ PS (90/ 10) blend with
its components. The blend curve falls between those of the components. But, a
comparison in case of 90/ 10/ 2 shows that the blend curve matches the curve for
polypropylene. Refer to Figure 5.6. The storage modulus has fallen due to
compatibilization. On increasing the concentration to 90/ 10/ 4, the blend curve rises
again and actually crosses the polystyrene at higher frequencies (Figure 5.7). Similar
observation was made for 90/ 10/ 6 as shown in Figure 5.8. A similar analysis was
carried out for 80/ 20 blends. Results are shown in Figures 5.9 thru 5.12. It is seen that
in the non compatibilized blend itself the blend curve crosses over the polystyrene curve.
This cross over behavior was observed in all the compatibilized 80/ 20 blends. The data
presented until now shows that there are two effects occurring simultaneously in the
blends. The effect of the polystyrene volume fraction and the effect of the concentration
of the compatibilizer. The next section discusses the events occurring at the interface and

explains the observations made above on the basis of these events.

138

5. 3.3 Physical Events at the interface and the value of interfacial tension

An important effect of adding the compatibilizer to the blend is to change the
nature of the interface between the dispersed and the continuous phase. The interface
goes from being ‘bare’ to ‘occupied’. The blocks of the compatibilizer hook up with the
individual components leading to compatibilization. It has been shown that if the
modulus of the compatibilizer is higher than that of the diffusing component
(polypropylene in this case), the net result is a lowering of the overall modulus. Now, the
diffusion of the component continues until the interface is saturated with the linkages
between the compatibilizer and the diffusing species. After this, firrther addition of the
compatibilizer leads to an enhancement in the elasticity of the blend. In the results of
Brahimi et. al. [1993], the interface with the tapered block copolymer did not get
saturated with theamount of compatibilzer added. This led to a continuous fall in the
storage modulus curve. But, for the straight di-block copolymer it did get saturated after

1% compatibilizer was added which led to an increase after the initial decrease.

Similar observations have been made in this study also. The storage modulus
curve shifts down for the 90/ 10/ 2 blend compared to the 90/ 10 blend but then the
curves move up. The reason is that the interface possible is saturated by the diffusing PP
chains into the -[ S- EP ] - copolymer. Further addition of compatibilizer does not
promote diffusion of the PP chains into the interface. The copolymer merely sits in the

interface, leading to an increased storage modulus. Now, the compatibilizer in the

139

interface has a surfactant effect also. This leads to reduced values of equilibrium

interfacial tension.

For non compatibilized polypropylene/ polystyrene system a value of 5 mN/ m
has been obtained experimentally [Sundraraj and Macosko, 1993]. This was checked by
using the dispersive and polar component of the components as well as the
thermodynamic theory of Helfand and Tagami [1971]. Table 5.3 shows the dispersive
and polar components for polypropylene and polystyrene [Paul, 1978]. Equation 5.4 was
used for determining the value of interfacial tension.

4rfrd 4rPrP

r°=rkufl— 2 — ‘ 2 5A
' 2 rf+rf rf+rf ( )

 

This gives a value of 6 mN/ m. The estimates from the thermodynamic theory using
Equation 5.5 lead to a value of 2.7 mN/ m. The values of the parameters used in this

theory are shown in Table 5.4 [Rudin; Immergut, 1989]. Thus, the experimentally

measured value of 5 mN/ m is reasonable.

1
r°=k1T1poz>V21§——‘2fl”+ +6(—-'j; —1’:‘; (5.5)

The parameter x is estimated from the Hilderbrand solubility parameters. The relation is

(5, —5,,)2 (5.6)

 

To estimate the interfacial tension values in this study, the storage moduli curves were
used. The storage modulus is the most sensitive to the values of the interfacial tension
[Graebling et. al., 1993]. Figure 5.13 shows a comparison of the experimentally obtained
storage moduli curve with that obtained from the model of Equation 5.2 and 5.3 for non
compatibilized PP/ PS ( 90/ 10) blend for an equilibrium interfacial tension value of 5
mN/ m. The agreement is good. For the non-compatibilized PP/ PS (80/ 20) system also
the agreement was reasonable for a value of 5 mN/ m for the equilibrium interfacial
tension, as shown in Figure 5.14. When a similar calculation was done for the PP/ PS /
1702 (90/ 10/ 2) blend, once again a value of 1 mN/ m was found to give a good estimate
as shown in Figure 5.15. That is, the interfacial tension has reduced from 5 mN/ m to 1
mN/ m while the particle size has fallen from 1.1 pm to 0.9 um. For PP/ PS / 1702 (90/
10/ 4) blend however, there was disagreement over the entire frequency range even for an
interfacial tension, values as low as 1 mN/ m, as shown in Figure 5.16. This observation
is similar to that of Brahimi et. al. [1991]. This behavior was noted in all the cases except
for the PP/ PS/ 1702 (80/ 20/ 4) blend. Thus, how does one estimate the value of the

interfacial tension in these blends ?

In order to estimate the interfacial tension values from the storage moduli data,
one has to focus on the low frequency plateau as suggested by Graebling et. al. [1993]. In
the observations of the present system, there is no clear cut evidence of such a plateau. In

order to delineate the low frequency transitions, it became crucial to abstract the

l4]

contributions due to the interface alone. To do this, an empirical relation shown in
Equation 5.7 was used. It subtracts the weighted average of the components from the

storage moduli of the blend.

Gin: = Glilend — [¢Gd +(1‘ (11)ij (5-7)

This was used for the storage moduli values obtained experimentally as well as from the
model. In order to determine the transition frequency, tangents were drawn on the G’int
obtained experimentally. An example is shown in Figure 5.17 for PP/ PS/ 1702 (90/ 10/
6) blend. Using this frequency and relation shown in Equation 5.8 [Graebling et. al.,

1993] the values of the equilibrium interfacial tension were obtained.

’10 =(Ram)1(19k+16)(2k+3—2¢(k—1))1 (5.8)

4F° 10(k + 1) - 2¢(5k + 2)
It is seen that the values do not follow a trend. Moreover, even these values did not give
a good agreement between the model and the experimentally obtained storage moduli
curves. The addition of the compatibilizing agent has changed the rheological behavior
such that it fundamentally differs from the non-compatibilized blends. This change can

be attributed to two factors:
0 The changes due to the interfacial product

0 The changes due to the formation of micelles

142

5. 3. 4 The issue of extent of coverage
Assuming all the block copolymer to be at the interface the interface area
occupied by the copolymer chains is given by Equation 5.9. It is similar to that defined in

directly reacted blends between nylon 6 and maleated polypropylene.

interface area/ volume _ Mchp
copolymer chains/ volume _ N A pc¢c

 

(5.9)

In Equation 5.9, SSp is the specific interfacial area ( m2/ m3) which is given by Equation

5.10.

s =— (5.10)

Table 5.5 shows the values calculated for 2 for the different blends. For a given
weight fraction of the dispersed phase, as the amount of copolymer is increased the value
of 2 falls as it should as the number of di block grafts formed in the interface also
increases. However, the values of equilibrium interfacial tension do not reduce
correspondingly. This has been shown in Figure 5.18. The equilibrium interfacial
tension falls with the addition of a small amount of the compatibilizing agent. Further
addition does not cause a decrease in the equilibrium interfacial tension. As the
observations show, after a critical point, the effect of adding the block copolymer is to
enhance the elasticity. This observation is similar to that in the directly compatibilized
blends. But, the fundamental difference is that this increase is not monotonic. In the
earlier stages of the addition in fact there is a reduction in the elasticity. This effect is

attributed to the dissolution of the matrix in the copolymer. Moreover, as observed from

143
Figure 5.2a , the storage moduli of the compatibilizer is much higher than those of the

components and hence the larger relaxation times as compared to those of the
components. That is, the relaxation effects due to the interface have been masked. In
addition to this, the ratio of the viscosities of the disperse phase to the matrix is around
3.5, and the ratio of the relaxation times is 1.5. The analysis by Graebling and 00-
workers [1993] shows that as the viscosity ratio increases, the position of the secondary
plateau also falls to lower values of the storage moduli. Also, the frequency of onset of
this secondary plateau also is reduced. A combination of this factor along with the high
elasticity of the compatibilizer have not allowed the appearance of the secondary plateau
in the observable frequency range. As shown in a related work in Chapters 3 and 4, to
capture the secondary plateau, it is essential that the materials be chosen such that their
relaxation times are very low. This permits the unmasking of the relaxation due to the

interface and brings it in the observable frequency range.

5.4 CONCLUSION

This study has investigated the role of progressive addition of an -S-EP- di block
copolymer to a blend of PP and PS on its rheology, with focus on the nature of the
interface and the value of interfacial tension. The data show that the equilibrium
interfacial tension is reduced from a value of 5 mN/ m to 1 mN/ m. But, in addition to
this reduction, an enhanced elasticity is imparted to the system due to the addition of the
compatibilizer. This elasticity is due to the two factors. Firstly, the compatibilizer which

is located at the interface and secondly the compatibilizer in the form of micelles. This

144

observation is similar to that observed in reactively compatibilized nylon 6 and maleated
polypropylene the results of which are shown in Chapters 3 and 4 respectively. However,
in directly reacted blends the increase was monotonic which is not the case in externally

compatibilized blends.

145

Table 5.1: The properties of the components used in the study.

 

 

Mn 710 0':
(kg/ kg-mole) (Pa-s) (s)

Polypropylene 54000 3700 3
Polystyrene 220000 14000 4.5

 

Table 5.2: The volume average radii of the disperse phase in the various blends.

 

 

Rv 8,1, x 10"
(pm) (312/ m3)
PP/ PS (90/ 10) 1.12 0
PP/ PS/ 1702 (90/ 10/ 2) 0.9 2.7
PP/HPS/ 1702 (90/ 10/ 4) 0.8 3.0
PP/ PS/ 1702 (90/ 10/ 6) 0.6 3.7
PP/ PS (80/ 20) 1.86 0
PP/ PS/ 1702 (80/ 20/ 2) 1.0 5.0
PP/ PS/ 1702 (80/ 20/ 4) 1.0 4.9
PP/ PS/ 1702 (80/ 20/ 6) 0.8 6.1

 

Table 5.3: The polar and dispersive component of polystyrene and polypropylene [Paul,
1978} '

 

l"° (mN/ m) I")P (mN/ m) [‘0D (mN/ m)
Polypropylene 21 .0 0 21 .0
Polystyrene 32. 1 5 .4 26.7

 

146

Table 5.4: The values of the parameters used in to estimate the interfacial tension using
the thermodynamic theory [Rudin 1993; Immergut et. al., 1989]

 

 

8i (rO/ Mm)i Ini density

(cal/ cc)“2 nm) (g/ mole) (g/ cc)
Polypropylene 8.3 835 x 10" 42 0.93
Polystyrene 9.0 670 x 10" 105 1.11

 

Table 5.5: The variation of the extent of coverage and the equilibrium interfacial tension
with progressive addition of the compatibilizing agent.

 

 

c 2 x 10“ I”
(g-mole SEP/ kg-PS) (m2/ no. of copolymer) (mN/ m)

PP/ PS (90/ 10) 0 0 5
PP/ PS (90/ 10/ 2) 0.001 0.35 1
PP/ PS (90/ 10/ 4) 0.003 0.18 1
PP/ PS (90/ 10/ 6) 0.004 0.15 1
PP/ PS (80/ 20) 0 0 5
PP/ PS (80/ 20/ 2) 0.0007 0.62 1
PP/ PS (80/ 20/ 4) 0.001 0.31 1
PP/ PS (80/ 20/ 6) 0.002 0.24 1

 

147

 

Figure 5.1a: Micrograph showing the morphology of PP/ PS (90/ 10).

148

 

 

Figure 5.1b: Micrograph showing the morphology of PP/ PS (80/ 20).

149

 

 

Figure 5.1c: Micrograph showing the morphology of PP/ PS/ 1702 (90/ 10/ 2).

150

 

 

Figure 5.1d: Micrograph showing the morphology of PP/ PS/ 1702 (90/ 10/ 4).

151

 

Figure 5.1c: Micrograph showing the morphology of PP/ PS/ 1702 (90/ 10/ 6).

152

 

 

 

Figure 5.1f: Micrograph showing the morphology of PP/ PS/ 1702 (80/ 20/ 2).

153

 

 

Figure 5.1g: Micrograph showing the morphology of PP/ PS/ 1702 (80/ 20/ 4).

154

 

Figure 5.1h: Micrograph showing the morphology of PP/ PS/ 1702 (80/ 20/ 6).

6', Pa

155

 

 

 

 

 

 

 

 

 

 

 

1.00E+05 l
l _
l M
M...’ . . O
‘ fl 0 . . . A ‘
1.0015104.___.~ ,r". ; 0 ’ . ... ‘ ‘
I’I’. i ' . A ‘
. A
‘ C) . . ‘ ‘
1 o . ‘ ‘ ‘
l o ‘
I . ‘ . 1
1.00903 .__-iVW L : ° . ‘*
I . ‘
1 , .
a ‘ ‘
. . l A A G (MI a
C .. ‘ ‘ O G (R)! E
100302 _______ _l - _._G(1702).Pa
- !
1 1
1.005101 1
1.00502 1.00501 1.005000 1.005001 ‘ 1.005002
Freq. radls

Figure 5.2a: Comparison of the storage moduli of the components of the blend.

6", Pa

156

 

 

 

 

 

 

 

 

 

 

1.00E+O5 I
l
|
\ Mfr;
1.005104 ._____ _w I=r""—:-‘::::$ .: z . 1
i M. o . . . A; ‘
A
(fr? 0 O . ‘ A
III I . . A ‘
I 0 A
I . ' 1
1 -° 1‘
o ‘ ‘
I . A
9 A
1.00E+03 .__ -__ . 01‘ ‘ A ‘ G"(FH. Pa
° 1 ‘ . G'1PS). Pa
A
i ‘ —-—--G" (1702). P8
. 1 ‘
. E I
1.00902 ' J . fl
1.00502 1.00E01 1.00E+00 1.00E+01 1.00E+02
Freq, radio

Figure 5.2b: Comparison of the loss moduli of the components of the blend.

6', Pa

Figure 5.3: Comparison of the storage moduli of the 90/ 10 blends with different

157

 

 

 

 

 

 

 

 

 

 

 

 

1.00505 I
I ...
. G' (PP/PS(90I10)). Pa WV ‘
A
. G'(PPIPS(90/10/2)).Pa x" “...
1.005+04._—._G'(PHPS(90/1014)),Pa . ,3” .:=.'
—.._G'(FPIPS(9011016)). Pa x" . ‘ 0 .
”a“ ‘ ‘ o
I ’3 5 O .
I ,. .
48 A ‘ O .
48 I . .
4‘ ..
. ,4‘ A '
1.00503 .WW 1 4‘ A "
T 0“ ‘ g .
.4“ .
.4“ 1
.4- . 1 .
,/. I ‘ o ‘
1.005102 .__.._ '7". ; ° L
i. . I
A . ‘
o I I
I I
I I
100501 I .
1.00502 1 00501 1.005100 1.005101
Freq, radls

amounts of the compatibilizing agent.

 

158

 

 

 

 

 

 

 

 

 

 

 

 

 

1.005105 I
. G'(FPIPS(80I20)). Pa
. G'(FPIPS(80/20I 2)). Pa
100504 __ _._G'(FPIFS(80I2014)).Pa
' T —._G' (PP! 53(30120/ 6)). Pa
I
I
I ,
. O
G I ’0’.
“2. 1.005103“. -2 _ _ _ __2 ___/:3?)
is I ,x; ‘ ‘
0’. A .’
O O O I,
/‘1/' e I
,x‘ '. v"
/ ° 4 '
. .,-’
1.005102 “___.” ___...‘A ___.-r‘
a’ I
I
I
I
1.005101 I
1.00502 1.00501 1.005100 1.005101
Freq. radls

 

1 .00E+02

Figure 5.4: Comparison of the storage moduli of the 80/ 20 blends with different
amounts of the compatibilizing agent

159

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.005105 I I
. can. Pa
O G ($)I a
_G' PPIPS 90/10 .Pa
1.00804 . _ fl —. I I I I)
I
I O
I 0
O
a I .
. 1.005403._ ____ e
a . ,- .
| O I
o . ’. ‘ ‘
A
O I ‘
O I ‘ I
I . I
1.005102 . _ _ * . I
. I
" . I I
A I I
1.005101 I I
1.00502 1.00501 1.00E+OO 1.005401
Froq,radls

1 .00E+02

Figure 5.5: Comparison of the storage moduli of the PP/ PS (90/ 10) blend with its

components.

160

 

  

 

 

 

 

 

 

 

 

1.00505 I
. G(PPI. Pa I
. 6' (PS). Pa . . .
_._G'(FPIPSI1702(90/1012)), Pa I . o ’
0
1.00504 .- __.__ __I -._ ' I
O
O
I . I
O
E 100503 - . I - ' I
6 I e ’(‘V
, o I-
1
0 I
, ,
. I / I
I, I
1.00802 1 ; 7’ !_ _____'{_______-7V I I
I I
// I I I
I I. I I
A I I
I I I
1.00501 fiI I .
1.00502 1.00501 1.00500 1.00501 1.00502
Froqnadlc

Figure 5.6: Comparison of the storage moduli of the PP/ PS/ 1702 (90/ 10/ 2) blend with
its components.

161

 

 

 

 

   

 

 

 

 

 

 

 

1.00505 I I
I
. G'(FP),Pa I
o G'(PS). Pa . o . .
_._G(PP/PSI1702(90/10/4)),Pa 1 ° ' ‘ . 1
100504 .__._._ I , 1 ‘
I . I. ‘
I ..
I o '
H I .
"t 1.00503 .5____fl__I. = '14 _ ‘I
:9 I o ’./ . I
i’ ,- . ‘ I
. I / , 1 I
. A
O / ‘ I
1.00502 ._- wan-LII ‘ I
I
. I I
I I I
1.00501 I I I
1.00502 1.00501 1.00500 1.00501 1.00502

Freq. radio

Figure 5.7: Comparison of the storage moduli of the PP/ PS/ 1702 (90/ 10/ 4) blend with
its components. I

162

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100505 I
A G (Mr H
, G'(PS), Pa
1.00804 «- _._G’ (PP/ PSI 1702(90/10l6)), Pa
I 0
lb
. O
I o
I O
a I . ./ A ‘
f; 1.00E+03 .__ 55 F _— I e ,- -‘-
C9 0 I 4
. A
I ,x’ . 1
o I,- ‘
o / A
. 1' .
_z' , .
1.00502 '1
‘ I
. I I
I
. I
1.00501 I -
1,005.02 1.00501 1.00500 1.00501 1.00502

Freq. rad/s

Figure 5.8: Comparison of the storage moduli of the PP/ PS/ 1702 (90/ 10/ 6) blend with
its components.

163

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.00E+05 I
I
O G (Ml a
. I
. G'(PS). Pa . - r '1'
I .1 A ‘ O O
1.OOBO4.__._‘G(PHPS(8OIZOII'Pa ..‘ 6..
I ‘ .. °
I o
I . °
I I ‘ ‘
I ‘ ,
. A ‘ °
I ‘ .
I A 0 °
“1 1.005403 .I_..._..I- - --_ ‘ e
in u ‘ 1
. O
I O
I A
I A
I 1 ,
‘ o
1.00E+02 ._ __--______ .m, I
1.00801 I I
1 DOE-02 1.00501 1 DOE-+00 1 .00E+01
Frequency, mil 8

 

1 .00E+02

Figure 5.9: Comparison of the storage moduli of the PP/ PS (80/ 20) blend with its

components.

1.00805

164

 

 

 

 

  

 

 

 

 

 

 

 

 

1.00802

I l
I
I
. G(PP). Pa
. G'(PS). Pa
1.00804 ~ ~——- _._G' (PF! PSI 1702(80120/2». Pa
I
I .
a ’ o . .l
{1. 1.00503 .W’ .— »- ~— e— I——-—0—— Q - ‘ H
0 . ’(I A
I o ‘
g I’- I
l ’ ‘
O
O I]. ‘ A I
./ . I
. |
1.00E+02 ._- _— - -l A ‘1' ‘
. I
I I .
1m I I I
1.00502 1 cos-01 1.00900 1.00801
Froq,rIdII

Figure 5.10: Comparison of the storage moduli of the PP/ PS/ 1702 (80/ 20/ 2) blend

with its components.

165

 

1.00805 I
I
s 30'")- PB
. G(%), a
100 .__._G'(PPIPSI1702(8012014)).Pa

 

 

 

 

 

 

I
I
I
I
I
I
I

 

 

 

 

 

 

0
°'_ 1.00503 ”___ __J --. ° {f2 “
b o / A
O
O I
0 ll" ‘
O I’- ‘
O 1., ‘
A
1.00E+02 ‘r— __ _ _'Z_I
.’ . I
‘ I
. I
1.00501 I I fl
1.00502 1.00501 1.00E+00 1.00E+01
Bundle

1 .00E+02

Figure 5.11: Comparison of the storage moduli of the PP/ PS/ 1702 (80/ 20/ 4) blend

with its components.

166

 

 

 

 

 

    
  

 

 

 

 

 

100505 I
I I
. I
. G'(PPI.PB I
. G'(PS).Pa I
_G‘ PPIPSI1702 80I20I6 .Pa '
1.00304 ... _' ( . ( )) . I
I o ‘ . I
I .
I Q . I
I o .1 ‘
I . . .I ‘ I
I . I’ ‘ ‘ I
n.
_.1.00E¥03.‘*IIII_,I 3 , a
0 I o ' 5 I
I A
,I',- .
O l ‘
O/III ‘ A I
1.00902.” _I _ I _ I ..IE___ ___ , I
. I I I
A I I
I I I
100901 I , I
1.00502 1.00601 1.00E+00 1.00E+01
Froq,rIdII

 

1.00E+02

Figure 5.12: Comparison of the storage moduli of the PP/ PS/ 1702 (80/ 20/ 6) blend

with its components.

6'. Pa

167

 

 

 

 

 

 

 

 

 

 

1.00305 I
I
. 6' (exp). Pa
1.00304 ._.____ o 6' (model). Pa . j
I
, , I: .
I ° .
0 A
I ° ‘
O A
0 A
I I: 1
0
1.00303 _“ I :1 f
I , 2
I , .
I ° ‘
I . ‘
I , A
I .
. O
I ‘ I
1.00302 III- 2 I I
o ‘ I
. I
I I
I I
1.00301 I f
1.00502 1.00501 1 00300 1.00501
Freq, radio

1 .00E+02

Figure 5.13: Comparison of the storage moduli obtained from the model with that

obtained experimentally for PP/ PS (90/ 10) (I0 = 5 mN/ m, R= 1.12 pm).

168

 

 

 

 

 

 

 

 

 

 

 

1.00305 .
I . 0
. G'(model). Pa . . ' ‘ .
. .G(eXP)9fi ~. .1..“‘
1.00m ..-—___ . _IL . .fi‘L". A
I . A I
I I I A
. O A
o A
I - .
. A
o ‘ ‘
I 1 .
. A
C I . A I
f‘: 1.00303I---_-_ """T . i . J
o I O A
I 0 ‘
I ' ‘
. A
I
. A
. l I
1.00E+02 .__.._ F. -0 :_I _.
. ‘ I I
A I I
I I
I I
1.00301 , I
1.00502 1 .OOE-01 1.00E+00 1 008-01 1 0051-02
Froq,radls

Figure 5.14: Comparison of the storage moduli obtained from the model with that
obtained experimentally for PP/ PS (80/ 20) (1"0 = 5 mN/ m, R= 1.86 pm).

169

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.005105 I
I .
G'(rmdel) 1 . :3 °
1.00304 Mm. ‘ 'pa = g .
A CD
I . 3 '
A O
I .-
_ :0
I A
A O
a I :'
. 1.00E003.___A7;_-I ”___ 1.
in T ‘ ; ‘I
I . 0
A O
I --
A . I
I A . I
I ‘ ' I
0
1.005102 ' .g m
. I I
A
' I
6
I
I
1.00301 .
1.00502 1.00501 1.00800 1.00301 1.00E+02
Froq,radls

Figure 5.15: Comparison of the storage and loss moduli obtained from the model with
that obtained experimentally for PP/ PS/ 1702 (90/ 10/ 2) (1'0 = 1 mN/ m, R= 0.9 pm).

170

 

 

 

 

 

 

 

 

 

 

1.00E+05 I I. I
I
.G‘(rmdel,5rerm),Pa
IG'blexlF’a -'
.G'I I.3nrNIm,Pa _-
xG‘mdel,1nNm,Pa I .I
I I
100304 . I .' .gi"
' ‘W”‘“‘W“‘"’I W I u o x "
I 'X
I I . .iil
. ' {‘x
| I.- Oi
I II °x
I .§X
I . ox
' I ' 9X
9:. 1.00303 - . _' ”12‘
0 I . 0X I
I ' °"
I 0X I
OX I
I' ox
I 0X
II.x
1.00302.___ ' ’I"
I 02
OXI
ox I
I
I I
100301 I I
1.00502 1.00501 1.00800 1.00E+01 1.00E+02
Froqnadls

Figure 5.16: Comparison of the storage moduli obtained from the model with that
obtained experimentally for PP/ PS/ 1702 (90/ 10/ 4) for different values of equilibrium
interfacial tension (R= 0.8 pm).

171

5.00E+03

 

 

 

 

 

 

 

 

 

 

 

I I
I .
I
I
400303 “ "“ . G'int(model). Pa _
. G'int (exp). Pa
I I
3.00303 ._ __ I 3-333y I
I I
I
I I .
“1 2.00303 .3 _ M -_ ._. In
in a
I I
I I
1.00303 .,_ _. ______ _____ I .
I n
I
i - I
. I . - -
0.006400._-._____ ‘— ——,--:--°—-I-O—O-—O-—o~o—.—. , . 0 . o . . . 0
I
4.00303 L
1.00302 1.00301 1.00300 1.00301

Freq, rad! s

Figure 5.17: Comparison of the G’int obtained from the model with that obtained
experimentally for PP/ PS/ 1702 (90/ 10/ 6) (1'0: 1 mN/ m, R= 0.6 pm).

172

 

['0th m)

 

 

 

0 0.001 0.002 0.003 0.004
c (g-mole SEPI kg-PS)

 

 

C
III-

Figure 5.18: The variation of the equilibrium interfacial tension with varying amount of
the compatibilizing agent.

CONCLUSIONS AND RECOMMENDATIONS

 

Chapter 6

 

6.1 CONCLUSIONS

The main questions investigated in this work center around the effect of the nature
of the interface the changes in which are brought about by the interfacial reaction.

. Does the equilibrium interfacial tension in reactively compatibilized blends fall due to
interfacial reaction?

The results in Chapters 3 and 4 respectively show that the value of equilibrium
interfacial tension in reactively compatibilized nylon 6 and maleated polypropylene fall
due to interfacial reaction. In Chapter 3 a study of nylon 6 (containing lubricants)
blended with neat polypropylene and maleated polypropylene containing 0.8 wt% of
I maleic anhydride was carried out. The value of equilibrium interfacial tension fell from 4
mN/ m to 1 mN/ m. In addition, it was seen that interfacial reaction imparts an additional
elasticity to the blend.

The study presented in Chapter 4 investigates the effect of progressive extent of
interfacial reaction on the equilibrium interfacial tension. It was carried out by blending
nylon 6 (containing no lubricants) with neat polypropylene and three different grades of
maleated polypropylene containing 0.15 wt%, 0.30 wt% and 0.80 wt% of maleic
anhydride respectively. The measurements show that the equilibrium interfacial tension
falls from 8 mN/ m to 7 mN/ m to 4 mN/ m for progressively increasing extents of

maleation.

173

174

o What are the accompanying effects (in rheology and morphology) besides the
reduction in equilibrium interfacial tension in reactively compatibilized blends?

The observations in Chapter 3 and Chapter 4 show that interfacial reaction leads
to a finer morphology. The particle size reduces fiom a few microns to sub-microns
range. It has been shown that one important effect of interfacial reaction is the
suppression of coalescence. Interfacial reaction imparts immobility to the interface and
thus a reduced rate of coalescence. In addition, a lower value of interfacial tension
reduces the thermodynamic drive to coalesce. It is seen that a very small amount of
reaction is required to attain a finer morphology. Increased extents of reaction does not
necessarily lead to a correspondingly finer morphology. The increasing extent of reaction
however did effect the rheological behavior. The reaction imparted an elasticity to the
blend. The model values under predicted the storage moduli in such blends. To account
for this additional elasticity surface shear modulus as defined by the model of Palieme
was employed. In absence of actual measurements of this modulus and the interfacial
thickness estimates of this parameter were made. On using this parameter the results
showed a considerable improvement.

0 How do the trends compare with externally compatibilized blends?
The results of Chapter 5 present the trends obtained in blends of polypropylene and
polystyrene compatibilized with -[S-EP]- di block copolymer. It is observed that the

equilibrium interfacial tension falls from 5 mN/ m in non compatibilized blends to l mN/

175

m in compatibilized blends. In addition, similar to the directly reacted blends the
compatibilized blends exhibited an enhanced elasticity.

o How good is the rheological model of Palieme applied in this work?

The results of Chapter 3, 4 and 5 show that the model of Palieme is applicable in reactive
blends for a given range of the concentration of the dispersed phase. For non
compatibilized blends however the agreement is good with the experimental values for a
more concentrated blend. An important limitation of the model is that the components
have to be chosen carefully so that the secondary plateau (a result of the equilibrium

interfacial tension) is accessible.

6.2 RECOMMENDATIONS AND FURTHER WORK

0 Quantify the extent of interfacial reaction

In this study the extent of interfacial reaction was not quantified. It was assumed that all
the maleic anhydride in the interface is reacted with the amine. This is a reasonable
assumption since there is 15. 74 g-mole of amine/ kg- nylon 6 compared to a maximum
of 0.008 g-mole of -MAH/ kg- PP-MA. However, in order to actually quantify this effect
it is important to make direct measurements of the extent of interfacial reaction and relate

it to the equilibrium interfacial tension.

0 Measure the interfacial thickness
The results in Chapter 4 indicate that in order to understand the issue of the extent of

coverage of the interface, thickness of the interface must be directly measured.

176
Theoretical work [0’ Shaughnessy and Sawhney, 1996 ] shows that interfacial reaction

should lead to a reduction in the interfacial thickness. This must be verified by using

techniques such as ellipsometry.

o Delineate the visco-elastic effects from the interfacial effects
In order to delineate the effects of the viscoelastic nature of the polymer from the
interfacial effect, the following experiments should be designed:
1. Synthesize polymers such that the viscosity stays constant and the number of
reactive sites is varied.
2. Synthesize polymers such that the number of reactive sites is a constant while

the viscosity is varied.

10.

11.

12.

13.

14.

15.

16.

BIBLIOGRAPHY

. “Polymer Handbook”, Ed. Brandrup J ., Immergut E.H., Third Edition, Wiley

Interscience Publication, New York, NY 1989.

“Reactive Extrusion. Principles and Practices”, Ed. M. Xanthos, Hanser Publishers,
New York, NY 1992.

. “The Elements of Polymer Science and Engineering”, A. Rudin, Academic Press,

New York, 1993.
Anastasiadis S.H., Gancarz I., Koberstein J .T., Macromolecules 1988, 21, 2980-2987.
Bentley B.J., Leal L.G., J. Fluid Mech., 167:219-240, 1986.

Bentley B.J., Leal L.G., J. Fluid Mech., 1672241-283, 1986.

Borggreve R.J.M., Gaymans R.J., Polymer, 30, 63(1989).

Brahimi B., Ait-Khadi A., Ajji A., Jerome R., Fayt R., J. Rheol., 35(6), 1991.

Broseta D., Fredrickson G.H., Helfand E., Leibler, Macromoleculesl990, 23, 132-
139.

Carriere C.J., Cohen A., Arends C.B., J. Rheology 33(5), 681-689(l989).

Cho K., Jeon H.K., Park C.E., Jim J., Kim K.U., Polymer, 37(7), 1117-1122 (1996).
Choi S.J., Schowalter W.R., Phys. Fluids 18, 420-427(1975).

Cohen A., Carriere C.J., Rheologica Acta 28:223-232 (1989).

Cox R.G., J. Fluid Mech. (1969), 37(3), 601-623.

Elemans P.H.M., Janssen J.M.H., Meijer H.E.H., J. Rheol., 34(8), 1311-1325(1991).

Elmendorp J.J.., Polym. Eng. Sci.26(6), March 1986, 418.
177

178

17. F ayt R., Jerome R., Ph. Teyssie, J. Polym. Sci., Polym. Lett. Ed., 24, 25(1986).

18. Ferry J .D. in “Viscoelastic Properties of Polymers”, John-Wiley and Sons, Inc., New
York, NY 1961.

19. Frankel N.A., Acrivos A., Chem. Eng. Sc., 22, 847-853(1967).

20. Germain Y.B., Ernst B., Genelot 0., Dhamani L., J. Rheol. 38(3), 1994.

21. Grace H.P., Chem. Eng. Commun. Vol. 14, p. 225.

22. Graebling D., Muller R., Colloids and Surfaces, (89-103)]991.

23. Graebling D., Muller R., J. Rheology 34(2), Feb. 1990.

24. Graebling D., Muller R., Palieme J .F ., Macromolecules, 26, 320-329(1993).
25. Gramespacher H., Meissner R., J. Rheol., 36(6), 1127(1992).

26. Helfand E., Sapse A.M., J. Chem. Phys., 62, 1327 (1975).

27. Helfand E., Tagami Y., J. Chemical Phys., 56(7), Apr. 1971.

28. Helfand E., Tagami Y., J. Chemical Phys., 62(4), Feb. 1975.

29. Helfand E., Tagami Y., Polym. Lett., 9(1971), 741-746.

30. Helfand S., Bhattacharjee S.M., Fredrickson G.H., J. Chem. Phys. 91, 7200(1989).

31. Hosoda S., Kojima K., Kanda Y., Aoyagi M., Polym. Networks Blends, 1,51-59
(1991)

32. Ide F., Hasegawa A., J. Appl. Polym. Sc., Vol. 18, 963(1974).

33. Janssen J .M.H., Meijer H.E.H., Polym. Eng. Sci.35(22), Nov. 1995, 1766.
34. Kennedy M.R., Pozrikdis C., Skalak R., Comp. Fluids, 23, 251-254(1994).
35. Lee J., Yang S., Polym. Eng. Sci., Mid-Dec. 1995, 35(23), p. 1821.

36. Lirn S., White J.L., Polym. Eng. Sci., Mid-Feb. 1994, 34(3), p. 221.

37. Loewenberg M., Hinch E.J., J. Fluid Mech., 321, 395-419(l996).

38.

179

Matsumoto T., Chiyoji H., Onogi S., Transactions of the Society of Rheology 19:4
541-555 (1975).

39. Nishio T., Suzuki Y., Kojima K., Kakugo M., J. Polym. Eng., Vol. 10, Nos.l-3, 1991.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

0’ ShaughnesSy B., Sawhney U., Macromolecules, 29, 7230(1996).
Okamoto M., Inoue T., Polym. Eng. Sci., Mid-Feb. 1993, 33(9), p. 175.
Oldroyd J .G., Proc. Roy. Soc. London Ser. A 218, 122-132 (1953).
Palieme J .F., Rheologica Acta 29:204-214 (1990).

Paul D.R., Barlow J .W., Keskkula H., in “Encyclopedia of Polymer Science and
Engineering”, 2"d Edition, Ed. Kroschwitz J .I., Wiley, New York, 1988, Vol. 12.

“Polymer Blends”, Paul D.R., Newman S., Academic Press, New York, 1978, Vol. 1.
Rayleigh Lord, Proc. London Math. Soc. 10, 4(1878).

Rumscheidt F.D., Mason 8.6., J. Colloid Sci., 17, 260(1962).

Sangani A.S., Mo G., Phys. Fluids, 6(5), 1994.

Scholz P., Froelich D., Muller R., J. Rheology 33(3), 481-499 (1989).

Scott C.E., Macosko C.W., Intern. Polym. Proc. X (1995) 1, p36.

Scott C.E., Macosko C.W., Polym. Eng. Sci.35, 1938(1995).

Serpe G., Jarrin J., Dawans F., Polym. Eng. Sci.30, 553(1990).

Sperling L.H. in “Introduction to Physical Polymer Science”, 2"d Edition, Wiley, New
York, 1992.

Stone H.A., Ann. Rev. Fluid Mech. 1994, 26:65-102.

Sundraraj U., Macosko C.W., Macromolecules, 28, 2647(1995).

56. Taylor G.I., Proc. Roy. Soc. London, A138, 41(1938).

57. Taylor G.I., Proc. Roy. Soc. London, A146, 501(1934).

58. Tjahjadi M., Ottino J.M., Stone H.A., AIChE J., 40, 3(385).

180
59. Tomotika S., Proc. Roy. Soc. London, A150, 302(1936).

60. Tomotika S., Proc. Roy. Soc. London, A150, 322(1935).
61. Wu S. in “Polymer Interface and Adhesion”, Marcel-Dekker, New York, NY 1982.
62. Wu S., Polym. Eng. Sci.27, 335(1987).

63. Wu S., Polym. Sci. Polym Phys. Ed., 21, 699(1983).

111011an STATE UNIV. LIBRRRIES
IIIII IIIIIIIII IIII III IIIII IIII II IIIIIIII IIII IIIIII III III IIIIIII
31293017163597