‘ ‘nmfm
rW.

2:7,“... yth.
1.1.2713 . .
u. r.
36nd. “$1,.
“1.4%.. Erma
,. 2...»
. 4..

r4 ..

4%...
v. ,

a... he». 5:
. .. 0V2
.. ....:..»r...,
. .
. Li.- 21.14..
.3. Luna...

.3. ..

4,1 LL}
w? a 3.3%... .
.axmungmfiv
a. .W H.

La. .3 .,
ink

.v
u :71
‘91

.M
an

.0 ‘ Z: and»

. . L: :t a.

{it .0 “a. . .f .

. .u‘ .3»? . 1.,

1.. .7 ... .3? v..
. L

2.:

a. r
. 7‘7,
.. é

.,
:1 1.:
4 .‘

.

54.... PAP»: V ntmh.
u” £28.:
.\‘o >\ h v 4

.1

 

 

. . V1.1? ta

Jamil]

 

This is to certify that the

dissertation entitled

The Effect of Reaction Processing on
the Physical, Morphological, and Barrier Properties of
Poly(ethylene 2,6-naphtha1ene dicarboxylate)/
Poly(ethylene terephthalate) Blends

presented by

Rujida Uthaisombut

has been accepted towards fulfillment
of the requirements for

Ph.D . degree in Packaging

 

   

Major professr

mama? /I. 0:2 0&/

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

 

 

 

LIBRARY

Michigan State

Unlversity

 

 

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

 

DATE DUE

DATE DUE

DATE DUE

 

Zilll

 

060511

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DateDuepGS—p. 15

 

THE EFFECT OF REACTION PROCESSING ON THE PHYSICAL,
MORPHOLOGICAL, AND BARRIER PROPERTIES OF
POLY(ETHYLENE 2,6-NAPHTHALENE DICARBOXYLATE)/
POLY(ETHYLENE TEREPHTHALATE) BLENDS

By

Rujida Uthaisombut

A DISSERTATION

Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree of

DOCTOR OF PHILOSOPHY

School of Packaging
2001

ABSTRACT
THE EFFECT OF REACTION PROCESSING ON THE PHYSICAL,
MORPHOLOGICAL, AND BARRIER PROPERTIES
OF POLY(ETHYLENE 2,6-NAPHTHALENE
DICARBOXYLATE)/POLY(ETHYLENE TEREPHTHALATE) BLENDS
By

Rujida Uthaisombut

PET/PEN blends have been considered as a new material, combining the
economic advantages of PET with the high barrier and thermal properties of PEN, which
can extend the limitatiOn of PET packaging applications. However, PET and PEN are'
inherently incompatible. Extrusion of PET/PEN blends above the melting temperature of
both polymers results in transesten'fication, which was shown to enhance the miscibility
of the blend and improve both the clarity and barrier properties of the resultant films.
This research focused on the study of the effect of the processing conditions and the blend
composition on the transesterificatilon reaction. In addition, the effect of degree of
transesterification, the blend composition, and the orientation on the blend characteristics
was also studied.

The results showed that the primary factors controlling the transesterification
reaction were blending time and temperature, whereas the composition of the blends was
found to have little or no effect on the interchange reaction.

The PET/PEN blends with a degree of transesterification of at least 6% were

miscible, homogeneous and optically clear. Further the transesterification reaction had

less of an effect on the Tg, the barrier, and the mechanical properties. However, the Tm,
%crystallinity, density, and molecular weight average were dependent on the degree of
transesterification, regardless of how this degree of transesteiification was achieved.

The blend composition was a very important factor controlling the thermal,
mechanical, and barrier properties of the blends. A depression of the Tm, % crystallinity,
and molecular weight was found, when a small amount of PEN was added to the PET
rich phase, or vice versa. The Tm, % crystallinity and molecular weight average were
lowest, when the PEN composition was 40-50 mole %. Moreover, the density of the
blends decreased, as the PEN composition increased. However, the T8, the gas barrier
properties and the tensile strength of the blends improved when the PEN composition
increased.

Orientation process enhanced the barrier and the mechanical properties of the
blend films. The water vapor, the oxygen, and the carbon dioxide barrier properties of
the PET/PEN blend films improved 1.5-2 times, as the blend films were 3x3 biaxially
oriented. The tensile strength of the oriented blend films was 4 times greater than that of
the non-oriented blend films. Moreover, oriented PEN, and the oriented blend films

were flexible, while the non-oriented films were very brittle.

 

To the memory of Dr. Jack R. Giacin

(December 31, 1938 — December 12, 2000)

ACKNOWLEDGEMENT

I would like to take this opportunity to express my gratitude to many people who
contributed in many ways to make my dissertation possible.

First, I wish to express my deepest appreciation to Dr. Jack R. Giacin, who served
as my advisor. I thank him for his guidance, kindness, encouragement, and effort. He
always had an open door for me to ask any questions and to ask for any advises. During
the hard time for him, due to his health, he never stopped giving good comments to
improve this dissertation. He was the greatest teacher I have ever known.

I would like to thank Dr Bruce Harte, who served as my second advisor. He
always gave good advises and supported me in many ways. I thank Dr. Jayaraman, Dr.
Susan Selke, and Dr. Ruben Hernandez, the members of my committee for their valuable
contIibutions throughout my graduate study.

I would like to thank my parents, Kwan and Wannee Leepipattanawit, for their
love, support throughout my life, their encouragement to pursue higher education, and
their teaching and being great examples of how to be a good person. I wish to thank my
husband, A
Dr. Patchrawat Uthaisombut for his love, understanding, encouragement, continuous
support, and patience during my graduate study. I would like to thank Dr. Robert Rasche
and Mrs. Dorothy Rasche for their support and kindness, during my stay at Michigan
State University. I would like to thank many friends (Ubonrat Siripatrawan, Lee Youn

Suk, Cengiz Caner, and others) for their warm hospitality and support.

I thank the Center for Food & Pharmaceutical Packaging Research (CFPPR) for
their financial support throughout my graduate program. I thank Mr. Michael J. Rich
from the Composite Materials and Structures Center at MSU, and Mr. Johnson Kermit
and Dr. Long Le fiom NMR laboratory, Chemistry Department at MSU for their kind
help through my work on thermal and NMR analysis. I wish to thank Dr. Issam
Dairanich and Amoco Research Center, Amoco Chemicals Company for the technical
support and film production. I thank to Shell Company for the donation of PET and PEN

samples.

TABLE OF CONTENTS

LIST OF TABLES .......................................................................................... x

LIST OF FIGURES ..................................................................................... xiv

Chapter 1: Introduction 1

1.1. Need for Modified PET ................................................................................................ 1

1.2. Alternative Polymer ...................................................................................................... 2

1.3. Combining PET and PEN ............................................................................................. 3

1.4. Transesterification Reactions ........................................................................................ 4

1.5. Overview of the Experiment ......................................................................................... 5

1.5.1. Processing Conditions....................... ............................................................. 5

1.5.2. Blend Composition ........................................................................................ 5

1.5.3. Film Orientation ............................................................................................. 6

1.5.4. Blend Characterization ................................................................................... 6

1.6. Objectives .......................................................................................................... 7

Chapter 2: Literature Review 8

2.1. Interesting Issues of PET .............................................................................................. 8

2.1.1. PET Structure and Properties ..................................................................... ....9

2.1.2. PET Manufacturing ...................................................................................... 10

2.1.3. PET Applications ......................................................................................... 13

2.1.4. PET Modification ......................................................................................... 16

2.2. Alternative Polymer, PEN .......................................................................................... 17

2.3. Combining PET and PEN ........................................................................................... 23

2.4. PET/PEN Blends and Copolymers ............................................................................. 25

Chapter 3:Transesterification Reactions 30

3. 1. Introduction ................................................................................................................. 30
3.1.1. Miscibility of Polymer Blends ..................................................................... 30 .

3.1.2. Transesterification Reactions ...................................................................... 31

3.1.2.1. Mechanisms of transesterification reactions ................................. 33

3.1.2.2. Factors controlling transesterification reactions ........................... 36

3.2. Materials and Methods ................................................................................................ 41

3.2.1. Materials ...................................................................................................... 41

3.2.2. Reaction Processing Parameters and Blend Composition ........................... 41

3.2.3. Extrusion ....................................................................... , .............................. 44

3.2.4. DSC Analysis ........................................................................................ ' ....... 4 5

3.2.5. ‘H-NMR Analysis ........................................................................................ 47

3.2.6. Calculation of Degree of Transesterification and D value .......................... 49

3.3. Results and Discussion .............................................................................................. 52

3.3.1. DSC Analysis ............................................................................................... 52

3.3.2. Degree of Transesterification (B) ................................................................ 55

3.3.2.1. Effect of blending time on transesterification reactions ............... 56

Screw speed ................................................................................... 56

vii

Numbers of passes ........................................................................ 57
3.3.2.2. Effect of blending temperature on transesterification reactions ...58

3.3.2.3. Effect of blend composition on transesterification reactions ........ 60

3.3.3. D value ......................................................................................................... 62
Chapter 4: Characterization of PET/PEN Blends 65
4.1 . Introduction ................................................................................................................. 65
4.1.1. Thermal Properties ...................................................................................... 65

4.1.2. Molecular Weight and Molecular Weight Distribution ............................... 68

4.1.3. Density ......................................................................................................... 69

4.1 .4. Morphology .................................................................................................. 69

4.2. Materials and Methods ................................................................................................ 71
4.2.1. Film Fabrication Process .............................................................................. 71

4.2.2. Orientation Process ...................................................................................... 73

4.2.3. Analysis Methods ......................................................................................... 74
4.2.3.1. Thermal analysis ........................................................................... 74

4.2.3.2. Density measurement .................................................................... 75

4.2.3.3. Average sequence lengths of terephthalate and naphthalate units 79

4.2.3.4. Molecular weight measurement .................................................... 80

4.2.3.5. Morphological analysis ................................................................. 80

4.3. Results and Discussion .............................................................................................. 82
4.3.1. Thermal Properties of Blends ...................................................................... 82
4.3.1.1. Glass transition temperature (T3) .................................................. 83

4.3.1.2. Melting temperature (Tm) .............................................................. 89

4.3.1.3. Percent crystallinity of the blends ................................................. 93

4.3.2. Density of Blend Resins and Films .............................................................. 95

4.3.3. Average Sequence Lengths of Terephthalate and Naphthalate Units ........ 103

4.3.4. Molecular Weight of Blends ...................................................................... 103

4.3.5. Morphology ..... 110
Chapter 5: Barrier and Mechanical Properties of PET/PEN Blends 120
5.1 . Introduction ............................................................................................................ 120
5.1.1. Barrier Properties ....................................................................................... 121

5.1.2. Mechanical Properties ................................................................................ 123
5.1.2.1. Factors controlling mechanical properties ............................... 124

5.1.2.2. Tensile properties ........................................................................ 126

5.2. Materials and Methods .............................................................................................. 128
5.2.1. Measurement of Water Vapor Transmission Rate ..................................... 128

5.2.2. Measurement of Oxygen Transmission Rate ............................................. 130

5.2.3. Measurement of Carbon Dioxide Transmission Rate ................................ 131

5.2.4. Tensile Test ................................................................................................ 135
5.3.Results and Discussion .............................................................................................. 138
5.3.1. Barrier Properties ...................................................................................... 138
5.3.1.1. Water vapor permeability coefficient ......................................... 138

. 5.3.1.2. Oxygen permeability coefficient ................................................. 142

5.3.1.3. Carbon dioxide permeability coefficient .................................... 146

viii

5.3.2. Mechanical Properties ................................................................................ 150

5.3.2.1.Tensile strength ............................................................................ 150

5.3.2.2. Percent elongation at break ......................................................... 155

5.3.2.3. Young’s modulus of elasticity ................................................. 160

Chapter 6: Orientation Effect 164
6.1. Introduction 164
6.1.1. Uniaxial Orientation ................................................................................... 165

6.1 .2. Biaxial Orientation ..................................................................................... 165

6.1.3. Effect of Orientation on Polymer Properties ............................................. 166
6.1.3.1. Percent crystallinity .................................................................... 166

6.1.3.2. Barrier properties ........................................................................ 167

6.1.3.3. Mechanical properties ................................................................. 167

6.1.3.4. Effect of orientation on selected properties of PET/PEN blendsl68

6.2. Materials and Methods .............................................................................................. 169
6.3. Results and Discussion ............................................................................................ 170
6.3.1. Effect of Orientation on Thermal Properties ............................................. 172
6.3.1.1. Glass transition temperature ....................................................... 172

6.3.1.2. Melting temperature .................................................................... 173

6.3.1.3. Percent crystallinity .................................................................... 173

6.3.2. Effect of Orientation on Density of the Blends ....................................... 173

6.3.3. Effect of Orientation on Barrier Properties ............................................... 174

6.3.4. Effect of Orientation on Mechanical Properties ........................................ 174

6.4. Summary ................................................................................................................... 176
Chapter 7: Summary and Conclusions 177
7.1. Transesterification Reactions .................................................................................... 177
7.2. Blend Characteristics ................................................................................................ 179
7.2.1. Effect of Degree of Transesterification ..................................................... 179

7.2.2. Effect of Blend Composition .................................................................... 181

7.2.3. Effect of Orientation .................................................................................. 183
Appendices ................................................................................................. 186
Bibliography .............................................................................................. 219

ix

List of Tables

Tables page
Chapter 3
3.1. Processing parameters and blend composition. .......................................................... 43
3.2. The effect of the extruder screw speed on the degree of transesterification (B). ....... 57
3.3. The effect of blending temperature and time on the degree of transesterification

(B). ......................................................................................................................... 59
3.4. The effect of PEN composition on the degree of transesterification (B). ................... 61
3.5. The effect of the blending temperature and time on the degree of

transesterification (B) and D-value. ....................................................................... 63
3.6. The effect of PEN composition on the degree of transesterification (B) and

D-value. .................................................................................................................. 64
Chapter 4
4.1. List of the samples tested for Raman chemical imaging analysis. ............................. 82
4.2. The effect of the degree of transesterification (B) on the glass transition

temperature of the PET/PEN blends. ..................................................................... 84
4.3. The effect of PEN composition on the glass transition temperature on the

PET/PEN blends. ................................................................................................... 86
4.4. The comparison between calculated and measured T8 of the blend resins. ................ 88
4.5. The effect of the degree of transesterification (B) on the melting temperature (T...)

of the PET/PEN blends. ......................................................................................... 90
4.6. The effect of PEN composition on the melting temperature of the PET/PEN

blends. .................................................................................................................... 92
4.7. The effect of the degree of transesterification (B) on the percent crystallinity of

the PET/PEN blends. ............................................................................................. 94
4.8. The effect of PEN composition on the percent crystallinity of the PET/PEN

blends. .................................................................................................................... 96
4.9. The effect of the degree of transesterification on the density of the blends. .............. 98
4.10. The effect of PEN composition on the density of the PET/PEN blend resins. ....... 100
4.11. The density ofthe PET, PEN and blends ......... 101
4.12. The effect of PEN composition on the density of the PET/PEN blend films. ........ 102
4.13. The effect of the degree of transesterification (B) on the average sequence

lengths of the terephthalate (1.“) and naphthalate (LnN) unit. ............................. 104
4.14. The effect of PEN composition on the average sequence lengths of the

terephthalate (141T) and naphthalate (LnN) unit. .................................................. 105
4.15. The molecular weight average of the blend films ................................................... 106
4.16. The effect of the degree of transesterification on the polydispersity (MW/Mn) of

the blend films ...................................................................................................... 108
4.17. The effect of the PEN composition on the polydispersity (MW/Mn) of the blend

films ..................................................................................................................... 109

List of Tables (cont.)

Chapter 5
5.1. The effect of the degree of transesterification on the water vapor permeability
coefficient of PET/PEN blend fihns at 378°C. ................................................... 139
5.2. The effect of PEN composition on the water vapor permeability coefficient of the
PET/PEN blend films at 37 .8°C. .......................................................................... 141
5.3. The effect of the degree of transesterification on the oxygen permeability
coefficient of PET/PEN blend films at 25°C ....................................................... 143
5.4. The effect of PEN composition on the oxygen permeability coefficient of '
PET/PEN blend films at 25°C. ............................................................................. 145
5.5. The effect of the degree of transesterification on the carbon dioxide permeability
coefficient of PET/PEN blend films at 25°C. ...................................................... 147
5.6. The effect of PEN composition on the carbon dioxide permeability coefficient of
PET/PEN blend films at 25°C. ............................................................................. 148
5.7. The effect of the transesterification degree (B) on the tensile strength of
PET/PEN blend films ........................................................................................... 151

5.8. The effect of PEN composition on the tensile strength of PET/PEN blend films. ...152
5.9. The effect of the transesterification degree (B) on the % elongation of PET/PEN

blend films. .......................................................................................................... 156
5.10. The effect of PEN composition on the % elongation of PET/PEN blend films. ....157
5.11. The effect of the transesterification degree (B) on the modulus of elasticity of
PET/PEN blend films ........................................................................................... 161
5.12. The effect of PEN composition on the modulus of elasticity of PET/PEN blend
films. .................................................................................................................... 162
Chapter 7
7.1. Summary of effect of degree of transesterification and blend composition on
selected properties of PET /PEN blends. .............................................................. 180

7.2. The equations showing the relations between the blend composition and the
barrier and mechanical properties of the blends obtained at 300°C and 1 pass. ..182

Appendices
Appendix A
A1. Analysis variance of degree of transesterification of PET/PEN blend resins and

films. .................................................................................................................... 187
A2. Analysis variance of D-values of PET/PEN blend resins and films. ........................ 188
A3. Analysis variance of degree of transesterification of blends with compositions of

14.5 to 50.4 mole % PEN ..................................................................................... 189
A4. Analysis variance of D-values of blends with compositions of 14.5 to 50.4 mole

% PEN .................................................................................................................. 190
A5. Analysis variance of D-values of blends with compositions of 33.7 to 50.4 mole

% PEN .................................................................................................................. 191

xi

List of Tables (cont.)

Appendix B

Bl. The positions in the gradient column and the density values of the standard
calibration floats ................................................................................................... 192

32. The positions and the density of the blend resins. .................................................... 193

B3. The positions and the density of the non-oriented films. .......................................... 193

B4. The positions and the density of the oriented films. ................................................. 194

Appendix C

Cl. Analysis of variance of the T8 values of the blend resins with 25% (wt/wt) PEN. .. 195
C2. Analysis of variance of the Tg values of the oriented films with 25% (wt/wt)

PEN. ..................................................................................................................... 196
C3. Analysis of variance of the calculated and measured T3 values of the blend resins

processed at 300°C and 1 pass. ............................................................................ 197
Appendix D
D1. Analysis of variance of the water vapor permeability coefficient of the non-

oriented films obtained from blend resins with 25 % (wt/wt) PEN ..................... 198
D2. Analysis of varianCe of the water vapor permeability coefficient of the oriented

films obtained from blend resins with 25 % (wt/wt) PEN. .................................. 199
D3. Analysis of variance of the water vapor permeability coefficient of blend films

with compositions of 14.5 to 50.4 mole % PEN .................................................. 200
D4. Analysis of variance of the oxygen permeability coefficient of the non-oriented

films obtained from blend resins with 25 % (wt/wt) PEN. .................................. 201
D5. Analysis of variance of the oxygen permeability coefficient of the oriented films

obtained from blend resins with 25 °/o (wt/wt) PEN. ........................................... 202
D6. Analysis of variance of the oxygen permeability coefficient of the blend films

with composition of 14.5 to 50.4 mole % PEN. .................................................. 203
D7. Analysis of variance of the carbon dioxide permeability coefficient of the non- .

oriented films obtained from blend resins with 25 % (wt/wt) PEN ..................... 204
D8. Analysis of variance of the carbon dioxide permeability coefficient of the

oriented films obtained from blend resins with 25 % (wt/wt) PEN ..................... 205
D9. Analysis of variance of the carbon dioxide permeability coefficient of the blend

films with composition of 14.5 to 50.4 mole % PEN. ......................................... 206
Appendix E .
E1. Analysis of variance of the tensile strength along the machine direction of the

non-oriented films with 25 % (wt/wt) PEN. ........................................................ 207
E2. Analysis of variance of the tensile strength along the cross machine direction of

the non-oriented films with 25 % (wt/wt) PEN. .................................................. 208
E3. Analysis of variance of the tensile strength of the oriented films obtained from

blend resins with 25 % (wt/wt) PEN .................................................................... 209

xii

List of Tables (cont.)

Appendix F
F 1. Analysis of variance of the % elongation along the machine direction of the non-
oriented films with 25 % (wt/wt) PEN. ............................................................... 210
F2. Analysis of variance of the % elongation along the cross machine direction of the
non-oriented films with 25 % (wt/wt) PEN. ........................................................ 211
F3. Analysis of variance of the % elongation of the oriented films obtained from
blend resins with 25 % (wt/wt) PEN .................................................................... 212
Appendix G
G1. Analysis of variance of the Young’s modulus along the machine direction of the
non-oriented films with 25 % (wt/wt) PEN. ........................................................ 213
G2. Analysis of variance of the Young’s modulus along the cross machine direction
of the non-oriented films with 25 % (wt/wt) PEN. .............................................. 214
G3. Analysis of variance of the Young’s modulus of the oriented films with 25 %
(wt/wt) PEN. ........................................................................................................ 215
G4. Analysis of variance of the Young’s modulus along the machine direction of the
non-oriented films with composition of 14.5 to 50.4 mole PEN % ..................... 216
G5. Analysis of variance of the Young’s modulus along the cross machine direction
of the non-oriented with composition of 14.5 to 50.4 mole % PEN .................... 217
G6. Analysis of variance of the Young’s modulus of the oriented films with
composition of 14.5 to 50.4 PEN mole % PEN. .................................................. 218

xiii

List of Figures

Figures page
Chapter 2
2.1. Chemical structure of polyethylene terephthalate (PET). ............................................. 9
2.2. Polymerization of PET, (a) Ester interchange reaction and (b) Polycodensation ....... 12
2.3. Chemical structure of polyethylene naphthalate (PEN) .............................................. 17
2.4. Polymerization of PEN, (a) Ester interchange reaction and (b) Polycodensation. ..... 20
Chapter 3
3.1. NMR spectrum of the PET/PEN blend with 25% (wt/wt) PEN processed through

the twin-screw extruder at 300°C and 1 pass. ........................................................ 48
3.2. DSC scan of the PET/PEN blend 25% (wt/wt) PEN processed through the single-

screw extruder at 300°C and 1 pass. ...................................................................... 53
3.3. DSC scan of the PET/PEN blend 25% (wt/wt) PEN processed through the twin-

screw extruder at 300°C and 1 pass. ...................................................................... 54
3.4. The effect of blending temperature and time on the degree of transesterification

(B). ......................................................................................................................... 59
3.5. The effect of PEN composition on the degree of transesterification (B) .................... 61
3.6. The effect of the blending temperature and time on the degree of

transesterification (B) and D value. ....................................................................... 63
3.7. The effect of PEN composition on the degree of transesterification (B) and D

value. ...................................................................................................................... 64
Chapter 4
4.1. Apparatus for the gradient tube preparation. .............................................................. 78
4.2. The effect of the degree of transesterification (B) on the glass transition

temperature of the PET/PEN blends. ..................................................................... 84
4.3. The effect of PEN composition on the glass transition temperature on the

PET/PEN blends. ................................................................................................... 86
4.4. The comparison between calculated and measured T8 of the blend resins. ................ 88
4.5. The effect of the degree of transesterification (B) on the melting temperature (Tm)

of the PET/PEN blends. ......................................................................................... 90
4.6. The effect of PEN composition on the melting temperature of the PET/PEN

blends. .................................................................................................................... 92
4.7 . The effect of the degree of transesterification (B) on the percent crystallinity of

the PET/PEN blends. ............................................................................................. 94
4.8. The effect of PEN composition on the percent crystallinity of the PET/PEN

blends. .................................................................................................................... 96
4.9. The effect of the degree of transesterification on the density of the blends. .............. 98
4.10. The effect of PEN composition on the density of the PET/PEN blend resins. ....... 100
4.11. The density of the PET, PEN and blends ................................................................ 101
4.12. The effect of PEN composition on the density of the PET/PEN blend films. ........ 102

xiv

List of Figures (cont.)

4.13. The effect of the degree of transesterification (B) on the average sequence

lengths of the the terephthalate (LnT) and naphthalate(LnN) unit. ........................ 104
4.14. The effect of PEN composition on the the average sequence lengths of the
terephthalate (La) and naphthalateamn) unit. ..................................................... 105

4.15. The effect of PEN composition on the molecular weight average of blend films. .106
4.16. The effect of the degree of transesterification on the polydispersity mm") of

the blends. ............................................................................................................ 108
4.17. The effect of PEN composition on the polydispersity (MW/Mn) of the blend

films. .................................................................................................................... 109
4.18. Dispersive Raman Spectroscopy of 300P1 25% (wt/wt) PEN polyblend

samples ................................................................................................................. 113
4.19. Dispersive Raman Spectroscopy PET/PEN blend resins with 25% and 40%

(wt/wt) PEN processed at 300 °C and 1 pass, and 315°C and 2 passes. .............. 114
4.20. Raman Chemical Imaging with 20X Magnification of 25% (wt/wt) PEN blend

resins processed through single-screw extruder at 300 0C and 1 pass. ............... 115
4.21. Raman Chemical Imaging with 20X Magnification of 25% (wt/wt) PEN blend

resins processed through twin-screw extruder at 300 0C and 1 pass ................... 116
4.22. The brightfield (a), polarized light (b), and Raman images (c) fi'om 20x

magnification of the blend resins. ........................................................................ 117
4.23. The brightfield (a), polarized light(b), and Raman images (c) from 100x

magnification of the blend resins. ........................................................................ 118

4.24. The brightfield (a), polarized light(b), and Raman images (c) from 20X and
100x magnification of the blend films with 25% (wt/wt) PEN processed

through twin-screw extruder at 3000C and 1 pass. .............................................. 119
Chapter 5
5a. A typical stress-strain curve for elastic materials. .................................................... 127

- 5b. Example of Permatran CIV graphical data. .............................................................. 134

5.1. The effect of the degree of transesterification on the water vapor permeability

coefficient of PET/PEN blend films at 37. 8°C. ................................................... 139
5.2. The effect of PEN composition on the water vapor permeability coefficient of the

PET/PEN blend films at 37 .8°C. .......................................................................... 141
5.3. The effect of the degree of transesterification on the oxygen permeability

coefficient of PET/PEN blend films at 25°C. ...................................................... 143
5.4. The effect of PEN composition on the oxygen permeability coefficient of

PET/PEN blend films at 25°C. ............................................................................. 145
5.5. The effect of the degree of transesterification on the carbon dioxide permeability

coefficient of PET/PEN blend films at 25°C. ...................................................... 147
5.6. The effect of PEN composition on the carbon dioxide permeability coefficient of

PET/PEN blend films at 25°C. ............................................................................. 149
5.7. The effect of the transesterification degree (B) on the tensile strength of

PET/PEN blend films ........................................................................................... 151

XV

List of Figures (cont.)

5.8. The effect of PEN composition on the tensile strength of PET/PEN blend films. 152
5.9. The effect of the transesterification degree (B) on the % elongation of PET/PEN

blend films. .......................................................................................................... 156
5.10. The effect of PEN composition on the % elongation of PET/PEN blend films. l 57
5.11. The effect of the transesterification degree (B) on the modulus of elasticity of
PET/PEN blend films ........................................................................................... 161
5.12. The effect of PEN composition on the modulus of elasticity of PET/PEN blend
films. .................................................................................................................... 162
Appendix B
Bl. The calibration curve for the density gradient column at 23°C ................................. 192

xvi

Chapter 1

Introduction

Over the past decade poly(ethylene terephthalate) (PET) has become popular in
the food packaging industry because it is a clear, light weight material with good barrier
characteristics to water vapor and carbon dioxide (Davis and Howell 1993). In addition,
its processibility, resealability, shatter resistance, and recyclability have made PET the

packaging material of choice for various food and pharmaceutical applications.

1.1. Need for Modified PET

Even though PET has found widespread acceptance, improved barrier and thermal
properties are required for extending product shelf life and for packaging systems which
require high oxygen barrier performance, coupled with high thermal stability and UV
protection.

PET has fair oxygen barrier properties. For instance, carbonated drinks can be
packaged in PET bottles. However, PET also cannot be used for packaging beer. Beer
rapidly loses desirable flavor and aroma characteristics when packaged in PET (Po et a1.
1996), because such moieties are very vulnerable to any amount of oxygen, which

permeates through PET bottles.

PET is also a low UV barrier material. Therefore, it cannot be used for products
such as medicines, baby foods, and beer, which are sensitive to UV light. These
products are typically packaged in glass or metal packages.

The glass transition temperature (T3) of PET is approximately 74°C. Thus, a
typical PET resin is not suitable for use as hot fillable or refillable packages. Several
products, including hit juices, jams, tomato sauces, etc., require a hot-fill process at a
temperature above 85°C (McGee and Jones 1995). Therefore, a material with a
temperature resistance of greater than 85°C is required. In addition, returnable or
refillable PET bottles also need to be washed with an alkaline solution at a temperature
close to the T8 of PET. Thin wall PET bottles shrink under such conditions. A few
techniques to overcome these problems have been proposed. For example, bottles with a
wall thickness of 6-7 mm and a weight of 100 g as compared to the 3 mm and 45 g of
one-way bottles, have been used to overcome this problem (Po et a1. 1996). However,

cost and weight limit their use.

1.2. Alternative Polymer

A new type of material, which is qualified for use under such conditions, has,
therefore, been sought. One such candidate polymer, which meets the above
requirements is poly(ethylene 2,6-naphthalene dicarboxylate), an analog of PET. PEN
possesses oxygen permeability characteristics approximately one quarter to one-fifth that
of PET (Stewart et a1. 1993). PEN also has higher temperature resistance. The T8 of

PEN is approximately 120°C, 50 °C above that of PET. In addition, PEN has higher UV

light barrier properties and lower shrinkage at elevated temperatures than PET (Callander
and Sisson 1994). The tensile strength of PEN is also 35% higher that that of PET (Shi
1998).

While PEN has decided advantages over PET for packaging oxygen sensitive and
hot-fill products, PEN costs approximately five time as much as PET. In addition, the
limited amount of monomer production for PEN has restricted its use (Callander and

Sisson 1994; Hoffinan and Caldwell 1995).

1.3. Combining PET and PEN

A potential technique for combining the economic advantages of PET and the
outstanding banier and thermal properties of PEN is copolymerization and blending
process of PET and PEN. PET/PEN copolymer is an optically clear material with
intermediate barrier, mechanical, and thermal properties between PET and PEN.
However, McGee and Jones (1995) found that PET/PEN blends had advantages over
random copolymers, because blends can crystallize over a broader compositional range
than copolymers. Within 15-85 mole % of naphthalene dicarboxylate content (the mole
fraction of the naphthalate unit of the total terephthalate and naphthalate units, NDC),
blends are semicrystalline materials while copolymers are amorphous materials. Within
this range of NDC, blends also have improved oxygen barrier properties, relative to their
random copolymer counterparts (McGee and Jones 1995). Therefore, a PET/PEN blend
was selected for this study.

However, a drawback to the blend approach is that PET and PEN are inherently

incompatible. The mixture of PET and PEN yields an opaque material with non-uniform

properties, and shows two T3 values (glass transition temperatures) and Tm values

(melting temperatures), corresponding to one PET-rich phase and one PEN-rich phase.
Transesterification reactions between PET and PEN have been studied and found

to improve the compatibility and ultimately produce miscibility of the blends, which in

turn results in improved clarity and uniform properties.

1.4. Transesterification Reactions

Transesterification reactions are interchange reactions between hydroxyl and
carboxyl groups of the PET and PEN chains (see the details in Chapter 3).
Transesterification reactions are accomplished by melt mixing these two polymers at
temperatures that are higher than their melting temperatures. The properties and
composition of the blend change when transesterification is achieved. During this
reaction processing, graft and block copolymers are initially formed, followed by
formation of random copolymers. In addition, the resulting copolymers compatibilize
PET and PEN, resulting in miscible blends (McGee and Jones 1995; Po et al. 1996; Shi
1998; Stewart et a1. 1993).

The PET/PEN blends with a transesterification degree of greater than 5% showed
a significant improvement in the blend miscibility and had only a single T8 and Tm
(Hoffman and Caldwell 1995 ; McGee and Jones 1995). McGee and Jones (1995)
showed that there is an inverse linear relationship between the transesterification level
and the haze value (measurement of the film opacity). The authors reported that as the

degree of transesterification increases, the haze value decreases.

1.5. Overview of the Experiment

1.5.1. Processing Conditions

In any scheme to produce materials from melt blends of PET and PEN, a number
of factors have to be dealt with. Processing conditions, as well as types of processing
equipment, are primary factors controlling transesterification reactions, which in turn
affect the miscibility of the blends. Stewart et a1. (1993) studied the reactive processing
of PET/PEN blends with a PEN composition ranging from 50 to 80 weight % and found
that the primary factors connolling transesterification are the blending time and
temperature, while the composition of the blend and the residual polyester catalysts have
little or no effect on the transesterification reactions. Furthermore, processing the melt
polymer in a twin-screw extruder was found to achieve a higher degree of
transesterification than processing in a single screw extruder (Shi 1998).

This study, therefore, focuses on the processing parameters, including processing
time and temperature, for accomplishing the reaction by extruding the melt polymer blend

in a twin-screw extruder.

1.5.2. Blend Composition

Even though blend composition has little effect on the transesterification
reactions, blend composition is a primary factor controlling thermal, barrier and
mechanical properties of the resultant blends. Therefore, the effect of the blend
composition on blend properties was investigated. This study focused on PEN

composition ranging from 10 to 40 weight percent.

1.5.3. Film Orientation

Orientation, which involves some degree of stretching and heat setting, is known
to enhance the properties of polymer films. Rearrangement of the polymer chains, by
application of an external stress above the polymer glass transition temperature, tends to
increase the percent crystallinity of semicrystalline polymers and results in improvement
of mechanical, thermal and barrier properties.

The influence of orientation on the thermal, mechanical, and barrier properties of

blend films, with a certain degree of transesterification, was studied.

1.5.4. Blend Characterization

PET/PEN blend resins with varied transesterification degree and PEN
composition were processed though a twin screw extruder at temperatures between 275
and 325°C, and one and two passes though the extruder. Blend films were fabricated in a
single screw extruder at 300°C, and biaxially oriented on a TM. Long film stretcher. The
properties of blend resins, including mechanical, thermal, barrier, and morphology, and
films then were determined.

The effect of the transesterification degree, blend composition, and orientation on

these properties was then evaluated.

1.6. Objectives

The followings are the primary objectives for this study:

1. To study the effect of processing conditions, blending time and temperature, on the
transesterification reactions and determine the optimum processing conditions to
achieve the desired blend properties.

2. To establish the relationship between the transesterification degree of the PET/PEN
blends and the associated thermal, barrier, and mechanical properties, with
transesterification degree evaluated as a function of processing conditions.

3. To identify conditions where thermal, barrier, and mechanical properties of the blend
are affected by the blend composition and transesterification reactions.

4. To determine the relationship between orientation and the thermal, mechanical and

barrier properties of fihns fabricated from PET/PEN blends.

Chapter 2

Literature Review

In this chapter, issues dealing with PET, PEN and PET/PEN copolymers and
blends are reviewed. The chemical structures, manufacturing processes, selected
properties, and the applications of PET and PEN are described. Finally, a review of

studies dealing with PET/PEN copolymers and blends are included.

2.1. Interesting Issues of PET

Poly(ethylene terephthalate) (PET) is one of the most important commercial
polyesters presently used in the packaging industry. PET is also referred to
poly(oxyethyleneoxyterephthaloyl) which is its IUPAC name (Odian 1991). Some trade
names for PET are Mylar, Dacron, Terylene, and Cleartuf.

In 1946, PET was discovered in the UK by Whinfield and Dickson. It was first
known as a textile fiber and was introduced to the textile fiber market in 1953 (Adur and
Bonis 1994). Shortly thereafter, PET film was commercially introduced. Since then,
films based on PET were developed and became well known in many countries, e. g. the
UK, USA, France and Japan (Jadhav and Kantor 1988; J aquiss et a1. 1982; Margolis
1985). In 1966, injection-molding of PET resins was introduced to the market
(Anonymous 1985; Mock 1983). Recently, PET has been widely used in many
industries including textile, food and household packages, pharmaceutical and medical

packages, electrical and electronic devices, and automobile parts.

2.1.1. PET Structure and Properties
PET is a semicrystalline, thermOplastic polyester. It is a linear condensation

homopolyester, and has repeating units as shown in the following figure,

0 O
{CQW

Figure 2.1: Chemical structure of polyethylene terephthalate (PET)

n

Characteristics of PET are affected by ester groups (—coo-), distributed along the main
chain of the PET molecular structure. Moreover, PET chains contain aromatic rings
which contribute to high chain stiffness and a high melting temperature.

The molecular structure of PET is described by the fringed-micelle model (Bonart
and Hosemann 1960; Hess and Kiessig 1943). The polymer exists as an imperfect two-
phase system of interconnected crystalline and amorphous domains. The regular linear
structure of PET can form an Ordered structure through chain orientation and
crystallization. PET has a glass transition temperature around 78°C and a melting
temperature around 255°C (Jabarin 1984; Jabarin and Chandran 1993). PET is ...r
suited for the biaxial film orientation, resulting in films with outstanding mechanical and
barrier properties (Bilmeyer 1962). PET exhibits high mechanical strength and
toughness, fatigue resistance up to 150-1 75°C, good chemical resistance, good gas barrier
properties, and thermal stability (Bilmeyer 1962; Odian 1991; Werner et al. 1988).

PET with in the optimal range of molecular weight average (MW) provides good

processibility and low thermal shrinkage. Moreover, with this Optimal MW, it shows

excellent mechanical properties (Werner et al. 1988). PET can be used on a wide variety
of film-processing equipment, due to its high tensile properties. PET films stretched
along one direction (tensilized films) are used as audiocassette tapes, and typewriter or
computer printer ribbons. PET exhibits excellent electrical insulating properties.
Therefore, it is widely used in electrical applications. Moreover, PET exhibits high
chemical resistance, and good water, oxygen, and carbon dioxide barrier properties.
Therefore, it is generally used as a packaging material for food, pharmaceutical,
electronic, and medical products (Briston 1989; Margolis 1985; Jadhav and Kantor 1988;

Werner et al. 1988).

2.1.2. PET Manufacturing

There are three steps in the PET manufacturing process including: (i) PET
polymerization; (ii) film fabrication; and (iii) auxiliary processing or post-treatment
(Werner et al. 1988).

The PET synthesis is a two-stage ester interchange process. This is a basic
manufacturing process for PET which is similar for fibers, filaments, bottles, and films
(Odian 1991; P0 et al. 1996). The two-stage process involves transesterification of
dimethyl terephthalate with ethylene glycol, followed by polycondensation (Odian 1991;
Jadhav and Kantor 1988 ; Werner et al. 1988).

The first stage begins with solution polymerization of ethylene glycol (EG) and
dimethyl terephthalate (DMT) or terephthalic acid (TPA). The polymerization involves
an ester interchange reaction which produces bis (2-hydroxyethyl) terephthalate (BHET)

along with a small amount of larger-sized oligomers (Figure 2.2a). The reaction

10

temperature increases from 150 to 2100C. Methanol, which is a by product of this stage,
is distilled off.

The second stage is melt polymerization (Figure 2.2b). The reaction temperature
is held at 270-280°C, which is above the melting temperature of the polymer. With this
high temperature and a partial vacuum of 66-133 Pa, the ethylene glycol is removed
(Odian 1991; Jadhav and Kantor 1988 ; Werner et al. 1988).

Temperature control and catalysts used in this process are very important for
minimizing side reactions (Odian 1991; Werner et al. 1988). Common catalysts used for
the first stage include acetate salts of calcium, lithium, manganese, cobalt, and zinc. The
second-stage catalysts are antimony (III) oxide, germanium compounds, and titanium
compounds. An alkyl .or aryl phosphite or phosphate is often added to inactivate the .
first-stage catalysts.

Recently, terephthalic acid in high purity became available. Terephthalic acid and
ethylene glycol are also used in the first stage process to produce bis (2-hydroxyethyl)
terephthalate and the two-stage process is modified for this PET production process.

The two-stage process is carried out either as a batch or continuous process. The _
molten polymer is quenched to form solid extrudate. Then the extrudate is granulated to
small pellets. PET pellets can be stored, and used for film or bottle production, later.

PET film fabrication is carried out using a cast extrusion process. PET pellets
must be carefully dried in order to avoid hydrolytic degradation during the extrusion
process (McGee and Jones, 1995; P0 et al. 1996 ; Seo and Cloyd, 1991; Stewart et al.
1993). Molten polymer is east through a flat die connected at the end of the extruder,

and output sheet is biaxially stretched and heat set (Schwartz 1982; Werner et al. 1988).

ll

o 0
II II
CH3-O—C —O—CH3 + 2HOCH2CH20H

EG
DMH‘ {}

o 0
II II .
HOCHZCHz-O—C —O—CH2CH20H + 2 CH3OH

BHET

(a)

o o
H H
n HOCHZCHz-O—CQ—C—O—CHZCHZOH
{} BHET

O

o
' H H -
H HZCHZ—o—C«©{}4><HZCH20H + (n-1)HOCH2CH20H
n EG
PET

(b)

Figure 2.2: Polymerization of PET, (a) Ester interchange reaction and
(b) Polycondensation.

n

In some large production plants, the film-fabricating unit is continuously connected to the
polymer-manufacturing unit (Werner et al. 1988).

Auxiliary treatment is required for PET modification to provide special properties
of the PET films. The required properties include slip, adhesivity, scalability, abrasion
resistance, antistatic behavior, release and weatherability. Auxiliary processes include
in-line coating, off-line coating, etching, electrical discharge, and other hi gh-energy
treatments (Schwartz 1982; Werner et al. 1988).

A recycling process of PET is also available. However, it is limited to plain PET.
For recycling of coated or coextruded PET films, the coating materials and the secondary
layer are required to be compatible with virgin PET and they must withstand the
complete process without decomposition or deleterious effects on the new films (Werner
et al. 1988). In Australia, recycled PET is used as an inner layer in a three-layer
structure for 2-liter Coca-Cola bottles. The outer layers are virgin PET which provides a
barrier to the migration of any contaminants from the post consumer material (F orcinio,

1993)

2.1.3. PET Applications

PET is a thermoplastic polyester which is classified as an engineering polymer,
due to its excellent thermal, mechanical, optical, and electrical properties. These
advantageous characteristics of PET allow it to be used in a wide variety of application
areas. In addition, the combination of these properties and the reasonable cost of PET

has resulted in a rapid growth of PET usage in past decades. Currently, there are more

13

than 50 specific application areas for PET. Main application areas include fiber, plastics,
and engineering plastics (Odian 1991; Margolis 1985; Werner et al. 1988).

PET fiber applications include cloths and fabrics, upholstery, curtains, thread, and
tire cords, and account for approximately 80% of all PET applications (Adian 1991).
PET fibers exhibit high crease resistance, and good abrasion resistance. They can be
treated with crosslinking resins to improve wash and wear properties. Blending PET with
cotton or other cellulosic fibers can improve the texture of PET fibers.

PET plastic applications include fihns and containers. Photographic films are the
largest application area for PET films. PET films possess excellent optical clarity,
dimensional stability, chemical resistance, and durability (Werner et al. 1988). These
photographic films include medical, dental, and industrial x-ray films; 35 mm slide and
negative films; instant photography; color-proofing systems; and phototool and
photoresist films for the manufacturing of printed microcircuits (Werner et al. 1988).
These types of PET films are mainly commercialized by Eastman Kodak, Dupont, Fuji,
Agfa Gevaert, and 3M.

PET is also widely used for food packaging applications because of its clarity,
light weight, high strength, fair gas barrier, and good metallizability (Davis and Howell
1995; Werner et al. 1988). The US Food and Drug Administration (FDA) has approved
PET for use as food packages. PET films for food packages include metallized
polyester films for barrier packages of coffee, wine, and meat; poly(vinylidene chloride)-
coated PET films for meat and cheese packages; coextr'uded PET films for heat scalable
packages; surface-modified PET films which are printable with water-based ink systems;

and laminated PET-based films or flexible retort packages for replacing steel cans, etc

14

(Briston 1989; Werner et al. 1988). The PET—container market for consumer products is
rapidly growing. PET bottles are used for carbonated drinks, household products, and
food products (Briston 1989; Odian 1991; Werner et al. 1988).

PET pharmaceutical and medical packages include single unit tablet or capsule
packages and peelable packages for medical devices. PET has been used in this
application area because of its excellent strength and dimensional stability at high
temperature (Werner et al. 1988). These packages are exposed to high temperatures
during cavity formation and heat sealing.

The growth area for PET films in the electrical market includes flexible printed
circuits, membrane touch switches, light-emitting diodes and liquid crystal display
larninations for automobile instrument panels and light switches (Odian 1991). This is
due to its high dielectric constant, dielectric strength, high melting temperature, good
mechanical strength, and flexibility. Moreover, PET films are used as insulators and
thermal barriers between conductor wires and outer jackets, and wrappers for
telecommunications, power and specialty cables (Briston 1989; Werner et al. 1988).

PET is used as an engineering plastic which can replace steel, aluminum, and
other metals in electrical and electronic devices, office appliances, and automobile parts.
In these applications, PET reinforced with glass fiber or compounded with ‘silicon,
graphite, or teflon is used to improve strength and rigidity of the PET article (Odian
1991). However, using PET in engineering applications is limited because of the low
rate of crystallization of PET. This results in increasing processing cost as the mold

recycle time increases (Odian 1991).

15

2.1.4. PET Modification

Even though PET has found widespread acceptance, improved barrier and thermal
properties are required for extending product shelf life and for packaging systems which
require high oxygen barrier performance, coupled with high thermal stability and UV
protection.

PET has fair oxygen barrier properties. For instance, carbonated drinks can be
packaged in PET bottles. However, PET cannot be used for beer. Beer rapidly loses
desirable flavor and aroma characteristics when packaged in PET bottles (Po et al. 1996),
because such products are very vulnerable to any amount of oxygen which permeates
through the PET bottle.

PET is also a low UV barrier material. Therefore, it cannot be used for products
such as medicines, baby foods, and beer, which are sensitive to UV light. These
products are typically packaged in glass or metal packages.

The glass transition temperature (T8) of PET is approximately 74°C. Thus, a
typical PET resin is not suitable for use as hot fillable or refillable packages. Several
products, including fruit juices, jams, tomato sauces, etc., require a hot-fill process at a
temperature above 85°C (McGee and Jones 1995). Therefore, a material with a
temperature resistance of greater than 85°C is required. In addition, returnable or
refillable PET bottles also need to be washed with an alkaline solution at a temperature
close to the T3 of PET. Thin wall PET bottles shrink under such conditions. A few
techniques to overcome these problems have been proposed. For example, PET bottles
with a wall thickness of 6-7 mm and a weight of 100g, as compared to the 3 mm and 45g

weight of one-way bottles, have been used to overcome this problem (Po et al. 1996).

16

 

However, cost and weight limit their use. Other alternative is a heat-set PET bottle. It
has 25-28% crystallinity and resists temperatures up to 85°C. In order to increase the
temperature tolerance to 95°C, the % crystallinity of PET needs to be increased to 40%.

This requires doubling cycle time and increases overall cost (Forcinio 1993).

2.2. Alternative Polymer, PEN

A new type of material, which is qualified under use with such conditions, has,
therefore, been sought. One such candidate polymers is poly(ethylene 2,6-naphthalene
dicarboxylate) or PEN.

PEN is a thermoplastic polyester. It is an analogue to PET. PEN has 2,6-
naphthalenedicarboxylate rings in the main chain, while PET has aromatic rings in the

main chain (Figure 2.3).

(l): _

O C-O-CHZCH20——

' l H O J n
0

Figure 2.3: Chemical structure of polyethylene naphthalate (PEN)

 

PEN is a condensation polymer which can be amorphous, partial crystallized, or
highly crystallized. PEN has a glass transition temperature around 120°C, and a melting
temperature around 265°C (Mencik 1976). The crystal unit cells were determined as
triclinic. The dimensions of the crystal units are a = 0.651 nm, b = 0.575 nm, and c = 1.32

nm, with the crystal angles of a = 81 .33°, [3 = 144°, and y = 100° (Buchner et al. 1989;

17

Mencik 1976). The PEN chains lie parallel to the c-axis and one chain passes through
each unit cell (Buchner et al. 1989).

In comparison to PET, PEN has higher chain stiffness (Buchner et al. 1989;
Cakmak et a1. 1990; P0 et al. 1996). This accounts for its high glass transition
temperature, high chemical resistance, high mechanical and high barrier properties
(Anonymous 1995, Forcinio 1993, Light and Seymour 1982, and Pidgeon 1996). PEN
also has higher capability of absorbing UV radiation (Pidgeon 1996).

PEN possesses oxygen permeability characteristics approximately one quarter to
one-fifth that of PET (Stewart et al. 1993). PEN also has higher temperature resistance.
The T8 of PEN is approximately 120°C, which is 50 °C above that of PET. In refillable
applications, PEN can be washed by a cleaning cycle which is used for glass returnable
bottles with a 2% caustic solution, at least 20 times at 85°C (Forcinio 1993). In addition,
PEN has higher chemical resistance than PET (F orcinio 1993). PEN has higher UV light
barrier properties and lower shrinkage at elevated temperatures than PET (Callander and
Howell 1994; Callander and Sisson 1994). PEN has 50% more stiffness (Forcinio 1993;
Toth et al. 1994) and 35% higher tensile strength than that of PET (Toth et al. .1994).
Moreover, PEN processing permits 50% shorter cooling time compared to PET
processing.

The polymerization process for PEN is similar to the PET manufacturing process.
PEN is synthesized from the polymerization of ethylene glycol and 2,6 dimethyl
naphthalene dicarboxylate (Wang and Sun 1994). There are two steps of PEN formation
(Figure 2.4). The first step is the ester interchange process to form 2,6-bis

(hydroxyethyl) naphthalate (BHEN) (Figure 2.4a). The transesterification reaction

18

between the ethylene glycol (EG) and 2,6 dimethyl naphthalene dicarboxylate (DMN)
first yields p-methoxycarbonyl-Z-hydroxyethyl naphthalate (P-MCHEN), which then
undergoes further reaction with another molecule of ethylene glycol. Finally, BHEN is
formed.

The second step is polycondensation of BHEN (Figure 2.4b). This reaction is
generally carried out at a temperature 10-15 °C higher than those of PET production (Po
et al. 1996), under reduced pressure or in an open system. Therefore, the ethylene glycol,
which is the byproduct of this step, can be distilled out.

PEN can be injection stretch blow-molded on the same equipment used for PET
(Anonymous 1994‘”; Forcinio 1993). Due to the higher viscosity of melted PEN, higher
power equipment (but within the parameters of existing equipment used for PET) and
longer resonance times are required for PEN processing (Forcinio 1993).

Regarding environmental issues, PEN may be recycled in a PET steam or in a
separated process. Thus, PEN may be seen as a replacement of the existing multilayer
PET structure, which is not suitable for recycling process (F orcinio 1993). The Shell
Company has conducted a series of studies and found that PEN does not have a
significant impact on the PET recycle stream (Davis and Howell 1995). However, no
report has confirmed that PEN can be recycled separately.

PEN was known to industry since the early 19403 as a polymer with superiority of
its main properties relative to PET. However, PEN was not widely commercialized until
1995, because of the lack of a secure source for 2,6 dimethyl naphthalene dicarboxylate.
The Amoco Company and Eastman Chemical Company announced their plants to

commercialize naphthalate dicarboxylate (NDC) monomer in mid 1995

19

 

0
ll
—o—<:H3 + 2 Hocnzcuzon
CH3—O—fi cc
0 DMN

i}

0
II

CD —o—CH3 + CH3OH
HOCHZCHZ 0 05 + HOCHZCHZOH

P-MC HEN & EC

0
ll
—O—CH2CH20H + 2 CH3OH
HOCHzCHz—O—f'l
O
BHEN (a)
9
—O—CH2CH20H
n HOCHZCHz—O—fi
O
BHEN

0
ll ,
H2CH20H + ("'1)HOCH2CH20H
H CHZCHz—O—fi n m
' o
PEN

(b)

Figure 2.4: Polymerization of PEN, (a) Ester interchange reaction and

(b) Polycondensation.

20

(Anonymousl994‘a); Anonymous 1994“”; Baker 1995; Castle 1996). Moreover
HiPERTUF, a trade name for a homopolyrner 2,6 dimethyl naphthalate-based
polyethylene naphthalate resins was commercialized by Shell Company in 1995
(Anonymous 1995; Crawshaw and Sisson 1995; Davis and Howell 1995).

PEN is categorized into many types which vary in molecular weight, additives
added, and naphthalate content (Anonymous 1999; Crawshaw and Sisson 1995; Davis
and Howell 1995). The different types of PEN include coating materials with low
intrinsic viscosity (IV), films with medium IV, fiber color with Ti02 additive, medical
materials, and resin grades with medium IV for food and beverage packaging, etc.

For food packaging applications, the European Union and the US. Food and Drug
Administration (FDA) approved PEN for use in a variety of food and beverage packages
(Castle 1996). These include beer and some pasteurized and sterilized food packages,
which require either hot filling condition or high oxygen banier properties. The Shell
Company noted that PEN homopolyrner was designed to be specifically used for
beverage, returnable, medical, and retort packages (Davis and Howell 1995). PEN
bottles have been intensively used in countries other than the US. For instance, PEN is
used as 1.5 liter returnable bottles for mineral water by Coca-Cola Co. in Uruguay
(Gerding et al. 1995; Pidgeon 1996). Small, wide mouth PEN containers, for individual
serving jams and jellies, have been commercialized by Skillpack in Holland (Tibbitt
1994). Moreover, PEN containers are used for 1.5-liter bottles of lemonade, peach and
orange juices in Japan (Gerding et a1 1995). PEN containers were selected for those

applications, due to their high resistance to the heat exposure during the washing and hot

21

filling processes, and the high UV (385 nm) protection which results in a longer shelf life
for the natural juices and vitamin C (Gerding et al. 1995; Pidgeon 1996).

In addition to food packaging applications, PEN has also been used in markets
including electrical films and magnetic tape, fibers for industrial applications, and
engineering resins for automotive and other uses (Forcinio, 1993). For instance, oriented
PEN fihns are used as professional videotapes (Anonymous 1994‘”). Moreover,
Kaladex, a trade name for PEN, is commercialized in Europe and is currently used for
electrical, flexible printed circuits and graphic applications (Anonymous 1994‘”).

While PEN has decided advantages over PET for packaging oxygen sensitive and
hot-fill products, some drawbacks limit PEN use. One drawback is its high melt
viscosity (Po et al. 1996); therefore higher processing times and temperatures are
required for the extrusion of PEN. Difficulties in achieving crystallinity may occur and a
fast crystallization rate is needed (Buchner et al. 1989). Appel (1996) reported that the
output rate of the blow molding process of PET bottles was approximately 20% faster
than PEN battles (12,000 bottles per hour for PET and 10,000 bottles per hour for PEN).
Another disadvantage is the limited amount of monomer production for PEN (Callander
and Sisson 1994; Hoffinan and Caldwell 1995; Po et al. 1996). This results in the higher
cost of dimethyl naphthalene dicarborboxylate, approximately three to four times as
much as that of dimethyl terephthalate. PET cost is approximately $0.70 per pound,

while PEN cost is $2 to $2.95 per pound (Kimmel 2000; Forcinio 1993).

22

2.3. Combining PET and PEN

A potential technique for combining the economic advantage of PET and the
outstanding barrier and thermal properties of PEN is copolymerization and blending of
PET and PEN.

PET/PEN copolymer is a random dispersion of the PET and PEN monomers
achieved by the melting reactions of these individual monomers (Davis and Howell
1995). PET/PEN copolymer is an optically clear material with barrier, mechanical, and
thermal properties falling between PET and PEN. Blending of PET and PEN is achieved
by the physical mixing, without any chemical reaction, of these individual polymers.

Pervious reports showed that thermal transitions of PET/PEN copolymers and
blends vary with the naphthalene dicarboxylate content, NDC (Hoffman and Caldwell
1995; McGee and Jones 1995; P0 et al. 1996). The NDC is defined as the mole fraction
of the naphthalate unit in the total terephthalate and naphthalate units (Shi 1998).
Hoffman and Caldwell (1995) indicated the boundaries of the serrricrystalline and
amorphous regions of the PET/PEN copolymers and blends by the absence of the melting
point. Within the range of 15-85% NDC, PET/PEN copolymers are amorphous
materials, while PET/PEN blends are semicrystalline materials. At less than 15% NDC
and greater than 85% NDC, copolymers and blends both behave like pure PET and PEN,
respectively. Both copolyrrrers and blends can crystallize over this range.

PET /PEN blends had advantages over random copolymers because blends can
crystallize over a broader compositional range than copolymers. Moreover, blends have
improved oxygen barrier properties, relative to their random copolymer counterparts

(McGee and Jones, 1995).

23

 

However, a drawback to the blend approach is that PET and PEN are inherently
incompatible. The mixture of PET and PEN yields an opaque material with non-uniform
properties, and shows two T8 values (glass transition temperatures) and Tm values
(melting temperatures), corresponding to one PET-rich phase and one PEN-rich phase.
Transesterification reactions between PET and PEN have been studied and found to
improve the compatibility and ultimately produce miscibility of the blends, which in turn
results in improved clarity and uniform properties.

Transesterification reactions are interchange reactions between hydroxyl and
carboxyl groups of the PET and PEN chains (see the details in Chapter 3).
Transesterification reactions are accomplished by melt mixing these two polymers at
temperatures, which are higher than their melting temperatures (Tyne et al. 1995). The
properties and composition of the blend change when transesterification is achieved.
During this reaction processing, graft and block copolymers are initially formed,
followed by formation of random copolymers. The formed copolymers compatibilize
PET and PEN, resulting in miscible blends (Ihm et al. 1996; McGee and Jones 1995; P0
et al. 1996; Shi 1998; Stewart et a1. 1993).

The PET/PEN blends with a transesterification degree of greater than 5% showed
a significant improvement in the blend miscibility and had only a single T8 and Tm
(Hoffman and Caldwell 1995; McGee and Jones 1995). McGee and Jones (1995)
showed an inverse linear relationship between the transesterification level and the haze
value (measurement of the film opacity). The authors reported that as the degree of

transesterification increased, the haze value decreased.

24

 

2.4. PET/PEN Blends and Copolymers

PET/PEN blends were first studied by Shepherd et al. in 1991 (U .8. Pat.
5,006,613). They reported that PET/PEN blend was a tri-component polymer of PET,
PEN and PET/PEN copolymers.

Cruz et al. (1992) studied the mechanical properties and structure of PET/PEN
random copolymers. For copolymers with the 0-30 mole % of PEN, PET sequences
formed a crystalline phase after annealing at a temperature 10°C below the melting point,
while PEN sequences remained in the amorphous phase. For copolymers with 80 mole
% of PEN, the PEN sequences crystallized after annealing at a temperature 20°C below
the melting point, while PET sequences were in an amorphous phase.

Guo and Zachmann (1993) studied the miscibility of PET/PEN blends using
intermolecular cross-polarization nuclear magnetic resonance analysis. They found that
blends of PEN and PET/PEN copolymer were miscible, while blends of PET and PEN
exhibited phase separation.

Callander and Howell (1994) and Callander and Sisson (1994) studied the thermal
transition behavior of PET/PEN blends and copolymers. The authors found that blends
can crystallize over a broader compositional range than copolymers. In the range of 15-
85% NDC, PET/PEN copolymers are amorphous materials, while PET/PEN blends are
semicrystalline materials.

Ihm et al. (1996) studied the effect of transesterification on the miscibility of
blends obtained from solution precipitation. Physical blends without transesterification
were immiscible. Blends with a 50% transesterification did not crystallize and exhibited

only a single glass transition temperature. The transesterification degree was significantly

25

influenced by the processing temperature and time, but not by blend composition. The
authors used nuclear magnetic resonance (NMR) analysis and calculated randomness
using the Yamadera and Murano equation (Yamadera and Murano 1967). They
concluded that transesterification reactions led to the formation of block or graft
c0polymers, and finally proceeded to form random copolymers. As the processing time
increased, the melting temperature decreased and the cold crystallization temperature
increased. A phase separation of the PET and PEN domains of PET/PEN blend
structures was observed by optical microscopy, with hot stage samples. Moreover, they
reported that the T8 values of the blends with various compositions fell in the same range
as T8 values determined by Fox equation.

Stewart et al. (1993) studied the reactive processing of PET/PEN blends. The ‘
blends were melt mixed through a Brabender single-screw extruder. They concluded that
processing time and temperature were the primary factors controlling the
transesterification reactions, while the PEN composition and residual polyester catalysts
had little effect. Moreover, they indicated that transesterification led to the formation of
a single amorphous phase of the polymer.

McGee and Jones (1995) studied the effect of the processing parameters on the
physical properties of PET/PEN blends for bottle applications. The blends were
processed through a Killion extruder with varied processing temperatures and times.

The levels of transesterification of the blends of PET and PET/PEN copolymers were
higher than those of blends of PET and PEN homopolymers. This was due to the
preexisting transesterification level in PET/PEN copolymer. The results showed

increasing extrusion temperature and time, and using PET/PEN copolymer as a

26

naphthalate source seemed to favor faster production of clear blends. The ultimate
properties of the blends, such as planar stretch ratio, oxygen permeability, and
crystallinity, were investigated. However, the extrusion conditions, by which the
transesterification level was achieved for a given blend system did not significantly affect
the ultimate properties. The melting temperature and the % crystallinity decreased
linearly as the transesterification level increased. The oxygen permeability of the blends
seemed to be slightly lower than that of copolymers with sirrrilar % NDC. However, the
authors suggested further experimentation to confirm this result. The shrinkage of
biaxially oriented films seemed to increase as the transesterification level increased.

Lu and Windle (1995) studied the melting temperatures of PET/PEN random
copolymers and found that the copolymer with approximately 60 mole % PET
composition had the lowest melting temperature. The authors mentioned that the lowest
melting temperature in the middle composition region was attributed to the statistical
limitation of the crystallite size inherent in the sequence-matching model (Hanna and
Windle 1988).

Jenkins (1995) studied the orientation and barrier properties of PET/PEN blends
and copolymers. The author reported that the glass transition temperatures of PET/PEN
blends and copolymers increased as the % NDC increased. The copolymers with low %
NDC exhibited a similar stress-train curve to PET. However, the strain-induced
crystallization was significantly lower than that of PET. The oxygen permeability
decreased as the % NDC increased. However, the oxygen permeability seemed to be

independent of draw ratio, stretch rate, and temperature in this particular test range.

27

Heisey et al. (1996) indicated that the transesterification reactions affected the
clarity and crystallinity of the blends. Blends were prepared using a Killion twin-screw
extruder. All blends that were processed through the extruder three times were clear. For
the blend with 63% NDC, when the number of extruder passes increased, the
crystallization rate decreased. However, for blends with 5-15% NDC, the crystallization
rates were not affected by the number of passes or blending time. Copolymers processed
three times through the extruder crystallized faster than those one processed only once.
This was due to the decrease in the molecular weight of the copolymer during extrusion.

Po et al. (1996) studied PET/PEN copolymers and blends for container
manufacturing. The authors suggested that the properties of the PET/PEN copolymers
and blends depended on PEN composition and their rnicrostructure. The crystallization
rate was a major parameter to control in bottle manufacturing, including melt
polymerization, solid-state polymerization, drying, molding, and blowing.

Bauer (1997) studied the effect of orientation on the density, shrinkage and
oxygen permeability of PET/PEN blends. PET/PEN fihns were biaxially stretched
(2.5x2.5, 3.0x3.0 and 3.5x3.5) and heat set at 120, 150, and 180°C for 5 minutes. The
results showed that density of the blend films increased as the degree of stretching and
heat set temperature increased. The increase of density indicated an increase in the
degree of crystallinity. The shrinkage of the biaxially oriented blend films (for all blend
compositions) was minimal (less than 2%) at 85°C. The shrinkage increased with the
degree of stretching but decreased with the heat set temperature. The higher % NDC, the
lower the shrinkage that was found. In addition, the oxygen permeability was reduced as

the % NDC, degree of stretching, and heat set temperature increased.

28

Shi (1998) studied the transesterification and properties of PET/PEN blends.
The author indicated that PET/PEN blends with most blend compositions were
immiscible and opaque. The transesterification reactions can improve the miscibility of
the blends. Blending time and temperature were the primary factors controlling the
transesterification reactions, while blend composition and PEN sources were secondary
factors. The composition seemed to be the primary factor controlling the thermal
properties of the blends. The glass transition temperature of the blends increased as the
PEN composition increased. In addition, the glass transition temperature and PEN
composition were correlated using the Fox equation (Fox 1956) and the Gibbs-DiMarzio

equation (DiMarzio and Gibbs 1959).

29

Chapter 3

Transesterification Reactions

3.1. Introduction

3.1.1. Miscibility of Polymer Blends

PET and PEN polymer mixtures exhibit immiscibility or incompatibility.
However, with the recent commercialization of polyblends, the miscibility of polymer
blends has been preferred. Polymer miscibility is important in terms of simple polymer
mixtures, the physical nature of block and graft copolymers, as well as in interpenetrating
networks and the thermosetting network of polymer mixtures. Miscibility is described as
a single-phase polymer blend. This characteristic is required to assure uniform polymer
properties including mechanical, barrier, and thermal properties. A property compromise
between the constituents of a polymer blend, such as PET and PEN, is therefore achieved
by preparing miscible blends.

The initial stage in development of blends and copolymers was studied by Olabisi
et al. (1979) and Po et al. (1996). The physics associated with the basic properties of
homopolymers will dictate the miscibility or imnriscibility of newly formed polyblend
structures. When selected homopolymer blends are mixed, a homogeneous structure is
observed and single-phase behavior is found. However, some blends, including

PET/PEN blends, can exhibit phase separation or immiscibility.

30

Miscibility of PET/PEN blends can be obtained by initiating transesterification
reactions between the two polymers. Extrusion of PET/PEN blends at the proper
conditions (above melt temperatures of both homopolymers and sufficient time) will
enhance transesterification reactions.

The investigation of the thermal properties of the blends by DSC analysis is one
of several techniques to identify the miscibility of blends. Miscible blends exhibit only
one T8 and Tm, and the TE; and Tm of these blends appear at a temperature between the T g
values of PET and PEN, and Tm values of PET and PEN, respectively. Irnmiscible blends
have 2 T8 and 2 Tm values. The first T8 and Tm correspond to the PET rich phase (T8
appears at 75°C and Tm at 250 oC), and the second values correspond to the PEN rich
phase (T8 appears at 120 °C and Tm at 265°C). Therefore, the miscible blends exhibit a

single phase and the immiscible blends show phase separation.

3.1.2. Transesterification Reactions

Transesterification reactions can be obtained when processing PET/PEN blends at
a temperature above their melting temperatures for sufficient time (Tyne et al. 1995).
Transesterification reactions of PET/PEN blends are interchange reactions between
hydroxyl, carboxyl and ester groups of the PET and PEN chains. The kinetic model for
this transesterification was found to be a second-order equilibrium reaction (Shi 1998).
The properties and composition of the blend change when transesterification is achieved
(MacDonald et al. 1991; Montaudo et al. 1992; Porter and Wang 1992; Ramjit and
Sedgwick 1976). During this reaction processing, graft and block copolymers are initially

formed, followed by formation of random copolymers. In addition, formed copolymers

31

 

compatibilize homopolymers, resulting in the miscible blends (Devauxet al. (a'b'c‘d) 1982;
Godard et a1. 1986; Pilati et al. 1985; Smith et al. 1981).

By using lHNMR technique, the occurrence of the interchange reaction of PET
and PEN chains can be identified and the degree of transesterification achieved can be
determined.

The degree of transesterification (B) and degree of randomness (RD) were
originally described and calculated by Yamadera and Murano (1967). The degree of
transesterification is defined as the ratio of the number of reactions between naphthalate
groups and terephthalate groups resulting in naphthalate-terephthalate (N -T) bonds, and
the number of N-T bonds expected for a random copolymer of the particular molar
composition. For blends of PET and PEN homopolymers, the degree of
transesterification (B) is equal to the degree of randomness, when degree of randomness
is measured as the ratio of the number of N-T bonds and the number expected for a
random copolymer. However, for blends of PET/PEN copolymer and PET or PEN
homopolymer, the randomness degree is greater than the transesterification degree
because of the presence of initial N-T bonds in the copolymer (McGee et a1. 1995).

The randomness degree (RD) is a parameter for characterization of the
microstructure of copolymer and polymer blends with some degree of transesterification.
The RD is independent of composition, and inversely proportional to block length (Po et
al., 1991; P0 et al., 1996). Therefore, the randomness degree can range from 0 in a
physical mixture of homopolymers, 1 in a random copolymer, to 2 in an alternating

copolymer (Po et al. 1991; P0 et al. 1996; Yamadera and Murano 1967).

32

The D value has been proposed as a new and more meaningful term for
determination of the level of the transesterification reactions. Details for calculation of
the transesterification degree (B) and the D values are included in the Materials and
Methods section 3.2.6.

PET/PEN blends with a transesterification degree of greater than 10% showed a
significant improvement in the blend miscibility (Hoffman and Caldwell 1995; McGee
and Jones 1995). The miscible blends had only a single T8 and Tm value, and exhibited
optical clarity. This is because domain sizes in the miscible blends were too small to
scatter light. McGee and Jones (1995) also indicated an inverse linear relationship
between the transesterification level and the haze value (measurement of the film

opacity); i.e. as degree of transesterification increases, the haze value decreases.

3.1.2.1. Mechanisms of transesteriflcation reactions

In polyesters, there are three possible mechanisms for interchange reactions
occurring during processing. These reactions take place at the ends of the polymer chains
I where the functional groups are hydroxyl and carboxyl groups. The mechanisms include
acidolysis, alcoholysis, and transesterification reactions (Kotliar 1981; Porter and Wang
1992). Transesterification is used as a general term for the interchange reactions as well
as a specific term for the mechanism described as transesterification. The interchange

reactions are:

33

 

Intermolecular alcoholysis

-O-EO- I -OH + C|O-
<— 0.
HO-

lnterrnolecular acidolysis

-O-EO- I -O + HOCO-
<— -
-COOH CO
Transesterification
-O-CO- I -O C0-
+ I + l
_CO_O_ I ‘CO 0"

Therefore, the interchange reaction processes that can take place between PET and PEN

chains, as related to the above three mechanisms, are assumed to be as follows:

34

 

(i) Alcoholysis between a hydroxyl end-group of PET and an ester group of PEN and vice

versa,
8
PETa-COOH PETa-C-O-CHZCHz-PENb
o —’I o
l I I l
PENa-C-O-CHZCHz-PENb PENa-C-OH
‘3.
PENa-COOH PENa-C-O-CHZCHz-PETb
o _’l o
l l I I
PETa-C-O-CHZCHz-PETb PETa-C-OH

(ii) Acidolysis between-a carboxyl end-group of PET and an ester group of PEN and vice

versa,
0

PETS-OH P ETa-O-a-PENa

o —"’l

l I

PENa-C-O-CHZCHZ-PENb H-O-CH2CH2-PENo
‘3.

PENa-OH PENa-O-C-PETa

O —>

<-——

ll '
PETa'C‘O‘CH2CH2 ’PETb H'O'CH2CH2‘PETb

35

and (iii) Transesterification between ester groups of PET and PEN.

O O
II II
PETa'C‘O‘CH2CH2'PETb PETa'C'O'CH2CH2'PENb
fl
PENa-lC-O-CHZCHz-PENb ‘—" PENa-fi-O-CHZCHz-PETb
0

These initial interchange reactions lead to the formation of block and /or graft
copolymer chains. Continued interchange reactions of the initially formed block and graft
copolymers result in the formation of random copolymer chains, which accounts for the

enhanced miscibility of the PET/PEN blends.

3.1.2.2. Factors controlling transesteriflcation reactions

In any scheme to produce materials from melt blends of PET and PEN, a number
of factors have to be deal with. These include identifying the effects of processing
parameters, as well as the specific type of the processing equipment, on the interchange

reactions and properties of the blends.

Type of extruders and the screw speed

From the results of previous studies (Shi 1998; Morton-J ones 1989) and our
preliminary experiments, it was concluded that processing the blends though a twin screw
extruder yielded a higher degree of transesterification than that obtained from processing
though a single screw extruder, using similar processing parameters, i.e. temperature and

residence time. This can be attributed to the fact that the twin-screw extruder provides

36

better mixing than a single screw extruder. Better mixing optimizes contact between the
PET and PEN molecules and results in higher shear generated heat, which leads to
temperature increases and enhanced rates for the transesterification reactions (Shi 1998;
Morton-J ones 1989). In addition, the twin-screw extruder offers better heat transfer and
a more uniform temperature, which increases the transesterification reactions (Morton-
J ones 1989). In this study, we focused on processing the PET/PEN blends though the
twin-screw extruder in order to achieve a varied transesterification levels.

The extruder screw speed is also an important factor in controlling the
transesterification reactions. Increasing the screw speed causes an increase in the shear
rate and mixing effect (Shi 1998). However, increasing the screw speed also results in a

decrease of the blending time or residence time.

Processing parameters

The effects of the processing parameters, including blending temperature,
blending time, catalyst and blend composition, on the transesterifcaiton reaction have
been investigated by several researchers. Blending temperature and time are the primary
factors controlling the transesterification reactions, whereas blend composition and
catalysts have less effect (lhm et al., 1996; McGee and Jones 1995; P0 et al.1996; Shi
1998; Stewart et al., 1993).

Previous researchers found a relationship between the processing temperature and
time, and the transesterification degree. Stewart et al. (1993) showed that the
transesterification degree at a given temperature is a linear function of blending time,

corresponding to a reaction rate constant. The authors also proposed a first-order reaction

37

rate constant model. In addition, the model showed that the reaction rate constant
increases by a factor of approximately 5, as the blending temperature increases from 286
to 309 °C. An activation energy of 110 kJ mol'1 was reported.

Several researchers also studied the effect of catalysts on the transesterification
degree of PET/PEN blends. Stewart et al. (1993) found that adding 0-40 ppm of
antimony-based catalyst to the PET/PEN blends had little effect on the transesterification
reactions. Po et al. (1996) also reported similar results, in that the transesterification
degree varied fiom 0.14-0.2, when 160 and 360 ppm. of Sb, 100 and 160 ppm. of Ge , 2
and 20 ppm. of Ti, and 160 and 320 ppm. of Sn were used as the catalysts. The results of
this study showed only a slight effect of the type of catalyst and the quantity of catalyst on
the transesterification reactions, as compared to the effect of processing time and
temperature. Po et al. (1996) stated that significant effects of the catalysts might be found
if substantially higher amounts of catalysts were added.

In PET/PC blend systems, the catalysts were found to be a primary factor in
controlling the interchange reaction. Joyce and Berzinis (1991) stated that residual
. polymerization catalysts did not significantly affect the transesterification reactions of
PET/PC blends, while the addition of titanium catalysts resulted in a significant increase
in the transesterification degree. The transesterification rate of polyester/polycarbonate
blends without titanium was found to be limited (Devaux et al.(°’°’°’°) 1982; Godard et al.
1986; Pilati et al. 1985; Smith et al. 1981). In contrast, the rapid transesterification of
other polyester/polyester blends, without any catalyst, has also been reported (Kugler et
al. 1987; MacDonald et al. 1991; Montaudo et al. 1992; Rarnjit and Sedgwick 1976).

Stewart et al. (1993) proposed that PET/PEN blends and other polyester/polyester blends

38

had hydroxyl and carboxyl end-groups that differentiated them from the polyester/PC
blends. Even though they involved similar reaction processes (alcoholysis, acidolysis,
and transesterification), the controlling mechanisms for polyester/polyester blends and
polyester/PC blends in the absence of titanium are different.

The effect of the blend composition on the transesterification reactions was also
studied. A slight effect of the blend composition on the transesterification degree was
reported (Ihm et al. 1996; P0 et al. 1996; Shi 1998; Stewart et al. 1993). Ihm et a1. (1996)
concluded that there was no preference between PET and PEN, when transesterification
reactions occurred. However, Shi (1998) found a slight effect of blend composition on
the transesterification degree. The blends with lower and higher blend compositions
showed a higher transesterification degree, while the blends with intermediate levels of
blend compositions (30-40% naphthalene dicarboxylate content (NDC), which is a
percent mole fraction of naphthalate units based on the total moles of terephthalate and
naphthalate units, had the lowest transesterification degree. This behavior was explained
by the reaction kinetics theory, which showed that the activation energy should be
constant for various compositions and the change of entropy should be a minimum at the
equivalent blend composition of 33% NDC or 50 mole % PEN (Shi 1998). In contrast,
for the PET/PC blend system, the blend composition strongly influenced the

transesterification reactions observed in the PC-rich phase (Joyce and Berzinis 1991).

39

Types of PET and PEN resins

Shi (1998) indicated that types of PET and PEN influenced the transesterification
reactions. PET and PEN polymers made by different processing methods and parameters
with different catalysts were found to differ in their molecular weight distribution and
viscosity. The PET/PEN blends with higher viscosity had a lower transesterification
degree, due to less mixing between reactant components. No report of the effect of
molecular weight average and distribution of the PET and PEN on the transesterification

reactions was shown.

Sources of naphthalate.

McGee and Jones (1995) proposed that the sources of the naphthalate, including
PEN homopolymer and PET/PEN copolymer, influenced the time required to achieve
miscible blends. Using the PET/PEN copolymer as the naphthalate source resulted in a
shorter processing time to obtain clear blends. This was attributed to the pre-existence of
interlinkage of PET and PEN chains from the PET/PEN copolymer.

In the present study, the focus of the processing parameters is on the blending
temperature and time. In addition, the composition of the blends was also studied. Even
though the blend composition has little or no effect on the transesterification reactions, it

is a major parameter controlling selected properties of the blends.

40

3.2. Materials and Methods
3.2.1. Materials

Two homopolymers used in this study were Poly(ethylene terephthalate) (PET)
and Poly(ethylene 2,6-naphthalene dicarboxylate) (PEN). PET (CTF 8406) and PEN

(VRF 40046) were provided by the Shell Company (Apple Grove, WV).

3.2.2. Reaction Processing Parameters and Blend Composition

For the preliminary experiment, a 25 weight % PEN blend was processed through
a single-screw extruder at 303-307°C for 1 pass. The DSC and NMR analysis showed
that the blend was immiscible and no transesterification reaction occurred. Blends had 2
T8 and Tm values. The 1H-NMR spectra showed a peak for PET occurring at
approximately 4.8 ppm, a peak for PEN occurring at approximately 4.9 ppm, and no
interaction peak occurring between the PET and PEN peaks. The blend exhibited phase
separation and it appeared to be hazy or white in color (see Results Sections 3.3.1 and
3.3.2). '

Therefore, for the further experiments, all the samples were processed through a
twin-screw extruder in order to increase the degree of transesterification and achieve
miscible blends. The effects of blending temperature, time, and blend composition on the
transesterification reactions were investigated. Blending temperatures ranging fi'om 275
to 325 °C were studied. The blending time was varied by the number of passes through a
twin-screw extruder (one and two passes) and extruder screw speeds of 150 and 250 rpm.
The residence times corresponding to the screw speed of 150 rpm were approximately 75

and 150 seconds for one and two passes through the extruder, respectively. The residence

41

times corresponding to the screw speed of 250 rpm were not determined. Blend
compositions were varied from 10 to 40 weight percent PEN or 8.1 to 34.6 mole percent
of Naphthalene Dicarboxylate (% NDC). The mole percent of naphthalene
dicarboxylate content (% NDC) is the mole fraction of naphthalate units based on the
total moles of terephthalate and naphthalate units expressed as percent % NDC was
calculated by using the molecular weight of the naphthalate unit (242 g/mol) and
terephthalate unit (192 g/mol). Since functional groups (hydroxyl and carboxyl groups) at
the end of the polymer chain are the active sites of the interchange reactions, the PEN
composition was also expressed as the mole percent of the end groups of the PEN in the
blend (% e). The mole percent of the end groups of PEN was calculated as follows;

%e = 1%WtOfPEN/MWEEE*2)
(% Wt OfPEN / MWPEN * 2) + (°/o Wt OfPET / MWPET * 2)

Mole percent of PEN (% PEN) was also calculated as follows;

% PEN = (% Wt of PEN / Mmel
(°/o wt of PEN / MWpEN) + (% Wt of PET / MWpET)

Therefore, the mole percent of the end groups of PEN (% e) is equal to the mole % PEN
in the blends. The weight average molecule weight of PET (MWpET) was 65,400 gram
/mole, and the weight average molecule weight of PEN (MW war) was 42,900 gram /mole.
The weight average molecular weights of PET, PEN, and blends were measured by gel
permeation chromatography (see Chapter 4).

The combinations of processing parameters and blend composition used in this
study are listed in Table 3.1. These processing parameters were selected by considering

the results of the previous studies as well as equipment available.

42

Table 3.1: Processing parameters and blend composition

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Blending Blending time Screw speed Blend composition
temperature (°C) No. of passes RPM % (wt/wt) PEN
275 1 150 25
275 1 250 25
285 1 150 25
300 1 150 25
315 1 150 25
325 1 150 25
285 2 150 25
300 2 150 25
315 2 150 25
300 l 150 10
300 1 150 30
. 300 1 150 40

 

 

For the preliminary experiment, PET/PEN blend with 25% PEN (wt/wt) were

processed though a single screw extruder at 303-3070C, and 1 pass.

All of the experiments were performed on a twin-screw extruder.

43

 

3.2.3. Extrusion
3.2.3.1. Drying conditions

PET and PEN resins were weighed, mixed, and dried. In order to avoid
hydrolytic degradation during the extrusion process, the moisture content of the resins
needs to be very low (McGee and Jones 1995; Seo and Cloyd 1991; Stewart et al. 1993).
Therefore, the resins were dried at 82°C for at least 20 hours in a DBD25 vacuum dryer
(Premier Pneumatics Inc., Salina, Kansas). In addition, the extrudate was redried under
the same conditions before the second pass. The moisture contents of the dry resins were
measured by the 720 KFS titrino Karl Fischer titration (Brinkrnann Instruments, Inc.,
Westbury, New York) to confirm that the samples had low moisture content (lower than

100 ppm).

3.2.3.2. Twin-screw extruder

Dried resins were automatically fed through the extruder at 10 lbs per hour using
an MDII weight feeder (Acrison Inc.). A Baker Perkins Model ZSK-30 co-rotating twin-
screw extruder (Werner & Pfleilderer Corporation, Ramsey, New Jersey) was used to
carry out the reaction processing.

The dimensions of the extruder were:

Outer screw diameter 30.7 mm
Inner screw diameter 21.3 mm
Thread depth 4.7 mm
L/D ratio 26:1

44

The extruder consists of three basic heating zones: (i) the feeding zone below the feed
hopper; (ii) the compression zone where polymers start to melt; and (iii) the metering
zone connected to the die. There are a total of six ports along the three heating zones of
the extruder. The temperatures of all ports were set equally.

The melt blends left the extruder through a die head with 2 cylindrically shaped
outlets. The extrudate was then immediately cooled in a water bath at approximately
50°C. After that, it was pelletized using a Conair pelletizer (J etro Division Inc., Bay
City, Michigan). The blend pellets were optically clear, rod shaped, and approximately 1
mm in diameter and 4 mm long.

For the preliminary experiment, the single screw extruder was used for initiating
transesterification reactions. Dry PET and PEN resins (75: 35 wt/wt) were fed though a
Killion single-screw extruder. Processing temperatures were 303-307°C, with 1 pass

though the extruder. The residence time was 4.5 nrinutes.

3.2.4. DSC Analysis

The thermal properties of the blend resins were determined by using lyfl3SC 2920
Modulated Differential Scanning Calorimetry (TA instrument Inc., New Castle,
Delaware).

All samples were precisely weighed with a Sartorious analytical balance
(Sartorius GMBH Gottngen (Germany), Westbury, New York). Optimum weights are
normally in the range of 5-10 milligrams. Whole pellets of PET/PEN blend resins were
used, if their weights were in the optimum range. The blend fihns were cut and stacked

to achieve the optimum weight. Use of low weight thin samples results in minimal

45

thermal gradients, which facilitates attaining steady state condition and gives accurate
data. Each sample was placed in an aluminum bottom dish and a top dish was put on.
The dishes were sealed by using the clamp. An empty dish was used for the
compensation of the heat flow corresponding to the dishes. The polymer blends were
equilibrated at 10°C for 5 minutes before being heated to 290°C (320°C for PEN) with the
heating rate of 4°C per minutes. The modulated mode with an imposed modulation rate
of +/- 1°C every 60 seconds was used. The modulated DSC has a superimposed
sinusoidal temperature oscillation, while the conventional DSC has a linear change in
temperature. Therefore, modulated DSC can separate the total heat flow into reversing
and nonreversing components, which is permits accurate and easy interpretation of the
results.

The glass transition temperature (T8) was determined from the transition
temperature of the reversing heat flow profile. Melting temperature (Tm), and heat of
fusion (AHf) were determined from the endothermic event of the total heat flow profiles.

The percent crystallinity of the blends was determined fi'om the heats of fusion of
100% crystalline blends, which were calculated from weight average (fi'om the mole
fraction of the blend composition) of the heats of fusion of 100% crystalline PET and
PEN homopolymers. The heats of fusion are 121.42 J oules/gram (29.0 calories /gram)
and 103.41 Joules/gram (24.7 calories/ gram) for PET and PEN, respectively (Cheng et a1.

1988; Roberts 1995). The percent crystallinity was calculated as follows:

46

The heat of fusion (AHf) of 100 % crystalline blend with X mole % NDC
= (X/100 "' 103.41) + (100-X)/100 * 121.42 Joules/gram
Thus;
AHf of 100 % crystalline blend with 8.1 mole % NDC = 119.62 Joules/gram
AHf of 100 % crystalline blend with 21 mole % NDC = 116.92 Joules/gram
AH; of 100 % crystalline blend with 25.4 mole % NDC = 116.02 Joules/gram
AHf of 100 % crystalline blend with 34.6 mole % NDC = 114.22 Joules/gram

% crystallinity of 8.1 mole % blend = AHfGoules/gram) *V 100 / 119.62

3.2.5. ‘H-NMR Analysis

The transesterification degree of the blends was measured using proton nuclear
magnetic resonance spectroscopy (lH-NMR INOVA 300, Varian VNMR). Samples were
prepared by dissolving 0.25 g of the blend in 5 ml of 70/30 (by weight) mixture of
deuterated chloroform and trifluoroacetic acid. 1H-NMR spectra were accumulated and
transformed using the standard operating software. A typical NMR spectrum for a
PET/PEN blend is shown in Figure 3.1. The degree of transesterification of the blends
was determined by spectra analysis using the region of the 1I-INMR spectrum between
4.8-4.9 part per million (PPM). PET and PEN peaks appear at approximately 4.8 and 4.9
PPM, respectively. In PET/PEN blends with transesterification, there are heterolinkages
between the naphthalate and terephthalate units. The 1H peak corresponding to these
heterolinkages appears at 4.85 PPM, between those of the PET and PEN homopolymers
(Figure 3.1). The curve of the NMR spectrum was fitted by 3 normal curves using the

least squares method. The intensities of the PET, PEN, and interaction peaks were

47

 

 

 

 

 

 

PET/PENpeak . l
l l
PENpeak\ \ I i.
actualspectrum
fullfit
individual olot , A
‘IT'P'1""71 r' IT I V1"
5.2 5.1 5.0 4.9 4 a 4 7 as.

Figure 3.1: NMR spectrum of the PET/PEN blend with 25% (wt/wt) PEN processed

through the twin-screw extruder at 300°C and 1 pass.

48

determined by the area under the corresponding curves. The degree of transesterification
was then calculated from these intensities using the Bemoullian statistics technique

(Y arnadera and Murano 1967, and Po et al. 1991), which is described in the next section.

3.2.6. Calculation of Degree of Transesterification and D value
Degree of transesterification (B) was originally described and calculated by
Yamadera and Murano (1967). By assuming that the lI-INMR spectrum is a measure of
the relative intensities of homolinkages and heterolinkages, the average sequence length
of polymer chains and the degree of transesterification can be determined.
Molar fractions of terephthalate (MT) and naphthalate (MN) are obtained from the
intensities of the NMR spectrum:
MT = AT-G-N /2 + Ar-G-r (1)
MN = AT-G-N /2 + AN-G-N (2)
where Au}; , AMLN, and A1204»; are the ratios of the intensities of the PET peak, the
PEN peak, and the interaction peak of PET and PEN, respectively, and the total
intensities of these three peaks. Assume that the polymer chains of PET/PEN blends are
extremely long. Therefore, the number of end units is negligible compared to the number
of all monomer units. Suppose that a monomer unit is randomly picked from the
PET/PENiblend chains. If the selected unit is a naphthalate (N) unit, then Pm is defined
as the probability that the next unit proceeding down the chain will be a terephthalate (T)
unit. Pm is defined similarly, except that the roles of N and T are switched. Pm and Pm

can be calculated fi'om the intensities of the NMR peaks as follows:

49

PM = ATG-N/ZMN (3)

Pm = ATGN/ZMT (4)
The degree of transesterification (B) is defined as:

B = PNT + PTN (5)
Yamadera and Murano (1967) also indicated that the T and N units obtained a random
distribution when B = 1 and the probability of finding a copolyrrrer unit is based on
Bernoulli statistics. Ifthese units tend to form in blocks of each unit, B is below 1. In
addition, if a homopolymer mixture is found and no transesterification reaction is
achieved, B is zero. On the other hand, if the homopolymer sequence length becomes
shorter than would be expected by random reaction, B is greater than 1. When an
alternating capolymer is formed, B is equal to two. However, a true alternating
copolymer can be formed only if PET and PEN are present in equimolar amounts.

For blends of PET and PEN homopolymers, percent randomness and
transesterification are equal and equivalent to the transesterification degree (B) times 100.
Therefore, a polymer blend without interaction or transesterification has 0 % randomness
(transesterification) and B = 0, while a random copolymer has 100% randomness and
B = 1.

The average sequence length of T units (La) and N units (LnN) are calculated by

1411' = ZMT/ AT-G-N = l/PTN ‘ (6)
Iain = 2MN / Ar-o-N = NEW (7)
The fraction of the T-G-T, N-G-N and T-G-N linkages can be obtained from the

intensities of the PET, PEN and PET/PEN peaks of the ‘HNMR spectrum. The

50

parameters MT, MN, Pm, Pm, LnT, LnN, and B can then be calculated from equations
(1) - (7)-

One should note that the B value is calculated by simply adding Pm to Pm
without weighing the molar fraction of PET and PEN. Therefore, the B value is most
meaningful only for a blend with equimolar amounts of PET and PEN.

Therefore, the D value is proposed. Suppose that a monomer unit is randomly
picked from the PET/PEN blend chains. D is defined as the probability that the next unit
is different from the selected unit; i.e. the selected unit is N and the next unit is T or vice
versa. D can be calculated as follows:

D = I’m MN + PTN Mr (8)
By substituting (3) and (4) into (8), it becomes
D = (Ar-c.N/2MN)(MN) + (AT-G-N/ZMTXMT) = AT—G-N (9)

The D value ranges from 0 to 1. When no transesterification reaction occurs, the
D value is equal to 0. When an alternating copolymer is obtained, the D value is equal to
l. Ifa random copolymer is formed, the D value is 0.5, regardless of the composition
ratio.

Let Dm (M) be the maximum achievable D where M is the smaller mole fraction
of these two polymers. Dmax (M) can be calculated as follow:

Dmax (M) = 2M (10)

For instance, if the PET and PEN compositions are equal (50% mole), Dm (M)
is l,and it is possible that an alternating copolymer is formed. If the PET and PEN
compositions are not equal, Dm (M) is less than 1, and the alternating copolymer cannot

be formed.

51

In the literature, the B value is widely used to represent the degree of
transesterification. Therefore, in this study, the B value is used as well. However, the

more meaningful D value is also presented.

3.3. Results and Discussion

3.3.1. DSC Analysis

The blends processed though the single screw extruder (300°C and 1 pass) had
two T8 values, which appeared at 73 and 119 °C, respectively (Figure 3.2). These T8
values corresponded to the PET and PEN rich phases, respectively. The melting peaks
appeared as split peaks, which included the major peak at 251°C and a relatively small
peak at 270°C (Figure 3.2). These melting peaks corresponded to the PET and PEN rich
phases, respectively, and they seemed to merge during the DSC scan, as shown by the
split peaks. Thus, the blend processed though the single screw extruder at 300°C
exhibited immiscibility or phase separation. In addition, 1HNMR analysis also confirmed
that no measurable amount of transesterification reactions was obtained.

The DSC analysis showed that the blends processed though the twin-screw
extruder had only one T8 and Tm value, which were between those of PET and PEN
homopolymers. Therefore, miscible blends were obtained. A typical DSC scan of a
miscible blend is shown in Figure 3.3. A summary of T8, T m and % crystallinity values
for the respective blends is presented in Chapter 4, Section 4.3. 1.

Therefore, for all further experiments, the twin-screw extruder was selected for

processing the blends, so as to achieve varied levels of transesterification.

52

smlo: 25:75 9574:9131 cowouuo DSC FIIoz CtPEN-PET.500

 

 
    
     
 

 

 

 

 

 

Size: 93000 mg Operatorz.MJR .
Method: moo 4.0 CMN TO 175 c Run Dm- 274111397 15.25
Commonlzuloouurso osc raw 303-307 c. RESIDENCE TIME 4.51m
0.00
-0.05J
73.07 cm
, 69.27 °C
0
gr 77.ss°c "8'82 c (1) 221.61 °c
g -0.10« “Hate seem/g
3 117.10’c ‘—
m
> 015-
0
o:
-0.20-1
251.01 °c
-025 ' T T l ' r T f V r
0 50 100 150 200 250 300

Temerature(°C)

Figure 3.2: The DSC scan of the PET/PEN blend with 25% (wt/wt) PEN processed

through the single-screw extruder at 300°C and 1 pass.

53

Sample: blend#3- 30001 onented film We C:...1DSC\ru;idal300010I

 

 

 

 

 

 

8120: 5.7000 mg DSC Operator. Rullda
Method: IUflda memod1 Run Date; 1680099 15 38
Comment 81000113 (30001 )-onented film 1 mil
- 02
a
g 3
3 E
E i
i :l:
I 3
0:
--0 3
.02 i r I f l T I ' r T I ' '0 4
0 50 100 150 200 250 300
Em Up Tm (°C) UI'IW v3.0a TAInsmm

Figure 3.3: The DSC scan of the PET/PEN blend with 25% (wt/wt) PEN processed

through the twin-screw extruder at 300°C and 1 pass.

54

3.3.2. Degree of Transesterification (B)

The results showed that no measurable transesterification reactions occurred
during processing the blends though the single-screw extruder at 300 0C. The lH-NMR
spectra showed a peak for PET occurring at approximately 4.8 ppm, a peak for PEN
occurring at approximately 4.9 ppm, and no interaction peak occurring between the PET
and PEN peaks. However, the blends processed through the twin-screw extruder at the
blending temperatures of 275 to 315°C and l and 2 passes resulted in varied degrees of
transesterification. The interaction peak occurred at 4.85 ppm (between the PET and PEN
peaks). The degree of transesterification ranged from 0.004 to 0.53. Blend resins and
blend films (non-oriented films) obtained at the same processing times and temperatures
had similar ranges of degree of transesterification. Statistical analysis using single factor
AN OVA (analysis of variance) showed no significant differences between the degree of
transesterification of blend resins and of blend films at the confidence level of 0.05,
except for 40 % of PEN by weight (Table A1 in appendix A). Therefore, it was
concluded that no significant transesterification reaction occurred during film production.

In addition, the effects of the processing factors and the blend composition on the
degree of transesterification were studied. As the PEN composition increased, the degree
of transesterification appeared to be constant. Statistical analysis was performed, and the
result showed no significant difference in the degree of transesterification among blend
samples with 10 to 40% PEN by weight (Table A3 in appendix A). The processing
conditions of blending time and temperature appeared to be the primary factors
controlling the transesterification reactions. The results are discussed in detail in the

following sections.

55

3.3.2.1. Effect of blending time on transesterification reactions
The blending time was varied by the extruder screw speed and the number of

passes through the extruder.

Screw speed

The extruder screw speed is a factor controlling the blending time, as a faster
screw speed yields a shorter residence time. The results showed that as the screw speed
of the extruder increased, the degree of transesterification decreased (Table 3.2). The
transesterification degree of the blends obtained at a 150 rpm screw speed (one pass at
275°C) was 0.004 while that of blends obtained at a 250 rpm screw speed (one pass at
27 5°C) was not measurable by NMR analysis. This is due to the limitation of lI-INMR
spectroscopy. In addition, blends processed at 250 rpm and one pass at 275°C were
white or opaque, which indicated the irnnriscibility of the blends. However, DSC
analysis of these blends (250 rpm) showed only one T8 and one Tm value, which indicated
that these blends achieved some minimal degree of transesterification.

Shi (1998) found the same phenomena and concluded that the transesterification
reactions lead to the creation of partial copolymers in the blend structure. This copolymer
behaves like a compatibilizer, and blends appear to be more compatible, compared to a
physical mixture or 0% transesterification. Stewart (1993) also showed that firrther
reaction occurs during the DSC analysis. Thus, the degree of transesterification may
increase during performing the DSC test, resulting in an; increase in miscibility, which

leads to show only one T g and Tm value.

56

From the results obtained in the present study, it can be concluded that when the
blends were processed at a screw speed of 250 rpm, the degree of transesterification
achieved was minimal (see Table 3.2). Therefore, a screw speed of 150 rpm was selected

as the base processing condition for the study of other processing variables.

Table 3.2: The effect of the extruder screw speed on the degree of transesterification (B).

The blends with 25% of PEN (wt/wt) were processed at 275°C for 1 pass.

 

 

 

Screw speed The degree of transesterification (B)
150 rpm 0.004
250 rpm less than 0.002"

 

 

"' The transesterification degree cannot be detected by NMR analysis.

Numbers of passes

The average degree of transesterification for the resin blends achieved from one
pass and two passes at melting temperatures of 285 to 315°C and at a screw speed of 150
rpm, ranged from 0.053 to 0.513. From NMR results, as shown in Table 3.3 and Figure
3.4, the degree of transesterification increased as the number of extrusion passes
increased. This confirmed the results reported by Stewart et al. (1993). The authors
processed PET/PEN blends through a single-screw extruder at 295 to 315°C for 1 to 3
passes and reported a linear relationship between the number of passes and the
transesterification level.

Our results showed higher degrees of transesteri fication were achieved than those

reported by Stewart et al. (1993), who found that degrees of transesterification ranging

57

 

from 0.08 to 0.14 and 0.18 to 0.21 were achieved by extruding PET/PEN blends through
a single screw extruder at 305°C for one and two passes, respectively, as compared to
level of transesterification between being achieved by processing though the twin screw
extruder at 300°C for one and two passes. The differences in degree of transesterification
observed are attributed to the fact that the twin-screw extruder used in this study gave

better mixing than a single-screw extruder.

3.3.2.2. Effect of blending temperature on transesterification reactions

Blends of 25% PEN (wt/wt) processed at 275, 285, 300, 315 and 325°C (150 rpm,
and one pass and two passes) were investigated. The results of NMR analysis of the
respective samples are shown in Table 3.3 and Figure 3.4. These results clearly
demonstrate that a significant increase in transesterification was achieved at the higher
processing temperatures. The relationship between the blending temperatures and the
degree of transesterification appeared to follow a linear expression. From linear
regression, the equations are Y = 0.0087X -2.4058 for 1 pass, and Y = 0.0117X —3.1395
for 2 passes through the extruder, where X is the temperature (°C) and Y is the degree of
transesterification (Figure 3.4). The R2 values are 0.9933 and 0.9812, respectively.

This conclusion is in agreement with the results of McGee and Jones (1995), and
Stewart et a1. (1993). Stewart et al. (1993) also demonstrated the kinetic model of
transesterification as a first-order reaction rate-constant.

The PET/PEN blend resins obtained from reaction processing at a temperature of
275°C did not appear to be uniform and were mixed between clear and hazy zones. This

indicated that a blending temperature of 275°C is the minimum limit to achieve a

58

Table 3.3 : The effect of blending temperature and time on the degree of
transesterification (B)

 

 

 

—Temp B of blend resins* B of blend films"
(0 C) lst pass 2nd pass lst pass 2 nd pass
275 0.053 - - -
285 0.061 0.151 0.076 0.192
300 0.219 0.383 0.222 0.405
315 0.309 0.513 0.356 0.525
325 0.429 - - -

 

 

 

 

 

 

 

"‘ Blends with 25 %PEN (wt/wt) processed through twinescrew extruder( 150 rpm)

 

 

 

 

0.7
5 2 asses ,
a 0.6 it p /.
3 y = 0.0117x - 3.1395
E 05 ._ R2 = 0.9812 ‘
g 0.4 7* g//// ,/i
g E / g/
'45 0.3 a? ’1/ /,/
/ /

E 0.2 -- y’ //!/ 1 pass
8 / " ' o/ y = 0.0087x - 2.4058
2 0-1 t /’. R2=0.9933
1- §(//

0 i *r i— : 1

270 280 290 300 31 0 320 330
Temperature ( °C)

 

O l pass-resins I lpass-films o 2 passes-resins D 2 passes-films i

 

 

Figure 3.4 :The effect of blending temperature and time
on the degree of transesterification (B)

59

measurable degree of transesterification. In addition, the PET/PEN resin blend obtained
from reaction processing at 325 °C was brittle and difficult to pelletize because of low
melt viscosity. Degradation might occur during processing blends at this temperature or
at temperatures greater than 325°C. Therefore, the optimum blending temperature is

considered to be in the range fi'om 285 to 315°C

3.3.2.3. Effect of blend composition on transesterification reactions

The degree of transesterification achieved among PET/PEN blends with
compositions ranging fiom 10% to 40% PEN (wt/wt) (14.5 to 50.4 mole % PEN) showed
no significant difference (Table 3.4 and Figure 3.5). These findings were in agreement
with the results of previous studies. Stewart et al. (1993), McGee and Jones (1995), and
lhm et al. (1996) reported that the blend composition had little or no influence on the
degree of transesterification. However, in studying blends with PEN compositions
ranging from 10 to 80 weight percent, Shi (1998) found that the blends with low and high
PEN levels had a slightly higher degree of transesterification than those with medium

PEN levels (approximately 32 mole % NDC or 50 mole % PEN).

60

Table 3.4 : The effect of PEN composition on the degree of

 

 

transesterification (B)
avg B avg B
Sample ID PEN mole% resins’“ films" Bavg
8 14.5 0.189 0.204 0.197—
3 33.7 0.219 0.222 0.221
9 39.5 0.181 0.208 0.195
10 50.4 0.13 0.192 0.161

 

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through twin-screw extruder.

 

.0

NP

0100
I

P
N
I

 

The transesterification degree (8)

 

 

 

\

0.15 4- E
0.1 ._
0.05 4~

0 1 i i i t
- 0.0 10.0 20.0 30.0 40.0 50.0
PEN mole %
o resins I films + Bavg

 

Figure 3.5 : The effect of PEN composition on the degree of
transesterification (B)

61

3.3.2. D value

The D value of blends ranged from 0.01 to 0.181. Statistically, there was no
significant difference between the D values of blend resins and blend films obtained at
the same processing conditions (Table A2 in Appendix A).

The relationship between the D value and the blending temperature and time were
found to be similar to that between the B value and the blending temperature and time
(Table 3.5 and Figure 3.6). However, the PEN composition slightly influenced the D
value when the PEN mole % was less than 25. Where the PEN mole % was higher than
35, the D value appeared to be constant(Table 3.6 and Figure 3.7). Statistical analysis
showed no significant difference between the D values of blends with 35 to 50 mole %
PEN. However, the D value of the blends with 14.5 mole % PEN was less than the D

values of the other blend samples (Tables A4 and A5 in appendix A).

62

Table 3.5: The effect of blending temperature and time on
the degree of transesterification and D value.

 

 

 

 

 

 

 

 

Temp D average B average
(0 C) D-l st pass D-2nd pass B- 1 st pass B-2 nd pass
275 0.0164 0.053
285 0.022 0.054 0.069 0.1715
300 0.074 0.130 0.221 0.394
315 0.113 0.184 0.333 0.519
325 0.148 0.450

 

 

"' Blends with 25 %PEN (wt/wt) processed through twin-screw extruder( 150 rpm)

 

 

 

 

 

 

0.7 0.7
0.6 ~r «e 0.6
a0 5 o 5
‘6§ ' '
§ 80.4 -~ -» 0.4
8%
'3 %03 «r -~ 0.3
B
"' 50.2 4- -~ 0.2
E
0-1 1- -_ 0.1
0 1 l l # fit 0
270 280 290 300 310 320 330
Temperature ( °C)
1 A B-lst pass A B-2 nd pass 0 D-lst pass 0 D-an pass 1

 

Figure 3.6 :The effect of blending temperature

and time on the degree of transesterification and D value

63

The D value

Table 3.6 : The effect of PEN composition on the degree of

transesterification and D value

 

 

 

 

 

Sample ID PEN mole% Davg Bavg
8 14.5 0.023 0197—“
3 33.7 0.073 0.221
9 39.5 0.072 0.195
10 50.4 0.072 0.161

 

 

 

* Blends processed at 300°C and 1 pass through twin-screw extruder.

 

 

 

 

 

0.250 0.250
/_,,,a
0.200 .. 1/ ; «~ 0.200
E \\
u- C
0 30150 a. + 0.150
t 8
8%
1’301004» +0100
O a '
fi 3 *_§___{
C
fl // ,
5 0.050 .. / ~~ 0.050
e/'
0.000 t . t 4. 0.000
0.0 10.0 20.0 30.0 40.0 60.0
PEN mole °/e
+Bavg '+Davg J

 

Figure 3.7 : The effect of PEN composition on
the degree of transesterification and D value

The D value

Chapter 4
Characterization of PET/PEN Blends

4.1. Introduction

Chemisivity, molecular weight and molecular weight distribution are crucial
parameters controlling other blend properties, such as thermal properties, density,
morphology, gas banier properties and mechanical properties. In this chapter, we will
focus on the relationships between the blend composition and the degree of
transesterification (or the processing condition), and the important properties of the
PET/PEN blends including thermal properties, density, molecular weight and molecular
weight distribution, and morphology. In Chapter 5, the relationships between the blend
composition and the degree of transesterification and the blend performance, including
gas barrier and mechanical properties, will be evaluated. By combining this information,
a better understanding of the basic characteristics of PET/PEN polyblends and blend
performance will be provided, and predictions of barrier and mechanical properties of the

resultant blend filrns can be made.

4.1.1. Thermal Properties
The physical state of a polymer determines the characteristics of that polymer.
The physical state in which a polymer can exist can be ideally considered as its regular

polymer chains. There are three possible states:

65

(i) completely free rotation, as in a polymer at the melting state,

(ii) no rotation, as in a polymer in the glassy state, at very low temperature, where its
chains are frozen, and

(iii) packing, as a crystalline polymer (Rodriguez 1996).

Two major critical temperatures that divide the physical states of a polymer are
the glass transition temperature (T8) and the melting temperature (Tm). The
crystallization is also described as how polymer molecule can fit together in such a way
so as to stabilize the chain arrangement in a regular lattice. A polymer at a temperature
greater than its T8 is in the rubber state in which polymer chains have segmental mobility.
A polymer at a temperature below its T8 is in the glassy or rigid state, in which polymer
chains cannot rotate and are frozen in a specific conformation. Polymers exhibit fi'ee
rotation in the melt state or at temperatures greater than its Tm. Morphology indicates
whether a polymer exhibits both thermal transitions or only one thermal transition (Odian
1991). Completely amorphous polymers exhibit only a glass transition temperature,
whereas semicrystalline polymers have both glass transition and melting temperatures.
Crystallizable blends were selected for this study. Therefore, these blends exhibit both a
glass transition temperature (T3) and a melting temperature (T m)-

The physical state of a polymer is a significant parameter controlling the
performance of that polymer. For instance, a polymer with a high T8 is a very rigid and
strong material, and has very high gas barrier properties, due to its non (or partial)
rotating polymer chains, which does not allow the gas molecules to penetrate the inner
structure. The melting temperature is also an important factor controlling the

processibility and flow properties of polymers, and indicates the maximum temperature

66

resistance of those polymers. There are the relationships between degree of crystallinity,
and barrier and mechanical properties of polymers, as well. The crystalline matrix of a
crystallized polymer has a very oriented microstructure, which results in high thermal
properties and very strong materials (Odian 1991).

F urtherrnore, the glass transition temperature and the melting temperature of
polymer blends have been used for measuring the numbers of the blend components.
PET/PEN blends without any transesterification reaction are inherently immiscible.
Irnrrriscible blends exhibit both PET and PEN properties, which are not uniform, and
phase separation of PET and PEN is found. The blends exhibit two T8 and two Tm
values, corresponding to PET and PEN rich phases.

When transesterification reactions are obtained, interlinkages between
terephthalate and naphthalate units (T-N bonds) occur. This causes changes in the
microstructure of the blends which, in turn, results in changes in the physical state of the
blends. The transesterification reactions lead to miscible blends, which exhibit minimal
sizes of the PEN domains, resulting in uniform properties of the blends. Miscible blends
exhibit only one T8 and Tm, with the T8 and Tm values falling between those of PET and
PEN homopolymers (Olabisi et al. 1979; MacKnight et al. 197 8; Karasz 1985; Utracki
1990).

The observation of the thermal properties of the blends was done by differential
scanning calorimetry (DSC). The effect of the transesterification reactions and blend

composition on T1; ,Tm, and % crystallinity were studied.

67

4.1.2. Molecular Weight and Molecular Weight Distribution

The molecular weight and the molecular weight distribution are measures of the
average size and the distribution of the sizes of the polymer chains/molecules,
respectively. The molecular weight and the molecular weight distribution are important
characteristics which are responsible for several basic properties of the polymer. For
thermoplastic materials, the molecular weight and molecular weight distribution strongly
control the processing behavior at elevated temperatures and the mechanical properties of
the therrnoplastics (Rudin 1982). In industry, plastic grades vary in the average sizes and
distribution of the average sizes of the polymer chains. Polymers with high molecular
weight averages exhibit higher resistance to deformation at elevated temperatures.
However, this type of polymer needs an intensively higher temperature for processing or
forming desired shapes (Rudin 1982).

From randomness theory, Paul and Bucknall (1999) concluded that when
transesterification reactions occur, the rrricrostructures of the polymer blends change. For
example, for PET/PEN blends, the average sequence lengths of terephthalate and
naphthalate units decrease, whereas the average length of the interlinkage T-N bond
increases.

The average sequence lengths can be calculated from lHNMR analysis using
Yamadera and Murano’s equations (Y amadera and Murano 1967). The relationship
between the degree of transesterification and the average sequence lengths of
terephthalate and naphthalate units can also be determined fiom Yamadera and Murano’s

equations, as shown in Section 4.2.3.2.

68

As the number of interchange reactions of the terephthalate and naphthalate units
increases. the molecular weight and the molecular weight distribution of the blends will
change. In the present study, determination of the molecular weight and molecular
weight distribution of the blends with selected degrees of transesterification was
performed by using gel permeation chromatography. The information obtained fi'om
molecular weight analysis can be used to evaluate the ultimate properties of the blends,

such as their melting temperatures (Brennan 1978).

4.1.3. Density

The density is a typical characteristic that determines the ultimate properties of
polymeric materials. For example, polymers of the same generic type with higher density
tend to exhibit higher performance, such as higher gas barrier and mechanical properties.
The change of the microstructure of the blends due to transesterification reactions may
cause the density of the blends to change.

The density of the blends was measured using the density gradient technique. The

effect of transesterification reactions and blend composition on the density was studied.

4.1.4. Morphology

Morphology is an important characteristic of polymers. It controls other
characteristics, such as the thermal properties of the blends. Most blends, such as
PET/PEN blends, consist of two or more phases. Immiscible blends exhibit two phases,
a PET and PEN rich phase. Typically, PET has a lower intrinsic viscosity and exists as a

continuous phase. PEN has a higher intrinsic viscosity and, therefore, exists as a

69

dispersed phase in the PET continuous phase (McGee and Jones 1995). In addition,
PET/PEN immiscible blends exhibit poor optical properties, because the sizes of the PEN
domains are large, and the domains scatter light.

When transesterification reactions are obtained, interlinkages between
terephthalate and naphthalate units (T—N bonds) occur, and the lengths of PET and PEN
chains become shorter. The interchange reactions between PET and PEN chains enhance
the miscibility of the blends (McGee and Jones 1995; Stewart et al. 1993). Partially
miscible blends consist of PET, PEN, and terephthalate-naphthalate interlinkage phases.
The blends are optically clear, due to the small size of the PEN domains, which are too
small to scatter light (McGee and Jones 1995).

Continued interchange reactions of the initially formed block and graft
copolymers lead to the formation of random copolymer chains. Random copolymers
exhibit only one phase, the T-N interlinkage phase. Therefore, changes in the
morphology of the blends result in changes in the characteristics of the blends.

One of the newest techniques to study the morphology of polymers is Raman
chemical microscopy. Raman chemical microscope is a new type of optical microscope
which is a laser confocal scanning microscope (LCSM) coupled with a filter or
spectrometer (Sawyer and Grubb 1996). The Raman analysis is a powerful technique
which can reveal the chemical architecture of polymeric materials without using stains,
dyes, or contrast agents which might cause the interference or artifact to the analytical
species. Using this technique, microstructures of immiscible blends without
transesterification reaction and miscible blends with some degree of transesterification

were investigated.

70

4.2. Materials and Methods

4.2.1. Film Fabrication Process

Film fabrication was done by the Amoco Research Center, Amoco Chemicals
Company, Illinois. The PET, PEN, and PET/PEN blends (11 samples) obtained from
reaction processing were dried at 120°C overnight in a forced air-desiccating oven. The
samples were crystallized and clumped together during the drying period. Therefore, the
pellets were separated using a grinder after crystallization, and returned to the oven. The
moisture content of the blend resins needs to be lower than 50 ppm. to avoid hydrolytic
degradation. The dried resins were fed through a Killion model KL-125 single screw
extruder (Werner & Pfleiderer Corporation, Ramsey, New Jersey). The extruder was
connected to a 6-inch flexible lip sheet die and 3-roll temperature controlled take-off.
The dimensions of the extruder were as follows:

Screw diameter 31.25 mm

L/D ratio 24:1

Extruder conditions for film fabrication
Sample 1 (Shell 8406 PET):
Screw speed = 75 rpm, Tm," = 537 °F, P = 1670-1700 psi, drive current = l3-l4amps,
take-up speed = 14.9-15.0 fpm
T profile (°F) = 515/525/530/530/530/500, Trolls (°F) = 100/ 100/ 100
Sample 2 (25% (wt/wt), processed at 285°C and 1 pass):
Screw speed = 75 rpm, Tm.“ = 534 °F, P = 310-320 psi, 4.5-5.5 amps, 14.2-14.3 fpm

r profile (F) = 515/525/530/530/530/500, Tm". (°F) = 100/100/100

71

Sample 3 (25%, processed at 300°C and 1 pass):
Screw speed = 75 rpm, Tm,“ = 534 °F, P = 540-560 psi, 5-6 amps, l4.2-14.3 fpm
T profile (°F) = 515/525/530/530/530/500, Trolls (°F) = 100/100/ 100
Sample 4 (25% (wt/wt), processed at 315°C and 1 pass):
Screw speed = 75 rpm, Tm." = 534° F, P = 430-460 psi, 3.5-4.5 amps, 14.2-14.3 fpm
T profile (°F) = 515/525/530/530/530/500, Trolls (°F) = 100/100/ 100
Sample 5 (25% (wt/wt), processed at 285°C and 2 passes):
Screw speed = 75 rpm, Tm," = 532 °F, P = 260 psi, 4 amps, 14.2-14.3 fpm
T profile (°F) = 515/525/530/530/530/500, Trolls (°F) = 100/100/ 100
Sample 6 (25% (wt/wt), processed at 300°C and 2 passes):
Screw speed = 75 rpm, Tm," = 530-532 °F, P = 240-260 psi, 3.5-4.0 amps, l4.2-l4.3
fpm. r profile (°F) = 515/525/530/530/530/500, Tm". (°F) = 100/100/100
Sample 7 (25% (wt/wt), processed at 315°C and 2 passes):
Screw speed = 75 rpm, Tm,“ = 532-533 °F, P = 300-320 psi, 3.5-4.5 amps, l4.2-14.3
fpm. T profile (°F) = 515/525/530/530/530/500, Trolls (°F) = 100/100/ 100
Sample 8 (10% (wt/wt), processed at 300°C and 1 pass):
Screw speed = 75 rpm, Tm." = 538 °F, P = 560-620 psi, 3-4 amps, 14.2-14.3 fpm
T profile (°F) = 515/525/530/530/530/500, Tron, °(F) = 100/ 100/ 100
Sample 9 (30% (wt/wt), processed at 300°C and 1 pass):
Screw speed = 75 rpm, Tm," = 536-537° F, P = 670-740 psi, 4-5 amps, 14.2-14.3 fpm
T profile (°F) = 515/525/530/530/530/500, Trolls (°F) = 100/ 100/ 100

Sample 10 (40% (wt/wt), processed at 300°C and 1 pass):

72

Screw speed = 75 rpm, Tm." = 530-532 °F, P = 740-780 psi, 5-6.5 amps, 14.6-14.7
fpm. T profile (°F) = 515/525/530/530/530/500, Trolls (°F) = 100/ 100/ 100
Sample 11 (Shell 40046 PEN):
Screw speed = 75 rpm, Tm," = 580-591 °F, P = 1590-1640 psi, 14-15 amps, 14.4-14.5
fpm. T profile (°F) = 560/575/580/580/580/545, Trolls (°F) = 100/100/ 100
Melting temperature (Tmcn ) and pressure (P) were measured at the end of the extruder
barrel. Temperature profiles are listed fiom the feed throat to the die.
The final film thickness was approximately 10 mils (0.01 inch). The non-oriented
films were found to be uniform and optically clear. The blend films were then biaxially

oriented.

4.2.2. Orientation Process
The eleven 10 mil films were biaxially oriented on a T.M. Long film stretcher,

utilizing a unit located at the Amoco Company facility. The elongation was 300%.
The 10 mil films were conditioned at 23°C and 50% RH for 40-48 hours. All of
. the samples were stretched 3x3 simultaneously in both directions at a rate setting of 300%
per second (6 inches per second). The stretch temperature was 100°C for all samples,
except for the PEN film, which was stretched at 135°C. All samples had a soak time of
15 seconds. Twelve biaxially oriented film specimens were prepared for each sample.
The final oriented film thickness was approximately 1 mil (0.001 inch). The film
samples (non oriented and oriented films) were then characterized with respect to
thermal, mechanical, barrier, and morphological properties as a function of composition

and orientation conditions.

73

4.2.3. Analysis Methods
4.2.3.1. Thermal analysis

The thermal properties of the blend resins were determined using a MDSC 2920
Modulated Differential Scanning Calorimetry (TA Instruments Inc., New Castle,
Delaware).

All samples were precisely weighed with a Sartorious analytical balance
(Sartorius GMBH Gottngen (Germany), Westbury, New York). Optimum weights are
normally in the range of 5-10 milligrams. For the blend resins, whole pellets of PET/PEN
blend resins were used, if their weights were in the optimum range. For the blend films,
films were cut and stacked to achieve the optimum weight. Use of low weight thin
samples results in minimal thermal gradients, which facilitates attaining steady state
conditions and gives a00urate data. Each sample was placed in an aluminum bottom dish
and a top dish was put on. The dishes were sealed using the clamp. An empty dish was
used for compensation for the heat flow corresponding to the dishes. The polymer blends
were equilibrated at 10°C for 5 minutes before being heated to 290°C (320°C for PEN),
with the heating rate 4°C per minute. The moderated mode with an imposed modulation
rate of +/- 1°C every 60 seconds was set. A modulated DSC has a superimposed
Sinusoidal temperature oscillation, while a conventional DSC has a linearchange in
temperature. Therefore, a modulated DSC can separate the total heat flow into reversing
and nonreversing components, which gives the benefits of accuracy and ease of result
interpretation.

The glass transition temperature (T8) was determined from the transition

temperature of the reversing heat flow profile. The melting temperature (Tm), and the

74

heat of firsion (AHr) were determined from the endothermic events of the total heat flow
profiles.

The % crystallinity of blends was determined from the heats of fusion of 100%
crystalline blends, which were calculated fi'om the weight average (of the mole fraction of
the blend composition) of the heat of fusion of 100% crystalline PET and PEN
homopolymers. The heat of firsions are 121.42 J oules/ gram (29.0 calories / gram) and
103.41 Joules/ gram (24.7 calories/gram) for PET and PEN, respectively (Cheng et al.
1988; McGee and Jones 1995). The calculation of the % crystallinity is shown in the
following equations.

The heat of fusion (AHf) of 100% crystalline blend with X mole % NDC
e (X/100 * 103.41) + (100-X)/100 * 121.42 Joules/gram
Thus,
AHr of 100% crystalline blend with 8.1 mole % NDC = 119.62 Joules/gram
AHf of 100% crystalline blend with 21 mole % NDC = 116.92 Joules/ gram
AHf of 100% crystalline blend with 25.4 mole % NDC = 116.02 Joules/gram
AHr of 100% crystalline blend with 34.6 mole % NDC = 114.22 Joules/gram

% crystallinity of 8.1 mole % blend = AHfGoules/gram) * 100 / 119.62

4.2.3.2. Density measurement
The density of blend resins, non-oriented films, and oriented films was
determined by the density-gradient technique (ASTM Procedure D1505-85, reapproved

1990). The density analysis was performed at room temperature (23°C).

75

The apparatus consisted of a 2-inch diameter and 32-inch height density gradient
column, a set of standard calibrated floats (Cole-Palmer Instrument Company, Vernon
Hills, Illinois), toluene-carbon tetrachloride solutions, and a set of glass apparatus for

gradient tube preparation (Cole-Parmer Instrument Company, Vernon Hills, Illinois).

The densities of the set of 5 calibrated floats ranged fiom 1.2494 to 1.4199 g/cc, as

 

 

 

 

 

 

follows:
Code Color Density (g/cc)
Bead 1 Large blue 1.2494
Bead 2 Blue 1.3000
Bead 3 Blue 1.3500
Bead 4 Yellow 1.3997
Bead 5 Red 1.4199

 

 

 

 

 

Solution preparation

To obtain a density gradient ranging from 1.25 to 1.455 g/cc, a solution A with the
density of 1.248 g/cc and a solution B with the density of 1.455g/cc were prepared fi'om
the toluene and carbon tetrachloride. The density of the toluene and carbon tetrachloride
is 0.87 and 1.59 g/cc, respectively. Solution A was a premixed solution of 375 ml of
toluene and 425 ml of carbon tetrachloride, and solution B was a premixed solution of

150 ml of. toluene and 650 ml of carbon tetrachloride.

Preparation of the gradient column
The glass apparatus for gradient tube preparation was assembled by method C-

Continuous Filling in ASTM D1505-90 (Anonymous 1990“”), as shown in Figure 4.1.

76

800 ml of solution A (the lighter solution) and 800 ml of solution B (the denser solution)
were placed in volumetric flasks A and B, respectively. The calibrated floats were gently
put in the empty gradient column. The gap between the end of the siphon tube and the
bottom of the gradient column is very critical and should be minimal, in order to obtain
the most uniform density gradient and limit bubble formation. Solution A and B were
introduced to the gradient column by turning stopcocks 1 and 2, consecutively. A
propeller-type stirrer in flask A was turned on and the stirring speed was adjusted to
achieve the continuous mixing of solution A and B. The total flow rate of solutions A
and B to the gradient column was approximately 20 ml per rrrinute, and was controlled by
monitoring the level of the solutions in flask A and B, and adjusting the stopcocks as
needed. The first solution delivered was the lighter end of the gradient (pure solution A
with density of 1.25 g/cc). The density gradient of the solution built up as the solution
being introduced into the column became progressively denser. After two thirds of the
column was filled, the stopcocks were turned off, and the siphon tube was slowly
removed from the column. The filled gradient column remained undisturbed for 24 hours
to allow the gradient to reach steady state. The positions of the calibrated floats were

then measured and the blend samples were tested.

Calibration of the standard glass floats

The calibrated floats, placed in the column previously, floated to constant
positions.
These positions corresponded to the density of each float. A calibration curve was

obtained by plotting the densities of the calibrated floats against their positions in the

77

gradient column. There was a linear relationship between the positions of the floats in the
gradient column and the density of the floats, with a correlation value of 0.99 (see Figure
Bl , Appendix B).
Testing the specimens

The blend resins were dropped directly into the center of the gradient column.
The non-oriented and oriented films were cut into different shapes with Size
approximately 2X2 mm, for convenience of identification. The specimen samples
traveled along the gradient column until they stopped at constant positions. The positions
were measured and remeasured again 24 hours later to confirm that there was no further
change. The density values of the samples were calculated using the calibration equation

(Appendix B).

 

 

Siphon tube

w ./

 

 

 

 

 

 

 

 

 

 

 

Stopcock l Stopcock 2

 

 

 

—
-

l—
—(
—
‘

1'1 IIIL

 

 

 

Figure 4.1: Apparatus for the gradient tube preparation

78

4.2.3.3. Average sequence lengths of terephthalate and naphthalate units.

The estimation of the average sequence lengths of terephthalate and naphthalate
units was done using the lHNMR technique, and the Yamadera and Murano equations
(see detail in section 3.2.5 ‘H-NMR analysis).

Molar fractions of terephthalate units (MT) and naphthalate units (MN) are

obtained from the intensities of the NMR spectrum:

Mr = Ar-G-N /2 + Arc-r (1)
MN - Ann: /2 + AN-G-N (2)
MT + MN = 1 (3)

The average sequence length of terephthalate units (La) and naphthalate units (LnN) were

calculated by
Lnr = ZMT / AT-o-N = l/PTN (4)
LnN = 2MN / AT-G-N = “PM (5)
where B = PNT + PTN (6)
B = l/LnT + l/LnN (7)

From equations (1) to (7), the following expressions were derived.
14.7 = 1/(1 + Mr)B (8)
LnN = 1/(1 + M1013 (9)
where Ann , AN-G-N, and A1204»; are the ratios of the intensities of the PET peak, the
PEN peak, and the interaction peak of PET and PEN, respectively, with the total

intensities of these three peaks. The fiaction of the T-G-T, N-G-N and T-G-N linkages

79

can be obtained from the intensities of the PET, PEN and PET/PEN peaks of the 1HNMR
spectrum. B is the degree of transesterification. Therefore, from (8) and (9), the average
sequence length of terephthalate units (Lnr) and naphthalate units (LnN) decreases, when

the degree of transesterification increases.

4.2.3.4. Molecular weight measurement

The molecular weight measurements were performed using the size exclusion
chromatography/multi-angle laser light scattering (SEC-MALLS) technique.

Chromatographic system consists of a Waters model 510 pump (0.6 ml/min), a
Validyne model 7025 injector, 2 PLgel C columns, a Wyatt Dawn DSP-F multi-angle
laser light scattering detector, an applied Biosystems model 757 variable wavelength
ultraviolet detector, and a Waters model 410 differential refiactometer (DRI).

Approximately 25 milligrams of each sample were weighed and mixed with 2.0
milliliters of a solvent of chloroform and trichloroacetic acid. Chloroform and
trichloroacetic acid were used as the solvent or eluent in all experiments. The solutions
' were gently mixed overnight to ensure homogeneity. Only the solution of PEN
homopolymer was heated during mixing. The solutions were then injected directly into

the SEC-MALLS. The data were collected using Wyatt’s Astra software (version 4.2).

4.2.3.5. Morphological analysis
The morphological analysis was performed using Raman Chemical Imaging
microscopy. A Rarnan image microscope is a new type of instrument. It is an optical

microscope or laser confocal scanning microscope (LCSM) with a filter or spectrometer

80

(Sawyer and Grubb 1996). The Raman imaging micr0800pe was diagrammed and
described in detail elsewhere (Schaeberle and Treado 1997; Schaeberle et al. 1999;
Morris et al. 1999).

In this study, the Raman analysis was performed by ChemIcon Company,
Pittsburgh, PA. A ChemIcon FALCON Rarnan microscope with Liquid Crystal Tunable
Filter (LCTF) imaging spectrometer and Silicon Charge-Coupled Device (CCD) detector
(Chemlcon Company, Pittsburgh, PA) was used to conduct the experiment.

Four resin and two film samples, listed in Table 4.1, were tested. The samples
were manually cross-sectioned. Final thickness was approximately 25 um. Each sample
was then immobilized on a glass microscope slide and the side was placed under the
Raman microscope.

Laser excitation was achieved by a 532 nm Nd: YAG laser which yielded 0.2 to
1.6 watts of power at the exiting laser head. Sample excitation and Raman scatter
collection was performed using a 20X (0.46 NA) and 100X (0.95 NA) objectives on the
Rarnan microscope. Once the laser scatter was ejected, the Raman emission was filtered
with LCTF. Raman images were collected using a 16-bit dynamic range, liquid nitrogen-
cooled (-40°C), slow scan charge—coupled device (CCD) detector. The detector has 512
x 512 pixels. Pixel size is 9 pm. The CCD exposure times were 5 seconds for PET, 2

seconds f0r PEN, and 10-30 seconds for blends.

81

Table 4.1: List of the samples tested for Raman chemical imaging analysis

 

Sample ID PEN Type of Blending Blending Types of
composition extruder Temperature Time sample
(% Weight) (°C) (Number
of asses)

 

. 300P1 single 25 Single 300 l Resin

 

. 300P1 25 Twin 300 1 Film and resin

 

.315P2 25 Twin 315 ' 2 Resin

 

 

. 300Pl-40% 40 Twin 300 1 Film and resin

 

 

 

 

 

 

The brightfield reflectance, polarized light, and Raman chemical images of each
sample with 20X and 100X magnification were collected. The dispersive Raman spectra

collected from approximately 600-3200 cm“were also acquired.

4.3. Results and Discussions

4.3.1. Thermal Properties of Blends

The blend resins and films obtained from one and two passes through the extruder
at temperatures of 285, 300 and 315°C showed one glass transition temperature and one
melt temperature. Moreover, these T8 and T m values fell between the T8 and Tm values of
pure PET and PEN, respectively. All blend samples were optically elear, which
indicated that the blends were miscible. However, fiom the DSC analysis, the peaks
corresponding to the melting temperatures of the blends were broader than those of PET
and PEN homopolymers. According to Andresen and Zachmann (1994), the results

suggest that the blends were only partially miscible.

82

 

The following sections will discuss the effects of the transesterification reactions

and the blend composition on the T8 and Tm values and the % crystallinity of the blends.

4.3.1.1. Glass transition temperature (Tg)

The results showed that the Tg values for PET/PEN blends were independent of
the degree of transesterification, but were directly proportional to the molar compositions
of PEN in the blend.

This conclusion agrees with Stewart et al. (1993), Hoffman and Caldwell (1995),

McGee and Jones (1995), P0 et al.(1996) and Shi (1998).

TS and degree of transesterification

DSC analysis showed that the T8 values of the blends with 25% (wt/wt) PEN were
in the range of 77 to 84°C. Those blends, which were processed at 285 to 315°C, and one
pass and two passes through the extruder, had a degree of transesterification ranging fi'om
0.061 to 0.525. I

There were no statistically significant differences between these T8 values (Table
4.2 and Figure 4.2). The statistical analysis is shown in Appendix C, Table C3.

Shi (1998) and McGee et al. (1995) found that after PET/PEN blends achieved a
certain degree of transesterification, a further increase in the degree of transesterification
has little effect on the resultant T8 value. Stewart et al. (1993) suggested that the
broadened range of the T8 values occurred due to the partial miscibility of the polymer
blends and the lack of sensitivity of DSC measurement. The partial miscibility can

account for both the broad range of T8 values obtained and the film haziness. Based on

83

Table 4.2 : The effect of the degree of transesterification (B) on the glass
transition temperature (T g) of the PET/PEN blends.

 

 

 

 

 

 

 

 

T. (°C)
sample B oreinted films non-oriented films resins
PET - 86.07 73.66 76.82
#2 0.069 84.56 82.79 76.87
#3 0.221 83.55 82.79 82.04
#4 0.333 84.79 80.97 78.63
#5 0.172 85.49 80.48 82.51
#6 0.394 83.98 81.33 84.28
#7 0.519 84.38 81.51 81.77
PEN - 114.01 120.87 119.52

 

 

* Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

 

 

 

 

0.6

 

 

 

100
3
E 95 -.
g 90 ~r
5 5'" 85 T 3 § 0 g
% l-a' g D '3
c v 80 7” a
3 3
3 75 ..
3
a
e 70 -.
:5
65 % T7 1 i i
0.0 0.1 0.2 0.3 0.4 0.5
The degree of transesterification (B)
l o oriented films 0 non-oriented films A resins

Figure 4.2: The effect of the degree of transesterification on the
glass transition temperature of the PET/PEN blends

84

the T8 values obtained, the blends with lower levels of transesterification exhibited
greater partial miscibility, while the Tg values for the polyblends which achieved a higher
level of transesterification fell within a narrower temperature range. In addition, Stewart

et al. (1993) found that the polymer blends further reacted during the DSC measurement.

T8 and blend composition

The average Tlg values of the blends with 10 to 40% (wt/wt) PEN and processed at
300°C and 1 pass were in the range of 78 to 89°C. The average T8 values of PET and
PEN resins were 77 and 120°C, respectively.

The average T8 values of PET/PEN blends with various blend compositions are
tabulated in Table 4.3. Figure 4.3 shows the plot of T8 versus PEN composition. The
average T8 values of the blend resins and non-oriented films were in the same range.
Moreover, there is a linear relationship between those T8 values and the PEN
composition. However, for the case of oriented films, the linear expression does not fit
well. Instead, a second degree polynomial expression gives the best fit. The orientation
effect will be discussed in Chapter 6.

The widely known expression based on the free volume model was used 'to
estimate the T8 of the copolymers and blends (DiMarzio and Gibbs 1959; Gordon and

Taylor 1952). The expression is:

T = wl(Tg)1+kw2(Tg)2
g wl+kw2

 

(10)

85

Table 4.3 : The effect of PEN composition on the glass transition temperature

 

 

 

of the PET/PEN blends.
T. (°C)
sample PEN mole % oriented films non-oriented films] resins
PET 0.0 86.07 73.66 76.82
#8 14.5 84.27 76.78 77.48
#3 33.7 83.55 82.79 82.04
#9 39.5 86.46 84.23 85.89
#10 50.4 84.39 88.88 89.08
PEN 100.0 117.00 120.87 119.52

 

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

 

The glass transition temperature
(T .°C)

non-oriented films
y = 0.4767x + 72.841

R2=0.9989

resins

y = 0.4438x + 74.481

R2 = 0.9921

   

  
 

 
  
  
 

 

 

 

go .
0
8° . oriented films
70 ' y = 0.004518 - 0.1712x + 85.881
R2 = 0.9918
60 r i ‘r i i
0 20 40 60 80
PEN mole %
i 0 oriented films 0 non-oriented films A resins

 

 

Figure 4.3 : The effect of PEN composition on the glass

86

transition temperature of The PET/PEN blends

100

where w), w, Ty, and T82 are the weight fractions and T1; values of pure-component l
and 2, respectively, and k is a fitted constant. When k is equal to l, a linear relationship
between T8 and blend composition is found (Rodriguez, 1996). The expression can be

simplified as;

g x l g 2

However, the above expression is applied specifically to the blends in which polymer-
polymer interactions are weak. A comparison between the calculated T8 values, based on
this expression and the measured Tg values were shown in Table 4.4 and plotted
graphically in Figure 4.4. Even through, both calculated and measured T3 values seem to
be in the same range, statistical
analysis showed significant differences between both T8 values, at the confidence level of
5% (Appendix C, Table C3).

The modified expression, which accounted for the effect of the polymer-polymer
interaction was studied by Kwei (1984) and Kwei et al. (1987). They proposed the

extended expression, such that

T _ w,(Tg)l +kw2(Tg)2 +qwlw2 (12)

g wl +lrw2

 

where q is the nature and magnitude of the polymer-polymer interactions. However, to

apply this expression to the PET/PEN blend system, more information is required.

87

Table 4.4:

The comparison between calculated and measured Tg

 

 

 

of the blend resins
sample PEN mole% Calculated Tg Measured Tg

PET 0.0 76.8 76.8

#8 14.5 79.7 77.5

#3 33.7 84.4 82.0

#9 39.5 86.0 85.9
#10 50.4 89.6 89.1
PEN 100.0 119.5 119.5

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

140

(To. °C) _.
8 8

8

. J

Glass transition Temperature

0

..A
N
O

3.3

N
O

 

IL

Calculated T9
y = 0.4324x + 72.18
R2 = 0.9375

Measured T9 .
y = 0.4429x + 70.892
r:2 = 0.9277

 

 

O

20

40
PEN mole %

60

 

80 100

 

l

o Calculated Tg

A Measured Tg l

Figure 4.4: The conparison between the calculated and
measured T8 of the blend resins

88

4.3.1.2. Melting temperature (Tm)

The results showed that the degree of transesterification and the blend
composition were the primary factors controlling the melting temperature of the blends.
The melting temperature decreased as the degree of transesterification increased.
Furthermore, depression of the Tm was found, when the polymers were blended as was
expected. The minimum Tm values of the PET/PEN blends were obtained when the

blend composition was 40 to 50 mole % PEN.

T... and degree of transesterification

The Tm values of blends with 25% (wt/wt) PEN ranged from 226 to 247°C.

These blends were processed at 285 to 300°C, and l and 2 passes, and transesterification
degree ranged fi'om 0.061 to 0.525. The results showed a linear relationship between the
degree of transesterification and the Tm of the PET/PEN blends studied (Table 4.5 and
Figure 4.5). As shown, the Tm of the blends appears to decrease as the transesterification
degree increases.

This melting temperature depression can be explained by the polymer interaction
mode (Paul and Bucknall 1999). Miscible blends have lower chemical potentials,
compared to their pure components. Thus their crystals are in equilibrium with the
miscible amorphous phase, at a lower temperature. Therefore, the melting temperature of

blends decreases.

89

Table 4.5 : The effect of the degree of transesterification (B) on
the melting temperature (Tm) of the PET/PEN blends.

 

   

 

 

 

 

 

Tm (°C)
sample B oriented films non-orientfims resins
PET - 248.71 250.07 248.54
#2 0.069 244.95 245.80 247.33
#3 0.221 237.57 237.12 240.45
#4 0.333 233.33 234.10 233.01
#5 0.172 239.66 241.38 242.23
#6 0.394 231.10 232.30 231.86
#7 0.519 226.16 225.95 226.85
PEN - 268.65 268.59 261.64

 

 

 

" Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

 

   
 
    

250
c oriented films
245 ._ y = 40.82x + 247.54
R2 = 0.9946
240 t resins
y = 46.723x + 250.78

235 , R2 = 0.9937

230 :—
non-oriented films

y = 42.366x + 248.65

The melting temperature (Tm,°C)

 

 

 

 

22 «—
5 R2 = 0.9732
220 . ' .
0 0.1 0.2 0.3 0.4 0.5 0.6
The degree of transesterification (B)
. O oriented films 77 D noLQriented films A gins

Figure 4.5 : The effect of the degree of transesterification on
the melting temperature of the PET/PEN blends

90

T... and blend composition

The average T... values of the blends with 10 to 40% (wt/wt) PEN, processed at
300°C and l and 2 passes, were in the range of 231 to 246°C. The average T... values of
PET and PEN resins were 249 and 263°C, respectively. The results showed that the blend
composition has a strong influence on the T... of the blends. The relationship between
PEN molar composition and the T... is shown in Figure 4.6 and summarized in Table 4.6.
The T... values of the blends were lower than those of the PET and PEN homopolymers.
The minimum T... values were found to appear with the blends consisting of 40 to 50
mole % PEN. When the PEN composition was lower than 50 mole % PEN, the T...
decreased with an increase in PEN composition. When the PEN composition was greater
than 50 mole % PEN, the T... increased with an increase in PEN composition.

Shi (1998) found a similar relationship between T... and PEN compositions, with
the lowest T... occurring at 33 mole % NDC or 50 mole % PEN. The depression of the T...
could be explained by the polymer interaction model (Paul and Bucknall 1999). In this
model, the critical blend composition is approximately 50% wt/wt for blends with similar
molecular weight, or closest to the component with the lower molecular weight. At the
critical composition, blends have the lowest interaction energy. This causes the crystals
to be in equilibrium with the amorphous phase at the lowest temperature. Shi (1998), and
Lu and Windle (1995) also explained this phenomenon by proposing that when a small
amount of PEN is added to PET, the crystalline structure of the PET rich phase is
disrupted and the crystallite size is limited, resulting in the depression of T... Above 50

mole % PEN, PEN can now be considered the continuous phase, and adding PET can be

91

Table4.6: Theefi'ectofPEl‘lconpositiononthennltingtenperatureof

 

 

 

the PET/PEN blends
Tm (°C)

sarrple PEN Mile % oriented filrm non-oriarted films resins
PET 0.0 248.71 250.07 248.54
#8 14.5 245.06 245.91 246.42
#3 33.7 237.57 237.12 240.45
#9 39.5 235.72 234.93 235.81
#10 50.4 232.27 231.03 232.62
PEN 100.0 268.65 268.59 261.64

 

 

 

 

 

 

 

*Blerxlsproeessedat300°Candlpessduoughahwin—screwexmxla.

 

non-orientedfilms
y=0.0111x2-0.9378x+251.33
. R2=0.9943

oriented filrrs E
y = 0.010118 - 0.8259x + 249.83

 

 

 

 

The melting temperature (Tm,°C)
i 9 a e a e i 3 a: e «3

 

J 3 y=0.W-0.7676x+250.34
R2=0.9723
0 20 40 60 80 100
- PENmole%
l Oorientedfilrrs Dnon-orientedfilrrs Aresins l

 

fimlfifl'heefieaotPENeonpoeldononthennltim
Mn ofthePETlPBlblende

92

viewed as disrupting the crystalline structure. The disruption of the crystalline structure

might cause a decrease in the interaction energy, which results in reduction of T...

4.3.1.3. Percent crystallinity of the blends

The results showed that the degree of transesterification and the blend
composition strongly affected the % crystallinity of the blends. The % crystallinity
decreased as the degree of transesterification increased. Furthermore, a reduction of the
% crystallinity was found, when adding a polymer to another (Shi 1998). The minimum

% crystallinity appeared in the blends with 40 to 50 mole % PEN.

Percent crystallinity and degree of transesterification

The % crystallinity of blends with 25% (wt/wt) PEN ranged from 23% to 38%.
These blends were processed at 285 to 300°C, and 1 and 2 passes, respectively, and the
transesterification degree of these blends ranged from 0.061 to 0.525. The results showed
a linear relationship between the degree of transesterification and the % crystallinity of
the PET/PEN blends (Table 4.7 and Figure 4.7). As shown, the % crystallinity of the

blends appeared to decrease, as the transesterification degree increased.

When the transesterification reactions increase, PET and PEN chains are
shortened, and the number of heterolinkages of terephthalate (T) and naphthalate (N)
units increases. This new structure of T-N chains would not easily form a crystalline
phase, and would interfere with the formation of crystals. The shortening of the

sequence length of the crystallizable components can account for the decrease in the

93

 

Table 4.7 : The effect of the degree of transesterification (B) on

the percent crystallinity of the PET/PEN blends

 

 

 

%Crystallinity
sample B oriented films non-oriented films resins
PET 40.94 32.75 50.45
#2 0.069 37.84 35.11 33.40
#3 0.221 30.83 25.60 28.88
#4 0.333 28.57 27.19 23.77
#5 0.172 34.66 31.53 28.66
#6 0.394 27.67 29.36 26.85
#7 0.519 25.70 24.46 22.88
PEN - 46.47 46.53 64.54

 

 

 

 

 

 

"' Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

 

 

    

 

 

 

45
4o -- oriented films

5 35 y = -27.149x + 38.599
5 “i R2 = 0.9297
3:; 30
5‘ 25 .
‘5'
§ 20 "
a 15 1
.3 10 «L resins non-oriented films
I- y = -21.289x + 33.459 y = 48.82:! + 34225

5 «r R2 = 0.3169 R2 = 0.5937

0 t : i t .

0.0 0.1 0.2 0.3 0.4 0.5

The degree of transesterification (B)

 

r 0 oriented films

0 non-oriented films

A resins l

Figure 4.7: The effect of the degree of transesterification on
the percent crystallinity of the PETIPEN blends

94

 

0.6

ability of the blends to crystallize (Eersels and Groeninckx 1996; Eersels and Groeninckx
1997; Fakirov et al. 1996 (8" (b); Lenz and Go 1973). Therefore, crystallinity is limited
when random copolymers are achieved. Random copolymers with 15 to 85 mole %
NDC appear to be in the amorphous phase and no crystalline phase is found (Callander

and Howell 1994; Hoffman and Caldwell 1995; McGee and Jones 1995; Shi 1998).

Percent crystallinity and blend composition

The % crystallinity of the blends with 10 to 40% (wt/wt) PEN and processed at
300°C and l and 2 passes is in the range of 20.5% to 36.5%. The % crystallinity of PET
and PEN resins were 50% and 64.5%, respectively. The results Showed that the blend
composition is a primary factor in determining the % crystallinity of the blends. The
relationship between the PEN composition and the % crystallinity was expressed as a
polynomial equation (Table 4.8 and Figure 4.8). This relationship is similar to the one
between T... and the PEN composition. From our data, the % crystallinity of the blend is a
minimum, when PEN composition is 40 to 50 mole % PEN. The % crystallinity of PET
and PEN is the highest at both ends (PET and PEN homopolymers). By adding the PET
to the PEN rich phase, the crystalline structure was disrupted and the % crystallinity

decreased. A similar behavior occurs when adding PEN to the PET-rich phase.

4.3.2. Density of Blend Resins and Films
The density of the blend resins and films was determined by the density gradient
technique. The results showed that the density of the blends is dependent on the degree

of transesterification and the blend composition. The density of the blends was

95

 

Table 4.8 : The effect of PEN composition on the percent crystallinity

 

 

 

of the PET/PEN blends
%Crystallinity

sample PEN mole% oriented films non-oriented films resins
PET 0.0 40.94 32.75 50.45
#8 14.5 34.54 36.49 34.24
#3 33.7 30.83 25.60 28.88
#9 39.5 27.07 27.79 23.97
#10 50.4 23.39 21.02 20.48
PEN 100.0 46.47 46.53 64.54

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

 

 

 

 

100.0

 

70
resins oriented films

60 y = 0.0147x2 - 1.3068x + 48.089 y = 0.008x2 - 0.7422x + 40.987
3. R2 = 0.9817 R2 = 0.9816
g 50 -
f.
z- 40 °
0
*c"
0 30 .
2
8
g 20 ~
1- non oriented films

10 _ y = 0.007):2 - 0.5947x + 36.222

R2 = 0.8789
0 . a: i .
0.0 20.0 40.0 60.0 80.0
PEN mole °/e
l 0 oriented films a non-oriented films A resins i

 

Figure 4.8 : The effect of PEN composition on the percent crystallinity

of the PETIPEN blends

96

considerably lower than that of the PET and PEN homopolymers. However, the density
values for the blend films were slightly higher than those of the blend resins, and alter the

orientation process, the density of the oriented films increased Significantly.

Density and degree of transesterification

The density of blends with 25% (wt/wt) PEN ranged from 1.3379 to 1.3503 g/cc
(Table 4.9). These blends were processed at 285 to 300°C, and 1 and 2 passes,
respectively, and the transesterification degree ranged from 0.061 to 0.525. The results
showed that there is a linear relationship between the degree of transesterification and the
density of the PET/PEN blends (Table 4.9 and Figure 4.9). The density of the blends

appeared to decrease, as the transesterification degree increased.

These results were in agreement with the effect of the transesterification reactions
on the % crystallinity. Both density and % crystallinity respond to the packing ability of
the polymer molecular chains. It should also be pointed out that as transesterification
reactions occur, the formation of the T-N interlinkages increases. These N-T bonds
would not be conductive to polymer chains packing as tightly together as those of the

individual polymers. Therefore, the density of the blends decreases.

97

 

Table 4.9: The effect of the degree of transesterification on

 

 

 

the density of the blends
sample ID Tansesterification Density (g/cc)
(B) resins non-oriented films oriented films
2 0.068 1.3389 1.3395 -
5 0.172 1.3383 1.3390 1.3503
3 0.217 1.3385 1.3390 1.3491
4 0.329 1.3384 1.3388 1.3488
6 0.389 1.3385 1.3390 1.3460
7 0.522 1.3379 1.3384 1.3459

 

 

 

 

 

 

* Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

Density (glee)

 

 

 

 

 

 

1.3520

oriented films
1-3500 ~~ y = -0.0129x + 1.3522

2 -
1.3480 + R 0.8455
1.3460 4L
non-oriented films

13440 «i y = -0.002x +1.3395

R2 = 0.7894
1.3420 «r -
1.3400 -- _

resins

13380 a. y = -0.0017x '1' 1.3389

R2 = 0.6433
1.3360 1 : e T.

0 0.1 0.3 0.4 0.5 0.6
Transesterification degree (8)
l A resins U non-oriented films 0 oriented films i

 

Figure 4.9: The effect of the dree of transesterification on the

density of the blends

Density and blend composition

The density of PET and PEN resins was 1.4041 and 1.3640 g/cc, respectively.
The density values of the blends with 10 to 40% (wt/wt) PEN, processed at 300°C and 1
and 2 passes, were in the range of 1.3367 to 1.3535 g/ce.

The density of the blended resins was significantly lower than the density of PET
and PEN resins. The density of the resins seemed to decrease very slightly, as the PEN
composition increased (Table 4.10, and Figure 4.10).

The density of the blends decreased tremendously, following the fihn processing.
The results showed that the densities of the PET and PEN non-oriented films were much
lower than those of PET and PEN resins (Table 4.11 and Figure 4.11). However, afier
the biaxial orientation process, the density of the PET, PEN and blend oriented films
increased. There is a linear relationship between the density of non-oriented films and
PEN composition (Table 4.12 and Figure 4.12). However, the relationship between the
density of the oriented films and the PEN composition can be best described by a
polynomial equation with a correlation coefficient (r-square value) of 0.9 (Table 4.12 and
Figure 4.12). The reduction of the density of the blends by adding PEN molecules to
the PET rich phase is due to the interference of PEN molecules in the regularity of the
PET structure. This would result in the decrease in the ability of the polymer chains to

pack close together.

99

 

Table 4.10: The effect of PEN composition on the density of

 

 

the PET/PEN blend resins
Density
Sample ID PEN wt % PEN mole % of blend resins (g/cc)
PET o 0.0 1.4041 ‘
8 10 14.5 1.3385
3 25 33.7 1.3385
9 30 39.5 1.3371
10 40 50.4 1.3367
PEN 100 100.0 1.3640

 

 

 

 

 

"‘ Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

1.4100

1.4000 3

.8

1.3900 ._

1.3800 J-

Density (glee)
is {3
8 8

 

 

 

..—
1.3500 4-
1.3400" I; ,,,,,,,,,, a... ..... 15"".
1.3300 . . .1 ;
0 20 40 60 80 100
PEN mole %

Figure 4.10 : The effect of PEN composition on density of the
PET/PEN blend resins

100

Table 4.11: The density of the PET, PEN and blends

 

 

 

 

 

 

 

 

 

 

 

Blending temp. Blending time Density (g/cc)
sample 1D % PEN mole (0C) (No. pass) resins non-oriented films oriented films
PET 0 - - 1.4041 1.3429 1.3621
2 33.7 285 1 1.3389 1.3395 1.3481
3 33.7 300 1 1.3385 1.3390 1.3491
4 33.7 315 1 1.3384 1.3388 1.3488
5 33.7 285 2 1.3383 1.3390 1.3503
6 33.7 300 2 1.3385 1.3390 1.3460
7 33.7 315 2 1.3379 1.3384 1.3459
8 14.5 300 1 1.3385 1.3421 1.3535
9 39.5 300 1 1.3371 1.3381 1.3467
10 50.4 300 1 1.3367 1.3371 1.3483
PEN 100 - - 1.3640 1.3316 1.3446
1.4050
1.3950
1.3850
73
2 1.3750
9
z. 1 3650
g 1.3550
‘3 1.3450
1.3350 f'
oriented ilms
1'3250 7 7 non-oriented films

sample ID

   
  

resins

Figure 4.11: The density of the blend resins and films

101

 

Table 4.12: The effect of PEN composition on the density of

 

 

 

 

 

 

 

the PET/PEN blend film
Densitiy (_g/ee)
Sample ID PEN wt % PEN mole % non-oriented films oriented fihns
PET 0 0.0 1.3429 1.3621
8 10 14.5 1.3421 1.3535
3 25 33.7 1.3390 1.3491
9 30 39.5 1.3381 1.3467
10 40 50.4 1.3371 1.3483
PEN 100 100.0 1.3316 1.3446

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

    

 

 

 

 

1.3650
1;
1.3600 \ oriented films
‘\. . y = 4E-06x2 - 0.0006x + 1.3599
1.3550 -- ‘ R2 = 0.8935
1\ \

3 1.3500 + \\§\\
32 §\
E 1.3450 1
a
c
8 1.3400 «-

1.3350 ..

non-oriented films
1.3300 .1 y = -0.0001x + 1.3419
R2 = 0.939
1.3250 1 1r i .
0 20 40 60 80 100
‘ PEN mole%
1 u non-oriented films 0 oriented fihns }

1

 

Figure 4.12: The effect of PEN composition on density of
the PET/PEN blend films

102

4.3.3. Average Sequence Lengths of Terephthalate and Naphthalate Units.

The average sequence lengths of the terephthalate and naphthalate units are
reported in Table 4.13 and Figure 4.13. Yamadera and Murano (1967) proposed a
relationship between the degree of transesterification and the average sequence lengths,
which was discussed in Section 4.2.3.3. The results confirmed that as the
transesterification reactions occur, the lengths of PET and PEN chains are shortened. In
addition, the number of interlinkages of terephthalate and naphthalate bonds increases.

The effect of the blend composition on the average sequence lengths was also
studied, as shown in Table 4.14 and Figure 4.14. The results showed that as the PEN
composition increased (up to 50 mole % PEN), the average sequence length of

terephthalate units decreased significantly, while those of naphthalate units increased

slightly.

4.3.4. Molecular Weight of Blends

Only 6 samples of the blend films (non-oriented films) were selected for
molecular weight measurement. The weight average (MW), number average (Mn) and 2-
average molecular weight (M2) of PET, PEN, and selected PET/PEN blends are reported
in Table 4.15.

The results showed that the molecular weight of PET was higher than that of
PEN. This is in agreement with the density analysis showing that the PET density was
higher
than the density of PEN (Table 4.11). Furthermore, the molecular weight averages for

the blends were lower than those of PET and PEN. This is due to the effect of the

103

Table 4.13: The efeect of the degree of transesterification on the average sequence length

of the terephthalate (LIT) and naphthalate(L.N) units.

 

 

sample [D B-an [airman LnN—ruins LnT-films LnN-fiims
2 0.076 84.7 21.3 63.9 16.7
3 0.222 21.8 5.8 21.6 5.7
4 0.356 15.3 4.1 12.8 3.6
5 0.192 34.4 8.2 26.7 6.6
6 0.405 12.6 3.3 12.0 3.1
7 0.525 8.1 2.6 8.7 2.4

 

 

 

 

 

 

 

 

* Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

The average sequence length
(La and Lu». units)

 

 

a:
.0
o

 

 

 

 

 

 

0-0 . . ‘F9—‘1'.
0 0.1 0.2 0.3 0.4 0.5 0.6
The degree of transeuerlfleatlon (B)
1 A LnT-resins A LnN-resins - LnT-fihns o LnN-films 1

 

Figure 4.13: The effect of the degree of transesterificaiton on the

average sequence length of the terepahthalate and
naphthalate units

104

 

Table 4.14: The effect of PEN composition on the average sequence
length (1.“) of the terephthalate and naphthalate (LnN) unit

 

 

sample ID PEN mole% LnT-resins 1mm... LnT-films Imam
8 14.5 69.8 5.9 81.4 5.3
3 33.7 21.8 5.8 21.6 5.7
9 39.5 23.0 7.3 19.1 6.4
10 50.4 24.9 1 1.3 15.0 8.0

 

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

110

 

100~

The average sequence length
(La and L"... units)
8 8 8 8 8 5‘ 8 8

10+

 

 

t

 

I

 

 

0.0

f

10.0

20.0

30.0

PENmole%

40.0

50.0 .

 

[ a LnT-resins

A LnN-resins

I LnT-films

i
1
l
J

 

Figure 4.14 : The effect of PEN composition on the average
sequence length of terephthalate and naphthalate units

105

60.0

 

Table 4.15: The molecular weight average of the blend films

 

 

 

 

 

 

 

 

 

 

 

 

   
  

 

 

 

 

 

Figure 4.15: The effect of PEN composition on the molecular
weight average of blend film (non-oriented film)

106

Blend Blend time
sample ID PEN mole% Temp. (0C) no. of pass B Mn Mw Mz
PET 0 46400 65400 82900
3 33.7 300 1 0.074 26600 32400 37200
4 . 33.7 315 1 0.123 27100 34400 42800
6 33.7 300 2 0.133 18200 27000 35100
10 50.4 300 1 0.192 22000 32700 40500
PEN 100 31300 42900 56100
90000 1
30000 .. M2
y =13.522x2 -1600.7x + 81619
‘ 70000 t .. R2 =0.9515 MW
:5. 60000 . i y = 9.6902x2 -1182.7x + 64653 4‘
2 =
E 50000 ,_ . R 0.9701
g "\g - - '
E 40000 -~ g 4
o .- 7‘ I I
a 30000 \X/
20000 —» Mn ‘
y = 6.6629x2 - 817.81x + 46435
10000 " R2 = 0.9999
0 1 1 ‘r a
0 20 40 60 80 100
PEN mole "lo
1 0 Mn - MW 3 Mz y

 

transesterification reactions. The transesterification reactions between the hydroxyl or
carboxyl groups of PET and PEthauses a decrease in the lengths of the polymer chains.

When comparing the average molecular weight values of blends processed at
300°C, the results showed that blends processed through the extruder twice had lower
average molecular weight values than the blends processed through the extruder only
once.

The plot between the mole % PEN and the'respective molecular weight of the
blends is shown in Figure 4.15. Polynomial expressions were found to give the best fit.
This shows a similar behavior as the effect of PEN composition on Tm, % crystalinity,
and the density. The minimum molecular weight values of the blends were found at a
blend composition of 45-50 mole % PEN. This confirmed that adding one polymer to.
another polymer rich phase would disrupt the blend structure and cause a decrease in
molecular weight. The effect of the blend composition on the molecular weight will be
crucial only when the transesterification reactions exist.

The polydispersity, or the ratio of M“, and Mn, which is a measure of the breadth
of the molecular weight distribution, is reported in Tables 4.16 and 4.17, and presented
graphically in Figures 4.16 and 4.17 . The polydispersity of the PET and PEN films were
1.41 and 1.37, respectively. The blend films processed at 300 and 315°C and 1 pass had a
lower polydispersity than those of PET and PEN homopolymers. However, the
polydispersity of blend films processed at 300 OC and 2 passes was higher than those of
the PET and PEN. This might be due to the degradation of the blends during the

processing of the second pass through the extruder.

107

 

Table 4.16: The effect of the degree of transesterification 0n
the polydispersity (MW/Mn) of the blend films

 

 

Blending Blend time
sample ID Temp. (0 C) no. of pass B Man
PET 1.41
3 300 1 0.074 1.22
4 315 1 0.123 1.27
6 300 2 0.133 1.48
PEN 1.37

 

 

 

 

 

 

* Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

 

r"
O)
O

 

 

.3.;
N3
CO

9

.0
o:
o

The polydispersity ("w/Mn)
p o
«5 on
O 0

.°
N
o

 

 

 

 

 

 

 

8

 

 

0.074 0.123 0.133
The degree of transesterification (8)

Figure 4.16: The effect of the degree of transesterification 0n the
polydispersity of the blend films

108

Table 4.17: The effeet of PEN composition on the polydispersity (Winn)

 

 

 

of the blend films
PEN
sample 10 mole% Mwan Mn Mw
PET 0 1.41 46400 65400
3 33.7 1.22 26600 32400
10 50.4 1.49 22000 32700
PEN 100 1.37 31300 42900

 

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

.4.-tee
810:503
C00

The Polydispersity (WM)
9

9.0.0.0
8888

 

 

 

 

 

 

 

 

 

 

 

0 33.7 50.4 100

PEN mole %

Figure 4.17: The effect of PEN composition on the
polydispersity (MW/Mn) of the blend films

109

Comparing the polydispersity of the blends with the blend composition of 33 and 50 mole
% PEN, the results showed no relationship between those values. Therefore, the
polydispersity of the blends appears to be independent of the blend composition.

However, replicate experiments are required to confirm this conclusion.

4.3.5. Morphology

The dispersive spectra of PET, PEN and selected blends are shown in Figures
4.1.8 and 4.19. The brightfield reflectance, polarized light and Raman chemical images
of each sample with 20X and 100X magnification were obtained and are shown in
Figures 4.20 to 4.24, respectively.

Dispersive Raman spectroscopy provided unique information for differentiation
between PET and PEN homopolymers. This included a shift of the aromatic C=C feature
from 1608 cm'1 (PET) to 1629 em"(PEN). Raman spectroscopy of the blend samples
showed the combined spectral features from both PET and PEN. Moreover, they
revealed a variation in the relative spectral peak height in localized PET and PEN
domains. The samples were tested as fihn and resin form. Testing on film and resin
samples seems to give similar results.

For the sample 300P1-sing1e (blend processed through single screw extruder at
300°C and 1 pass), the brightfield reflectance, polarized light and Raman chemical image
revealed an agreeable result and showed evidence of sample heterogeneity (Figure 4.20).
The images from both low (20X) and high (100X) magnification yielded molecular
contrast between PET and PEN components. From Figure 4.20, the blue region,

corresponding to the Raman shifi of 1608 em'l, was indicative of the PET domain, and

110

the green region, corresponding to the Raman shift of 1629 cm", was indicative of the
PEN domain. These images suggested that PEN is a minor or dispersive component with
a submicron average domain size within the PET continuous phase. This result is
consistent with previous results, in that the blend processed through the single screw
extruder was immiscible and exhibited phase separation. Thermal analysis of this sample
indicated the phase separation of the blends by showing 2 T g and Tm values. This also
confirmed that PEN has a higher viscosity than PET (McGee 1995), therefore PEN is
likely to be the dispersive domain within the PET continuous domain.

For other blend samples processed through the twin-screw extruder, 300P1, 315P1
and 300P1-40% (Figures 4.21 to 4.24), the brightfield reflectance, polarized light and
Raman chemical images showed evidence of sample homogeneity, at the spatial resolving
power of this microscope technique (approximately 200 nm). The images from both low
and high magnification indicated the same results. The results are in the agreement with
the thermal analysis indicating that those blends were miscible and they exhibited only
one T8 and Tm. From the NMR analysis, the transesterification degree of samples 300P1,
315P1, and 300P1-40% varied fiom 0.161 to 0.520.

In conclusion, the morphological analysis revealed that the transesterification
reactions yielded miscibility of the blends, and caused a reduction in size of the PET and
PEN chains. At the spatial resolving power of this microscope technique (approximately
200 nm), an occurrence of total miscibility or a single phase cannot be concluded. This is
because Utracki (1990) reported that phase separation could still be found when the
domain sizes of blend components are smaller than 10 A°. Moreover, blends with a

single T8 do not always imply that blends are totally miscible (U tracki 1990). Single Tm

111

peaks from the DSC analysis indicated that miscibility is obtained. However, the peaks
were broad, which suggest that blends were only partially miscible, not totally miscible.
In this experiment, the size of the PEN domain of the blends processed through
the twin-screw extruder cannot be determined. In order to improve the resolution of the
Raman Chemical Imaging microscope, a microtomed sample is required. The
microtomed sample can minimize the contribution from out-of-plane components.
Another recommendation is to use the Transmission Electron Microscope (TEM), which

can give a higher spatial resolving power.

112

Offset Intensity

300P1 25%PEN
MM

300P1 25%PEN
Bead
inf __ g . N
M

 

 

 

 

 

1000 1500 2000 2500 3000

Raman Shift (cm")

Figure 4.18: Dispersive Raman Spectroscopy of 300P1 25% (wt/wt)

PEN polymer samples.

113

Offset Intensity

 

315P2 (25% PEN)
- - _ AW
300P1-Single (25% PEN)
MM . A —M—

 

 

 

300P1-40% PEN
1“?” m... , 5 -‘W
1100 1600 2100 2600 3100
Raman Shift (cm-1)

 

Figure 4.19: Dispersive Raman Spectroscopy PET/PEN blend resins with.25% and

40% (wt/wt) PEN processed at 300 °C and 1 pass, and 315°C
and 2 passes.

114

Polarized Light Image

Raman Chemical

    

Brightfield Image

      

LCTF Raman Spectra

PER Blue = 1608 cm‘1
\A

Green = 1629 cm“

PET

'D
0
.t‘
E
22
.9?
39
>2

 

I l l l l I

15% 1590 1a!) 162) 1640 1%0

Raman Shllt (cm")

Figure 4.20: Raman Chemical Imaging with 20X magnification of 25% (wt/wt) PEN
blend resins processed through single-screw extruder at 300°C and 1 pass.

115

Brightfield Image Polarized Light Image Raman Chemical Image

  
   

      

LCTF Raman Spectra

 

3 Blue = 1608 cm"
g Green = 1629 cm'1
E
25
9 E
>3 e , , , , . fl

1580 1600 1620 1640 1660 1680

Raman Stilt (em'l)

Figure 4.21: Raman Chemical Imaging with 20X magnification of 25% (wt/wt)
PEN blend resins processed through twin-screw extruder at
300°C and 1 pass.

116

300P1single c

 

Figure 4.22: The brightfield (a), polarized light (b), and Raman images (0)
from 20x magnification of the blend resins
Blue region (PET) = 1608 cm", and green region (PEN) = 1629 cm'1

117

300P1$1ng|e c

 

Figure 4.23: The brightfield (a), polarized light(b), and Raman images (0)
from 100x magnification of the blend resins .
Blue region (PET) = 1608 cm], and green region (PEN) = 1629 cm'1

118

 

Figure 4.24: The brightfield (a), polarized light (b), and Raman images (c) from 20X and
100x magnification of the blend films with 25 % (wt/wt) PEN processed
through twin-screw extruder at 300°C and 1 pass.

Blue region (PET) = 1608 cm", and green region (PEN) = 1629 cm‘1

119

Chapter 5

Barrier and Mechanical Properties of PET/PEN Blends

5.1. Introduction

Many foods and beverages were once packed in glass containers or metal cans.
However, many of them are now packed in plastic containers due to their light weight,
convenience of production, use and disposal, low cost of storage and handling,
recyclability, etc. A main conversion has been the replacement of glass by polymers such
as PET. However, a limitation of PET applications is its only fair barrier properties, and
developing techniques for extending applications for PET has been the subject of recent
studies. A modified PET with a higherbarrier performance has been sought for
extending product shelf life, and for packaging systems which require high gas barrier
performance.

PET/PEN blends are an altemative which combines the high barrier and
mechanical performance of PEN with the economic advantage of PET. Previous studies
show that the oxygen barrier of PEN is approximately four times higher than that of PET
(Stewart et a1. 1993) and the tensile strength of PEN is also 35% higher than that of PET
(Shi 1998). However, there are few reports on the barrier and mechanical properties of
PET/PEN blends. Therefore, in this chapter, these important properties of the PET/PEN

will be studied.

120

5.1.1. Barrier Properties

The barrier properties are very important parameters in selecting packaging
materials for consumer products, especially in the food industry. This is due to the fact
that the barrier properties, including water, oxygen, carbon dioxide, and organic vapor
barrier, are primary factors controlling the shelf life and the quality of the packaged
products.

The general theory of gas permeation through a polymer matrix involves two
basic parameters, namely the diffusion and solubility coefficients. The diffusion and
solubility coefficients are often independent of one another (Nemphos et a1. 1986; Salarne
1967). There are four steps in the gas permeation process, which include solubility of
the permeant in the polymer matrix, diffirsion though the polymer matrix, building up of
a concentration gradient across the polymer matrix, and desorption of the permeant in the
packaging system from the inner wall.

The permeability of a polymer in the packaging system is dependent on several
internal and external factors. The internal factors include the chemical structure and the
. functional groups of the polymer, magnitude of the interchain forces (secondary forces),
the packing ability of the chain segments, the degree of crystallization, and orientation
(Brennan et al. 1996; Holsti-Miettinen et al. 1995; Nemphos et al. 1986). The external
factors include the nature of the permeant, and the environmental condition such as
temperature and relative humidity. In this study, the external factors were kept constant,
while the internal factors were varied by adjusting the transesterification reactions and the

blend composition.

121

When blending PET and PEN, the physical, mechanical, and barrier properties of
this blend change, compared to its pure components. As discussed in the previous
chapter, the transesterification reactions cause morphological changes. In particular, the
average sequence lengths of the terephthalate and naphthalate units decrease, and the
number of N-T bonds increases. Consequently, this causes changes in the volume
fraction of both the matrix phase (PET) and the dispersed phase (PEN). From the
volume fractions of all components, the permeability coefficient of the polymer blend can
be predicted (Barrer et al. 1983; Fricke 1924; Holsti-Miettinen et al. 1995; Kit at e1. 1995;
Michaels and Bixler 1961; Paul and Buclcnall 1999). Paul and Bucknall (1999)
mentioned that the presence of a impermeable dispersed phase in the continuous matrix
domain increases the tortuosity of the path a permeant molecule must travel through the
polymer matrix. Maxwell proposed an expression for the tortuosity factor (7), which is
the effective path length divided by actual thickness of the film. The expression is based
on the volume fraction (¢d) of spherical, impermeable or nonconducting particles. The

expression is:
¢d

TEl+-2— (1)

The permeability coefficient of a blend can be calculated as follows (Barrer et a1.

1983; Micheals and Bixler 1961):

(2)

where P; and P”. are the permeability coefficients of the blend and matrix polymer,

respectively, and ¢,,. is the volume fraction of the matrix polymer.

122

Another parameter controlling the barrier properties of the blends is the blend
composition. As the PEN composition increases, the barrier properties have been shown
to improve. This behavior can be explained as follows. PET and PEN have similar
chemical structures. However, PEN has naphthalate rings or double-aromatic rings,
while PET has single-aromatic rings. The barrier properties of PEN are better than those
of PET since the aromatic rings of naphthalates are stiffer than the single aromatic rings
of terephthalates. The high stiffness of the PEN chains results in less segmental
mobility of the polymer chains. Therefore, the free volume in the polymer matrix is
fixed, which is difficult for a permeant to travel across the polymer matrix.

Therefore, by increasing the PEN composition in the blend, the number of
naphthalate units in the main chains increased. This results in improvement of the gas
barrier properties of the blends (McGee and Jones 1995; Stewart et a1. 1993). However,
as discussed in Chapter 4, as a small amount of PEN is added to the PET rich phase, the
PEN will interfere with the crystalline formation of the PET. This will reduce the degree
of crystallinity of the blends and might decrease the barrier properties of the blends.

In this chapter, the effect of the transesterification reactions and PEN composition .

on the barrier properties will be discussed.

5.1.2. Mechanical Properties
The study of the mechanical properties of polymers is an irnportant area in
applied polymer science. By understanding the mechanical behavior and the factors

controlling it, polymeric materials can be properly and efficiently utilized.

123

5.1.2.1. Factors controlling mechanical properties

The mechanical properties of the polymer blends are dependent on internal and
external factors. The internal factors refer to the morphological structure or the chemical
nature of the polymer blends. The primary internal factors are the following (Paul and
Bucknall 1999; Rudin 1982):
(a) for matrix materials

(i) molecular weight average

(ii) branching, cross-linking, entanglement density

(iii) degree of crystallinity

(iv) plasticizers, fillers, and additives

(v) orientation

(vi) other consequences of the processing history, and the thermal history of the
polymer
(b) for dispersed-phase materials

(i) composition

(ii) particle size

For example, as the molecular weight, entanglement density, and % crystallinity
of the polymer increase, the tensile strength and the modulus of elasticity of the polymers
increase, while the flexibility or % elongation decrease.

The thermal properties are also important factors in determining the brittleness
and the ductility of a polymer. T8 relates to the segmental mobility of the molecular

chains, and it also determines whether chain-straightening and/or molecular slippage will

124

occur. The chain-straightening refers to the elongation of the molecular chain from the
equilibrium position to a new dimension under stress, while molecular slippage refers to
the molecule movement past adjacent molecules due to the application of stress
(Schwartz 1982). As polymers are placed under stress, they may exhibit both
phenomena, depending on their morphological structure, such as branching, cross-
linking, degree of crystallinity, etc. At a temperature below T8, those phenomena do not
occur appreciably. Therefore, the polymers are hard and brittle. At a temperature above
T3, polymers exhibit the chain-straightening and molecular slippage. Therefore, the
polymers are soft and flexible (Rudin 1982).

The external factors which affect the mechanical properties of polymers include
test temperature, annealing, and hydrostatic pressure (Paul and Bucknall 1999; Rudin
1982). As ambient temperature increases, the stress at a given strain always decreases,
while the elongation at break increases. As the hydrostatic pressure increases, Young’s
modulus, and the yield stress of semicrystalline materials increase (Schultz 1974).
Furthermore, the annealing process enhances the strength of semicrystalline polymers
(Rudin 1982; Schultz 1974).

In this study, all external factors were fixed. The study focused on the effect of
the transesterification reactions and the blend composition on the mechanical properties.
As discussed above, the morphology of PET/PEN blends depend on the
transesterification reactions and the blend composition, and the mechanical properties are

dependent on the morphology.

125

5.1.2.2. Tensile properties

The tensile test is the most important technique for measuring the mechanical
properties of polymers. It determines the force necessary to pull the specimen apart, and
the stretch of the deformed plastics before breaking (Schwartz 1982). The general terms
used in the tensile test are stress and strain. The stress is the amount of load applied to a
unit area, and the strain is the ratio of the extension to original length. The proportional
limit is the greatest stress that a material can withstand, without deviating from
proportionality of stress to stain (Hooke’s law), and the elastic limit is the greatest stress
the polymer membrane can be placed under, without exhibiting permanent deformation
(ASTM D638-9land D882-90).

A typical stress-strain curve for a polymeric material is shown in Figure 5a. This
plot introduces 5 descriptive parameters:

1) tensile strength: the maximum stress that the material can withstand before
breaking

2) yield strength: the stress beyond which the material exhibits a deviation from

the proportionality of stress to strain, and becomes nonelastic

3) elongation: the extension under stress

4) modulus of elasticity: the ratio of the stress and strain in the interval of linear

proportionality.

5) toughness: the total energy absorbed by the material under stress before

breaking.

126

Stress Modulus of Elasticity (slope of the linear portion)

Tensile Strength

............
...............
..............
.................
.................
....................
....................
.............................
...............................
.....................................
........................................

..........................................

V

...........................................

..........................................
...........................................
..........................................
............................................
..........................................
............................................
...........................................
............................................
...........................................
..............................................
...........................................

..............................................

...............................................
...............................................

................................................
..............................................

................................................
...............................................

................................................
...............................................

 

 

Elongation at Break

Figure 5a. A typical stress-strain curve for elastic materials

127

In this chapter, the effect of the transesterification reactions and blend composition on the

tensile strength, % elongation, and the modulus of elasticity will be described.

5.2. Materials and Methods

5.2.1. Measurement of Water Vapor Transmission Rate

The Permatran-W 3/31 developed by the MoCon Co. (7500 Boone Avenue North,
Minneapolis, Minnesota 55428) was employed to measure the water vapor transmission
rates. The isostatic testing procedure was performed following the ASTM F -1249-90
standard test method (Anonymous 1990 m) for water vapor transmission rate through
plastic film and sheeting using a modulated infrared sensor. The water vapor
transmission rate tests were performed at 378°C (100°F) and 90% relative humidity
(RH).

The Permatran-W3/31 consists of two diffusion cells (Cell A and Cell B). Two
film samples can be tested at the same time. Each diffusion cell is divided into two
chambers. The front chamber behaves as the water vapor environment, and the rear
chamber simulates the dry environment (0% RH). The water vapor is generated by
introducing the carrier gas (nitrogen) through a water reservoir located next to the
diffusion cells. The water vapor is then delivered directly to the front chamber producing
a concentration gradient. The level of the relative humidity is controlled by adjusting the
pressure of the carrier gas. Dry carrier gas (0 % RH) is circulated through the rear
chambers. Film samples are inserted between the front and rear chamber of the diffusion
cell. The water vapor traveling from the front chamber is absorbed, diffused through the

film wall, and is then desorbed to the rear chamber.

128

The following test parameters were used;

The exposed area 50 cm2
The examination time 10 minutes
The number of rezero cycles 4 cycles

The partial pressure of the water vapor in the front chamber was 44.24 mmHg at 378°C
and 90% RH. The rear chamber was dry; thus it was assumed to have partial pressure
for water vapor of 0 mmHg. The conditioning period for the non-oriented films (10 mil
films) and oriented films (1 mil films) were 5 hours and 1 hour, respectively.

During the examination period of 10 minutes, the rear chamber of a diffiision cell
is connected to a fixed wavelength infrared sensor, which is interfaced to a computer
system. The computer software receives the signal from the infrared sensor, and converts
the signal output to the transmission rate value. For each test cycle, the sample in a
diffusion cell istested for 10 minutes and the Permatran system will then automatically
switch to test the sample in another diffusion cell. The rezero stage is the testing with dry
nitrogen or cganier gas for 10 minutes every 4‘h cycle, in order to confirm that no water
vapor is contaminating the carrier gas system. The film samples were tested until
equilibrium was reached. The water vapor transmission rate values were the average of
duplicate samples. The water vapor permeability coefficients were calculated using the

following equation:

WVTR*l
P40 =7

where Pup , WVT R, I, and Ap are the permeability coefficient (g mth m2 mmHg), the

(3)

water vapor transmission rate (g/h m2), fihn thickness (mil), and the differential pressure

129

or the driving force (mmHg). The water vapor permeability coefficients were later

converted to SI units (kg m/s m2 Pa).

5.2.2. Measurement of Oxygen Transmission Rate

The oxygen transmission rate was determined with an Oxtran 200 permeability
tester (MoCon Co., 7500 Boone Avenue North, Minneapolis, Minnesota 55428). The
isostatic testing procedure was performed following the ASTM D3985-81 stande test
method (Anonymous 1981) for oxygen gas transmission rate through plastic film and
sheeting using a coulometric sensor. The oxygen transmission rate tests were performed
at 25°C (70°F) and dry conditions (0 % RH).

The Oxtran 200 permeability tester is similar to the Permatran-w3/31, in that the
Oxtran 200 system consists of two diffusion cells and two samples can be tested at the
same time. Instead of the infrared sensor, a coulometric sensor is connected to the
Oxtran 200. In this experiment, compressed air with 21% oxygen was used as the test
gas or source of the oxygen, and dry nitrogen was used as the carrier gas. The film
samples were placed between the upper and lower chamber of the diffusion cell. The test
gas was introduced through the upper chamber, while the carrier gas was purged through
the lower chamber. A computer system is connected to the Oxtran 200, therefore one can
control the system via the computer. The computer converts the signal strength from the

sensor to the oxygen transmission rate.

130

The following test parameters were used;

The exposed area 50 cm2
The examination time 20 minutes
The number of rezero cycles 4 cycles

The driving force was 0.21 atmosphere of oxygen between the upper and the
lower chambers. The conditioning period for the non-oriented films (10 mil films) and
oriented films (1 mil fihns) were 15 hours and 1 hour, respectively. The film samples
were tested until equilibrium was reached. The oxygen transmission rate values were the
average of duplicate samples. The oxygen permeability coefficients were calculated

using the following equation:
(4)

where P02, 027‘ R, l, and Ap are the oxygen permeability coefficient (cc mil/h m2 atrn),

the oxygen transmission rate (cc/h m2), film thickness (mil),land the differential pressure
or driving force (atrn). The oxygen permeability coefficients were later converted to SI

units (m3 m/ s m2 Pa and kg m/s m2 Pa).

5.2.3. Measurement of Carbon Dioxide Transmission Rate
The carbon dioxide transmission rate was measured using a Permatran CIV
permeability tester (MoCon Co., 7500 Boone Avenue north, Minneapolis, Minnesota

55428). The dynamic accumulation technique was used to perform the carbon dioxide

131

transmission rate test. (See the manufacturers manual for the Permatran CIV). The tests
were performed at 25°C and dry conditions (0% RH).

The Permatran CIV permeability tester consists of five diffusion cells and an
infrared sensor. Each diffusion cell is divided into upper and lower chambers. The upper
chambers are purged with 100% carbon dioxide (test gas). The lower chambers are
purged with nitrogen gas (carrier gas). The lower chambers are connected to the infrared
sensor with sensitivity at a particular wavelength that can detect carbon dioxide (C02).

The first diffusion cell is used for calibration. The other four diffusion cells are
test chambers. A film sample is inserted between the upper and lower chambers of each
test cell. The sample is conditioned for a sufficient amount of time to establish a steady
state of carbon dioxide transmission rate. The oriented l-mil films were conditioned at
25°C for 24 hours. The thicker film samples were conditioned at the accelerated
condition (45°C) for a longer times since these samples required a longer time to reach
steady state. The non-oriented 10-mil films were conditioned for 72 hours at 45°C. In
addition, the non-oriented 10—mi1 films were kept at the test temperature (25°C) for 24
hours after the conditioning period and before the testing to allow the sample to acclimate
to the test environment.

When the test begins, the lower chamber of a particular diffusion cell and the
sensor are connected and form a recirculating loop. The gradual accumulation of the
amount of carbon dioxide permeated through the film samples is monitored by the sensor.
The chart recorder displays the resultant signal. As the concentration of C02 increases

with time, the corresponding output voltage increases. The carbon dioxide transmission

132

rate can be calculated from the rate of change of the CO2 concentration in the
recirculating loop.

Each cell was allowed to accumulate C02 in this manner for 15 minutes, before
the loop was opened. The sensor was then purged with dry carrier gas for 3 minutes to
remove residual C02 from the sensor. The instrument was then prepared for the next test
cycle. It should be noted that while a chamber is being tested, the other three chambers
can be conditioned at the same time.

The calibration cell (the first diffusion cell) is loaded with an aluminum sheet or a
very low-permeated film. The very low-permeated fihn will prevent the transfer of C02
from the upper chamber to the lower chamber. During the calibration step, the lower
chamber of the calibration cell and the sensor are connected and form a recirculating '
loop. A known amount of C02 is then transferred from a reservoir directly into the loop.
The output from the chart recorder corresponds to the known amount of C02, and is used
for calibration (the amount of C02 (cc) per response height (cm.) of the chart recorder).

The following test parameters were used;

The exposed area 50 cm2
The accumulation time 15 minutes
The purging time 3 minutes

The driving force was 1 atmosphere of C02 at the upper chamber. The samples
were tested every 6 or 12 hours until there was no further increase in the transmission
rate, which means that the samples had reached steady state. The carbon dioxide

transmission rate values were calculated using the following equations (see Figure 5b):

133

 

 

pu--¢----

 

     
 

Vd

 

 

 

 

steward (0.0213 cc.)
calibration
I L RP
V J \W—j
accumulation period purging period
15 minutes 2 minutes

 

 

Figure 5b. Example of Permatran CIV graphical data '

134

,5
Rv (5)

V
COZTR = t—j (6)

where t stands for the time for 0.0213 cc of carbon dioxide to have permeated. RP and R.
stand for the recorder paper advance (cm), and the chart recorder speed (20 cm/h),
respectively. CO2T R is the carbon dioxide transmission rate (cc/m2 h). V4 and A are the
standard amount of C02 (0.0213 cc) and the test areas (50 cmz), respectively.

The carbon dioxide permeability coefficient was calculated using the following equation:

C0 TR *1
Pco. = 4A7— (7)

where PCO2 , C02T R, I, and Ap are the carbon dioxide permeability coefficient (cc mil/m2

h atrn), the carbon dioxide transmission rate (cc/m2 h), film thickness (mil), and the
differential pressure or driving force (1 atrn). The carbon dioxide permeability

coefficients were later converted to SI units (m3 m/ s m2 Pa and kg m/s m2 Pa).

5.2.4. Tensile Test

Tensile tests were performed using an Instron Universal Tester (Instron Inc.,
Canton, Massachusetts) following the ASTM D882-90 and ASTM D638-91 standard
method (Anonymous 1990‘“); Anonymous 1991). The tensile strength, Young’s modulus,
and percent elongation at break were determined. The tests were performed at room
temperature (approximately 25 °C), and 50% RH.

The tensile test is a technique which involves placing a sample under tensile

stress, and monitoring the resistance to the deformation of that sample. This test method

135

is used to measure mechanical properties such as the strength and flexibility of the

samples.

5.2.4.1. Test samples

All 11 samples of the non-oriented 10-mil films and the oriented l-mil films were
tested. Each oriented sample was tested with at least 8 replications. For the non-
oriented films, each sample was tested along the machine direction (MD, the direction in
which film samples came out from the flat die) and cross machine direction (CMD, the
direction perpendicular to the machine direction) with at least 8 replications. All
samples were cut to 1X5 inches by a JDC Precision Sample cutter (Thwing-Albert
Instrument Company, Philadelphia). The thickness of each sample was measured by a
model 549M micrometer (Testing Machines Inc., Amityville, New York). The results
showed that thickness of the non-oriented and oriented film samples were 0.01 i 15%
and 0.001 1 15% inches, respectively. Therefore, cross sectioned areas of 0.01 and 0.001

in2 were used for non-oriented and oriented samples, respectively.

5.2.4.2. Instron machine

The Instron Universal tensile tester is a testing machine which employs a constant
rate of crosshead movement. It conSists of a fixed member carrying a grip to hold one
side of the sample, and a movable member carrying a second grip to hold the opposite
side of the same sample. Samples are held vertically by those grips. A gap between the
two grips is very important and is always set to a fixed length. The sample should be

aligned perfectly along the pulling direction, and not slip fiom the grips. When the load

136

is applied, the film sample is pulled in the vertical direction, and is stretched under the
applied stress. It will deform and eventually break. The following parameters were

used for all tests;

Load capacity 1000 pounds
cross-head speed 20 inches per minute
gap length 2 inches.

The load and extension profile is recorded by a chart recorder connected to the
Instron machine. A typical stress-strain chart is showed in Figure 5a. All the important
parameters, including the tensile strength, % elongation, and Young’s modulus of
elasticity, can be calculated as follows:

tensile strength = maximum load / cross sectioned area (lb/m2)

% elongation = the extension/ original gap length *100

modulus = the slope of the linear portion of the stress-strain chart (lb/m2)
Tensile strength and modulus were reported in SI units (MPa), using the following unit
conversion:

PSI (lb/in2) * 6.8948/1000 = MPa

137

5.3. Results and discussion

5.3.1. Barrier Properties

The respective permeants, including water vapor, oxygen and carbon dioxide,
showed the same trends. The barrier properties were found to be dependent on the PEN
composition. In contrast, the degree of transesterification had little or no effect on these
properties.

The permeability coefficient values were feund to decrease as the PEN
composition increased. The relationship between the permeability coefficient values
and the PEN composition of the blends appeared to be linear. However, there was no
significant difference between the permeability coefficients of the blend fihns with
various degrees of tranSesterification.

The results are summarized in Tables 5.1-5.6 and presented graphically in Figures
5.1-5.6, respectively. A detailed discussion of the water vapor, oxygen and carbon
dioxide permeability coefficient values of the PET/PEN blend fihns is included in the

following sections.

5.3.1.1. Water vapor permeability coefficient
Degree of transesterification and water vapor permeability coefficient

The water vapor permeability coefficient (Pup )values of the blend films with

various degrees of transesterification and 25% (wt /wt) PEN are shown in Table 5.1 and

Figure 5.1. The Pup of the non-oriented and the oriented blend fihns with 25% (wt/wt)

PEN ranged 66m 2.29 x 10'l5 to 2.42 x 10'”, and 1.38 x 101510 1.64 x 10'15 kg m/s

m2 Pa, respectively. The degree of transesterification and the processing condition had

138

 

Table 5.1: The effect of the degree of transesterification on the water
vapor permeability coefficient of PET/PEN blend films at 37.8°C.

 

 

 

average PHZO x 10''5 (Kg m/ m2 5 Pa)
Sample ID PEN wt % B non-oriented films oriented films
2 25 0.076 2.42 1.43
3 25 0.222 2.40 1.52
4 25 0.356 2.42 1.38
5 25 0.192 2.41 1.49
6 25 0.405 2.29 1.43
7 25 0.525 2.35 1.64

 

 

 

 

 

 

"' Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

3.00E-15

 

2505-15 ._
2005-15 «~
1505-15» a g: a 1 r:
1.005-15~~

5005-16 4:

The water vapor permeability
(kg mlsee-mz-Pa)

 

0.00E+00 r ‘ %

 

 

0 0.1 0.2 0.3 0.4 0.5
The degree of transesterlflcatlon (B)

 

1

I D non-oriented films 0 oriented fihns j

 

Figure 5.1: The effect of the degree of transesterification on the water
vapor permeability coefficient of PET/PEN blend films at 37.8°C

139

little or no effect on the Pup of the blend fihns. There was no statistically significant
difference between the Pup values of the blend films with different degrees of
transesterification. The details of the statistical analysis of the Pup values as performed

are shown in Tables D1 and D2, in Appendix D.

PEN composition and water vapor permeability coefficient

The Pup values of the PET and PEN non-oriented fihns were 2.96 x 10'15 and
1.15 x 10'15 kg m/s m2 Pa, respectively (Table 5.2). The Pap value of non-oriented PET

film was approximately 2.6 times as high as that of the non-oriented PEN fihn. The

PHI, values of the oriented PET and PEN fihns were 1.65 x 10.15 and 0.494 x 10'15 kg
m/s m2 Pa, respectively (Table 5.2). The P” 10 value of the oriented PET film was

approximately 3.3 times as high as that'of the oriented PEN fihn. The lower Pup value

of the PEN film is due to the naphthalate rings in the main chains of the PEN. The
naphthalate rings cause a high stiffness of the molecular chains, and this results in less
segmental mobility and higher barrier properties (Brennan et al., 1996; Holsti-Miettinen
et al. 1995; Nemphos et al. 1986).

The P” 10 values of the blend fihns processed at 300°C and 1 pass through the

extruder, with a PEN concentration ranging from 10 to 40% (wt/wt) PEN, were in the
range of 2.76 to 2.12 x 10‘15 kg m/s m2 Pa for non-oriented films, and 1.70 to 1.28kg m/s

m2 Pa for oriented films, respectively. There is a statistical difference between

140

Table 5.2: The effect of PEN composition on the water vapor permeability
coefficient of the PET/PEN blend films at 37.8°C

 

 

 

DI Pnzo x 10"5 (Kg m/ m2 5 Pa)
Sample 1 PEN mole% non-oriented films oriented films
1 0.0 2.96 1.65
8 14.5 2.76 1.70
3 33.7 2.40 1.52
9 39.5 2.37 1.26
10 50.4 2.12 1.28
l 1 100.0 1.15 0.494

 

 

 

 

 

 

'1' Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

 
     

 

 

 

3.505-15
3.005-15
g y = -2E-17x + 35-15
D 2 _
g a. 2.505-15 «I R ‘ 0'99“
E 9.-
8} 2.005-15 1»
8 3 «\-
g 1' 1505-15 -~ {\0\
b E- n
3 3" 100515 \ P
a 4.- - 4- xxx,
; .
g y.= -1E-17x + 2E-15 \\
.— 5.00E‘16 “” R2 = 0.9358 9
0.005+00 . r .
0 20 40 60 80 100
PEN mole %

 

 

F a non-oriented films 0 oriented films

 

Figure 5.2: The effect of PEN composition on the water vapor
permeability coefficient of PET/PEN blend films at 37.8°C

141

the P1120 values of blend films with various PEN compositions at a confidence level (p) of

0.05 (see Table D3 in Appendix D).
The results showed that the water vapor barrier properties of the PET/PEN blend
films improved as the PEN composition increased. The relationship between the PEN

composition and the Pflzo values for the non-oriented and oriented films was found to be

linear, as shown graphically in Figure 5.2. This behavior can be attributed to the fact
that as the‘PEN composition increases, the numbers of the naphthalate units in the main
chains of the blend increases, resulting in an increase in chain stiffness and a decrease in

the PH20 values.

As shown in Table 5.2, the Pflzo values of the oriented films were 50% of those

of the non-oriented films. The orientation effect is discussed in Chapter 6.

5.3.1.2. Oxygen permeability coefficient
Degree of transesterification and oxygen permeability coefficient

The oxygen permeability coefficient (P02 ) values of the blend films with various

degrees of transesterification and 25% (wt/wt) PEN are shown in Table 5.3 and presented

graphically in Figure 5.3. The P02 of the non-oriented and oriented films with 25%
(wt/wt) PEN ranged from 4.99 x 10‘19 to 5.31 x 10'”, and 2.82 x 10'19 to 3.90 x 10'19 kg
m/s m2 Pa, respectively. The P02 of

these non-oriented blend films and oriented blend films were approximately 20% and
35% lower than those of non-oriented and oriented PET films, respectively. There was

no significant difference between the P02 of the blend films with different degree of

142

 

Table 5.3: The effect of the degree of transesterification on the oxygen permeability
coefficient of PET/PEN blend films at 25° C

 

 

 

average Po2 x 10"9 (kg 111/1112 5 Pa) P02 x 10''9 (m3 m/mz s Pa)
Sample ID B non-oriented films oriented films non-oriented films oriented films
2 0.076 5.22 3.32 3.66 2.33
3 0.222 5.31 2.82 3.72 1.98
4 0.356 5.22 3.55 3.66 2.49
5 0.192 5.11 3.45 3.58 2.42
6 0.405 4.99 3.51 3.50 2.46
7 0.525 5.07 3.90 3.55 2.73

 

 

 

 

 

 

 

" Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

 

 

 

 

6.E-19
5‘ 5.5-19- a: §§ i i ¥
g ..
8 .. £ 4.5-19 -- ,
c"
3.5 5 1 ' ' i
E ., 35-19 0
5 o 9
a 3 E
i‘ 5': 25-19 L
0 e.
O
S
I- 1.5-19-
0.5+00 1 1 1 1 5
0 0.1 0.2 0.3 0.4 . 0.5

The degree of transesterification (B)

 

o non-oriented films

0 oriented films j

 

Figure 5.3 : The effect of the degree of transesterification on the
oxygen permeability coefficient of the PET/PEN blend films at 25°C

I43

0.6

transesterification. Statistical analysis of the P02 values was performed and the results are

summarized in Tables D4 and D5, in Appendix D. Based on the results of statistical

analysis, it was concluded that degree of transesterification, or the processing conditions,

had little or no effect on the P02 of the blend films.

PEN composition and oxygen permeability coefficient

The P01 values of the non-oriented PET and PEN films were 6.53 x 10'19 and 3.00
x 10‘19 kg m/s m2 Pa, respectively (Table 5.4). The P02 values of non-oriented PET films

were approximately 2.2 times as high as those of the non-oriented PEN films. The Po2 of

the oriented PET and PEN films were 4.23 x 10'19 and 8.66 x 10’" kg m/s rn2 Pa,

respectively (Table 5.4). The P02 value of oriented PET film was approximately 5 times

as high as that of the oriented PEN film.

The P02 values of the blend films processed at 300 °C and 1 pass through the

extruder, with PEN composition ranging from 10 to 40% (wt/wt) PEN, were in the range
of 5.89 x 10'19 to 4.69 x 10"9kg m/s m2 Pa for non-oriented films, and 3.76 x 10"9to
2.96 x 10'19kg m/s in2 Pa for oriented films, respectively. Statistical analysis was

performed and the results showed a significant difference between the P02 values of the
blend films with various PEN compositions (see Table D6 in Appendix D). The P02
values of the blends fall between the P02 values of PET and PEN films. A linear

relationship between the PEN composition and the P02 values of the non-oriented and

oriented films was found, as shown in Figure 5.4. The results showed that as the PEN

144

Table 5.4: The effect of PEN composition on the oxygen permeability
coefficient of PET/PEN blend films at 25°C

 

 

 

 

Po2 x 10"9 (kg rn/rn2 5 Pa) Po2 x 10"9 (rrr3 m/rn2 s Pa)
Sample ID PEN mole% non-oriented films oriented films non-oriented films oriented films
1 0.0 6.53 4.23 4.58 2.96
8 14.5 5.89 3.76 4.13 2.64
3 33.7 5.31 2.82 3.72 1.98
9 39.5 5.06 3.05 3.55 2.14
10 50.4 4.69 2.96 2.29 2.07
11 100.0 3.00 0.87 2.10 0.61

 

 

 

 

 

 

 

"' Blends processed at 300°C and 1 pass through a twin-screw extruder.

8.E-19

 

7.5-19 1-
y = -3E-21x + 6E-19
R2 = 0.9985

  

6.E-

A

CD
1
V

  

 

 

 

 

a
C
2
5
§ ..
.e a:
E “5 5.5-19 1-
“ I
E 3 3

0 4.5-19
a E.

D
5 5 3.5-19 «-
2
’0‘ 2.5-19
.2 y = -3E-21x + 45-19
'- 1'5'19 “ R2 = 0.9618 7*

OE+00 -‘ ¢ 1 : ; Y 1 t i
0 10 20 30 40 50 60 70 80 90 100
PEN mole %
D non-oriented films 0 oriented fihns

 

 

 

Figure 5.4: The effect of PEN composition on the oxygen
permeability coefficient of PET/PEN blend films at 25° C

145

composition increased, the oxygen permeability coefficient of the PET/PEN blend films
decreased.

Furthermore, the P02 values of the oriented films were 50% of those of the non-

oriented films. The orientation effect is discussed in Chapter 6.

5.3.1.3. Carbon dioxide permeability coefficient
Degree of transesterification and carbon dioxide permeability coefficient

The carbon dioxide permeability coefficient (PCO2 ) values of the blend films with
various degrees of transesterification and 25% (wt/wt) PEN are shown in Table 5.5 and

presented graphically in Figure 5.5. The Pa,2 values of the non-oriented and oriented
films with 25% (wt/wt) PEN ranged from 3.71 x 10'18 to 4.03 x 10'”, and 2.17 x 10'18 to
2.70 x 10'18 kg m/s m2 Pa, respectively. The Pa,2 of these non-oriented blend films and
oriented blend fihns were approximately 27% and 30% of those of non-oriented and
oriented PET films, respectively. There was no significant difference between the FCC:
values of the blend films obtained from various processing conditions. Statistical

analysis of the Pa,2 values as a function of processing conditions was performed and the

results are summarized in Table D7 and D8, in Appendix D.

Therefore, the degree of transesterification, or the processing conditions, had little

or no effect on the Pa,2 of the blend films.

146

 

Table 5.5: The effect of the degree of transesterification on the

carbon dioxide permeability coefficient of PET/PEN blend films at 25°C

 

 

 

sample In B PC02 x 10'18 (kg ml m2 5 Pa) PC02 x 10'18 (m3 m/ m2 s Pa)
non-oriented films oriented films non-oriented films oriented films

PET 0 5.49 3.08 2.80 1.57
#2 0.076 3.71 2.52 1.89 1.29

#3 0.222 4.03 2.17 2.06 1.1 1

#4 0.356 3.98 2.41 2.03 1.23

#5 0.192 3.84 2.66 1.96 1.35

#6 0.395 3.71 2.70 1.89 1.38

#7 0.53 3.94 2.45 2.01 1.25
#PEN 0 1.42 0.54 0.73 0.28

 

 

 

 

 

 

 

* Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

5.E-18

 

“+—

4.E-18

4.5-18

9’
r.“
—l
a)

1.E-18

5.E-19

Carbon diox1de permeablllty coefficient
(kg.mlm2.sec. Pa)
N N (a)
m m rp
a 66 63
—+——a —k——e —1———1

1——+~1—

0.E+00

9
o

0.1

1
1'

0.2

1

0.3 0.4 0.5
Degree of Transesterification (B)

 

: .
1 D non-onented films

0 oriented films

 

Figure 5.5: The effect of the degree of transesterification on the
C02 permeability coefficient of PET/PEN blend films at 25°C

147

PEN composition and carbon dioxide permeability coefficient

The Pco, values of the non-oriented PET and PEN films were 5.49 x 10'18 and
1.42 x 10'18 kg m/s m2 Pa, respectively (Table 5.6). The FCC: value of non-oriented PET

film was approximately 4 times as high as that of non-oriented PEN film. The Pm:

values of the oriented PET and PEN films were 3.08 x 10'18 and 0.54 x 10'18 kg m/s m2

Pa, respectively (Table 5.6). The P092 value of oriented PET film was approximately

5.7 times as high as that of the oriented PEN film.

The PCO2 values of the blend films processed at 300°C and 1 pass through the

extruder, with the composition ranging from 10 to 40% (wt/wt) of PEN, were in the range
0f4.62 to 3.25 x 10'18 kg m/s rn2 Pa for non-oriented films, and 2.92 to 1.63 x 10"“kg
m/s m2 Pa for oriented films, respectively. Statistical analysis was performed and the

results showed a significant difference between the Pa,2 values of the blend films with
various PEN composition (see Tables D9, appendix D). The Pm2 values of the blends
fall between the Pa,2 values of PET and PEN films, respectively. A linear relationship
between the PEN composition and the Pco, values of the non-oriented and oriented films

was found, as shown in Figure 5.6. The results showed that the barrier properties of the
PET/PEN blend films improved as the PEN composition increased.

The PCO2 values of the oriented films were approximately 50% of those of the

non-oriented films. The orientation effect is discussed in Chapter 6.

148

Table:5.6: The effect of PEN composition on the C02 permeability coefficient
of PET/PEN blend filns at 25°C

 

 

 

sample 11) pm mol% Pcozx 10‘18 (kg m/ mfs Pa) Pcozx 10'1§(m3 ml m2 s Pa)
non-oriented films oriented films non—oriented films oriented films

PET 0.0 5.49 3.08 2.80 1.57

#8 14.5 4.62 2.92 2.35 1.49

#3 33.7 4.03 2.17 2.06 1.11

#9 39.5 3.83 1.95 1.95 0.99

#10 50.4 3.25 1.63 1.66 0.83

#PEN 100.0 1.42 0.54 0.73 0.28

 

 

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

  
    
 
 

 

 

 

6.E-18
E!
g E 5-5‘13 ‘ y = -4E-20x + 5E-18
g "5 R2 = 0.9938
g . 4.E-18 «-
h 0
3 8
g d, 3.E-18
X x
.9 r:
‘O C
c .92 2.E-18 ~~
o .2
1
0 0 1-E-18 -- y = -3E-20x 1 35-18
R2 = 0.9814
0.5100 1 1 1 1 .
0 20 40 60 80 101

PEN mole °/o

 

 

D non-oriented films 0 oriented films

 

 

Figure 5.6: The effect of PEN composition on the
C02 permeability coefficient of PET/PEN blend films at 25 °C

149

5.3.2. Mechanical Properties

The tensile tests of non-oriented and oriented films were performed at room
temperature. The results showed that the non-oriented PEN and PET/PEN films were
very brittle, whereas the non-oriented PET films were very flexible. After biaxial
orientation, PEN and blend films were more flexible, and the %elongation at break of
these samples increased dramatically.

There was no relationship between the degree of transesterification and the tensile
strength, %elongation, and Young’s modulus of elasticity of the non-oriented, and
oriented blend films. However, the blend composition seems to control the tensile
strength and %elongation of the non-oriented and oriented films. The Young’s modulus
of elasticity of the non-oriented films was independent of the blend composition, while
the Young’s modulus of elasticity of the oriented films seems to be slightly influenced by
the blend composition.

A detailed discussion of the effect of the degree of transesterification and blend
composition on the mechanical properties of the non-oriented and oriented films is

included in the following section.

5.3.2.1. Tensile strength

The tensile strengths of the PET, PEN, and PET/PEN blend films are summarized
in Tables 5.7 and 5.8, respectively. The results showed that the tensile strength along the
machine direction (MD) and cross machine direction (CMD) of non-oriented PET films
were 45.6 and 41.8 MPa, respectively (Table 5.8). The tensile strength of the non-

oriented PEN films was 61.5 MPa (Table 5.8). Therefore, the tensile strength of non-

150

Table 5.7: The effect of the transesterification degree (B) on the tensile strength

 

 

 

 

 

of PET/PEN blend films
Tensile strength (MPa)
non-oriented -films oriented films
Sample ID Bavg MD CMD
0.076 50.89 45.35 153.2
3 0.222 47.33 48.08 182.7
4 0.356 50.27 40.84 170.0
5 0.192 47.05 44.74 148.1
6 0.405 44.89 44.90 126.0
7 ' 0.525 51.42 50.99 148.8

 

 

 

 

 

 

* Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

 

 

 

 

250
. 200 --
(I
E E I
g db
*2 150 § § {
g f
3 100 +
'- 50 “ W ‘ I I E I I
0 1 1 1 1 1
0 0.1 0.2 0.3 0.4 0.5 - 0.6

Degree of transesterlflcation (B)

 

 

1:1 non-oriented films-MD , A non-oriented films-CMD o oriented films

 

Figure 5.7: The effect of the transesterification degree
on the tensile strength of PET/PEN blend films

151

 

Table 5.8 : The effect of PEN composition on the tensile strength

 

 

 

 

of PET/PEN blend films
Tensile strength (MPa)
non-oriented -fi1ms oriented films
Sample ID PEN mole % MD CMD
1 O 45.6 41.8 241.4
8 14.5 42.2 45.5 186.7
3 33.7 47.3 48.1 182.7
9 39.5 50.7 45.3 190.7
10 50.4 53.9 50.3 201.2
11 100 61.5 61.5 269.3

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

 

 

 

 

120 300
110 oriented films 270
100 y = 0.0245x’ - 2.021411 + 29.25
1; ., ‘ R2 = 0.8908 240 ..
n. E 90 a
- 210 °-
E 1: 80 E E
3 70 180
g 5 60 15 g
g .g 0 g
. , 55,--
2 £50 /../' 120 3
'3 c 40 .. 90 =
5 3 30 ~- - 3
1- .. MD CMD .. 50 .2
20 y = 0.187311 1 42.753 y = 0.1917x 1 41.149
10 1 R2 = 0.8943 R2 = 0.9306 30
0 1 1L 1 1 o
0 20 40 60 80 100

PEN mole %

 

1

L

o oriented films 0 non-oriented films-MD A non-oriented films-CMD

 

Figure 5.8: The effect of PEN composition on the tensile strength
of PET/PEN blend films

152

of oriented films

oriented PEN film was approximately 40-50% higher than those of non-oriented PET
film. Shi (1998) also studied the tensile strength of the PET/PEN blends, and reported a
similar result, but with a slightly lower percent difference (35%) between the PET and
PEN films. The average tensile strength along the machine direction of the non-oriented
films was slightly higher that those along the cross machine direction.

The orientation process enhanced the tensile strength of the PET, PEN, and blend
films. The tensile strengths of the non-oriented PET, PEN and blend films were in the
range of 41 to 62 MPa while those of oriented films were in the range of 108 to 269 MPa,
(Tables 5.7 and 5.8). Therefore, the tensile strength of the PET, PEN, and blend films
improved approximately 4 to 5 times afier orientation. The orientation effect on the

tensile strength of blend films will be discussed in Chapter 6.

Degree of transesterification and tensile strength

The results showed no relationship between the tensile strength of the blend
films processed at the various degrees of transesterification (Figure 5.7). This was in
agreement with the permeability and the T8 studies, in that these parameters were also
independent of the degree of transesterification.

Statistical analysis of the tensile strength of the non-oriented and oriented films
with the same PEN composition of 25% (wt/wt) of PEN and processed at various
conditions was performed as shown in Appendix E. There was no statistically
significant difference between the MD tensile strength values of the non-oriented blend
films, whereas a significant difference between the CMD tensile strength values of non-

oriented blend films was found (Table El and E2 in Appendix E). Moreover, statistical

153

analysis showed a significant difference between the tensile strength values of the

oriented films processed at different conditions (Table E3 in Appendix E).

PEN composition and tensile strength

A relationship between the tensile strength of the blend films and the PEN
composition was found. The MD and CMD tensile strength values of the non-oriented
films were directly proportional to the PEN composition (Figure 5.8). For the oriented
films, the relationship between the PEN composition and the tensile strength was best
described by a second order polynomial expression (Figure 5.8). This enhancement of
the tensile strength occurred due to the inclusion of a number of the naphthalate groups in
the main structure of the blends, resulting in an increase in the stiffness of the polymeric
chains and a corresponding increase in the glass transition temperature of the blends,
even though, when the PEN composition was increased, a depression of melting
temperature and percent crystallinity were found. Above the glass transition
temperature, the tensile strength and the modulus is dependent on the percent crystallinity
of the polymer, while below the glass transition temperature, the effect of crystallinity on
the tensile strength and modulus is relatively small (Paul and Bucknall 1999)
Therefore, in this study, the effect of the naphthalate rings seems to be the predominant
factor controlling the tensile strength.

Moreover, the results showed that the tensile strength values of the non-oriented
blend were in the range between those of PET and PEN. Therefore, the tensile strength
of the non-oriented blend films improved by adding PEN to PET. However, the tensile

strength values of the oriented blend films were lower than those of PET and PEN

154

oriented films. The effect of orientation on the tensile strength of PET and PEN films
seems to be stronger than on blend films. A detailed discussion of the orientation effect

on the tensile strength of the blends is included in Chapter 6.

5.3.2.2. Percent elongation at break

The non-oriented PEN and blend films were very brittle, relative to PET films.
The % elongation values of the non-oriented films with 10 to 40 % (wt/wt) PEN and
processed at 285 to 315 °C, and l and 2 passes ranged from 2.7 to 3.4 %, and those of
PEN were in the range of 4% to 5% (Tables 5.9 and 5.10). The PET non-oriented films
had a considerably higher elongation 500 to 600% (Table 5.10). I

For the PET and PEN films, the % elongation along the machine direction was
greater than that along the cross machine direction. For the non-oriented blend films, the
% elongation along both directions appeared to be similar, while non-oriented films
typically showed differences in % elongation between the machine and cross machine
direction (DeVries et a1. 1977; Park and Mount 1988). This might be due to the very low
percent elongation of the blends and limitations in the sensitivity of the Instron machine.

After the orientation process, the % elongation of the PET fihns decreased to
213% (Table 5.10). Ray et a1. (1977) studied the mechanical properties of PET films
and reported similar results, where the % elongation of oriented PET films (60 to 100%)
was lower that that of non-oriented films (250%). The decrease in the % elongation of
the PET films might be the outcome of the orientation effect, which in leads to an
increase in the % percent crystallinity. As presented in Table 4.8 (Chapter 4), afier the

orientation process, the % crystallinity of PET increased. From the literature,

155

Table 5.9: The effect of the transesterification degree (B) on
the % elongation of PET/PEN blend films

 

% Elongation

 

non-oriented films

 

oriented films

 

 

Sample ID Bavg MD CMD
2 0.076 3.4 2.99 148.8
3 0.222 3.03 3.06 130.4
4 0.356 3.5 2.7 149.5
5 0.192 3.0 2.87 148.6
6 0.405 3.0 2.86 127.7
7 0.525 3.0 3.14 140.2

 

 

 

 

 

 

* Blends with 25% PEN (wt/wt) processed through a twin-screw extruder( 150 rpm)

 

 

 

 

 

5 300
4.5 -~ v 4 280
a 4 -_ 1- 260
E .
c E 3.5 ~- In 0 - 240
O
a g 3 .. g i i fl g i «- 220
gg 2.5 «~ ~» 200
E g 2 «~ L 180
° 5 1.5 —» 1- 160
e 1
1 ~ i } «~ 140
0.5 - } ~~ 120
0 1 4 1 1 a 100
0 0.1 0.2 0.3 0.4 0.5 0.6
Transesterificatlon (B)

 

1
1

Figure 5.9: The effect of the transesterification degree

on the % elongation of PET/PEN blend films

156

[El non-oriented films-MD A non-oriented films-CMD 0 oriented films}

%Elongation
for oriented films

Table 5.10: The effect of PEN composition on the % elongation of

 

 

 

 

PET/PEN blend films
%Elongation
non-oriented films oriented films
Sample ID PEN mole % Nfl) CMD
l 0 604.00 506.43 213.1
8 14.5 3.28 3.0 181.6
3 33.7 3.03 3.06 130.4
9 39.5 3.08 2.9 144.1
10 50.4 3.28 3.26 134.6
11 100 5.33 4.07 107.8

 

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

  
   
  

     

 

 

 

 

 

 

1
a 14 lb oriented films 1_ 200
g .. y=0.0135x’-2.3691x+211.27
g 12 1’ R2=0.9489
c ”D 1— 150
§ 10 fl y=0.0005x2-0.0341x+3.6487
é 8 I R2=0.9984
o «1.
E CMD 100
.2 6 1 y=0.0002:8-0.0054x+3.0041 1=
5 R2=0.9456
C
2
m 2 ._
a!
0 1 1 ; 1 o
0 20 40 60 80 100
PEN mole%

 

 

D non-oriented films-MD A non-oriented films-CMD <> oriented films

 

 

Figure 5.10: The effect of PEN composition on the % elongation

of PET/PEN blend films

157

%Elongatlon for oriented films

crystallinity seems to reduce the ability of chain-straightening, and molecular slippage,
which corresponds to a decease in the extensibility of the polymer chains (Rudin 1982;
Schultz 1974; Schwartz 1982).

In contrast to the PET films, the % elongation of the oriented PEN and blend
films improved tremendously. The elongation value of the oriented PEN film was 108%
and those of oriented blend films were in the range of 128 to 182 %, respectively (Table
5.9 and 5.10). The results indicated that the PET, PEN and blends exhibited different
stress-strain behaviors, when they were oriented under the following orientation
conditions: a stretch temperature of 100°C for PET and blend samples, 135°C for the
PEN films, and a soak time of 15 seconds for all samples. Moreover, the results fiom
Chapter 4 (Table 4.8) showed that following the orientation process, the % crystallinity
of PEN films remained the same, and those of the blends increased slightly, relative to
that of PET. This suggests that PET, PEN, and blend films have a different mechanism
of strain-induced crystallization. This mechanism involves the promotion of chain-
straightening, and molecular slippage (Jenkins 1995; Shi 1998). The details of the effects
of orientation on the % elongation of the PET, PEN and blend films are presented in

Chapter 6.

Degree of transesterification and percent elongation at break
The results showed no relationship between the % elongation of the blend films
with various degrees of transesterification (Figure 5.9). These results are in agreement

with the tensile strength study and can be explained as mentioned previously.

158

Statistical analysis of the % elongation of the blend films with the same PEN
composition of 25% (wt/wt) of PEN and processed at the various conditions was
performed, as shown in Appendix F. There was no statistically significant difference in
the % elongation along the machine direction of the non-oriented blend films, whereas a
significant difference in the % elongation along the cross machine direction of the non-
oriented films was found (Table F1 and F2 in Appendix F). Additionally, statistical
analysis showed no significant different in the % elongation of the oriented films

processed at different conditions (Table F3 in Appendix F).

PEN composition and percent elongation at break

The results showed that there was a relationship between the % elongation of the
film samples and the PEN composition (Table 5.10 and Figure 5.10). For the non-
oriented films, the % elongation appeared to increase slightly as the PEN composition
increased. Due to the very low range of the % elongation of these samples, it is
recommended that the experiment be repeated to confirm the results. However, these
results seem. to be reasonable. The results are in agreement with the effectof the PEN
composition on % crystallinity, density, and molecular weight average, which are the
primary factors controlling the elongation of the polymers. These properties decreased as
the PEN composition increased (Chapter 4). The reduction in % crystallinity, density,
and molecular weight average promote the flexibility of the polymers (Paul and Bucknall
1999; Rudin 1982; Schultz 1974; Schwartz 1982).

In contrast to the non-oriented films, the % elongation of the oriented films

decreased as the PEN composition increased. This might be due to the fact that the

159

orientation effect outweighs the effect of the PEN composition, and results in a decrease

in the ability for extension under the application of stress.

5.3.2.3. Young’s modulus of elasticity

Young’s modulus of elasticity values for the PET, PEN, and PET/PEN blend
films are shown in Tables 5.11 and 5.12. The results showed that the modulus of non-
oriented PET, PEN, and blend films were in the range of 1554 to 1739 MPa (Tables 5.11,
and 5.12), respectively.

There was no statistically significant different between the modulus of those
samples (Figures 5.11 and 5.12). However, after orientation, the modulus of PET, PEN,
and blend films increased (Table 5.12 and Figure 5.12). The average moduli of oriented
PEN films (2574 MPa) was slightly higher than that of oriented PET films (2304 MPa).
However, the moduli of oriented blend films were lower than either of PET or PEN

oriented films.

Degree of transesterification and Young’s modulus of elasticity
The results showed no relationship between the modulus of elasticity of the
blend films with various degrees of transesterification (Figure 5.11). These results were
in agreement with the barrier studies and the tensile and elongation results, in that those
properties of the blend films did not depend on the degree of transesterification.
Statistical analysis was performed and showed that there was no significant

difference between the moduli of the blend films with 25% (wt/wt) of PEN, processed

160

 

Table 5.11: The effect of the transesterification degree (B) on
the modulus of elasticity of PET/PEN blend films

 

 

 

 

 

 

Young's Modulus (MPa)
non-oriented films oriented films
Sample ID 331% Nfl) CMD

2 0.076 1569.95 1549.95 1785.8

3 0.222 1718.18 1618.90 1823.6

4 0.356 1714.74 1560.98 1679.5

5 0.192 1688.54 1549.95 1693.9

6 0.405 1627.86 1520.30 1697.8

7 0.525 1739.07 1586.49 1673.6

 

 

 

 

* Blends with 25% PEN (wt/wt) processed through a twin-screw extruder (150 rpm)

The modulus of elasticity (MPa)

2200

 

2000 +

1800 .1

1600 1-

1400 .-

1200 1-

 

1000

 

 

0.1

0.2

0.3 0.4
Transesterificatlon (B)

0.5

 

i U non-oriented films-MD A non-oriented films-CMD o oriented filmsj

 

Figure 5.11: The effect of the transesterification degree

on the modulus of elasticity of PET/PEN blend films

161

0.6

 

Table 5.12: The effect of PEN composition on the modulus
of elasticity of PET/PEN blend films

 

 

 

 

Young's Modulus (MPa)
non-oriented films oriented films
Sample ID PEN mole % MD CMD
1 0 1648 155.4 2304.3
8 14.5 1684 1653 1932.4
3 33.7 1718 1619 1823.6
9 39.5 1556 1617 1901.1
10 50.4 1692 1532 2049.2
11 100 1619 1685 2573.5

 

 

 

 

 

 

* Blends processed at 300°C and 1 pass through a twin-screw extruder.

 

 

 

 

 

3000
a“! 2500
E
.é‘
o 2000
'5
ll)
2
.2 1500
O
n
2
g 1000 ..
E oriented films
53 500 1- y = 0.1958x2 - 15.874x + 2223.1
R2 = 0.8993
0 r $ 1 1r
0 20 40 60 80 100

. PEN mole %

— a — non-oriented films-MD ----A--- non-oriented filrns-CMD _
o oriented films 1

Figure 5.12 : The effect of PEN composition on the modulus of
elasticity of PET/PEN blend films

 

 

162

under various conditions of temperatures and blending times (Table G1, G2 and G3, in

Appendix G).

PEN composition and Young’s modulus of elasticity

There was no relationship between the MD and CMD moduli of the non-oriented
blend films, and the PEN composition. Statistical analysis showed (Appendix G) no
significant difference in MD moduli of the non-oriented blend films with different
compositions, whereas statistically significant differences between the CMD moduli of
non-oriented films and PEN composition were found (Tables G4 and G5, in Appendix
G). The significant difference between the CMD modulus values for the non-oriented
films might be due in part to the low sensitivity of the Instron machine, and may also ,
reflect a slight effect of the PEN composition on the modulus of elasticity of non-oriented
films.

However, a relationship between the modulus values of the oriented films and the
PEN composition was found, and was best described by a second order polynomial
expression (Figure 5.12). Statistical analysis showed a significant difference between
these samples as well (Table G6 in Appendix G). This relationship was similar to the
relationship found between the tensile strength and the PEN composition. A non-
equivalent effect of the orientation process on PET, PEN and blends was found, in that
the orientation effect on the modulus of PET and PEN films seems to be greater than for
the blend films. Thus, the modulus of the oriented blend films was lower than those of
PET, and PEN oriented films. A detailed discussion of the orientation effect on the

modulus of elasticity is included in Chapter 6.

163

Chapter 6

The Orientation Effect

6.1. Introduction

Orientation is an important process to enhance the performance of fabricated
polymers. Most commercial semicrystalline polymers are partially oriented in order to
yield advantageous properties, e. g., mechanical strength, barrier properties, and dielectric
polarizability (Peterlin 1988). Products with designed-in orientation, including
fabricated sheets and filaments with uniaxial orientation, and films and bottles with
biaxial orientation, are increasingly important in the polymer industry (White 1988).
Therefore, in-line processes for uniaxial and biaxial orientation have been developed for
a variety of molding operations.

Molecular orientation generally refers to the alignment and rearrangement of
segments of polymeric chains by an application of an external stress at a particular
temperature. Hence, it involves some degree of stretching and heat setting. In general,
the orientation process is a rapid stretching of a polymer at a temperature above the T‘ for
amorphous polymers and at a temperature between the T8 and the Tm for semicrystalline

polymers, and a subsequent quenching of the polymer (Park and Mount 1988).

164

6.1.1. Uniaxial Orientation

Uniaxial orientation promotes high performance of the polymers in one direction,
usually along the machine direction. Melt polymer or extrudate is released from a flat
die and passed through a set of chill rolls which force the hot plastic to stretch under a
controlled temperature. This film fabrication process produces some degree of uniaxial
orientation. In addition, the quenching rate is controlled to ensure that the orientation is
not lost by molecular relaxation (Park and Mount 1988) Uniaxial orientation along the
cross machine direction can be obtained using the tenter frame or melt inflation.
However, these two techniques are seldom employed (Park and Mount 1988).

The relationship between physical properties and uniaxial orientation of the
polymer has been studied. As the orientation occurs, the c-axis of the molecular chains in
the amorphous and crystalline phases of the semicrystalline polymer is aligned in the
orientation direction. This alignment of the c-axis chains allows the strong covalent
bond along the chain backbone to carry the load applied in the orientation direction. In
the perpendicular direction, the weaker intermolecular Van der Waals forces are
predominant, which can only carry a lower load (DeVries et al. 1977; Hoshino et. al.
1962; Seferis and Samuels 1979). Therefore, uniaxial orientation promotes the

mechanical properties only along the orientation direction.

6.1.2. Biaxial Orientation
Biaxial orientation substantially improves the physical properties of polymers in
both the machine and the cross machine directions. Commercial equipment for biaxial

orientation includes tenter-frame, double-bubble, and blown film machines (Park and

165

Mount 1988). In the tenter-frame and double-bubble process, polymers are biaxially
oriented in the solid state at a temperature below the crystalline melting point. In the
blown film process, polymers are oriented in the melt state and are rapidly quenched
(DeVries et al. 1977; Middleman, S. 1977; Samuels 1974). Biaxially oriented films
achieve superior improvement in physical properties compared to uniaxially oriented
films. This superior improvement is due to the redistribution of the c-axis chains by the

biaxial orientation (DeVries et a1. 1977).

6.1.3. Effect of Orientation on Polymer Properties
Orientation influences the morphological, physical and thermal properties of
polymers, and consequently improves the mechanical, barrier, optical, and electrical '

properties.

6.1.3.1. Percent crystallinity

Strain-induced crystallization is one of the results of the orientation (White 1988).
Orientation tends to enhance. crystallinity by bringing the long axis of the molecular
chains close together and parallel to each other (Rudin 1982). The amount of
crystallization induced by the orientation process is controlled by the stretch rate, stretch
temperature, and the stretch ratio (J abarin 1992).

However, the difference between orientation and percent crystallinity is that
crystallinity requires a regular placement of molecular chains, whereas orientation

requires only the alignment of molecular chains, regardless of the location of the

166

molecules (Rudin 1982). Therefore, orientation can occur in both amorphous and

semicrystalline polymers.

6.1.3.2. Barrier properties

In general, orientation promotes an enhancement of the barrier properties of
glassy polymers (El-Hibri and Paul 1985; O’Brien et al. 1987). This improvement in the
barrier properties of semicrystalline polymers is due to the rearrangement of the larnellar
crystalline domains, in a direction perpendicular to the direction of the penetrant flow. In
addition, simple crystallinity acts as a tortuous blocking which lengthens the path of
penetrants in their diffusion across the film bulk phase. This behavior, therefore, leads to
a significant reduction of the gas permeability of polymeric films. Furthermore, the
effect of orientation on highly crystalline materials is stronger than that on an amorphous
or less crystalline material (Brady et a1. 1974; Koros and Hellums 1988).

O’Brien et a1. (1987) studied the effect of orientation on the C02 and CH4
permeability of uniaxially and biaxially oriented polyirnide samples of pyromellitic
dianhydride and oxydianiline. Biaxial orientation showed a stronger effect on the
solubility, diffusivity, and permeability, than that of the uniaxial orientation. El-Hibri
and Paul (1985) also agreed that orientation produces a significant reduction in solubility

and diffusivity, resulting in a decrease in the permeability of polymeric materials.
6.3.1.3. Mechanical properties

Uniaxial orientation strongly affects mechanical properties, in that the tensile

strength and the Young’s Modulus of elasticity increase whereas the elongation at break

167

decreases (Nadella et a1. 1978). Therefore, uniaxially oriented fihns are brittle in the
cross machine direction (Matsumoto et a1. 1981). In contrast to uniaxial orientation,
biaxial orientation enhances the tensile strength, Young’s modulus, and elongation at
break, in all directions in the plane of the films (White 1988). For instance, biaxially
oriented bottles exhibit high mechanical properties in both the bottle-axis direction and
the hoop direction.

However, this relationship between the tensile strength and Young’s modulus, and
orientation are only found at a low degree of orientation. When crystallites are
completely oriented, further stretching enhances the tensile strength and Young’s
modulus (Nadella et al. 1978; White et al. 1974; Samuels 1974).

Amorphous and semicrystalline polymers behave differently, when uniaxially. -
oriented. Semicrystalline polymers exhibit necking or an abrupt change of the thickness.
In contrast, amorphous polymers can be stretched to any thickness, if properly supported

(Samuels 1974).

6.1.3.4. Effect of orientation on selected properties of PETIPEN blends

A difference between the stress-strain behavior of PET and PEN was found
(Cakrnak et 31.1996; Murakami et al. 1996; Shi 1998) PEN shows neck formation,
when it is uniaxailly oriented at temperature between T8 and he (cold crystallization
temperature), at a stretch ratio of less than 5 mm/min (Shi 1998). However, for PET the
neck behavior usually occurs at T8 or below. Therefore, a difference in the mechanism
and the temperature dependency of strain-induced crystallization of PET and PEN was

considered (Shi 1998).

168

Bauer (1997) showed that the density of biaxially oriented films of PET/PEN
blends increased with an increase in the stretching ratio (2.5x2.5, 3.0x3.0 and 3.5x3.5)
and heat setting temperature (120, 150 and 180°C). The increased density indicates an
increase in the level of crystallinity. Oxygen permeability also decreased as the
stretching ratio or orientation temperature was increased. The oxygen permeability of
the 3.5x3.5 biaxially oriented films of a polymer blend with 25% wt/wt of PEN was
about 40% lower than the corresponding PET films. Hoffman and Caldwell (1995)
studied the oxygen permeability of PET/PEN blends and copolymers of compositions
varying from 0-100 weight percent PEN, and found that the permeability coefficient
decreased to approximately half when fihns were 4x4 biaxially oriented at 20°C above
T8. In addition, Bauer (1997) mentioned that for pre-heated samples, the density was
independent of the stretching ratio. At the same heat set temperature but different draw
ratios, samples reached almost the same range of density.

In this chapter, the effect of biaxial orientation on selected properties of the PET,
PEN and PET /PEN blends will be discussed. A comparison of the orientation effect on
the blend films with various degree of transesterification and various compositions will

be discussed.

6.2. Materials and Methods

The eleven PET, PEN, and PET /PEN blend films (IO-mil, non-oriented films)
were biaxially oriented on a T.M. Long film stretcher. All of the samples were stretched
3x3 simultaneously in both directions. The rate setting was 300% per second. The

stretch temperature was 100°C for all samples, except for the PEN films, which were

169

stretched at 135 °C. Twelve biaxially oriented film specimens were prepared from each
sample. The final oriented film thickness was approximately 1 mil (0.001 inch). The
details of the orientation process are discussed in Section 4.2.2.

The thermal properties of the oriented blend films were determined using a DSC
2920 Modulated Differential Scanning Calorimetry (MDSC). The glass transition
temperatures (T8), melting temperatures (Tm), and the % crystallinity were determined.
The details of the thermal analysis procedure are presented in Section 4.2.3. 1.

The density gradient technique was used to determine the density of the non-
oriented and oriented fihns (Section 4.2.3.2).

The Permatran-W 3/31 was employed to measure the water vapor transmission
rates. The oxygen and carbon dioxide permeability values were determined with an
Oxtran 200 permeability tester and the Permatran CIV permeability tester, respectively.
All procedures had followed ASTM standards, where applicable (Sections 5.2.1, 5.2.2,
and 5.2.3). .

Tensile tests were performed using an Instron Universal Tester, following the
ASTM standard methods (Section 5.2.4). Tensile strength, Young’s modulus, and %

elongation at break were determined.

6.3. Results and Discussion

The T8, Tm, and % crystallinity of all non-oriented and oriented film samples were
reported in the Results and Discussion section of Chapter 4 (Section 4.3.1). The density

of all samples was presented in Section 4.3.2. The mechanical and barrier properties of

170

the oriented and non-oriented film samples were summarized in Sections 5.3.1 and 5.3.2,
respectively.

To facilitate the discussion, the term “orientation-induced increase (011)” of a
particular numerical property of a polymer has been defined to be the difference between
the value of that property of a 3x3 biaxially oriented polymer and that for the non-
oriented one. For instance, the tensile strength of 3x3 biaxially oriented PET film was
45.6 MPa, and the tensile strength of the non-oriented PET film was 241.4 MPa (Table
5.8). Therefore, the 011 of the tensile strength of this PET film was 241.4 - 45.6 = 195.8
MPa.

The result showed that orientation affected the T3, % crystallinity, density, barrier
properties, and mechanical properties of both PET and PEN, as well as their respective
blends. However, orientation seems to have less of an effect on the Tm of the blends,
compared to the effects of degree of transesterification and PEN composition.

For the PET films, the orientation-induced increase (011) of all properties, except
for the Tm and the Young’s modulus, was higher than that of PEN and blend films. In
PET rich phase systems (PEN composition lower than 50 mole % PEN), the 011 of all
properties, except for Tm and Young’s modulus, were higher than that of blends with a
higher PEN composition. There was no significant difference of the 011 of all properties
(except for the % crystallinity and the density) of the blends with different degrees of
transesterification. However, the 011 of the % crystallinity and density of the blends

slightly decreased as the degree of transesterification increased.

171

Therefore, the degree of transesterification had less effect on the 011 of most
properties of the blends than PEN composition. The effect of OH on particular

properties of the PET, PEN and blend films is discussed in the next section.

6.3.1. Effect of Orientation on Thermal Properties
6.3.1.1. Glass transition temperature

The results showed that the T, of the PET and the blend films increased about 13
°C after 3x3 biaxial orientation. In contrast, the T, of the PEN film decreased 6°C after it
was oriented (Table 4.3 and Figure 4.3).

For the blends with different degrees of transesterification, the T, increased
between 2-5°C after orientation (Table 4.2 and Figure 4.2). Moreover, the T, values of
oriented blend films were approximately the same for different degrees of
transesterification.

These blend films had a composition ranging from 0'to 50.4 mole % PEN. No
experiments for the blends with PEN compositions of more than 50.4 mole % PEN were
performed. Therefore, the predicted T, values of blend films with high PEN
compositions were obtained by best fitting the data with a mathematical expression. The
relationship between the composition of the oriented blend films and the T, values was
found to be a second order polynomial, while that of the non-oriented blend films and the
T, values was linear. The result showed that in PET rich phase systems (PEN
composition lower than 50 mole % PEN), the 011 of the Tg of the blend with the lower
PEN composition was higher than the 011 of blend with the higher PEN composition

(Table 4.3 and Figure 4.3).

172

6.3.1.2. Melting temperature

The results showed that orientation had little effect on the Tm of the film samples
(Tables 4.5 and 4.6, and Figures 4.5 and 4.6). The Tm of the PET and the blend films
decreased slightly after orientation. The Tm of PEN films did not change after

orientation.

6.3.1.3. Percent crystallinity

After orientation, the % crystallinity of the PET films increased fi‘om 33% to 41%
while the % crystallinity of the PEN films did not change (Table 4.8 and Figure 4.8).

For the blends, regardless of the degree of transesterification, the % crystallinity
increased slightly afier. orientation (Table 4.8 and Figure 4.8).

The result showed that in PET rich phase systems, the 011 of the % crystallinity of
blends with a lower PEN composition was higher than that of blends with a higher PEN
composition (Table 4.8 and Figure 4.8). In the PEN rich phase systems, the %
crystallinity did not change afier the blends were oriented. The difference in the
orientation effect on PET, PEN and the blends is due to the difference in the mechanism
of strain-induced crystallization and the temperature dependency of the strain-induced

crystallization for the pure components and the blends (Jenkins 1995; Shi 1998).

6.3.2. Effect of Orientation on Density of the Blends
Orientation slightly but predictably affected the density of the PET, PEN and
blend films (Tables 4.10 and 4.12, and Figures 4.10 and 4.12). The 011 of the density of

blends with lower degees of transesterification was higher than that of the ones with

173

higher degrees of transesterification (Table 4.10 and Figure 4.10). Moreover, the 011 of
the density of the PET fihn was higher than both those of PEN and blend films. In PET
rich phase systems, the 011 of the density of the blends with lower PEN composition was

higher than those with higher PEN compositions (Table 4.12 and Figure 4.12).

6.3.3. Effect of Orientation on Barrier Properties

The water vapor, oxygen, and carbon dioxide barrier properties of PET, PEN and
blend films improved considerably when these films were 3x3 biaxially oriented (Tables
5.1 to 5.6, and Figures 5.1 to 5.6).

The results showed no significant difference in the permeability coefficients of
the oriented blend films with different degree of transesterification. The 011 of the
permeability coefficients of PET film were higher than those of PEN and blend films. In
addition, the OH in the permeability coefficients of the blends with a lower PEN

composition were higher than those with a higher PEN composition.

6.3.4. Effect of Orientation on Mechanical Properties
6.3.4.1. Tensile strength

The tensile strength of PET, PEN, and PET/PEN blends significantly increased
after the films were biaxially oriented (Tables 5.7 and 5.8, and Figures 5.7 and 5.8). The
OH of the tensile strength of the blends, regardless of the degree of transesterification,
was approximately the same (Table 5.7 and Figure 5.7). The 011 of the tensile strength
of the PET and the PEN films were higher than those of the blend films. A linear

relationship between the PEN composition and the tensile strength of the non-oriented

174

film was found. However, for the oriented films, a polynomial expression was found to
best describe the relationship between the PEN composition and tensile strength.
Moreover, there was a slight difference of the OH of the tensile strength of the blend

films with different PEN compositions (Figure 5.8).

6.3.4.2. Percent elongation at break

The % elongation at break of the PEN and the PET/PEN blend films significantly
increased as the films were biaxially oriented (Tables 5.9 and 5.10, and Figures 5.9 and
5.10). However, the % elongation at break of the PET films decreased after the PET
films were oriented (Table 5.10). The OH in the % elongation of the blends, regardless
of the degree of transesterification, was approximately the same. The OH in the %
elongation of the blends decreased as the PEN composition increased (Table 5.10 and
Figure 5.10). Moreover, the OH in the elongation of the blend films were higher than

that of the PEN film.

6.3.4.3. Young’s modulus

The Young’s modulus of elasticity of the PET and the PEN films improved
dramatically afier the films were 3x3 biaxially oriented. The OH of the modulus of the
PEN film was higher than that of the PET film (Table 5.12 and Figure 5.12). The
Young’s modulus of the blend films increased slightly after they were oriented (Tables
5.11 and 5.12, and Figures 5.11 and 5.12). Both degree of transesterification and blend

composition had a little or no influence on the OH of the modulus of the blends.

17S

6.4. Summary

The degree of transesterification does not influence the orientation-induced
increase (OH) of most blend properties. After blends achieved a certain degree of
transesterification, the blend properties do not depend on the degree of transesterification.

However, blend composition is a major factor influencing the OH of most blend
properties. The 011 of PET was greater than that of PEN and the blends. In addition,
the OH of the blends with the lower PEN composition were greater than that of the higher
PEN composition.

I This might be due to the difference between the temperature dependency and the
mechanism of orientation-induced crystallization of PET and PEN. In this study, PET,
PEN and blends were 3x3 biaxailly oriented under the same stretch rate but different
stretch temperatures. The stretch temperature for PET and the respective blends was
100°C, and for PEN was 135°C. The stretch temperatures of PEN and blends might not
be sufficient or equivalent to that of PET. Moreover, the stretch rates for PET, PEN and
blends to achieve sufficient orientation results might be different.

Shi (1998) reported similar results. PET and PEN exhibited different stress-
strain behaviors when PET and PEN were uniaxially oriented under equivalent ‘
conditions, at a temperature of 15°C above the polymer T, (Shi 1998). Therefore, future
investigation of the orientation conditions for PET, PEN and blends to achieve the

maximum orientation effect is recommended.

176

Chapter 7

Summary and Conclusions

7.1. Transesterification Reactions

PET and PEN are inherently immiscible. The mixture of PET and PEN yields an
opaque material with non-uniform properties, and shows two T, and Tm values
corresponding to one PET-rich phase and one PEN-rich phase. Transesterification
reactions are interchange reactions between hydroxyl and carboxyl groups of the PET and
PEN chains. T ransesterification reactions are accomplished by melt mixing these two
polymers at a temperature which is higher than their melting temperatures. The
transesterification reactions enhance the miscibility of the blend, and cause a reduction in
size of the PET and PEN chains. Morphological analysis confirmed that blends which
achieved a minimal degree of transesterification exhibited sample homogeneity, at the
spatial resolving power of Dispersive Raman spectroscopy technique (approximately 200
nm), while the blends without a minimal degree of transesterification showed phase
separation. The properties and composition of the blends change when transesterification
is achieved. During this reaction processing, graft and block copolymersvare formed. The
PET/PEN blends with a transesterification degree of at least 6% showed a significant
improvement in the blend miscibility and had only a single T, and Tm. These blends
were optically clear and uniform. However, the DSC analysis of the blends showed

single, broad Tm peaks, suggesting the occurrence of partial miscibility.

177

The processing conditions including the blending temperature and the blending
time are primary factors controlling the degree of transesterification, while blend
composition had little or no effect. There was a linear relationship between the degree of
transesterification and the blending temperature. Moreover, the degree of
transesterification increased as the blending time (number of pass) increased. However,
there was no relationship between the degree of transesterification of the blends and the
blend composition.

In the range of the processing conditions used in this study, the optimum blending
temperature was found to be between 285 and 315°C. The PET/PEN blend resins
obtained fiom reaction processing at a temperature of 275°C did not appear to be uniform
and were mixed between clear and hazy zones. According to DSC and lHNMR analysis,
a blending temperature of 27 5°C is the minimum limit to achieve a measurable degree of
transesterification. In addition, the PET/PEN blend resin obtained fi'om reaction
processing at 325 °C was brittle and had low melt strength. Degradation might occur
during processing at this temperature, or at temperatures greater than 325°C.

In addition, the D value was proposed to represent the degree of
transesterification instead of the existing B-value. The D value is more meaningful than
the B-value. However, in this dissertation the B-value is used to represent the degree of

transesterification, so that the results can be compared with other studies.

178

7.2. Blend Characteristics

Table 7.1 shows a summary of the effects of transesterification and blend
composition on the blend characteristics including: (i) thermal properties, (ii) %
crystallinity, (iii) density, (iv) molecular weight, (v) banier properties, and (vi)

mechanical properties.

7.2.1. Effect of Degree of Transesterification

The PET/PEN blends with a degree of transesterification of at least 6% were
miscible, homogeneous and optically clear. ’ Further, the transesterification reactions had
little effect on the T,, the banier, and the mechanical properties. However, the Tm, %
crystallinity, density, and molecular weight average were dependent on the degree of
transesterification, regardless of how this degree of transesterification was achieved.

The "1’m , % crystallinity, and density of the blends decreased linearly, as the
degree of transesterification increased. This is due to the fact that the transesterification
reactions lead to the formation of terephthalate-naphthalate (N -T) bonds, which might
interfere with crystal formation.

The T, did not change as the degree of transesterification increased. This is in
agreement with the water vapor, oxygen, and carbon dioxide barrier properties and the
mechanical properties, including tensile strength, % elongation at break, and Young’s

modulus of elasticity, in that they were independent of the degree of transesterification.

179

Table 7.1: Summary of the effect of the degree of transesterification and blend

composition on selected properties of the PET/PEN blends.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The effect of degree of The effect of degree of PEN
transesterification (B) composition (mole % PEN)
Resins Films Oriented Resins Films Oriented
films films
T, N N N L+ L+ P+
Tm L- L- L- U U U
%Crystallinity L- L- L- U U U
Density L- L- L— n/a L— P-
Permeability N/a N N n/a L- L-
PH,01P0,,1PCO,
Tensile strength N/a N N n/a L+ U
%Elongation N/a N N n/a n/a P-
at break
Young’s N/a N N n/a N U
modulus
L linear relationship
N no relationship
P second order polynomial expression
U parabola expression
+ the preperty increases as B or mole % PEN increases

the property decreases as B or mole % PEN increases

180

 

7.2.2. Effect of Blend Composition

The blend composition is an important parameter which controls all the
characteristics and pr0perties of the blends. A depression of the T,,, , % crystallinity, and
molecular weight was found, when a small amount of PEN was added to the PET rich
phase, or vice versa. The Tm , % crystallinity and molecular weight average were lowest
when the PEN composition was 40-50 mole %. Moreover, the density of the blends
decreased, as the PEN composition increased. The depression of these properties is a
consequence of the disruption of the crystalline structure, as a small amount of one
polymer is added to another. However, the decrease in Tm , % crystallinity, molecular
weight and density seems to have a less influence on the barrier and mechanical
properties of the blends.

Since the blends were only partially miscible, characteristics of PET and PEN are
still exhibited by the blends. When the PEN composition increased, the T,, the gas
barrier properties, and the tensile strength of the blends improved. Increase of the
number of naphthalate units leads to a decrease in the segmental mobility of the
molecular chains, resulting in higher chain stiffness and improvement in gas barrier
properties and tensile strength.

The relationship between the blend composition and the above properties can be
expressed as a series of mathematical equations as shown in Table 7.2. Therefore, the
properties of the blends with different blend compositions can be predicted using those
equations. However, it should be pointed out that the equations presented apply only to

the blends obtained by the processing conditions used in this study.

181

Table 7.2: The equations showing the relations between the blend composition and the
barrier and mechanical properties of the blends obtained at 300°C and 1 pass.

 

 

 

Properties Equation
(C = PEN composition, % mole of PEN)
Non-oriented films Oriented films

Barrier properties

Pup =-2E-17C+3E-15 Pup =-lE-17C+ 213-15
1. Water vapor permeability
coefficient, P ”20

:: - .. + - = - - -

2. OXYgen permeability P°= 3E 21C 65 19 P0. 3E 21C+ 4E 19
coefficient, P0,
3- Carbon dioxide Pa, = 415200 + 513.1 8 Pa, = -3E-20C + 312-1 8

permeability
coefficient, P €02

 

Mechanical Properties
1, Tensile strength, 7, rsavm) = 0.1873C +42.753 Ts = 0.0245C’-2.0214C+229.25
Ts(CMD) = 0.1917C +4l.149

2. %elongation at break, 5,0110) = 0.000502 —0.034rc E.= 0.013502 -2.3591c +211.27

E +3.6487
B E,(CMD) =0.0002C’ —0.0054C
_ +3.0041
N/A _ g=0.1958C" —15.874C+2213.l

3. Young’s modulus of
elastics, a

 

 

 

 

 

The unit of permeability coefficients-is kg ,m / s m2 Pa.
The unit of the tensile strength and the Young’s modulus is kPa.

182

7.2.3. Effect of Orientation

Orientation is a significant process employed to enhance the barrier and
mechanical properties of polymer blends. The water vapor, oxygen, and carbon dioxide
barrier properties of the PET/PEN blend films improved 1.5-2 times, when the blend
films were 3x3 biaxially oriented. The tensile strength of the oriented blend films was 4
times greater than that of the non-oriented blend films. Moreover, oriented PEN, and the
oriented blend films were flexible, while the non-oriented films were very brittle.

To facilitate the discussion, the term “orientation-induced increase (011)” of a
particular numerical property of a polymer has been defined to be the difference between
the value of that property for a 3x3 biaxially oriented polymer and that of the non-
oriented one. For instance, the tensile strength of 3x3 biaxially oriented PET film was
241.4 MPa, and the tensile strength of the non-oriented PET film was 45.6 MPa (Table
5.4). Therefore, the OH of the tensile strength of this PET film was 241.4 - 45.6 = 195.8
MPa.

The orientation-induced increase (OH) values of the T,, Tm, barrier, and
mechanical properties were independent of the degree of transesterification. However,
the 011 of the %crystallinity and the density showed a slight decrease as the degree of
transesterification increased.

Orientation had a non-equivalent effect on selected properties of PET, PEN and
blend fins. For example, for the PET films, the relative % increases in the T,, %
crystallinity, density, barrier properties, tensile strength, and % elongation were higher
than those of the PEN and blend films. In contrast, the OH of the Young’s modulus of

the PEN films was higher than that for the PET films. In addition, the OH of those

183

properties (except for the Young’s modulus) of the blend films appeared to be dependent
on the blend composition. The OH of the respective properties of blends with a higher
PET rich phase seemed to be greater than those of the blends

with a higher PEN rich phase. This might be due to the difference in the mechanism of
strain-induced crystallization and the temperature dependency of the strain-induced
crystallization for PET, PEN, and their respective blends.

The results from this study showed that PET/PEN blends are a fair alternative for
modified PET. However, transesterification reactions between PET and PEN are
required to enhance the miscibility of the blends. The blend composition is a very
important factor in controlling the thermal, mechanical, and barrier properties of the
blends. Therefore, selection of the blend composition is very crucial, and a compromise
between the requirements of the blend performance and the cost of the PEN needs to be
considered. The results of this study showed that the barrier properties and the tensile
strength of the blends were significantly improved over the corresponding PET values.
The blends were as brittle as PEN. However, after biaxial orientation, the blends are
much more flexible.

It is recommended that further studies on PET/PEN blends for bottles and other
containers be carried out. Thermal treatment and orientation are important processes in
bottle production. These processes are also important factors controlling the
characteristics of the blends. In this study, the film fabrication process for PET, PEN and
blends was very similar in terms of the equipment and technique, except for the
processing temperature and time. For bottle production, it is expected that similar

equipment and techniques are also applicable for PET, PEN, and blends. Moreover, for

184

the blow molding process, the viscosity of the melted resins and heat resistance of the

blend bottles are important and should be considered as well.

185

Appendices

186

Appendix A

Statistical analysis of the degree of transesterification

The analysis of variance (ANOVA) was performed by using a single factor

AN OVA on the microcomputer statistical program in Microsoft 2000.

Table A1: Analysis of variance of degree of transesterification of the PET/PEN blend

resins and films.

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Blend resins Blend films F-value Fem“.
Mean Mean
Sample ID
2 285°C
1 pass 0.061 0.076 2.6052" 10.128
3 ' 300°C
1 pass 0.219 0.222 0.0526* 6.6079
4 315°C
1 pass 0.309 0.356 9.8932" 16.2581
5 285°C
2 passes 0.151 0.192 5.1675“ 98.502
6 300°C
2 passes 0.383 0.405 4.4185" 5.9874
7 315°C
2 passes 0.513 0.525 3.6570“ 7.7086
8 10%
PEN wt. 0.189 0.204 0.3065" 10.128
9 30%
PEN wt . 0.181 0.208 5.2177* 10.128
10 40%
PEN wt 0.13 0.192 40.9731 10.128

 

 

 

 

 

 

 

* Insignificant difference at the confidence level (p) of 0.05, where F -value is less
than Fmtjca].

** Insignificant difference at the confidence level (p) of 0.01, where F—value is less
than Fmfica].

187

Table A2: Analysis of variance of D-values of the PET/PEN blend resins and films.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Blend resins Blend films F-value Fen-“ca.
Sample ID Mean Mean
2 285°C
1 pass 0.020 0.025 2.8285“ 10.1279
3 300°C
1 pass 0.073 0.074 0.0126* 6.60787
4 315°C
1 pass 0.103 0.123 8.0671" 21.1975
5 285°C
2 passes 0.047 0.061 3.92* 18.5127
6 300°C
2 passes 0.126 0.133 0.6508“ 5.98737
7 315°C
2 passes 0.187 0.180 2.5312“ 7.7086
8 10%
PEN wt. 0.027 0.024 0.5504‘ 10.1279
9 30%
PEN wt 0.066 0.079 4.6035“ 10.1279
10 40%
PEN wt 0.056 0.087 44.3538 10.1279

 

 

 

* Insignificant difference at the confidence level (p) of 0.05, where F-value is less
than Fcritical- ‘
"Insignificant difference at the confidence level (p) of 0.01, where F-value is less
than Fcritical-

188

Table A3: Analysis of variance of degree of transesterification of blends with

varied composition of 14.5 to 50.4 mole % PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean

Sample ID

8 14.5 mole % PEN 0.197

3 33.7 mole % PEN 0.221

9 39.5 mole % PEN 0.195

10 50.4 mole % PEN 0.161

ANOVA: single factor

Source of Degree of Mean

Variation SS freedom square F P-value Fain-ca.
Between

Groups 0.008291 3 0.002764 4.666937"' 0.01394 5.09192
Within

Groups 0.010659 18 0.000592

Total 0.01895 21

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.01, where F-value is less

than Fcritical-

189

 

Table A4: Analysis of variance of D-values of blends with varied composition of 14.5 to

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

50.4 mole % PEN.
Source of variance Mean
Sample ID
8 14.5 mole % PEN 0.025
3 33.7 mole % PEN 0.074
9 39.5 mole % PEN 0.072
10 50.4 mole % PEN 0.072
ANOVA: single factor
Source of Degree of Mean
Variation SS freedom square F P-value Fcn'fica]
Between
Groups 0.009503 3 0.003168 31.0989“ 2.47E-07 5.09192
Within
Groups 0.001833 18 0.000102
Total 0.011336 21

 

 

 

 

 

 

 

*Significant difference at the confidence level (p) of 0.01, where F -va1ue is larger

than Fcritical-

190

 

Table A5: Analysis of variance of D-values of blends with
varied composition of 33.7 to 50.4 mole % PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample ID
3 33.7 mole % PEN 0.074
9 39.5 mole % PEN 0.072
10 50.4 mole % PEN 0.072
AN OVA: single factor
Source of Degree of Mean
Variation SS freedom square F P-value Fen-“cal
Between
Groups 2.48E-05 2 1.24E-05 0.099373“ 0.906038 3.73889
Within
Groups 0.001747 14 0.000125
Total 0.001772 16

 

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than Fcritical-

191

 

Appendix B

Calibration curve for the density analysis
Calibration curve for the standard calibration floats
The solvent system: Toluene and Carbon tetrachloride.
The calibration equation: Y = -0.0038X + 1.4336, where Y is the density (g/cc) and X is
the position hour the bottom of the gradient column.

Table B1 : The positions in the gradient column and the density values of the standard
calibration floats.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

 

 

Code Color Density Position
lg/cc)
Bead 1 Large blue 1.2494 49.5
Bead 2 Blue 1 1.300 34.5
Bead 3 Blue 2 1.3500 22.0
Bead 4 Yellow 1.3997 9.0
Bead 5 Red 1.4199 4.0
i
; Figure 13.1. The calibration curve
I for the standard floats at 23° C
1 1.4400 g
1.4200 -
1.4000 1
’13 1.3800 4
g 9 1.3600 .
Q 1.3400 1
§ 1.3200 4 .
'3 1.3000 1 1
.1: .
I" 1:200 ’ y = -0.0038x + 1.4336 I
' 0° ‘ R’ = 0.999 1
1 1.2400 « !
' 1.2200 . . . . . i
0 10 20 30 40 50 60

The Position :

192

The positions of the blend samples and the calculated density based on calibration

equation.

Table B2.The positions and the density of the blend resins

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample ID Position Posiiton Posiiton Position Average density

test 1 test 2 test 3 test 4 (g/cc)

PET 8 8 8 8 1.4041
2 25.6 25.6 25.65 25.6 1.3389
3 25.75 25.75 25.7 25.75 1.3385
4 25.75 25.75 25.75 25.75 1.3384
5 25.75 25.8 25.8 25.8 1.3383
6 25.75 25.7 25.75 25.75 1.3385
7 26 26 25.8 25.8 1.3379

8 25.75 25.75 25.7 25.7 1.3385
9 26.1 26 26.15 26.15 1.3371
10 26.2 26.2 26.25 26.25 1.3367
PEN 18.8 18.8 18.9 18.9 1.3640

 

Table B3. The positions and the density of the non-oriented films

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

sample ID Position Position Average density

test 1 test 2 (g/cc)

PET 24.5 24.6 1.3429
2 25.6 25.3 1.3395

3 25.6 25.6 1.3390
4 25.6 25.7 1.3388

5 25.6 25.6 1.3390
6 25.6 25.6 1.3390
7 25.75 25.75 1.3384

8 24.8 24.7 1.3421
9 25.8 25.9 1.3381
10 26.1 26.1 1.3371
PEN 27.6 27 .6 1.3316

 

 

193

 

Table B4. The positions and the density of the oriented films

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample ID Position Position Average Density

test 1 test 2 (g/cc)

PET 18.5 19 1.3621
#2 22.25 22.4 1.3485
#3 22.4 21.75 1.3494
#4 21.8 22 1.3501
#5 21.2 21.6 1.3520
#6 23 23 1.3459
#7 22.8 23 1.3463
#8 20.4 21.3 1.3541
#9 23 22.6 1.3467
#10 22 23.25 1.3473
PEN 23.2 23.3 1.3450

 

 

 

 

194

 

Appendix C
Statistical analysis for the glass transition temperature of the blends

The analysis of variance (AN OVA) was performed by using a single factor
AN OVA on the microcomputer statistical program in Microsoft 2000.

Table Cl: Analysis of variance of the T g values of the blend resins

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

with 25 % (wt/wt) PEN
Source of variance Mean
Sample Processing Degree of
condition transesterification T, (°C)
(B)
2 285°C /1 pass 0.069 76.87
3 300°C /1 pass 0.221 82.04
4 315°C /1 pass 0.333 78.63
5 285°C /2 pass 0.172 82.51
6 300°C /2 pass 0.394 84.28
7 315°C /2 pass 0.519 81.77
ANOVA
Degree
Source of of
Variation SS freedom Mean square F P-value Fen-m1
Between Groups 77.96479 5 15.59296 1.766286" 0.207538 3.325837
Within Groups 88.28105 10 8.828105
Total 166.2458 15

 

 

 

 

 

 

 

 

 

V *Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than Fcritical~

195

Table C2: Analysis of variance of the Tg values of the oriented films

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

with 25 % (wt/wt) PEN.
Source of variance Mean
Sample Processing Degree of
condition transesterificatio
n T0 (°C)
(B)
2 285°C /1 pass 0.069 84.56
3 300°C /1 pass 0.221 83.55
4 315°C /1 pass 0.333 84.79
5 285°C /2 pass 0.172 85.49
6 300°C /2 pass 0.394 93.98
7 315°C /2 pass 0.519 84.38
ANOVA
Degree
Of
Source of Variationl SS freedom lMean square F P-value F "ML
Between Groups 4.488835 5 0.897767 0.505588" 0.764399 4.387374
Within Groups 10.65414 6 1.77569
Total 15.14297 1 l

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than Fcritical-

196

 

Table C3: Analysis of variance of the calculated and measured T, values of the blend
resins processed at 300°C and 1 pass.

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample % (wt/wt) PEN T, (°C)
8 10 78.6
3 25 83.2
9 30 86.0
10 40 89.4
. Degree
Source of Of Mean
Variation SS freedom square F P-value F min-ca.

 

Bet-ween Groups 124.5782 3 41.52606 31.66289“ 0.00302 6.591392
Within Groups 5.246023 4 1.311506

 

 

 

 

 

 

 

 

 

 

 

Total 129.8242 7

 

*Significant difference at the confidence level (p) of 0.05, where F-value is greater
than Fcritical-

197

Appendix D

Statistical analysis of the permeability coefficient

The analysis of variance (ANOVA) was performed by using a single factor
AN OVA on the microcomputer statistical program in Microsoft 2000.

Table D1: Analysis of variance of the water vapor permeability coefficient (Pflzo ) of the
non-oriented films obtained from blend resins with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of P” 10
condition transesterification (Kg ml, m2 Pa)
(B)
2 285°C /1 pass 0.069 2.42E-15
3 300°C /1 pass 0.221 2.40E-15
4 315°C /1 pass 0.333 2.42E-15
5 285°C /2 pass 0.172 2.41E-15
6 300°C /2 pass 0.394 2.29E-15
7 315°C glass 0.519 2.35E-15
ANOVA
Degree
Source of of '
Variation SS freedom Mean square F P-value Fmfic,_.__

 

Between Groups 2.457E-32 5 4.91E-33 5.29312“ 0.033139 8.745928
' Within Groups 5.569E-33 6 9.28E-34
Total 3.014E-32 1 l

 

 

 

 

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.01, where F -value is less
than F critical- -

198

Table D2: Analysis of variance of the water vapor permeability coefficient (PHI, ) of the
oriented films obtained from blend resins with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of P1120
condition transesterification (Kg ml, m2 Pa)
(B)
2 285°C /1 pass 0.069 1.43E-15
3 300°C /1 pass 0.221 1.52E-15
4 315°C /1 pass 0.333 1.38E-15
5 285°C /2 pass 0.172 1.49E-15
6 300°C /2 pass 0.394 . 1.43E-15
7 315°C /2 pass 0.519 1.64E-15
ANOVA
Degree
Source of of
Variation . SS freedom Mean square F P-value Fain-cap,

 

Between Groups 1.47E-31 5 2.93E-32 1.283578" 0.333081 3.105875
Within Groups 2.74E-31 12 2.29E-32

 

 

 

 

 

 

 

 

 

 

Total 4.21E-3l l7

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less
than Fcritical-

199

Table D3: Analysis of variance of the water vapor permeability coefficient (PHI, ) of
blend films with composition of 14.5 to 50.4 mole % PEN.

 

Source of variance

Mean of Pfizo (kg m/ m2 s Pa)

 

 

 

 

 

 

 

 

 

Sample ID Non-oriented films Oriented films
1 PET 2.96E-15 1.65E—15
8 14.5 mole % PEN 2.76E-15 1.7OE-15
3 33.7 mole % PEN 2.40E-15 1.52E-15
9 39.5 mole % PEN 2.37E-15 1.26E-15
10 50.4 mole % PEN 2.12E-15 1.28E-15
11 PEN 2.15E-15 0.49E-15

 

 

 

 

Blends processed at 300°C and 1 pass though the extruder.

AN OVA: single factor/non-oriented films

 

 

 

 

 

Source of Degree of Mean
Variation SS freedom square F P-value melfi
Between
Groups 8.9E-31 4 2.224E-31 296.7477 3.98E-06 5.192163
Within
Groups 3.7SE-33 5 7.494E-34
Total 8.93E-31 9

 

 

 

 

 

 

 

*Significant difference at the confidence level (p) of 0.05, where F-value is greater than

I:"critical-

AN OVA: single factor/oriented films

 

 

 

 

 

Source of ' Degree of Mean ‘
Variation SS fi'eedom square F P-value Fmfl
Between

Groups 4.06E-3l 4 1.02E-31 7.178281 0.007013 3.63309
Within

Groups 1.27E-3l 9 1.42E-32

Total 5.34E-31 l3

 

 

 

 

 

 

 

*Significant difference at the confidence level (p) of 0.05, where F -value is greater than

Fcritical-

200

 

 

Table D4: Analysis of variance of the oxygen permeability coefficient (P02 ) of the non-
oriented films obtained from blend resins with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of P02
condition transestagfication (Kg ml, m2 Pa)

2 285°C /1 pass 0.069 5.22E-l9

3 300°C /1 pass 0.221 5.31E-19

4 315°C /1 pass 0.333 5.22E-l9

5 285°C /2 pass 0.172 5.11E-19

6 300°C /2 pass 0.394 4.99E-19

7 315°C /2 pass 0.519 5.07E-19
ANOVA

Degree

Source of of

Variation SS freedom Mean square F P-value Fain-0.1
Between Groups 2.80686E-39 5 5.6lE-40 0.71319* 0.621525122 2.77285
Within Groups 1.41682E-38 18 7.87E-40
Total 1.69751E-38 23 .

 

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than l:"critical-

201

Table D5: Analysis of variance of the oxygen permeability coefficient (P0z ) of the

oriented films obtained from blend resins with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of PO2
condition transestgfication (Kg ml, m; h)
2 285°C /1 pass 0.069 3.32E-19
3 300°C /1 pass 0.221 2.82E-l9
4 315°C /1 pass 0.333 3.55E-19
5 285°C /2 pass 0.172 3.45E-19
6 300°C /2 pass 0.394 3.51E-19
7 315°C /2 pass 0.519 3.90E-19
ANOVA
Degree
Source of of
Variation SS freedom Mean square F P-value Fcritical_
Between Groups 7.96641E-38 5 1.59E—38 2.3561* 0.163235 4.387374
Within Groups 4.05733E-38 6 6.76E-39
Total 1.20237E-37 l 1

 

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than Fcfiticglo

202

 

Table D6: Analysis of variance of the oxygen permeability coefficient (P02 ) of the blend

films with composition of 14.5 to 50.4 mole % PEN.

 

 

 

 

 

 

 

 

Source of variance Mean of P02 (cc mil/ m2 d atrn)
Sample ID Non-oriented films Oriented films
1 PET 157.83 102.16

8 14.5 mole % PEN 142.36 90.84

3 33.7 mole % PEN 128.32 68.21

9 39.5 mole % PEN 122.29 73.61

10 50.4 mole % PEN 113.29 71.48

1 1 PEN 72.45 20.91

 

 

 

 

 

Blends processed at 300°C and 1 pass though the extruder.

ANOVA: single factor/ non-oriented films

 

 

 

Degree
Source of of
Variation SS freedom Mean square F P-value Famed
Between
Groups 4915.988 4 1228.997 21.1345 5.05E-06 3.055568
Within
Groups 872.2683 15 58.15122
Total 5788.257 19

 

 

 

 

 

 

 

 

 

*Significant difference at the confidence level (p) of 0.05, where F-value is greater than

Fcritical-

ANOVA: single factor/oriented films

 

 

 

 

 

 

 

 

 

 

 

Degree
Source of of
Variation SS freedom Mean square F P-value Fem“.
Between ' '
Groups 1516.824 4 379.2061 23.6317 0.004811 6.388234
Within
Groups ' 64.18599 4 16.0465
Total 1581.01 8

 

*Significant difference at the confidence level (p) of 0.05, where F-value is greater than

Fcritical-

203

 

 

Table D7: Analysis of variance of the carbon dioxide permeability coefficient (PCO2 ) of
the non-oriented films obtained fi'om blend resins with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of Pm2
condition transesterification (Kc ml, m2 Pa)
(B)
2 285°C /1 pass 0.069 3.71E-18
3 300°C /1 pass 0.221 4.03E-18
4 315°C /1 pass 0.333 3.98E-18
5 285°C /2 pass 0.172 3.84E-18
6 ’ 300°C /2 pass 0.394 - 3.71E-18
7 315°C /2 pass 0.519 3.94E-18
ANOVA
Degree
Source of - of '
variation 88 freedom Mean square F P-value Fm'fiu]

 

Between Groups 1.89E-37 5 3.77E-38 824643" 0.01158 8.745928
Within Groups 2.74E-38 6 4.57E-39

 

 

 

Total 2.16E—37 l 1

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.01, where F-value is less
than Feritical-

204

Table D8: Analysis of variance of the carbon dioxide permeability coefficient (Pa, ) of
the oriented films obtained from blend resins with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of PCO2
condition transest(eBr;fication (Kg ml, m; Pa)
2 285°C /1 pass 0.069 2.52E-18
3 300°C /1 pass 0.221 2.17E-18
4 315°C /1 pass 0.333 2.41E-18
5 285°C /2 pass 0.172 2.66E-18
6 300°C /2 pass 0.394 2.70E-18
7 315°C /2 pass 0.519 2.45E-l8
ANOVA
Degree
Source of of
Variation SS freedom Mean square F P—value Fcriticpl_
Between Groups 2.79E-37 5 5.58E—38 5.23702“ 0.101738 9.013434
Within Groups 3.2E-38 3 1.07E-38
Total 3.11E-37 8

 

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than Fcritical- .

205

Table D9: Analysis of variance of the carbon dioxide permeability coefficient (Pm: ) of
the blend films with composition of 14.5 to 50.4 mole % PEN.

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean of Poo, (cc mil/ cm2 d atm)
Sample II) Non-oriented Oriented films
films

1 PET 9.65E-02 5.41E-02

8 14.5 mole % PEN 8.1 113-02 5.13E-02

3 33.7 mole % PEN 7.08E-02 3.81E-02

9 39.5 mole % PEN 6.74E-02 3.43E-02

10 50.4 mole % PEN 5.71E-02 2.87E-02

l 1 PEN 2.50E-02 9.52E-03

 

 

Blends processed at 300°C and 1 pass though the extruder.

ANOVA: single factor/ non-oriented films

 

 

 

 

 

 

 

 

 

 

 

Degree
Source of of
Variation SS fieedom Mean square F P-value Fem“.
Between
Groups 0.001787 4 0.000447 204.5253 5.192163
Within
Groups 1.09E-05 5 2.18E-06
Total 0.001798 9

 

 

 

*Significant difference at the confidence level (p) of 0.05, where F -value is greater than

Fcritical-

ANOVA:

    

Source of

V

Total

F

47.47457

5.192163

*Significant difference at the confidence level (p) of 0.05, where F-value is greater than

Fcritical-

206

Appendix E

Statistical analysis of the tensile strength

The analysis of variance (ANOVA) was performed by using a single factor
AN OVA on the microcomputer statistical program in Microsoft 2000.

Table E1: Analysis of variance of the tensile strength along the machine direction of the
non-oriented films with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of Tensile strength
condition transesterification (MPa)
(B)
2 285°C /1 pass 0.076 50.89
3 300°C /1 pass 0.222 47.33
4 315°C / 1 pass 0.356 50.27
5 285°C /2 pass 0.192 47.05
6 300°C /2 pass 0.405 44.89
7 315°C /2 pass 0.525 51.42
ANOVA
Degree
of
Source of Variation SS freedom Mean Square F P-value F m-ticL
Between Groups 4807141 5 9614282 3.0392"I 0.02136 3.557915
Within Groups 11704607 37 3163407
Total 1651 1748 42

 

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.01, where F-value is less

than Fcfifica].

207

Table E2: Analysis of variance of the tensile strength along the cross machine direction
of the non-oriented films with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of Tensile strength
condition transesterification (MPa)
(13)
2 285°C /1 pass 0.076 45.35
3 300°C /1 pass 0.222 48.08
4 315°C /1 pass 0.356 40.84
5 285°C /2 pass 0.192 44.74
6 300°C /2 pass 0.405 44.9
7 315°C /2 pass 0.525 50.99
ANOVA
Degree
Source of of
Variation SS freedom Mean square F P-value PM“,
Between Groups 8242377 5 1648475 30748” 1.52E-12 2.455828
Within Groups 2090876 39 53612.2
Total 10333253 44

 

 

 

 

 

 

 

 

 

‘Significant difference at the confidence level (p) of 0.05, where F-value is greater

than Feritical-

208

Table E3: Analysis of variance of the tensile strength of the oriented films obtained from

blend resins with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of Tensile strength
condition transesterification (MPa)
(B)
2 285°C /1 pass 0.076 153.2
3 300°C /1 pass 0.222 182.7
4 315°C /1 pass 0.356 170.0
5 285°C /2pass 0.192 148.1
6 300°C /2 pass 0.405 126.0
7 315°C /2 pass 0.525 148.8
AN OVA
Degree
Source of of
Variation SS freedom Mean square F P-value Fwy-c..—
Between Groups 12054.5 2410.89 15.086* 8.16E-08 2.49361
Within Groups 5433.47 34 159.808
Total 17487.9 39

 

 

 

 

 

 

 

 

 

*Significant difference at the confidence level (p) of 0.05, where F-value is greater

than Fcritical-

209

Appendix F

Statistical analysis of the Percent elongation

The analysis of variance (ANOVA) was performed by using a single factor
AN OVA on the microcomputer statistical program in Microsoft 2000.

Table Fl: Analysis of variance of the % elongation along the machine direction of the
non-oriented films with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of % Elongation
condition transesteri fication
(B)
2 285°C /1 pass 0.076 3.4
3 300°C /1 pass 0.222 3.03
4 315°C /1 pass 0.356 3.5
5 285°C /2 pass 0.192 3.0
6 300°C /2 pass 0.405 3.0
7 315°C /2 pass 0.525 3.0
ANOVA
Degree
of
Source of Variation SS freedom Mean Square F P-value F critical
Between Groups 1.63185 5 0.32637 2.1918" 0.076632 2.477165
Within Groups 5.360398 36 0.1489
Total 6.992248 41

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than Fcritical-

210

 

Table F2: Analysis of variance of the % elongation along the cross machine direction of
the non-oriented films with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of % Elongation
condition transesterification
(B)
2 285°C /1 pass 0.076 2.99
3 300°C /1 pass 0.222 3.06
4 315°C /1 pass 0.356 2.7
5 285°C /2 pass 0.192 2.87
6 300°C /2 pass 0.405 2.86
7 315°C /2 pass 0.525 3.14
ANOVA
Degree
Source of of
Variation SS freedom Mean square F P-value Fain-“L
Between Groups 0.77159 5 0.154318 7.9954“ 2.91E-05 2.455828
Within Groups 0.75273 39 0.019301
Total 1.52432 44

 

 

 

 

 

 

 

 

 

*Significant difference at the confidence level (p) of 0.05, where F-value is greater

than Fcritical-

211

Table F3: Analysis of variance of the % elongation of the oriented films obtained from

blend resins with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of % Elongation
condition transesterification
(B)
2 285°C /1 pass 0.076 148.8
3 300°C /1 pass 0.222 130.4
4 315°C /1 pass 0.356 149.5
5 285°C /2 E58 0.192 148.6
6 300°C /2 pass 0.405 127 .7
7 315°C /2 pass 0.525 140.2
ANOVA
Degree
Source of of
Variation SS freedom Mean square F P-value FmtjcaL
Between Groups 3074.44 614.888 2.6826" 0.038469 3.63048
Within Groups 7563.92 33
Total 10638.4 38

 

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.01, where F-value is less

than Fcritical-

212

Appendix G

Statistical analysis of the Young’s modulus

The analysis of variance (AN OVA) was performed by using a single factor
AN OVA on the microcomputer statistical program in Microsoft 2000.

Table G1: Analysis of variance of the Young’s modulus along the machine direction of
the non-oriented fihns with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of Young’s modulus
condition transesterification (MPa)
(13)
2 285°C /1 pass 0.076 1569.95
3 300°C /1 pass 0.222 1718.18
4 315°C /1 pass 0.356 1714.74
5 285°C /2 pass 0.192 1688.54
6 300°C /2 pass 0.405 1627.86
7 315°C /2 pass 0.525 1739.07
ANOVA
Degree
of
Source of Variation SS fieedom Mean Square F P-value. F crimp
Between Groups 2772.832 5 554.5664 25854" 0.04254 3.574399
Within Groups 7721.813 36 214.4948
Total 10494.64 41

 

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.01, where F-value is less

than Fcritical-

213

Table G2: Analysis of variance of the Young’s modulus along the cross machine
direction of the non-oriented films with 25 % (wt/wt) PEN.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample Processing Degree of Young’s modulus
condition transesterification (MPa)
(B)
2 285°C /1 pass 0.076 1549.95
3 300°C /1 pass 0.222 1618.90
4 315°C /1 pass 0.356 1560.98
5 285°C /2 pass 0.192 1549.95
6 300°C /2 pass 0.405 1520.30
7 315°C /2 pass 0.525 1586.49
ANOVA
Degree
Source of of
Variation _ SS freedom Mean square F P-value Fcfitica_l_
Between Groups 981.8469 5 196.3694 1.8280* 0.131293 2.469648
Within Groups 3974.564 37 107.4207
Total 4956.41 1 42

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than l:cr'itical-

214

Table G3: Analysis of variance of the Young’s modulus of the oriented films with 25 %

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(wt/wt) PEN.
Source of variance Mean
Sample Processing Degree of Young’s modulus
condition transesterification (MPa)
(B)
2 285°C /1 pass 0.076 1785.8
3 300°C /1 pass 0.222 1823.6
4 315°C /1 pass 0.356 1679.5
5 285°C /24>ass 0.192 1693.9
6 300°C /2 pass 0.405 1697.8
7 315°C /2 pass 0.525 1673.6
ANOVA
Degree
Source of of
Variation SS freedom Mean square F P-value Fain-m;
Between Groups 136299 5 27259.9 0.9082* 0.485661 2.45583
Within Groups 1170557 39 30014.3
Total 1306856 44

 

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F-value is less

than Fcriticai-

215

Table G4: Analysis of variance of the Young’s modulus along the machine direction of
the non-oriented films with composition of 14.5 to 50.4 mole % PEN.

 

 

 

 

 

 

 

 

 

Source of variance Mean

Sample ID Young’s modulus
(MPa)

1 PET 1648

8 14.5 mole % PEN 1684

3 33.7 mole % PEN 1718

9 39.5 mole % PEN 1556

10 50.4 mole % PEN 1692

ll PEN 1919

 

 

 

 

Blends processed at 300°C and 1 pass though the extruder.

ANOVA: single factor

 

 

 

 

 

 

Source of Degree of Mean
Variation SS freedom square F P-value Fem-“L
Between
Groups 2318.794 5 463.7587 1.69459“ 0.163425 2.502631
Within
Groups 9031.084 33 273.6692
Total 11349.88 38

 

 

 

 

 

 

 

*Insignificant difference at the confidence level (p) of 0.05, where F -value is less

than Fcritical-

216

 

Table G5: Analysis of variance of the Young’s modulus along the cross machine

direction of the non-oriented with composition of 14.5 to 50.4 mole % PEN.

 

 

 

 

 

 

 

 

 

Source of variance Mean
Sample ID Young’s modulus
(MPa)

1 PET 1554

8 14.5 mole % PEN 1653

3 33.7 mole % PEN 1619

9 39.5 mole % PEN 1617

10 50.4 mole % PEN 1532

11 PEN 1685

 

 

 

 

Blends processed at 300°C and 1 pass though the extruder.

ANOVA: single factor

 

 

 

 

 

 

Source of Degree of Mean
Variation SS freedom square F P-value Fcritiglr
Between . .
GI‘OUE 2763.005 5 552.601 7.68659" 3.58E-05 2.443429
Within
Groups 2947.554 41 71.89155
Total 5710.559 46

 

 

 

 

 

 

 

 

*Significant difference at the confidence level (p) of 0.05, where F-value is greater

than Fcritical-

217

Table G6: Analysis of variance of the Young’s modulus of the oriented films with
composition of 14.5 to 50.4 mole % PEN.

 

 

 

 

 

 

 

 

 

Source of variance Mean

Sample ID Young’s modulus
(MPa)

1 PET 2304.3

8 14.5 mole % PEN 1932.4

3 33.7 mole % PEN 1823.6

9 39.5 mole % PEN 1901.1

10 50.4 mole % PEN 2049.2

11 PEN 2573.5

 

 

 

 

Blends processed at 300°C and 1 pass though the extruder.

ANOVA: single factor

 

 

 

 

 

 

Source of Degree of Mean
Variation SS freedom square F P-value Fwy
Between
Groups 2693506 5 5387013 16.1716"I 1.55E-08 2.462549
Within
Groups 1265834 38 33311.43
Total 3959341 43

 

 

 

 

 

 

 

*Significant difference at the confidence level (p) of 0.05, where F -value is greater

than Feritical-

218

 

Bibliography

219

Bibliography

Adur, A. M., and L. J. Bonis. 1994. PET-LCP Compatibilized Alloys: A New Unique
Development. Future-Pak 1994. Proceedings 11th International Schroeder Conference
on Packaging Innovation, Westin Hotel, Chicago, IL. Nov 16-17.

Anonymous. 1981. ASTM D3985-81 Standard Test Method for Oxygen Gas
Transmission Rate Through Plastic Film and Sheeting Using a Coulometric Sensor.

Anonymous. 1985. PET Economical Choice for Engineering Uses. Plastics World.
August, p.122-123.

Anonymous. 1990 (a). ASTM D882-9O Standard Test Method for Tensile Properties of
Thin Plastic Sheeting.

Anonymous. 1990“). ASTM D1505-90 Standard Test Method for Density of Plastics by
the Density-Gradient Technique.

Anonymous. 1990“). ASTM F 1249-90 Standard Test Method for Water Vapor

Transmission Rate Through Plastic Film and Sheeting Using a Modulated Infrared
Sensor.

Anonymous. 1991. ASTM D638-91 Standard Test Method for Tensile Properties of
Plastics.

Anonymous. 1994 (a). PEN: Assessing the Potential. European Plastics News. 21(10):32.
Anonymous. 1994 0”. Enter a Super Polyester. Textile Month. September, p. 30.
Anonymous. 1995. PEN the Hero in Hot-fill Breakthrough. Packaging Week. 10(44): 1.

Anonymous. 1999. Online Literature: HiPERTUFTM Polyester Resin.
http://www2.shellchemical.com.

Appel, O. H. 1996. Barrier Properties of PET and PEN Bottles. Kunststoffe Plast Europe.
May, p. 13-14.

Bakkar M., Encyclopedia of Packaging Technology, 1986. Wiley and Sons Inc., New
York.

Barrer, R.M., J.A. Banie, and MG. Rogers. 1983. Journal of Polymer Science, Part A.
1:2565.

220

Bauer, C.W., 1997. The Effects of Degree of Stretching on the Properties of Heat Set
Biaxially Stretched Naphthalate-Containing Polyester Films. Plastic Film & Sheeting.
13:336-347.

Bilrneyer, F .W., J r., 1962. Textbook of Polymer Science. Wiley-Interscience, New York.
p. 237.

Bonart, R., and R. Hosemann. 1960. Zur Analyse der Langperiodeninterferenzen. Z.
Elelctrochem. 64:314.

Brady, T. E., S. A. J abarin, and G.W. Miller. 1974. Permeability of Plastic Films and
Coatings to Gases, Vapors, and Liquids. Plenum Press, New York

Brennan, D.J., J .E. White, A.P. Haag, S.L. Kram, M.N. Mang, S.Pikulin, and ON.
Brown. 1996. Poly(hydroxy amide ether): New High-barrier Thermoplastics.
Macromolecules. 29:3707-3716.

Brennan, W.P. 1978. Characterization and Quality Control of Engineering
Thermoplastics by Thermal Analysis in Thermal Analysis application. Perkin-Elmer
Corporation, Norwalk, Connecticut.

Briston, J. H. 1989. Plastics films. 3rd. Longrnan Scientific and Technical, Harlow Essex
CM20 2JE, England.

Buchner, S., D. Wiswe, and H. G. Zachmann. 1989. Kinetics of Crystallization and
Melting Behavior of Poly(ethylene naphthalene-2,6-dicarboxylate)‘. Polymer. 30:480.

Cakmak, M, J .C. Kim, and SW. Lee. 1996. Neck Formation and its Elimination in Uni
and Biaxial Defamation of PEN and its Blends. Polymer Preprint. American Chemical
Society, Division of Polymer Chemistry. p. 233.

Cakmak, M., Y.D. Wang, and M. Sirnharnbhatla. 1990. Processing Characteristics,
Structure Development, and Properties of Uni and Biaxially Stretched PEN Films.
Polymer Engineering and Science. 30(12): 721-733.

Callander, D and B. Howell, 1994. High Performance Naphthalate Based Packaging
Resins part H. Proceedings Bev-Pak Asia 1994, Hong Kong.

Callander, D and E. Sisson 1994. High Performance Naphthalate Based Packaging
Resins. Proceedings Bev-Pak America 1994, Tarpon Springs, FL.

Castle, D. 1996. PEN Homopolymer Safe for Food Contact. Packaging Week. 13(45):5.
Cheng, S.Z.D., R. Pan, H. S. Bu, M. Cao, and B. Wunderlich. 1988. Thermodynamic

Properties in the Liquid State of High Melting Polymers Containing Phenylene and
Naphthylene Groups. Macromolecular Chemistry. 189:1579-1594.

221

Crawshaw, R., and E. Sisson. 1995. “HiPERTUF” Polyester Resins-refining the
Packaging Envelope. Proceedings Inno Bev, Mainz, Germany, October 3.

Cruz, C.S., F. J. Balta Calleja, H. G. Zachmann, and D. Chen. 1992. Mechanical
Properties and Structure of Glassy and Sernicrystalline Random Copolymers of PET and
PEN. Journal of Material Science. 27:2161-2164.

Davis, V.A., and R.W. Howell, 1995. HiPERTUF 1'” Resins for High Performance
Containers. Polyester Technical Center, Shell Chemical Company, Akron, OH, 44305.

Devaux, J ., P. Godard, and J .P. Mercier. 1982(3). Bisphenol-A Polycarbonate-
Poly(butylene Terephthalate) Transesterification. , 1. Theoretical Study of the Structure
and of the Degree of Randomness in F our-Component Copolycondensates. J oumal of
Polymer Science, Polymer Physics Edition. 20:1875-1880.

Devaux, J ., P. Godard, J .P. Mercier, R. Touillaux, and J.M. Dereppe. 1982“”. Bisphenol-
A Polycarbonate-Poly(butylene Terephthalate) Transesterification. H. Structure Analysis
of the Reaction Products by IR and 1H and 13C NMP. Journal of Polymer Science,
Polymer Physics Edition. 20: 1881-1894.

Devaux, J ., P. Godard, and JP. Mercier. 1982“). Bisphenol-A Polycarbonate-
Poly(butylene Terephthalate) Transesterification. IH. Study of Model Reactions. Journal
of Polymer Science, Polymer Physics Edtion. 20:1895-1900.

Devaux, J ., P. Godard, and J .P. Mercier. 1982“”. Bisphenol-A Polycarbonate-
Polbeutylene Terephthalate) Transesterification. IV. Kinetics and Mechanism of the

Exchange Reaction. Journal of Polymer Science, Polymer Physics Edition. 20: 1901-
1907.

De Vries, A. J ., C. Bonnebat, and J. Beautempts. 1977. Uni-and Biaxial Orientation of
Polymer Films and Sheets. Journal of Polymer Science: Polymer Symposium. 58:109-
156.

DiMarzio, E. A. and J. H. Gibbs. 1959. Glass Temperature of Copolymers Journal of
Polymer Science. 40: 121-131.

Eersels, K.L.L., and G. Groeninckx. 1996. Influence of Interchange Reactions on the
Crystallization and Melting Behaviour of Polyarnide Blends as Affected by the
Processing Conditions. Polymer. 37(6): 983-989.

Eersels, K.L.L., and G. Groeninckx. 1997. Thermal Analysis of Polyarnide Blends as

Obtained by Reactive Melt-Extrusion: Influence of Blend Composition on Crystallization
and Melting Behavior. Journal of Applied Polymer Science. 63:573-580.

222

El-Hibri, M. J. and D. R. Paul. 1985. Effects of Uniaxial Drawing and Heat-Treatment on
Gas Sorption and Transport in PVC. Journal of Applied Polymer Science. 30:3649-3678

Fakirov, S., M. Sarkissova, and Z. Denchev. 1996‘“). Melting- and Crystallization-
induced Sequential Reordering in Immiscible Blends of Poly(ethylene terephthalate) with
Polycarbonate or Polyarylate. Macromolecular Chemistry and Physics. 197 :2837-2867.

F akirov, S., M. Sarkissova, and Z. Denchev. 1996“”. Sequential Reordering in
Condensation Copolymers, 3a) Miscibility-Induced Sequential Reordering in Random
Copolyesterarnides. Macromolecular Chemistry and Physics. 197:2889-2907.

Forcinio, H. 1993. Plastic Packaging, The Heat is on. Prepared Food. November, p. 86-
88.

Fricke, H. 1924. A Mathematical Treatment of the Electric Conductivity and Capacity of
Disperse Systems. Physical Review. 24:575-587.

Gerding T. K., F. Van Den Berg, and N. de Kruijf. 1995. Trend in Food Packaging:
Arising Opportunities and Shifting Demands. Proceeding 9th International Association of
Packaging Research Institutes (IAPRI) World ConferenCe on Packaging. September
1995. 1(1): 1-14.

Godard, P., J .M. Dekoninck, V. Devlesaver, and J. Devaux. 1986. Molten Bisphenol-A
Polycarbonate-Poly(ethylene Terephthalate) Blends. H. Kinetics of the Exchange
Reaction. Journal of Polymer Science: Part A: Polymer Chemistry. 24:3315-3324.
Gordon, M. and J. S. Taylor. 1952. Journal of Applied Chemistry. 2: 493.

Guo, M. 1995. ‘3 C NMR Studies on Macroinitiator and Block Copolymers Derived from
13 C enriched Difunctional Initiator. Polymer Preprint. American Chemical Society,
Division of- Polymer Chemistry. p. 825.

Guo, M., and HG. Zachmann 1993. Intermolecular Cross-polarization Nuclear Magnetic
Resonance Studies of the Miscibility of PET/PEN Blends. Polymer. 34(12): 2503-2507.

Hanna, 8., and A. H. Windle. 1988. Geometrical Limits to Order in Liquid Crystalline
Random Copolymers. Polymer. 29:207-223.

Heisey, C. L., DC. Hoffinann, and J .A. Zawada. 1996. Crytallization of NDC-containing
Blends and Copolyrner. Polymer Preprint. American Chemical Society, Division of
Polymer Chemistry. p. 231.

Hess, K., and H. Kiessig. 1943. Zur Kenntnis der Feinstruktur der Polymidfasem. Die
Naturwissenschaften. 3 1 : 171 .

Hoffman, D.C., J .K, Caldwell, 1995. Prediction of Barrier and Glass Transition
Temperature Enhancements in PET/PEN Blends. Proceedings Specialty Polyesters 1995.

223

Holsti-Miettinen, R.M., K.P. Perttila, J.V. Seppala, and MT Heino. 1995. Oxygen
Barrier Properties of Polypropylene/Polyarnide 6 blends. Journal of Applied Polymer
Science. 58(9) 1551-1560.

Hoshino, S., J. Powers, D. G. Legrand, H. Kawai, and R. S. Stein. 1962. Orientation
Studies on Drawn Polyolefins. Journal of Polymer Science. 58:185-204.

Ihm, D.W., S. Y. Park, C. G. Chang, Y. S. Kim, H. K. Lee. 1996. Miscibility of
Poly(ethylene terephthalate)/poly(ethylene naphthalate) Blends by Transesterification.
J oumal of Polymer Science: Part A: Polymer Chemistry. 34:2841.

J abarin, SA. 1984. Orientation Studies of Poly(Ethylene Terephthalate). Polymer
Engineering and Science. 24(5): 376-384.

J abarin, S.A., 1992. Strain-Induced Crystallization of Poly(Ethylene Terephthalate)
Polymer Engineering and Science. 32(18): 1341-1349.

J abarin, SA. and P.Chandran. 1993. Biaxial Orientation of PET, Part 1: Nature of the
Stress-strain Curves. Advances in Polymer Technology. 12(2): 119-132.

J adhav, J .Y., and SW. Kantor. 1988. Polyesters, Thermoplastic in Encyclopedia of

Polymer Science and Engineering. 2nd edition. John Wiley and Sons Inc., New York.
12:193-256.

Jaquiss, D.B.G., W.F.H. Borman, and R.W. Campbell, 1982. Encyclopedia of Chemical
Technology, M. Grayson ed. John Wiley and Sons Inc., New York. 18:549.

Jenkins, SD. 1995. PET/PEN Copolymers and Blends for Hot-fill and High Barrier
Packaging Applications. Proceddings Specialty Polyesters 1995. ICI Chemicals and
Polymers Ltd. p. 197.

Joyce, R. P. and A. P. Berzinis. 1991. Proceedings Compalloy 91. Schotland Business
Research, Princeton. p. 65.

Karasz, F. E. 1985. Glass Transitions and Compatibility, Phase Behavior in Copolymer
Containing Blends. In Polymer Blends and Mixtures. D. J. Walsh, J .S. Higgins, and
A.Maconnachie (eds). Martinus Nijhoff Publishers, Boston.

Kelley, F .N., and F. Bueche. 1961. Viscosity and Glass Temperature Relations for
Polymer-Diluent Systems. Journal of Polymer Science. 50:549-556.

Kimmel, R. 2000. The Challenge of the Plastic Beer Bottle. Presentation at 2000 [FT
Annaul Meeting Technical Program, held at Dallas, Texas, USA. June 11-14.

224

Kit, K.M., J .M. Schiltz, and R. Gohil. 1995. Morphology and Barrier Properties of
Oriented Blends of Poly(Ethylene Terephthalate) and Poly(Ethylene 2,6-Naphthalate)

with Poly(Ethylene-co-Vinyl Alcohol). Polymer Engineering and Science. 35(8): 680-
692.

Koros, W. J ., and M. W. Hellums. 1988. Transport Properties in Encyclopedia of
Polymer Science and Engineering. 2nd edition. John Wiley and Sons Inc., New York.
Supplement volume. p. 788-799.

Kotliar, A.M. 1981. Interchange Reactions Involving Condensation Polymers.
Macromolecular Reviews. 16:367-395.

Kugler, J ., J .W. Gilrner, D. Wiswe, H.G. Zachmann, K. Hahn, and E.W. Fischer. 1987.
Study of Transesterification in PET by Small-Angle Neutron Scattering.
Macromolecules. 20:1 1 16.

Kwei, T.K. 1984. The Effect of Hydrogen Bonding on the Glass Transition Temperatures
of Polymer Mixtures. Journal of Polymer Science: Polymer Letters Edition. 22(6): 307-
3 14.

Kwei, T.K., E.M. Pearce, J .R. Pennacchia, and M. Charton. 1987. Correlation Between
the Glass Transition Temperatures of Polymer Mixtures and Intermolecular Force
Parameters. Mracromolecules. 20:1174-1176.

Lenz, R.W., and S. Go. 1973. Crystallization-Induced Reactions of Copolymers. HI. Ester
Interchange Reorganization of Poly(cis/trans-l ,4-Cyclohexylenedimethylene
Terephthalate). Polymer Science: Polymer Chemistry Edition. 11:2927-2946.

Light, R. R. and R. W. Seymour. 1982. Effect of Sub-T, Relaxation on the Gas Transport
Properties of Polyesters. Polymer Engineering and Science. 22(14): 857-864.

Lu, X., and AH. Windle. 1995. Crystallization of Random Copolymers of PET and PEN.
Polymer. 36(3): 451.

MacDonald, W.A., A.D.W. McLenaghan, G. Mclean, R.W. Richards, and SM. King.
1991. A Neutron Scattering Investigation of the Transesterification of a Main-Chain
Aromatic Polyester. Macromolecules. 24: 6164.

MacKnight, W. J ., F. E. Karasz, and J. R. Fried. 1978. Solid State Transition Behavior of

Blends. In Polymer Blends. D.R. Paul and S. Newman (eds.). Academic Press. New
York.

Margolis, J .M. 1985. Engineering Thermoplastics: Properties and Applications. Marcel
Dekker, Inc., New York.

225

Matsumoto, K., J. F. Fellers, and J. L. White. 1981. Uniaxial and Biaxial Orientation
Development and Mechanical Properties of Polystyrene Films. Journal of Applied
Polymer Science. 26:85-96.

McGee, T.M., and AS. Jones. 1995. The Effect of Processing Parameters on Physical
Properties of PET/PEN Blends for Bottle Application. Eastman Chemical Company,
Kingsport, TN. .

Mencik, Z. 1976. Krystalicka Struktura Polykondenzatu Kyseliny Nafialen-2,6-
dikarbonove Etylenglykolem. Chemisky Prumysl. 17(2): 78-81.

Michaels, AS, and H.J. Bixler. 1961. Flow of Gases Through Polyethylene. Journal of
Polymer Science. 50:413-439.

Middleman, S. 1977. Fundamentals of Polymer Processing. McGraw-Hill Inc., New
York

Mock, J. A. 1983. Advancing Technology to Spur Big Growth for Plastic Composites
During the ‘808. Plastic Engineering. 39(2): 13-19.

Montaudo, G., M. S. Montaudo, E. Scamporrino, and D. Vitalini. 1992. Mechanism of
Exchange in Polyesters. Composition and Microstructure of Copolymers Formed in the
Melt Mixing Process of PET and Poly(ethylene adipate). Macromolecules. 2225099.

Morris, H. R., J. F. Turner II, B. Munro, R.A. Ryntz, and P. J. Treado. 1999. Chemical
Imaging of Thermoplastic Olefin (TPO) Surface Architecture. Langmuir. 15:2961-2972.

Morton-J ones, DH. 1989. Polymer Processing. Chapman and Hall Ltd.

Murakami, S., M. Yamakawa, M.Tsuji, and S. Kohjiya. 1996. Structure Development in
the Uniaxial-Drawing Process of Poly(ethylene naphthalate). Polymer. 37(17): 3945-
3951.

Nadella, H. P., J. E. Spruiell, and J. L. White. 1978. Drawing and Annealing of
Polypropylene Fibers: Structural Changes and Mechanical Properties. Journal of Applied
Polymer Science. 22:3121-3133.

Nemphos, S. P., M. Salame, and S. Steingiser. 1986. Barrier Polymers. In Encyclopedia
of Packaging Technology. M. Bakkar (eds). Wiley & Sons Inc., New York.

O’Brien, K.C., W.J. Koros, and GR. Husk. 1987. Influence of Casting and Curing
Conditions on Gas Sorption and Transport in Polyimide Films. Polymer Engineering and
Science. 27(3): 211-217.

Odian, G. 1991. Principles of Polymerization. 3rd. John Wiley and Sons Inc., New York.

226

Olabisi, O., L. M. Robeson, and MT. Shaw. 1979. Polymer-Polymer Miscibility.
Academic Press Inc, New York.

Park, H.C and EM. Mount HI. 1988. Films, Manufacture in Encyclopedia of Polymer
Science and Engineering. 2nd edition. John Wiley and Sons Inc., New York. 7:96-107.

Paul, DR. and CB. Bucknall. 1999. Polymer Blends. Volume 2: Formulation. John
Wiley and Sons Inc., New York.

Peterlin, Anton. 1988. Morphology in Encyclopedia of Polymer Science and
Engineering. 2nd edition. John Wiley and Sons Inc., New York. 10:72-94.

Pidgeon, R. 1996. Amoco Unviels Details of PEN Container Range. Packaging Week.
10(45): 937. '

Pilati, F., E. Marianucci, and C. Berti. 1985. Study of the Reactions Occurring during
Melt Mixing of Poly(ethylene Terephthalate) and Polycarbonate. Journal of Applied
Polymer Science. 30: 1267-1275.

Po, R., E. Occhiello, G. Giannotta, L. Pelosini, and L.Abis. 1996. New Polymeric
Materials for Containers Manufacture Based on PET/PEN Copolyesters and Blends. _
Polymers for Advanced Technologies. 7: 365-373.

Po, R., P. Cioni, L. Abis, E. Occhiello, F. Garbassi. 1991. Sequence Analysis of
Aromatic/ethylene Glycol Copolyesters Using 1H and 13C NMR of Oxyethylenic
Groups. Polymer Communications. 32(7): 208-212.

Porter, R.S., and L.Wang. 1992. Compatibility and Transesterification in Binary Polymer
Blends. Polymer. 33(10): 2019-2030.

Ramj it, H.G. and RD. Sedgwick. 1976. The Kinetics of Ester-Ester Exchange Reactions
by Mass Spectrometer. Journal of Macromolecular Science-Chemistry A. 10:815.

Ray, D. D., C. B. Shriver, and R. J. Gartland. 1977. Here’s Why Polyethylene
Terephthalate is the Maj or Competitor for Beverage Container Applications. Part 1-
Properties and Applications. Plastics Design & Processing. 17(8): 47-52.

Rodriguez, F. 1996. Principles of Polymer Systems. Taylor & Francis, Bristol, PA 19007.

Rudin A. 1982. Basic Principles of Polymer Molecular Weights. In: The Elements of
Polymer Science and Engineering. Acedemic Press, Inc., San Diego, California.

Salame,M. 1967. A Correlation between the Structure and Oxygen Permeability of High

Polymers. Polymer Preprint. American Chemical Society, Division of Polymer
Chemistry. 8:137.

227

Samuels, RI. 1974. Structured Polymer Properties. John Wiley & Sons Inc., New York,
New York.

Sawyer, LC, and D. T. Grubb. 1996. Polymer Microscopy, 2nd. Chapman & Hall.
London, UK.

Scarlett, AC. 1958. US. Pat. 2,823,421 (to E.I. do Pont de Nemours & co., Inc. on Feb
18, 1958).

Schaeberle, M. D., and P. J. Treado. 1997. Proceedings Pittsburgh Conference, Atlanta,
GA. March, p. 913.

Schaeberle, M. D., H. R. Morris, J. F. TurnerII, and P. J. Treado. 1999. Raman Chemical
Imaging Spectroscopy. Analytical Chemistry News & Features. March 1, p. 175A-181A.

Schultz, J. 1974. Polymer Material Science. Prentice-Hall, Inc., Englewood Cliffs, New
Jersey.

Schwartz, S. S. 1982. Plastics Materials and Processes. Van Nostrand Reinhold
Company, New York.

Seferis, J. C. and R J. Samuels. 1979. Coupling of Optical and Mechanical Properties in
Crystalline Polymers. Polymer Engineering and Science. 19(14): 975.

Sec, KS, and J .D. Cloyd, 1991. Kinetics of Hydrolysis and Thermal Degradation of
Polyester Melts. Journal of Applied Polymer Science. 42:845-850.

Shepherd, RA. and KL. Ronald. 1991. US. Pat. 5,006,613 (to Eastman Kodak
Company); -

Shi, Y. 1998. Study on PET/PEN blends: Transesterification Reaction and Properties.
Ph.D. dissertation. The University of Toledo. Toledo, Ohio.

Smith, W. A., J. W. Barlow, and DR. Paul. 1981. Chemistry of Miscible Polycarbonate-
copolyester Blends. Journal of Applied Polymer Science. 26: 4233-4245.

Stewart, M.E., A.J. Cox, and D.M. Naylor. 1993. Reactive Processing of PET/PEN
Blends. Polymer. 43(19): 4060-4067.

Swallow, J .C. and D.K. Baird. 1948. Brit. Pat. 609,797 (to ICI on Oct. 6, 1948).
Tibbitt, J. M. 1994. Isophthalic Acid and 2,6 Naphthalene Dicarboxylate Modified
Polyesters for Packaging Application. Proceedings 11th International Schroeder

Conference on Packaging Innovation, Future-Pak 1994, Westin Hotel, Chicago, Illinois.
November 16-17, 1994

228

Toth, J ., C. Rusch, and L. Desoulter. 1994. PET/PEN Blends, Enhancing the
Performance of Polyester Containers-Outlook. Proceedings Bev-Pak Asia 1994, Hong
Kong. September 22-23, 1994.

Tyne M., B. Howell, and E. Sisson. 1995. Redefining the Polyester Packaging Envelope
for High Temperature and Chemically Aggressive Applications, Bev-Pak Americas 1995,
Innisbrook, FL.

Utracki, L. A. 1990. Polymer Alloys and Blends; Thermodynamics and Rheology.
Hanser Publishers, New York.

Wang, CC. and Y.M. Sun. 1994 (a). Studies on the Formation of PEN. 1. Kinetic Aspects
of Catalyzed Reaction in Transesterification of Dimenthylnaphthalate with Ethylene
Glycol. Journal of Polymer Science: Part A: Polymer Chemistry. 32: 1295-1304.

Wang, CC. and Y. M. Sun. 1994“”. Studies on the Formation of PEN. H. Polymerization
of Bis(Hydroxyethyl)naphthalate by Various Metallic Catalysts. J oumal of Polymer
Science: Part A: Polymer Chemistry. 32: 1305-1315.

Werner, E., S. J anocha, M. J. Hopper, and K. J. Mackenzie. 1988. Polyesters, Films in
Encyclopedia of Polymer Science and Engineering. 2nd edition. John Wiley and Sons
Inc., New York. 12:193-216.

Whinfield, J .R. and J .T. Dickson. 1946. Brit. Pat. 578,079 (June 14, 1946).

White, J. L. 1988. Orientation in Encyclopedia of Polymer Science and Engineering. 2nd
edition. John Wiley and Sons Inc., New York. 7:595-616.

White, J. L., K. C. Dharod, and E. S. Clark. 1974. Interaction of Melt Spinning and
Drawing Variables on the Crystalline Morphology and Mechanical Properties of I-IDPE
and LDPE Fiber. Journal of Applied Polymer Science. 18:2539-2568.

Yamadera, R., and M. Murano. 1967. The Determination of Randomness in Copolyesters
by High Resolution NMR. Journal of Polymer Science. 5: 2259-2268.

229

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

muttWImultilwulyunull